Journals Library

Health Technology Assessment

Volume 28 • Issue 66 • October 2024
ISSN 2046-4924

Low-dose titrated amitriptyline as

second-line treatment for adults with
irritable bowel syndrome in primary care:
the ATLANTIS RCT
Alexandra Wright-Hughes, Alexander C Ford, Sarah L Alderson, Pei Loo Ow , Matthew J Ridd,
Robbie Foy, Felicity L Bishop, Matthew Chaddock, Heather Cook, Deborah Cooper,
Catherine Fernandez, Elspeth A Guthrie, Suzanne Hartley, Amy Herbert, Daniel Howdon,
Delia P Muir, Sonia Newman, Christopher A Taylor, Emma J Teasdale, Ruth Thornton,
Hazel A Everitt and Amanda J Farrin

DOI 10.3310/BFCR7986

a
Low-dose titrated amitriptyline as second-
line treatment for adults with irritable bowel
syndrome in primary care: the ATLANTIS RCT

Alexandra Wright-Hughes ,1 Alexander C Ford ,2,3*

Sarah L Alderson ,4 Pei Loo Ow ,1 Matthew J Ridd ,5
Robbie Foy ,4 Felicity L Bishop ,6 Matthew Chaddock,7
Heather Cook ,1 Deborah Cooper,4 Catherine Fernandez ,1
Elspeth A Guthrie ,4 Suzanne Hartley ,1 Amy Herbert ,5
Daniel Howdon ,4 Delia P Muir ,1 Sonia Newman ,8
Christopher A Taylor ,1 Emma J Teasdale ,6
Ruth Thornton ,8 Hazel A Everitt 8 and Amanda J Farrin 1
1
Clinical Trial Research Unit, Leeds Institute of Clinical Trials Research, School of
Medicine, University of Leeds, Leeds, UK
2
Leeds Institute of Medical Research at St. James’s, University of Leeds, Leeds, UK
3
Leeds Gastroenterology Institute, St. James’s University Hospital, Leeds, UK
4
Leeds Institute of Health Sciences, School of Medicine, University of Leeds, Leeds, UK
5
Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, UK
6
Centre for Clinical and Community Applications of Health Psychology, School of
Psychology, University of Southampton, Southampton, UK
7
Let’s Cure IBS, Leeds, UK
8
Primary Care Research Centre, Faculty of Medicine, University of Southampton,
Southampton, UK

Corresponding author
*

Disclaimer: This report contains transcripts of interviews conducted in the course of the research and
contains language that may offend some readers.

Published October 2024

DOI: 10.3310/BFCR7986

This report should be referenced as follows:

Wright-Hughes A, Ford AC, Alderson SL, Ow PL, Ridd MJ, Foy R, et al. Low-dose titrated
amitriptyline as second-line treatment for adults with irritable bowel syndrome in primary care:
the ATLANTIS RCT. Health Technol Assess 2024;28(66). https://doi.org/10.3310/BFCR7986
Health Technology Assessment
ISSN 2046-4924 (Online)

Impact factor: 3.6

A list of Journals Library editors can be found on the NIHR Journals Library website

Launched in 1997, Health Technology Assessment (HTA) has an impact factor of 3.6 and is ranked 32nd (out of 105 titles) in the
‘Health Care Sciences & Services’ category of the Clarivate 2022 Journal Citation Reports (Science Edition). It is also indexed by
MEDLINE, CINAHL (EBSCO Information Services, Ipswich, MA, USA), EMBASE (Elsevier, Amsterdam, the Netherlands), NCBI
Bookshelf, DOAJ, Europe PMC, the Cochrane Library (John Wiley & Sons, Inc., Hoboken, NJ, USA), INAHTA, the British Nursing
Index (ProQuest LLC, Ann Arbor, MI, USA), Ulrichsweb™ (ProQuest LLC, Ann Arbor, MI, USA) and the Science Citation Index
Expanded™ (Clarivate™, Philadelphia, PA, USA).

This journal is a member of and subscribes to the principles of the Committee on Publication Ethics (COPE)
(www.publicationethics.org/).

Editorial contact: journals.library@nihr.ac.uk

The full HTA archive is freely available to view online at www.journalslibrary.nihr.ac.uk/hta.

Criteria for inclusion in the Health Technology Assessment journal

Manuscripts are published in Health Technology Assessment (HTA) if (1) they have resulted from work for the HTA
programme, and (2) they are of a sufficiently high scientific quality as assessed by the reviewers and editors.

Reviews in Health Technology Assessment are termed ‘systematic’ when the account of the search appraisal and synthesis
methods (to minimise biases and random errors) would, in theory, permit the replication of the review by others.

HTA programme
Health Technology Assessment (HTA) research is undertaken where some evidence already exists to show that a technology can
be effective and this needs to be compared to the current standard intervention to see which works best. Research can evaluate
any intervention used in the treatment, prevention or diagnosis of disease, provided the study outcomes lead to findings that
have the potential to be of direct benefit to NHS patients. Technologies in this context mean any method used to promote health;
prevent and treat disease; and improve rehabilitation or long-term care. They are not confined to new drugs and include any
intervention used in the treatment, prevention or diagnosis of disease.

The journal is indexed in NHS Evidence via its abstracts included in MEDLINE and its Technology Assessment Reports inform
National Institute for Health and Care Excellence (NICE) guidance. HTA research is also an important source of evidence for
National Screening Committee (NSC) policy decisions.

This article
The research reported in this issue of the journal was funded by the HTA programme as award number 16/162/01. The
contractual start date was in September 2018. The draft manuscript began editorial review in July 2023 and was accepted for
publication in January 2024. The authors have been wholly responsible for all data collection, analysis and interpretation, and for
writing up their work. The HTA editors and publisher have tried to ensure the accuracy of the authors’ manuscript and would like
to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages
or losses arising from material published in this article.

This article presents independent research funded by the National Institute for Health and Care Research (NIHR). The views
and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS,
the NIHR, the HTA programme or the Department of Health and Social Care. If there are verbatim quotations included in this
publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect
those of the authors, those of the NHS, the NIHR, the HTA programme or the Department of Health and Social Care.

This article was published based on current knowledge at the time and date of publication. NIHR is committed to being inclusive
and will continually monitor best practice and guidance in relation to terminology and language to ensure that we remain relevant
to our stakeholders.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract
issued by the Secretary of State for Health and Social Care. This is an Open Access publication distributed under the terms of
the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, distribution, reproduction and adaptation
in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/.
For attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must
be cited.

Published by the NIHR Journals Library (www.journalslibrary.nihr.ac.uk), produced by Newgen Digitalworks Pvt Ltd, Chennai, India
(www.newgen.co).
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Abstract

Low-dose titrated amitriptyline as second-line treatment

for adults with irritable bowel syndrome in primary care:
the ATLANTIS RCT

Alexandra Wright-Hughes ,1 Alexander C Ford ,2,3*

Sarah L Alderson ,4 Pei Loo Ow ,1 Matthew J Ridd ,5 Robbie Foy ,4
Felicity L Bishop ,6 Matthew Chaddock,7 Heather Cook ,1
Deborah Cooper,4 Catherine Fernandez ,1 Elspeth A Guthrie ,4
Suzanne Hartley ,1 Amy Herbert ,5 Daniel Howdon ,4 Delia P Muir ,1
Sonia Newman ,8 Christopher A Taylor ,1 Emma J Teasdale ,6
Ruth Thornton ,8 Hazel A Everitt 8 and Amanda J Farrin 1

1
Clinical Trial Research Unit, Leeds Institute of Clinical Trials Research, School of Medicine, University of
Leeds, Leeds, UK
2
Leeds Institute of Medical Research at St. James’s, University of Leeds, Leeds, UK
3
Leeds Gastroenterology Institute, St. James’s University Hospital, Leeds, UK
4
Leeds Institute of Health Sciences, School of Medicine, University of Leeds, Leeds, UK
5
Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, UK
6
Centre for Clinical and Community Applications of Health Psychology, School of Psychology, University
of Southampton, Southampton, UK
7
Let’s Cure IBS, Leeds, UK
8
Primary Care Research Centre, Faculty of Medicine, University of Southampton, Southampton, UK

Corresponding author A.C.Ford@leeds.ac.uk

*

Background: Irritable bowel syndrome, characterised by abdominal pain and a change in stool
form or frequency, is most often managed in primary care. When first-line therapies are ineffective,
National Institute for Health and Care Excellence guidelines suggest considering low-dose tricyclic
antidepressants as second-line treatment, but their effectiveness in primary care is unknown and they
are infrequently prescribed by general practitioners.

Objective: To evaluate the clinical and cost-effectiveness of low-dose titrated amitriptyline as a second-
line treatment for irritable bowel syndrome in primary care.

Design: A pragmatic, randomised, multicentre, two-arm, double-blind, placebo-controlled trial. A

nested, qualitative study explored participant and general practitioner experiences of treatments
and trial participation, and implications for wider use of amitriptyline for irritable bowel syndrome in
primary care.

Participants, clinicians, investigators and analysts were masked to allocation.

Setting: Fifty-five general practices in three regions in England (Wessex, West of England,
West Yorkshire).

Participants: Patients aged ≥ 18 years meeting Rome IV criteria for irritable bowel syndrome with
ongoing symptoms after trying first-line treatments and no contraindications to TCAs.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, v
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Abstract

Intervention: Amitriptyline 10 mg once-daily, self-titrated by participants to a maximum of 30 mg

once-daily or matched placebo for 6 months. Participants randomised 1 : 1 with most having the option
to continue blinded treatment for a further 6 months.

Main outcome measures: The primary participant-reported outcome was the effect of amitriptyline on
global irritable bowel syndrome symptoms at 6 months, measured using the irritable bowel syndrome
Severity Scoring System, with a 35-point between-group difference defined as the minimum clinically
important difference. The key secondary outcome was the proportion of participants reporting
subjective global assessment of relief at 6 months, defined as somewhat, considerable, or complete
relief of symptoms. Other secondary outcomes included: effect on global symptoms, via the irritable
bowel syndrome Severity Scoring System, and subjective global assessment of relief of irritable bowel
syndrome symptoms at 3 and 12 months; effect on somatic symptom-reporting at 6 months; anxiety
an–d depression scores; ability to work and participate in other activities at 3, 6 and 12 months;
acceptability, tolerability and adherence to trial medication.

Results: Four hundred and sixty-three participants were randomised to amitriptyline (232) or placebo
(231). An intention-to-treat analysis of the primary outcome showed a significant difference in favour
of amitriptyline for irritable bowel syndrome Severity Scoring System score between arms at 6 months
[−27.0, 95% confidence interval (CI) −46.9 to −7.10; p = 0.008]. For the key secondary outcome of
subjective global assessment of relief of irritable bowel syndrome symptoms, amitriptyline was superior
to placebo at 6 months (odds ratio 1.78, 95% CI 1.19 to 2.66; p = 0.005). Amitriptyline was superior
to placebo across a range of other irritable bowel syndrome symptom measures but had no impact on
somatoform symptom-reporting, anxiety, depression, or work and social adjustment scores. Adverse
event trial withdrawals were more common with amitriptyline (12.9% vs. 8.7% for placebo) but most
adverse events were mild. The qualitative study thematically analysed 77 semistructured interviews
with 42 participants and 16 GPs. Most participants found the self-titration process acceptable
and empowering.

Conclusions: General practitioners should offer low-dose amitriptyline to patients with irritable bowel
syndrome whose symptoms do not improve with first-line therapies. Guidance and resources should
support GP–patient communication to distinguish amitriptyline for irritable bowel syndrome from use as
an antidepressant and to support patients managing their own dose titration.

Study registration: This trial is registered as ISRCTN48075063.

Funding: This award was funded by the National Institute for Health and Care Research (NIHR)
Health Technology Assessment programme (NIHR award ref: 16/162/01) and is published in full in
Health Technology Assessment Vol. 28, No. 66. See the NIHR Funding and Awards website for further
award information.

vi

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Contents
List of tables xi

List of figures xv

List of supplementary material xvii

List of abbreviations xix

Plain language summary xxi

Scientific summary xxiii

Chapter 1 Introduction 1
Background and rationale 1
Objectives 2

Chapter 2 Methods 3
Trial design 3
Trial objectives and outcome measures 3
Primary 3
Key secondary 3
Secondary 3
Cost-effectiveness objectives 4
Nested qualitative study objectives 4
Participants 4
Inclusion criteria 4
Exclusion criteria 5
Study settings 6
Screening, recruitment and registration 6
Randomisation 8
Blinding 8
Intervention 9
Assessments and data collection 10
Summary of changes to the protocol 13
Sample size 14
Statistical methods 14
Descriptive analysis 14
Primary outcome analysis 15
Secondary outcome analysis 15
Safety analysis 16
Exploratory analysis 16
Health economics methods 16
Qualitative study 16

Chapter 3 Clinical trial results 17

Study summary 17
Screening and recruitment 18

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, vii
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Contents

Protocol violations, withdrawals and follow-up 19

Analysis populations 22
Baseline characteristics 26
Demographics and clinical characteristics 26
Baseline questionnaires 26
Six-month treatment delivery and receipt 32
Treatment receipt and completion 32
Treatment adherence, dosage titration and modification 34
Further treatment summaries 34
Primary end point: 6-month global symptoms of irritable bowel syndrome (irritable
bowel syndrome Severity Scoring System) 36
Key secondary end point: subjective global assessment of relief of irritable bowel
syndrome symptoms at 6 months 41
Secondary end points 41
Global symptoms of irritable bowel syndrome (irritable bowel syndrome Severity
Scoring System) at 3 months 41
Subjective global assessment of relief of irritable bowel syndrome symptoms at 3 months 44
Hospital Anxiety and Depression Scale-A scores at 3 and 6 months 44
Hospital Anxiety and Depression Scale-D scores at 3 and 6 months 47
Ability to work and participate in other activities (work and social adjustment scale
scores) at 3 and 6 months 47
Irritable bowel syndrome-associated somatic symptom-reporting (patient health
questionnaire-12 scores) at 6 months 47
Participant-reported weekly adequate relief of irritable bowel syndrome symptoms 48
Acceptability of trial medication at 6 months 50
Adherence 51
Tolerability at 3 and 6 months 51
Further 12-month treatment delivery and secondary end points 51
Treatment receipt, adherence, dosage and titration 51
Secondary end points 57
Exploratory analysis 64
50-point reduction in irritable bowel syndrome Severity Scoring System at
3 and 6 months 64
Thirty per cent reduction in 6-month irritable bowel syndrome Severity Scoring
System abdominal pain and distention 64
Moderator analysis of 6-month irritable bowel syndrome Severity Scoring System score 64
Moderator analysis of 6-month subjective global assessment of relief of irritable
bowel syndrome symptoms 64
Safety 66
Serious adverse events and reactions 66
Other safety events 66
Blinding 68
Emergency unblinding 68
Treatment allocation exit survey 68
End-of-trial participation unblinding 68

Chapter 4 Nested qualitative study 73

Aims73
Methods 73
Design 73
Ethical considerations 73
Sampling and recruitment 74
Qualitative interviews 75

viii

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Participant interviews 75
General practitioner interviews 76
Qualitative data analysis 76
Findings 77
Participants 77
Barriers and facilitators to uptake of low-dose 78
Experiences of taking part in the ATLANTIS trial 86
Reflections on managing IBS during the coronavirus disease discovered in 2019 pandemic 89
Reflections on treatment allocation 89
Relating the qualitative and quantitative findings 93
Discussion 95
Overview 95
Barriers and facilitators 95
Experiences of the trial 96
Irritable bowel syndrome during the coronavirus disease discovered in 2019 pandemic 96
Treatment allocation 97
Strengths and limitations 97
Conclusion 98

Chapter 5 Discussion 99
Limitations 100
Generalisability 100
Interpretation 101
Overall evidence 102
Recommendations for future research 103
Equality, diversity and inclusion 103
Patient and public involvement 103
Conclusions 104

Additional information 105

References 113

Appendix 1 Additional clinical results tables and figures 119

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, ix
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

List of tables
TABLE 1 Participant study schedule 11

TABLE 2 Demographics of patients mailed-out to, responded, interested, eligible,

registered and randomised 21

TABLE 3 Withdrawals 21

TABLE 4 Monthly questionnaire completion 23

TABLE 5 Characteristics of participants completing and not completing 6-month

questionnaires 25

TABLE 6 Per-protocol and safety population 27

TABLE 7 Randomisation stratification factors 28

TABLE 8 Demographic characteristics of randomised participants 28

TABLE 9 Clinical characteristics of randomised participants 29

TABLE 10 Baseline questionnaires 30

TABLE 11 Six-month treatment receipt 32

TABLE 12 Time from randomisation to treatment end date 33

TABLE 13 Treatment adherence and dose for participants on treatment: at 3 weeks, 3

and 6 months 35

TABLE 14 Number of participants trying a new diet, changing their amount of exercise,
or trying other irritable bowel syndrome treatments during the study, and experienced
improvement in irritable bowel syndrome symptoms 37

TABLE 15 Total IBS-SSS score, IBS-SSS severity and change in IBS-SSS score from
baseline 38

TABLE 16 Six-month total IBS-SSS score: linear regression model – parameter

estimates in primary, complete case and per-protocol analysis 40

TABLE 17 Subjective global assessment of relief of IBS symptoms at 3 and 6 months 42

TABLE 18 Six-month SGA of relief of IBS symptoms: logistic and ordinal regression –
parameter estimates in primary, sensitivity and secondary analysis 43

TABLE 19 Treatment effect estimates of 3-month total IBS-SSS score and SGA of relief
of IBS symptoms secondary outcomes: primary, secondary and sensitivity analysis 44

TABLE 20 Hospital Anxiety and Depression Scale-A, HADS-D, WSAS and PHQ-12
scores at baseline and 3 and 6 months 45

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, xi
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 List of tables

TABLE 21 Treatment effect estimates of 3- and 6-month HADS, WSAS and PHQ-12
secondary outcomes: primary and sensitivity analysis 47

TABLE 22 Overall weekly relief summary 48

TABLE 23 Acceptability of trial medication at 6 months 50

TABLE 24 Six-month acceptability: logistic regression models – adjusted OR of

acceptability in primary and sensitivity analysis 50

TABLE 25 Adherence to therapy end-point summary – up to 6 months 52

TABLE 26 Participant-reported tolerability on the ASEC at 3 and 6 months 53

TABLE 27 Twelve-month treatment receipt 56

TABLE 28 Treatment adherence, dose and replenishment for participants on

treatment: at 6, 9 and 12 months 58

TABLE 29 Baseline and 12-month IBS-SSS and SGA outcomes 59

TABLE 30 Twelve-month treatment effect estimates in primary and sensitivity analysis

of irritable bowel syndrome Severity Scoring System, subjective global assessment,
Hospital Anxiety and Depression Scale and Work and Social Adjustment Scale
secondary outcomes 61

TABLE 31 Baseline and 12-month HADS and WSAS outcomes 62

TABLE 32 Participant-reported tolerability on the ASEC at 12 months 66

TABLE 33 Moderator analyses of the treatment effect on the 6-month total

IBS-SSS score 67

TABLE 34 Moderator analyses of the treatment effect on the 6-month SGA of

relief of IBS symptoms by IBS subtype 68

TABLE 35 Six-month SGA of relief of IBS symptoms score by IBS subtype 69

TABLE 36 Serious AE summary 70

TABLE 37 Exit survey of which treatment participants thought they received 71

TABLE 38 End-of-trial participation unblinding summary 72

TABLE 39 Baseline characteristics of interviewees at 6 and 12 months 78

TABLE 40 Demographic data for GP interviewees (n = 16) 79

TABLE 41 Mapping themes to key quantitative findings 94

TABLE 42 Patient screening and recruitment by hub 119

TABLE 43 Number and reasons for protocol violations 119

xii

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 44 Index of multiple deprivation decile by treatment allocation and

recruitment hub 122

TABLE 45 Treatment adherence and dose reported at last follow-up for participants
discontinued trial medication before month 6 126

TABLE 46 Treatment adherence and dose at trial medication discontinuation: after

month 6, all discontinuations 126

TABLE 47 Treatment replenishment for participants on trial medication 127

TABLE 48 Details of new diets, other IBS treatments, and attributed reasons
for any improvement in IBS symptoms 128

TABLE 49 Irritable bowel syndrome Severity Scoring System items 131

TABLE 50 Primary outcome missing data exploration – included in multiple

imputation model 136

TABLE 51 Three-month total IBS-SSS score: linear regression – primary (ITT) and
sensitivity (complete case, per protocol) analysis 138

TABLE 52 Three-month SGA of relief of IBS symptoms: logistic and ordinal regression –
primary, sensitivity and secondary analysis 139

TABLE 53 Hospital Anxiety and Depression Scale-A (3, 6 and 12 months): linear
regression – primary (ITT, 12-month ITT) and sensitivity (complete case) analysis 142

TABLE 54 Hospital Anxiety and Depression Scale-D (3, 6 and 12 months): linear
regression – primary (ITT, 12-month ITT) and sensitivity (complete case) analysis 143

TABLE 55 Work and Social Adjustment Scale (3, 6 and 12 months): linear regression –
primary (ITT, 12-month ITT) and sensitivity (complete case) analysis 144

TABLE 56 Patient Health Questionnaire-12 (6 months): linear regression – primary

(ITT) and sensitivity (complete case) analysis 145

TABLE 57 Number and proportion of participants reporting adequate relief each

week and model estimates 147

TABLE 58 Participant-reported tolerability on the ASEC: number of participants

reporting symptoms (mild, moderate or severe) at 3, 6 and 12 months for
participants on trial medication 149

TABLE 59 Twelve-month total IBS-SSS score: linear regression – primary

(12-month ITT) and sensitivity (complete case) analysis 152

TABLE 60 Twelve-month SGA of relief of IBS symptoms: logistic regression – primary

and sensitivity analysis 153

TABLE 61 ≥ 50-point reduction in total IBS-SSS score at 3 and 6 months: logistic

regression (n = 463, multiple imputation) 154

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, xiii
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 List of tables

TABLE 62 ≥ 30% reduction in IBS-SSS abdominal pain at 3 and 6 months: logistic

regression (n = 463, multiple imputation) 155

TABLE 63 ≥ 30% reduction in IBS-SSS abdominal distention: logistic regression

(n = 463, multiple imputation) 156

TABLE 64 Mean (SD) total IBS-SSS score at baseline and 6 months by IBS subtype
and recruitment hub 156

xiv

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

List of figures
FIGURE 1 Consolidated standards of reporting trial flow diagram 17

FIGURE 2 Screening and recruitment by hub 18

FIGURE 3 Recruitment graph 20

FIGURE 4 Estimated treatment effect on weekly relief: logistic regression:

443 participants with at least one weekly response 49

FIGURE 5 Participant-reported tolerability on the ASEC at 3 months

(safety population for participants on treatment) 54

FIGURE 6 Participant-reported tolerability on the ASEC at 6 months

(safety population for participants on treatment) 55

FIGURE 7 Participant-reported tolerability on the ASEC at 12 months

(safety population for participants on treatment) 65

FIGURE 8 Time from general practice mail-out to randomisation 120

FIGURE 9 Timing of questionnaire completion 120

FIGURE 10 Weekly adequate relief question completion 121

FIGURE 11 Years from IBS diagnosis to trial randomisation 123

FIGURE 12 Reported side effects as a reason for treatment discontinuation 124

FIGURE 13 Distribution of time from randomisation to treatment end date 124

FIGURE 14 Treatment adherence reported at each follow-up time point

(for participants on trial medication) 125

FIGURE 15 Treatment dose reported at each follow-up time point

(for participants on trial medication) 125

FIGURE 16 Total IBS-SSS score severity at each time point 129

FIGURE 17 Unadjusted total IBS-SSS score and item level scores with
95% CIs based on available data 130

FIGURE 18 Plots of residuals for 6-month total IBS-SSS score 135

FIGURE 19 Subjective global assessment of relief of IBS symptoms at 3 months, 6

months and 12 months 137

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, xv
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 List of figures

FIGURE 20 Plots of residuals for logistic regression of the 6-month SGA 137

FIGURE 21 Unadjusted HADS-A and HADS-D scores with 95% CIs based
on available data 140

FIGURE 22 Unadjusted WSAS scores with 95% CIs based on available data 141

FIGURE 23 Number and proportion of participants reporting adequate relief

each week 146

FIGURE 24 Moderating effect of recruitment hub on the total IBS-SSS score

treatment effect at 6 months 157

FIGURE 25 Total IBS-SSS score by treatment arm and recruitment hub 157

FIGURE 26 Moderating effect of IBS subtype on the total IBS-SSS score

treatment effect at 6 months 158

FIGURE 27 Total IBS-SSS score by treatment arm and IBS subtype 158

FIGURE 28 Moderating effect of baseline IBS-SSS score on the total IBS-SSS score
treatment effect at 6 months 159

FIGURE 29 Moderating effect of baseline HADS-A score on total IBS-SSS score at

6 months 159

FIGURE 30 Moderating effect of baseline HADS-D score on total IBS-SSS score at

6 months 160

FIGURE 31 Moderating effect of IBS subtype on SGA of relief of IBS symptoms

treatment effect at 6 months 160

FIGURE 32 Subjective global assessment of relief of IBS symptoms at 6 months by

treatment arm and stool type 161

xvi

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

List of supplementary material

Report Supplementary Material 1 Participant-facing materials used in the randomised
trial and nested qualitative study

Supplementary material can be found on the NIHR Journals Library report page (https://doi.
org/10.3310/BFCR7986).

Supplementary material has been provided by the authors to support the report and any files
provided at submission will have been seen by peer reviewers, but not extensively reviewed.
Any supplementary material provided at a later stage in the process may not have been
peer reviewed.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, xvii
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

List of abbreviations
AE adverse event NICE National Institute for Health
and Care Excellence
AR adverse reaction
NIHR National Institute for Health
ASEC Antidepressant Side-Effect
and Care Research
Checklist
OR odds ratio
CI confidence interval
PHQ-12 Patient Health
CONSORT Consolidated Standards of
Questionnaire-12
Reporting Trials
PI principal investigator
CRP C-reactive protein
PIC participant identification centre
CTRU Clinical Trials Research Unit
PPI patient and public involvement
DMEC Data Monitoring and Ethics
Committee QALYs quality-adjusted life-years
GP general practitioner RCT randomised controlled trial
HADS Hospital Anxiety and SAE serious adverse event
Depression Scale SAR serious adverse reaction
HTA Health Technology Assessment SD standard deviation
IBS irritable bowel syndrome SGA subjective global assessment
IBS-C irritable bowel syndrome with SUSAR suspected unexpected serious
constipation adverse reaction
IBS-D irritable bowel syndrome with TCA tricyclic antidepressant
diarrhoea
TMG Trial Management Group
IBS-M irritable bowel syndrome with
mixed bowel habits TSC Trial Steering Committee

IBS-U irritable bowel syndrome tTG tissue transglutaminase

unclassified WSAS Work and Social Adjustment
IBS-SSS irritable bowel Scale
syndrome Severity Scoring WCC white cell count
System

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, xix
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Plain language summary

Background

People with irritable bowel syndrome experience stomach (abdominal) pain and changes to their bowel
movements. Irritable bowel syndrome can have a serious impact on people’s lives. Previous small
trials suggest that a drug called amitriptyline used at a low dose may help irritable bowel syndrome.
Amitriptyline is already used to treat other conditions. It is available for irritable bowel syndrome but is
not used much by general practitioners.

Methods

We recruited adults aged ≥ 18 years with irritable bowel syndrome from UK general practices who did
not have any issues preventing the use of amitriptyline. Patients received either low-dose amitriptyline
or placebo (a dummy tablet) for 6 months. Patients could adjust the dose according to symptoms
and side effects. Neither the researchers nor the patients knew which treatment they were getting.
Participants recorded symptoms using a questionnaire containing an irritable bowel syndrome severity
score. We looked at the difference in average irritable bowel syndrome severity score between patients
receiving amitriptyline and placebo. We also looked at effects of amitriptyline on mood, ability to work,
and non-gut symptoms related to irritable bowel syndrome, as well as safety and acceptability. Some
patients and general practitioners were interviewed about their experiences.

Results

Four hundred and sixty-three patients took part. Participants receiving amitriptyline reported a bigger
improvement in their irritable bowel syndrome severity scores at 6 months, compared with patients on
placebo. Amitriptyline was better across a range of irritable bowel syndrome symptom measures but did
not impact anxiety, depression or ability to work. Forty-six people (19.8%) stopped taking amitriptyline
and 59 (25.5%) stopped the placebo before 6 months. Patients liked being able to adjust their dose and
valued contact with the research team.

Conclusion

This study showed that amitriptyline is more effective than a placebo and is safe. General practitioners
should offer low-dose amitriptyline to people with irritable bowel syndrome if symptoms do not improve
with other standard treatments. Patients should be supported and helped to adjust their dose as
needed. The dose adjustment sheet used in this trial will be made available.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, xxi
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Scientific summary
Background

Irritable bowel syndrome (IBS) affects 5% of the population, accounting for > 3% of all consultations
in primary care in England and Wales. Symptoms include abdominal pain in association with a change
in stool form or frequency. The condition impacts on quality of life and ability to work and limits
social activities. The medical management of IBS is unsatisfactory, with no therapy proven to alter
the long-term natural history and, at best, modest symptom reduction. Previous meta-analyses of
trials conducted in secondary and tertiary care suggest low-dose tricyclic antidepressants (TCAs) may
be efficacious, probably because of their pain-modifying properties, as well as their influence on gut
motility, rather than any effects on mood. Although National Institute for Health and Care Excellence
guidelines for the management of IBS in primary care suggest considering low-dose TCAs as second-line
treatment, their effectiveness in this setting is unknown and they are infrequently prescribed by general
practitioners (GPs).

Objectives

Our objective was to determine the clinical and cost-effectiveness of low-dose titrated amitriptyline
compared with placebo for 6 months as a second-line treatment in adults with IBS in primary care.

Methods

ATLANTIS was a pragmatic, randomised, multicentre, parallel-group, two-arm, double-blind, placebo-

controlled trial. A nested, qualitative study explored participant and GP experiences of treatments and
trial participation. A within-study cost-effectiveness analysis was planned but, due to the coronavirus
disease discovered in 2019 (COVID-19) pandemic, health economic analyses were removed after
obtaining additional funding to complete the trial to prioritise funds for participant recruitment. These
will be subject to further funding. Participants, their GPs, investigators, the research team, and the
analysis team were all masked to treatment allocation throughout the trial. Patients meeting Rome IV
criteria for IBS who had tried first-line treatments and with ongoing IBS symptoms [score of ≥ 75 on the
IBS Severity Scoring System (IBS-SSS)] were recruited via mail-out from 55 general practices in three
regions in England. Participants were randomised 1 : 1 to receive either low-dose titrated amitriptyline
or placebo. Both treatments were supplied for 6 months, with the dose commenced at 10 mg o.d.
and titrated to a maximum of 30 mg o.d. or a minimum of 10 mg alternate days. Dose titration was
participant-led according to IBS symptoms and side effects, with support from the trial team and a dose
titration document developed with input via patient and public involvement. Participants recruited
earlier to the trial had the option to continue blinded treatment for an additional 6 months.

The primary outcome was the effect of amitriptyline on global IBS symptom scores at 6 months. The key
secondary outcome was the proportion of participants with relief of IBS symptoms at 6 months. Other
secondary outcomes included effect on global IBS symptoms and relief of IBS symptoms at 3 and 12
months, effect on IBS-associated somatic symptoms at 6 months, effect on quality of life, anxiety, and
depression scores, and ability to work and participate in other activities at 3, 6 and 12 months, as well as
acceptability and tolerability of, and adherence to, treatment.

Patient-reported questionnaires at baseline and 3, 6 and 12 months post randomisation (unless

otherwise indicated) were used to assess IBS symptom severity (measured via the IBS-SSS), relief of IBS
symptoms [measured by subjective global assessment (SGA) of relief], adequate relief of IBS symptoms

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, xxiii
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Scientific summary

(measured by a weekly response to the question ‘Have you had adequate relief of your IBS symptoms?’),
IBS-associated somatic symptoms [using the Patient Health Questionnaire-12 (PHQ-12)], mood [using
the Hospital Anxiety and Depression Scale (HADS)], ability to work and participate in other activities
[using the Work and Social Adjustment Scale (WSAS)], quality of life (using the EQ-5D-3L), healthcare
use (using a bespoke health resource use questionnaire), and tolerability [using the Antidepressant Side-
Effect Checklist (ASEC)]. Numbers of participants reporting serious adverse events (SAEs), including
serious adverse reactions (SARs), were reported for each treatment group.

An evaluable sample size of 414 participants would provide 90% power to detect a minimum clinically
important difference of 35 points between amitriptyline and placebo at 6 months on the IBS-SSS. This
sample size provided at least 85% power to detect a 15% absolute difference in the key secondary
outcome of SGA of relief of IBS symptoms at 6 months. We planned to recruit 518 participants, allowing
for 20% loss to follow-up. Effectiveness outcomes were analysed in the intention-to-treat population,
defined as all participants randomised, regardless of adherence. All statistical testing used two-sided
5% significance levels. The primary outcome was analysed using a linear regression model, adjusted
for minimisation variables and baseline IBS-SSS score. Missing data were imputed by treatment arm,
via multiple imputation, and results were expressed as point estimates with 95% confidence intervals
(CIs). Secondary binary outcomes were analysed in logistic or ordinal regression models, with results
expressed as odds ratios (ORs) with 95% CIs. Continuous secondary outcomes, including PHQ-12,
HADS and WSAS scores, were analysed as for the primary outcome, adjusted for the respective baseline
score. All participants receiving at least one dose of trial medication, according to medication received,
were included in the safety analysis.

The nested, qualitative study aimed to identify factors that would facilitate or impede prescribing of,
acceptability of, and adherence to, low-dose amitriptyline in IBS, to identify participants’ and GPs’
perspectives on the broader impact of the trial, and to explore psychosocial and contextual factors
that might shape wider use of amitriptyline for IBS. Familiarity with amitriptyline may both hinder
uptake, given its association with depression, and facilitate it, given that it is a known drug, taken in a
low dose distinct from the antidepressant dose, already used for a range of other painful conditions
and has comparatively mild, and in some cases potentially beneficial, side effects such as on sleep.
Semi-structured audio-recorded telephone interviews were conducted with a diverse sub-sample
of trial participants and GPs involved in the trial and transcribed verbatim. The final sample size was
dependent on saturation, to achieve a rigorous, credible analysis in relation to the aims. Topic guides
allowed flexible exploration of all required topics, while remaining open to participants’ individual
experiences and perspectives. To enhance trustworthiness of the analysis, all qualitative study team
members contributed to avoid producing idiosyncratic interpretations, a negative case analysis was
undertaken, and an audit trail was produced to enhance transparency, including detailed coding manuals
and interviewer field notes. Reflexive thematic analysis, incorporating techniques from grounded theory,
was used to analyse the qualitative data. Data collection and initial analyses proceeded iteratively, and
informed subsequent interviews. Analysis was primarily inductive, with researchers identifying themes in
the data rather than imposing any pre-existing interpretive framework. Qualitative findings were related
to the main trial findings by comparing themes across subgroups and against the quantitative data.

Clinical results

In total, 15,672 potentially eligible patients were invited to take part, of whom 1253 were interested
and were screened. Of those screened, 463 (37.0%) were randomised {mean age 48.5 years [standard
deviation (SD) 16.1 years], 315 (68.0%) female}, to amitriptyline (n = 232) or placebo (n = 231). Six-
month follow-up was achieved for 401 (86.6%) participants, 204 (87.9%) in the amitriptyline arm, and
197 (85.3%) in the placebo arm. Participants were well balanced between treatment arms according to
demographics and baseline characteristics, IBS symptom severity, PHQ-12 scores, HADS-depression
and HADS-anxiety scores, and previous first-line treatments. Among participants, 80.4% had IBS-D or

xxiv

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

IBS-M, 84.2% had a normal HADS-depression score, and 84.7% had moderate to severe scores on the
IBS-SSS, with a mean IBS-SSS score in all participants of 272.8 (SD 90.3). The median duration of IBS
was 10 years.

In total, 338 (73.0%) participants completed 6 months of treatment, 173 (74.6%) randomised to
amitriptyline and 165 (71.4%) to placebo. Discontinuation of trial medication before 6 months occurred
in 105 (22.7%) participants, 46 (19.8%) allocated to amitriptyline and 59 (25.5%) to placebo. The most
common reason for discontinuing trial medication was adverse events (AEs) in 30 (12.9%) participants
allocated to amitriptyline and 20 (8.7%) to placebo, followed by lack of benefit in 7 (3.0%) randomised
to amitriptyline and 18 (7.8%) to placebo. There were a further 17 (3.7%) participants lost to follow-up
and 3 (0.6%) who did not commence trial medication. By 3 months, similar proportions of participants
randomised to amitriptyline had titrated their dose to 20 mg o.d. (35.2%) or 30 mg o.d. (37.8%),
although by 6 months this had increased to 42.8% taking 30 mg o.d. However, in the placebo arm,
57.0% of participants titrated their dose to 30 mg o.d. within 3 months and this proportion was similar
at 6 months.

For the primary outcome, amitriptyline was superior to placebo at 6 months in the intention-to-treat
analysis, with a significant difference in mean IBS-SSS score between arms (−27.0, 95% CI −46.9 to
−7.1; p = 0.008). For the key secondary outcome, SGA of relief of IBS symptoms, amitriptyline was also
superior to placebo (OR for relief of IBS symptoms = 1.78, 95% CI 1.19 to 2.66; p = 0.005). At 3 months,
the difference in mean change in IBS-SSS score also favoured amitriptyline (−23.3, 95% CI −42.0 to
−4.6; p = 0.014), as did the SGA of relief of IBS symptoms (OR = 1.70, 95% CI 1.15 to 2.53; p = 0.008). In
a sensitivity analysis using an alternative definition of SGA of relief of IBS symptoms, where only those
reporting considerable or complete relief of IBS symptoms at 3 or 6 months were classed as responders,
the effect size in the amitriptyline arm increased at both 3 (OR = 1.81, 95% CI 1.17 to 2.79) and 6
months (OR = 1.88, 95% CI 1.20 to 2.95). Other sensitivity analyses on the per-protocol population
for the primary outcome and on participants with complete data for the primary and key secondary
outcomes gave consistent results, albeit with larger estimated treatment effects.

In terms of adequate relief of IBS symptoms, amitriptyline was also superior to placebo with increased
odds of adequate relief across all 25 weeks (OR = 1.56, 95% CI 1.20 to 2.03; p < 0.001), and a higher
proportion of participants reporting adequate relief for ≥ 13 of 25 weeks [90/222 (40.5%) vs. 67/221
(30.3%)]. Significantly higher numbers of participants taking amitriptyline reported the drug to be
acceptable and would have been willing to continue taking it at 6 months (OR = 1.60, 95% CI 1.08 to
2.35; p = 0.018). Adherence at 3 months was identical in the two treatment arms, but it was higher
in the amitriptyline arm at 6 months [172/232 (74.1%) vs. 155/228 (68.0%)]. Amitriptyline had no
significant effect on PHQ-12 scores at 6 months, or HADS-anxiety, HADS-depression or WSAS scores
at either 3 or 6 months.

In terms of treatment-emergent AEs, there was a statistically significant increase in the total ASEC
score in those receiving amitriptyline compared with placebo at 3 months (1.39, 95% CI 0.29 to 2.50;
p = 0.013) but not at 6 months (0.26, 95% CI −0.98 to 1.51; p = 0.681). The AEs reported in participants
receiving amitriptyline in excess of those reported by the placebo arm mainly related to its known
anticholinergic effects, including dry mouth, drowsiness, blurred vision and problems with urination.
However, rates of treatment-emergent AEs fell between 3 and 6 months and few were severe. The
commonest AEs leading to discontinuation in the amitriptyline arm were drowsiness and deterioration
of mood. In total, there were five SARs, two in the amitriptyline arm and three in the placebo arm. There
were five SAEs unrelated to trial medication, of which four occurred in the amitriptyline arm and one in
the placebo arm.

In the subset of participants recruited to 12 months’ follow-up and with the choice to continue
treatment beyond 6 months, 44% of participants completed 12 months’ treatment. Despite the mixed
sample, in the 12-month ITT population, weak evidence of a significant effect in favour of low-dose

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, xxv
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Scientific summary

amitriptyline remained on the mean IBS-SSS (−22.6, 95% CI −49.35 to −4.16; p = 0.098) and the SGA
of relief of global IBS symptoms (OR = 1.58, 95% CI 0.94 to 2.64; p = 0.083). In contrast to 6-month
results, there was a statistically significant effect on the HADS-depression (−0.88, 95% CI −1.71 to
0.06; p = 0.036) and WSAS (−2.14, 95% CI −3.80 to −0.49; p = 0.011) scores in favour of low-dose
amitriptyline.

Qualitative results

The qualitative study conducted and thematically analysed 77 semistructured interviews with 42
participants and 16 GPs. A multidisciplinary team including patient collaborators explored multiple
aspects of participants’ and GPs’ experiences of treatments and participating in the ATLANTIS trial.

The qualitative analysis of barriers and facilitators suggests that low-dose amitriptyline for IBS is
acceptable to, and is often welcomed by, GPs and patients as an additional treatment option. Addressing
concerns and promoting facilitators could facilitate wider use of low-dose amitriptyline for IBS which
may be achieved through:

• Clear communication to clinicians, for example in clinical guidelines, that distinguishes low-dose
amitriptyline for IBS from amitriptyline use for other conditions (especially depression).
• Resources to support GP–patient communication to distinguish low-dose amitriptyline for IBS from
amitriptyline for other conditions (especially depression). This might include, for example, tips for GPs
when discussing amitriptyline for IBS with patients, online materials to support or reinforce messages
given during consultations, tailored packaging and patient inserts, and education for pharmacists.
• Clear guidance about low-dose amitriptyline for IBS and anticholinergic burden. This should highlight
that low-dose amitriptyline has lower potential risk and that currently anticholinergic burden risk
scores do not account for dose, so can overinterpret risk with low-dose amitriptyline.
• Guidance and resources for GPs and patients to support patients managing their own dose titration.
The dose-titration document used in ATLANTIS was well received by GPs and patients.

Conclusions

In the largest trial of a TCA in IBS ever conducted, titrated low-dose amitriptyline was superior to
placebo as a second-line treatment for IBS in primary care across multiple outcomes and was safe.
The results of ATLANTIS strongly support use of titrated low-dose amitriptyline in this setting. GPs
should offer low-dose amitriptyline to patients with IBS whose symptoms do not improve with first-
line therapies, with appropriate support to guide patient-led dose titration, such as the self-titration
document developed for this trial.

Trial registration

This trial is registered as ISRCTN48075063.

Funding

This award was funded by the National Institute for Health and Care Research (NIHR) Health Technology
Assessment programme (NIHR award ref: 16/162/01) and is published in full in Health Technology
Assessment; Vol. 28, No. 66. See the NIHR Funding and Awards website for further award information.

xxvi

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Chapter 1 Introduction
Background and rationale

Irritable bowel syndrome (IBS) is a common, chronic, disorder of gut–brain interaction, characterised
by abdominal pain in association with a change in stool form or frequency.1 Prevalence is 5% in the
community,2 and IBS accounts for > 3% of all consultations in primary care.3 The total cost to the
health service in the UK has been estimated to be > £1 billion/year.4 Quality of life of people with IBS
is impaired substantially, to a level comparable with that seen in some organic bowel disorders, such as
Crohn’s disease.5 Current first-line treatment of IBS in primary care includes dietary and lifestyle advice,
fibre supplements, laxatives, antispasmodic drugs, or loperamide, but if these are ineffective, general
practitioners (GPs) are often left with few treatment options, meaning people are frequently referred to
see a specialist in secondary care.6

Medical management of IBS is unsatisfactory, with no therapy proven to alter the long-term natural
history and, at best, modest symptom reduction. Previous meta-analyses have suggested tricyclic
antidepressants (TCAs) may be an efficacious treatment.7–9 The most recent of these identified 12 trials,
which included 787 patients.9 Beneficial effects on IBS symptoms may arise from their well-known
pain-modifying properties,10–13 as well as their influence on gut motility,14 rather than any antidepressive
effects, as the doses used in randomised controlled trials (RCTs) in IBS are considerably less than the
dose required to have any effect on mood. However, duration of follow-up was limited to 12 weeks,
all trials were conducted in secondary or tertiary care, where patients have more severe symptoms,
and most studies were small. These limitations are important. The clinical relevance of demonstrating
the effectiveness of a drug over a 12-week RCT in a condition that is chronic, and often lifelong,15 is
debatable. In addition, although there is evidence from pooling data from secondary and tertiary care-
based trials in a meta-analysis, it is not clear whether this effect would translate into a benefit in primary
care, and whether this will reduce resource use and referrals to secondary care or improve quality of life
and social functioning.

The National Institute for Health and Care Excellence (NICE) guideline for the management of IBS in
primary care states only that GPs should ‘consider’ TCAs as second-line treatment for IBS for their
analgesic effect,16 for example ‘amitriptyline at a dose of 10 mg to 30 mg’, if dietary changes, fibre
supplements, laxatives, antispasmodics or loperamide have not helped. However, this guideline also
acknowledges that there is limited evidence to support this statement and proposes that a large RCT be
conducted comparing a TCA with placebo in adults with IBS in primary care, with outcomes assessed at
3, 6 and 12 months, and including global improvement in IBS symptoms, effect on health-related quality
of life, and adverse effects.

At present, therefore, there is uncertainty as to whether TCAs are effective for the treatment of IBS in
primary care, and this may mean GPs are reluctant to consider using them. In a prior survey < 10% of
GPs used them often, and only 50% believed they were effective.17 Given that 95% of GPs use these
drugs for the treatment of insomnia in primary care,18 it is presumably uncertainty over their efficacy in
IBS, rather than concerns about side effects, which explains this reluctance. If a drug that is potentially
efficacious for IBS is being under-utilised, this will have a negative effect on both the health service and
society, in terms of worse control of IBS symptoms, which will lead to lower quality of life for people
with IBS, increased sickness absences from work, and higher costs of managing IBS in secondary care,
due to greater numbers of referrals and increased rates of investigation.

Given the recommendations of the NICE guideline,16 together with the fact that two of the trials
in the meta-analyses used amitriptyline,19,20 both of which were small, but positive, and its proven
pain-modifying properties,11,12 as well as its effects on gut motility14 and visceral hypersensitivity,21

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 1
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Introduction

we chose to assess amitriptyline. This study was funded successfully as part of a commissioned call
by the National Institute for Health and Care Research (NIHR) Health Technology Assessment (HTA)
programme (NIHR award ref: 16/162/01), which identified the need to address the short- and long-term
benefits of low-dose antidepressants for IBS in primary care, to help guide treatment decisions. Our
work with patients and the public prior to obtaining funding confirmed a perceived need for the study
but identified potential concerns about the use of a drug identified as an antidepressant for a condition
like IBS. This provided a rationale for both the trial and a nested qualitative study to explore potential
barriers to implementation, should low-dose amitriptyline prove to be effective.

Objectives

The objective of the AmitripTyline at Low-dose ANd Titrated for IBS as Second-line treatment
(ATLANTIS) trial was to determine the clinical and cost-effectiveness of low-dose amitriptyline for
IBS in primary care compared with placebo. A nested, qualitative study explored participant and GP
experiences of treatments and trial participation, and implications for wider use of amitriptyline for
IBS in primary care. We aimed to deliver a definitive assessment of the benefits, harms, and cost-
effectiveness of low-dose amitriptyline as second-line treatment for IBS in primary care, within the NHS,
to guide future adoption and implementation.

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Chapter 2 Methods
Trial design

ATLANTIS was a pragmatic, randomised, multicentre, parallel-group, double-blind, placebo-controlled,

superiority trial of low-dose amitriptyline as a second-line treatment for adults with IBS in primary care.
The majority of participants recruited consented to 12-month study participation, consisting of an
initial 6 months of trial medication with the option to continue this voluntarily for a further 6 months.
Treatment duration and follow-up was curtailed to 6 months for later recruits, due to protocol changes
made during the coronavirus disease discovered in 2019 (COVID-19) pandemic. Additionally, although
the data to inform a cost-effectiveness analysis were collected, the analysis was unable to be completed
and is planned as future work, if further funding becomes available. A nested, qualitative study explored
participants’ and GPs’ experiences of treatments and participating in the trial, including acceptability,
adherence, unanticipated effects, and implications for the wider use of amitriptyline for IBS.

Patient and public involvement (PPI) representatives were involved at all stages and provided
valuable contributions to trial design, documentation and outputs. The final protocol and subsequent
amendments were approved by Yorkshire and the Humber (Sheffield) Research Ethics Committee (19/
YH/0150) and published in full.22 The trial was conducted in accordance with the principles of Good
Clinical Practice and the Declaration of Helsinki and registered with the ISRCTN (ISRCTN48075063).

Trial objectives and outcome measures

Primary
The primary objective was to determine the effect of amitriptyline on global symptoms of IBS, as
measured by the IBS Severity Scoring System (IBS-SSS), 6 months after randomisation. The IBS-SSS is a
validated, participant-reported, five-item questionnaire used widely in IBS trials.23 It measures presence,
severity and frequency of abdominal pain, presence and severity of abdominal distension or tightness,
satisfaction with bowel habit and degree to which IBS symptoms are affecting, or interfering with, the
person’s life in general. The maximum score is 500 points: a score of < 75 points indicates symptoms
that are felt to be in remission, with normal bowel function; 75–174 points indicates mild IBS symptoms,
175–299 points moderate IBS and 300–500 points severe IBS.

Key secondary
The key secondary objective was to determine the effect of amitriptyline on global symptoms of
IBS, according to the proportion of participants with subjective global assessment (SGA) of relief
of IBS symptoms.24 Participants rate their relief from IBS symptoms on a scale of 1 to 5 ranging
from ‘completely relieved’ to ‘worse’. Scores are dichotomised so that those scoring from 1 to 3 are
considered responders and those 4 or 5 non-responders. At 6 months after randomisation, response
was, therefore, defined as reporting somewhat, considerable, or complete relief of IBS symptoms.

Secondary
Secondary objectives were to assess the effect of amitriptyline on:

• Global symptoms of IBS, as measured by the IBS-SSS at 3 and 12 months.

• Global symptoms of IBS, as measured by the proportion of participants with SGA of relief of IBS
symptoms,24 at 3 and 12 months, with response defined as above.
• Adequate relief of IBS symptoms via a weekly response to the question ‘Have you had adequate
relief of your IBS symptoms?’

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 3
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Methods

• IBS-associated somatic symptoms, as measured by the Patient Health Questionnaire-12 (PHQ-

12),25,26 at 6 months.
• Anxiety and depression scores, as measured by the Hospital Anxiety and Depression Scale (HADS),27
at 3, 6 and 12 months.
• Ability to work and participate in other activities, as measured by the Work and Social Adjustment
Scale (WSAS),28-30 at 3, 6 and 12 months.
• Acceptability of treatment, as measured by a response of ‘Yes’ to the participant-reported question
at 6 months: ‘On balance do you find this medication acceptable to take and would you want to
keep taking it?’. Non-acceptability defined for any participants reporting ‘No’, having discontinued
treatment before 6 months, not starting trial treatment, or lost to follow-up.
• Self-reported adherence to treatment, based on participant-report, via the question: ‘Since you
were last asked, which of the options best describes how often you have taken at least one tablet
of the trial medication daily?’: ‘Every day or nearly every day’, ‘Half of the days or more than half the
days’, ‘Less than half of the days’, ‘None or nearly none of the days’, with an additional response for
participants having discontinued or not started trial treatment or lost to follow-up, at 3, 6, 9 and
12 months.
• Tolerability of treatment, as measured by the Antidepressant Side-Effect Checklist (ASEC),31 and
mean ASEC total score [providing an approximate index of all reported adverse events (AEs) between
arms], at 3, 6 and 12 months.

Cost-effectiveness objectives
Cost-effectiveness objectives were to assess the effect of amitriptyline on:

• self-reported healthcare use, as measured by a health resource use questionnaire, at 3, 6 and

12 months
• health-related quality of life, as measured by the EQ-5D-3L,32,33 at 3, 6 and 12 months
• cost-effectiveness, as measured via the incremental cost-effectiveness ratio, and expressed in terms
of incremental cost per quality-adjusted life-year (QALY) at 6 and 12 months.

Data were collected but analysis is on hold, awaiting further funding.

Nested qualitative study objectives

These are provided in Aims.

Participants

Adult patients with IBS in primary care, who were still symptomatic despite first-line therapies, as
defined by NICE, were potentially eligible to take part in the trial. Eligible patients met all the listed
inclusion criteria, and none of the exclusion criteria. Eligibility waivers to inclusion and exclusion criteria
were not permitted.

Inclusion criteria

• A diagnosis of IBS [of any subtype [IBS with constipation (IBS-C), diarrhoea (IBS-D), mixed bowel
habits (IBS-M), or unclassified (IBS-U)]} in the patient’s primary care record, and fulfilling the Rome IV
criteria34 (see Report Supplementary Material 1).
• Age ≥ 18 years.
• Ongoing symptoms, defined as an IBS-SSS score of ≥ 75 at screening,23 despite having tried dietary
changes and first-line therapies as defined by NICE [fibre supplements (e.g. ispaghula husk), laxatives
(e.g. bisacodyl), antispasmodics (e.g. mebeverine) or anti-diarrhoeals (e.g. loperamide)],16 which was
assessed at screening via patient self-report.

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

• A normal haemoglobin, total white cell count (WCC), and platelets within the last 6 months prior
to screening.
• A normal C-reactive protein (CRP) within the last 6 months prior to screening.
• Exclusion of coeliac disease, via anti-tissue transglutaminase (tTG) antibodies, as per NICE guidance.16
• No evidence of active suicidal ideation, as determined by three clinical screening questions below,
and no recent history of self-harm (an episode of self-harm within the last 12 months prior to
screening). Any positive response on any of the three questions triggered urgent review by the
patient’s GP. These clinical questions were used in preference to a formal suicidal risk rating scale, as
such scales perform poorly in clinical practice:
◦ whether the patient had experienced any thoughts of harming themselves, or ending their life, in

the last 7–10 days

◦ whether the patient currently had any thoughts of harming themselves or ending their life

◦ whether the patient had any active plans or ideas about harming themselves, or taking their life, in

the near future.

• If female:
◦ post menopausal (no menses for 12 months without an alternative medical cause), or

◦ surgically sterile (hysterectomy, bilateral salpingectomy or bilateral oophorectomy), or

◦ using highly effective contraception (and had to agree to continue for 7 days after the last dose of

the investigational medicinal product).

• Able to complete questionnaires and trial assessments.

• Able to provide written informed consent.

Exclusion criteria

• Age > 60 years with no GP review in the 12 months prior to screening (to assess for organic
gastrointestinal disease as a cause of gastrointestinal symptoms, as this becomes more likely with
increasing age).
• Meeting locally adapted NICE 2-week referral criteria for suspected lower gastrointestinal cancer.35
• A known documented diagnosis of inflammatory bowel disease or coeliac disease.
• A previous diagnosis of colorectal cancer.
• Currently participating in, or within the 3 months prior to screening having been involved in, another
clinical trial of an investigational medicinal product.
• Pregnant or breastfeeding.
• Planning to become pregnant within the next 18 months.
• Currently using a TCA or using a TCA for another indication within the last 2 weeks prior
to randomisation.
• Allergy to TCAs.
• Other known contraindications to the use of TCAs, including patients with any of the following:
◦ taking monoamine oxidase inhibitors, or receiving them within the last 2 weeks

◦ already prescribed a TCA for the treatment of depression

◦ previous myocardial infarction

◦ recorded arrhythmias, particularly heart block of any degree, prolonged Q-T interval on ECG

◦ mania

◦ severe liver disease

◦ porphyria

◦ congestive heart failure

◦ coronary artery insufficiency

◦ receiving concomitant drugs that prolong the QT interval (e.g. amiodarone, terfenadine or sotalol)

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 5
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Methods

Study settings

Participants were recruited from 55 general practices [including two participant identification centres
(PICs)] within urban and rural settings, with a range of sociodemographic and diversity characteristics.
A further three general practices (two in West Yorkshire and one in Wessex) were opened to recruitment
but did not mail-out before the recruitment period had ended. Each practice was classed as a research
site with a GP as the principal investigator (PI). Practices were required to have obtained management
approval and to have undertaken a site initiation meeting prior to the start of recruitment into the trial.

The 55 involved general practices were in three geographical regions, 22 in West of England (including
one PIC), 13 in West Yorkshire (including one PIC) and 20 in Wessex. These three geographical regions
were referred to as ‘hubs’. Each hub research team included a hub lead clinician and research nurse(s) or
clinical study officer and was responsible for co-ordinating patient activity.

Screening, recruitment and registration

General practices willing to participate in the study were recruited with the assistance of the regional
clinical research networks (CRNs). They searched their patient registers for potentially eligible patients
aged ≥ 18 years with a diagnosis of IBS, using a SnoMed clinical terms search, which was developed
previously in the NIHR-funded ACTIB trial,36,37 and updated with the support of the Wessex CRN. A GP
at the practice checked the list of patients to be contacted prior to the invitation letters being sent out, to
ensure that it was appropriate to contact them. Potentially eligible individuals were then contacted by letter,
sent by the general practice via DocMail, informing them about the trial and inviting them to take part.

The postal invitation included a participant information sheet and informed consent form (see Report
Supplementary Material 1). Potential participants interested in taking part returned a reply slip in a
pre-paid envelope or contacted the study team directly via e-mail or telephone. The reply slip included
a section for the potential participant to agree to be contacted about the study and, following this,
for information to be requested from their GP to confirm their suitability to take part in the trial. This
agreement was obtained by e-mail or telephone if the initial contact was not via returning a reply slip.
The reply slip also included a ‘reason to decline’ section so that we could gather information on why
people chose not to participate in the trial. Recruiting general practices were asked to provide an
anonymised list of the age and sex of those invited so that the characteristics of those invited could be
compared with those entering the trial.

General practitioners could also provide information about the trial to potential eligible patients
opportunistically during their surgeries. Posters and leaflets were displayed in waiting rooms and the
trial was advertised on general practice websites, where possible. Thus, if a patient with IBS attended a
consultation, they were able to ask the GP about the study and to be given contact details for the study
team, if appropriate.

Patients with IBS could also be identified by general practices working as PICs. PICs were responsible
for the identification of potential patients for the trial and mailing out the invitation letter, participant
information sheet, and informed consent form. Patients were then directed to respond to the invitation
to the main hub research team. Patients identified from PICs were seen at the main general practice.

In an attempt to recruit an ethnically diverse sample, we reached out to minority ethnic organisations for
advice as to how to make the trial more attractive to people from minority ethnic backgrounds and to
publicise and raise the profile of the trial among these particular groups of potential participants.

The hub research nurse or clinical study officer contacted potential participants who replied to the
study invitation to arrange a screening call. At this call, they provided further information about the

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

trial and obtained verbal consent to telephone-screen the potential participants, using a screening form
consisting of the Rome IV criteria,34 the IBS-SSS,23 and questions about the inclusion and exclusion
criteria. All potential participants were assigned a unique screening identification number.

To allow generalisation of the trial results, and in accordance with Consolidated Standards of Reporting
Trials (CONSORT) guidelines, each recruitment hub, on behalf of each general practice, maintained
and provided to the Leeds Clinical Trials Research Unit (CTRU) an anonymised screening log of the age
and sex of all patients who were screened for entry into the study, including all those who confirmed
interest. Documented reasons for ineligibility or declining participation were recorded and were closely
monitored by the CTRU as part of a regular review of recruitment progress.

Patients who were potentially eligible, after telephone screening, were asked to attend a face-to-face
appointment at their general practice to complete full eligibility screening, provide written informed
consent, and obtain blood tests, if required. Patients were allowed sufficient time, and at least 24 hours,
unless they wished to participate sooner, to consider participation and were given the opportunity to
discuss the study with their family and healthcare professionals before they were asked whether they
would be willing to take part. The research nurses and clinical study officer were trained in both the
informed consent process and the ATLANTIS study and provided the patient with full and adequate oral
and written information about the study, including the background, purpose, and risks and benefits of
participation, as well as ensuring that the opportunity to ask questions concerning study participation
was given. A GP was available to answer any questions or concerns, if required. The research nurses
or clinical study officer also confirmed that the participant was free to withdraw from the study at any
time without it affecting their future care. The original copy of the signed, dated, informed consent form
was stored in the investigator site file. One copy was also filed in the medical records, one given to the
participant, and one returned to the CTRU.

Informed consent from participants also included a request to take part in qualitative interviews at 6
and 12 months (for participants who had consented to 12-month follow-up), or 6 months only (for later
participants who had only consented to 6-month follow-up), with information concerning the likely
duration of these interviews provided. Finally, permission was also sought to collect longer-term routine
data from electronic health records concerning amitriptyline and other IBS medication prescriptions,
GP consultations for IBS, and secondary care referrals, outside the time frame of the trial itself, should
further funding become available.

Following confirmation of written informed consent, participants were registered into the trial as soon
as possible by an authorised member of the hub research staff. Informed consent for entry into the
trial had to be obtained prior to registration. Registration was performed centrally using the CTRU
automated web-based registration and randomisation system. All participants were allocated a unique
trial identification number after they had been registered.

Blood test results were made available to the hub lead clinician (or delegate), the research nurse
or clinical study officer, and the participant’s GP. In accordance with the trial inclusion criteria, if
the blood tests showed an abnormal result (i.e. anaemia, raised or lowered total WCC, raised or
lowered platelet count, a CRP over the normal laboratory range, or a positive anti-tTG antibody),
the individual was not randomised into the trial, but was referred back to their GP for further
assessment. However, if there were marginal, and potentially clinically insignificant, abnormalities
of haemoglobin, WCC, or platelets, these were reviewed by the responsible GP and the hub lead
clinician for consideration for inclusion into the study. In the case of an abnormal blood result for
haemoglobin, WCC, platelets, or CRP that may have been a temporary abnormality (e.g. secondary to
a recent infection), or a marginal or minor abnormality, the blood test could be repeated 2 to 4 weeks
later, if the participant wished to undertake further screening for the study. If on repeat testing total
haemoglobin, WCC, platelet count, and CRP were acceptable clinically, the participant could continue
with screening.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 7
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Methods

If blood results were within acceptable limits, the GP from the research site was asked to confirm
eligibility and sign the study-specific trial medication prescription form. Participants were then
provided with web-based or postal questionnaires, depending on preference, to complete at baseline.
These had to be completed no more than 7 days prior to randomisation if online, and within 14 days
prior if postal. Participants were not randomised until the baseline questionnaires were completed. If
randomisation had not taken place within 4 weeks of the baseline questionnaires being completed,
these were repeated.

Prior to randomisation, women of childbearing potential were asked to confirm verbally that they were
not pregnant and were on reliable contraception due to the potential risks of amitriptyline in pregnancy.
If they were unable to do so, they were asked to perform a urine pregnancy test within the 7 days
prior to randomisation. They were provided with a pregnancy test to use at home, to facilitate a result
as close as possible to randomisation, and hence treatment commencing. Because this formed part of
the eligibility criteria, the lead clinician at each hub reviewed and signed the final eligibility for these
participants. If the test was positive or unclear the individual was not randomised into the trial but was
directed to their GP.

If the participant was ineligible, or no longer wished to take part in the trial, the local research team
withdrew the participant. They did not undergo any other study assessments and they were referred to
their GP. Reasons for non-randomisation were documented, where available.

Randomisation

Participants were randomised 1 : 1 to receive amitriptyline or placebo. Allocation, via a web

randomisation system at the University of Leeds CTRU, was performed using minimisation with a
random element, to ensure balance with respect to the presence of raised HADS-depression scores
(score ≥ 8), IBS subtype (IBS-C, IBS-D, IBS-M or IBS-U) and recruitment hub between treatment arms.

Blinding

As the trial was double-blind, neither the participant nor those responsible for their care and evaluation
(treating team and research team) knew the treatment allocation or coding of the treatment allocation.
Central pharmacy was also blinded to treatment allocation. This was achieved by identical tablets,
packaging, and labelling of both the amitriptyline and placebo. The process for dose titration was the
same for amitriptyline and placebo. Each bottle of amitriptyline or placebo was identified by a unique
kit code, generated randomly. Lists of the kit codes and their corresponding treatments were generated
by the CTRU and sent to Modepharma Limited, who supplied the kits and the code break envelopes.
Management of kit codes on the kit logistics application, which was linked to the 24-hour randomisation
system, were conducted by the unblinded CTRU safety statistician, in addition to maintaining the
back-up kit code list.

Access to the code break envelopes at CTRU was restricted to the safety statistician and designated
safety team. Interim emergency unblinding before 6- or 12-month assessments was strongly
discouraged and the blind was only to be broken if information about the participant’s trial treatment
was clearly necessary and would alter the appropriate medical management of the participant. Interim
unblinding could be requested on the grounds of safety by the chief investigator, local PI, an authorised
delegate, or a treating physician. It was anticipated that these requests would most likely originate at
the time of an AE or planned change in non-trial related drug therapy. In the event of a serious adverse
event (SAE), all participants were to be treated as though they were receiving the active medication (i.e.
amitriptyline). Any unblinded interim reports supplied to the Data Monitoring and Ethics Committee
(DMEC) were provided by the CTRU safety statistician and were password-protected securely.

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

At study entry, all participants were asked to provide their consent to be contacted by the CTRU to
send them their treatment allocation after they had reached the 12-month assessment point in the trial,
for participants who had consented to 12-month follow-up, or after they had reached the 6-month
assessment point in the trial, for later participants who had consented to only 6-month follow-up.
If they agreed, they and their GP were informed of their treatment allocation shortly after their final
follow-up at 6 or 12 months. This was done to facilitate post-trial treatment decisions. The information
was provided via e-mail, supported by an evidence-based unblinding leaflet (see Report Supplementary
Material 1) to deal with any potential questions that may arise, which was developed with input from
PPI representatives. Information concerning treatment allocation was provided only when CTRU had
confirmation that all study assessments and contacts were complete, and all required data had been
received. This information was only provided to the participant and their GP to protect and maintain
the overall blind for the research team. Where participants needed further support following provision
of treatment allocation and the evidence-based leaflet, they were directed to the ATLANTIS qualitative
researcher, who was independent of the day-to-day running of the trial, recruitment, data collection, and
treatment decisions.

Intervention

Participants were randomised 1 : 1 to receive titrated low-dose amitriptyline (Teva, the Netherlands) or
identical-appearing placebo for 6 months. All participants also received the British Dietetic Association
NICE-approved dietary advice sheet for IBS,38 and were provided with usual care for IBS from their GP
during the trial, with the exception that amitriptyline or other TCAs could not be commenced during
the trial. In addition, drugs contraindicated with TCAs, such as monoamine oxidase inhibitors or drugs
prolonging the QT interval, were prohibited during the trial. Following randomisation, participants were
offered an optional GP appointment at 1 month, in case of any questions, in addition to research nurse
and clinical study officer support.

All participants were provided with standardised written information about dose adjustment (see Report
Supplementary Material 1), developed with input from PPI representatives, to guide participant-led dose
titration. This advised participants to commence at a dose of amitriptyline or placebo of 10 mg (one
tablet) once-daily at night, with dose titration occurring over 3 weeks, up to a maximum of 30 mg (three
tablets) once-daily at night or down to a minimum of 10 mg on alternate days, depending on side effects
and response to treatment. Participants were supported throughout the titration phase, with telephone
calls from the research nurse or clinical study officer at weeks 1 and 3 to assess tolerability. After an
initial 3-week titration, it was expected the majority of participants would remain on a steady dose, but
they could modify their dose throughout the study in response to their symptoms and any side effects,
reflecting how amitriptyline would be used in usual care.

For safety purposes, due to the potential risks from amitriptyline in overdose, participants were provided
with an initial 1-month supply of trial medication following randomisation. Further trial medication
was dispensed at months 1 (2-month re-supply), 3 (3-month re-supply), 6 (3-month re-supply) and 9
(3-month re-supply), as appropriate, with the hub research nurse or clinical study officer contacting
participants by telephone prior to re-supply at weeks 1 and 3 and months 3, 6, 9 and 12 to assess both
for new evidence of suicidal ideation, adherence to trial medication, tolerability of trial medication
via the ASEC, and reasons for discontinuation of trial medication (if appropriate), as well as recording
concomitant medications and providing support as needed.

Participants were planned to be followed up for up to 12 months. Our PPI input prior to the grant
submission revealed that 6 months of blinded treatment was felt to be the maximum reasonable initial
commitment. Therefore, participants randomised in the first stage of recruitment (before 7 October
2021) received 6 months of trial medication initially and then were able to either continue blinded trial
medication for a further 6 months or stop trial medication; participants randomised in the later stages

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 9
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Methods

of recruitment (on or after 7 October 2021) received 6 months of trial medication only. Following trial
and outcome measure completion participants had the option to be unblinded and discuss amitriptyline
prescription for IBS under the care of their GP if they wished.

Trial medication was supplied by Modepharma Limited and dispensed by post to the participant’s
home by a central pharmacy at Leeds Teaching Hospitals NHS Trust. A copy of the study-specific trial
medication prescription form was sent to the central pharmacy to facilitate prompt dispensing when the
participant was randomised, although a wet ink copy was required before the study trial medication was
dispensed and shipped to the participant. The participant confirmed medication receipt with the hub
research team.

Assessments and data collection

Trial assessments are summarised in Table 1 with further details of assessment instruments provided
below. Participants were contacted by the researcher via telephone at 1 and 3 weeks and 3, 6, 9 and
12 months (as appropriate) to support titration and ongoing treatment, including assessments of suicidal
ideation, toxicity, adherence to, and acceptability of trial medication. Participants completed electronic
or postal questionnaires at baseline, and 3 and 6 months, and answered a weekly question ‘Have you
had adequate relief of your IBS symptoms?’ for the initial 6-month study duration. Participants recruited
to 12-month follow-up, before 7 October 2021, also completed electronic or postal questionnaires
at 12 months. Text message and e-mail reminders were sent at 1 week to prompt completion of
questionnaires. Non-responders were telephoned with a final reminder.

Suicidal ideation was assessed by the researcher during all planned telephone calls via three brief
questions. If yes to any, the participant was not issued with any further trial medication and their GP was
contacted immediately.

• Has the participant experienced any thoughts of harming themselves, or ending their life, in the last
7–10 days?
• Does the participant currently have thoughts of harming themselves or ending their life?
• Does the participant have any active plans or ideas about harming themselves, or taking their life, in
the near future?

Adherence to treatment was measured by the researcher during the planned telephone calls.
Participants were asked ‘Since you were last asked, which of the options best describes how often
you have taken at least one tablet of the trial medication daily?’ with response options: ‘Every day or
nearly every day’, ‘More than half of the days’, ‘Less than half of the days’ or ‘None or nearly none of
the days’.

Acceptability of treatment was measured by participant self-report during the researcher telephone
calls, as well as the decision to continue trial medication beyond 6 months. Participants were asked ‘On
balance do you find this medication acceptable to take and would you want to keep taking it?’.

Adverse events were collected via a validated self-completed questionnaire, the ASEC,31 which
consists of 21 potential AEs rated on a scale of 0 (absent) to 3 (severe), and also asks the individual
whether they deem the AE to be treatment-related. This has been shown to demonstrate
good agreement with a psychiatrist’s rating of the occurrence of treatment-related AEs with
antidepressants. The ASEC was completed as part of toxicity and tolerability assessments conducted
by the researcher at weeks 1 and 3, and month 9 telephone calls, and via participant-completed
questionnaires at months 3, 6 and 12.

10

NIHR Journals Library www.journalslibrary.nihr.ac.uk

 TABLE 1 Participant study schedule
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
Study period

Screening, recruitment,
registration Randomisation Follow-up

Week Week Week Month Month Month Month

Time point −4 weeks − 0 0 1 3 4 3 6 9a 12a

Enrolment

Verbal consent and eligibility screen X

Eligibility confirmation (including Rome IV criteria, blood and X

pregnancy tests)

Informed consent X

Sociodemographic details: medical history; duration of IBS X

symptoms; previous depression or anxiety; Bristol Stool Form Scale

Allocation X

Interventions

Amitriptyline

Health Technology Assessment 2024 Vol. 28 No. 66

Placebo

Study medication provision X X X X X

Optional GP review X

Assessments (research nurse/clinical study officer collected, while on treatment)

Suicidal ideation X X X X X

Current dose X X X X X X

Concomitant Medications Review X X X X X X

Treatment adherence X X X X X

Treatment acceptability X X X X

ASEC X X X

Exit survey X

continued
11
 12
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Methods
TABLE 1 Participant study schedule (continued)

Study period

Screening, recruitment,
registration Randomisation Follow-up

Week Week Week Month Month Month Month

Time point −4 weeks − 0 0 1 3 4 3 6 9a 12a

Assessments (self-completed questionnaire)

ASEC X X X

IBS-SSS X X X X

HADS X X X X

EQ-5D-3L X X X X

WSAS X X X X

PHQ-12 X X

SGA of relief of IBS symptoms X X X

Health resource use X X X X

Relief of IBS symptoms Weekly diary with SMS text

reminder

Optional participant interview (nested qualitative study) X X

a Months 9 and 12 time points applicable to participants recruited to 12-month follow-up, before 7 October 2021.
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

An exit survey was completed with the participant by the researcher during the 6-month telephone call
to record any changes participants had made to diet, exercise levels, or IBS treatments, their experience
of the ATLANTIS trial medication, and which treatment they thought they were allocated to and why.

The IBS-SSS is used widely in trials of medical therapies in IBS. It is a five-item self-administered
questionnaire, as described above.23

The HADS is a well-validated, commonly used, self-report instrument for detecting symptoms of anxiety
and depression in people with medical illnesses.27 It consists of a total of 7 items measuring anxiety, and
7 measuring depression, scored from 0 to 3, with a total score of 21 for each. Higher scores indicate
more severe symptoms of anxiety or depression.

The EQ-5D-3L is the most frequently used measure for generating QALYs.32,33 It has been demonstrated
to be appropriate in patients with IBS.

The WSAS measures the effect of chronic diseases on peoples’ ability to work and manage at home and
participate in social or private leisure activities and relationships.28–30 The WSAS has been shown to be
sensitive to change in IBS trials. It has five aspects scored from 0 (not affected) to 8 (severely affected),
with a total possible score of 40.

The PHQ-12 comprises 12 somatic symptoms from the full Patient Health Questionnaire-15. Each
symptom is scored from 0 (‘not bothered at all’) to 2 (‘bothered a lot’). Higher scores indicate the
presence of somatoform-type behaviour, which is a measure of psychological health.

The SGA of relief of IBS symptoms is frequently used in treatment trials in IBS to identify responders to
therapy as described above.24

Health resource use, including healthcare use, use of other medications for IBS, and need for referral to
secondary care, was self-reported by the participant via a resource use questionnaire, using a 3-month
recall period. If the participant consented to 12-month follow-up, then the recall period was extended to
6 months. This collected data concerning all resource use and medications in the community, in primary
and secondary care, social care, hospitalisation, outpatient specialist visits, and diagnostic investigations.
Because of the societal perspective, the questionnaire also included questions on out-of-pocket
expenses, employment status, and days lost due to illness.

Relief of IBS symptoms was measured by the yes/no response to ‘Have you had adequate relief of your
IBS symptoms?’ asked electronically, or via a paper-based diary. Participants were sent a weekly text
reminder from CTRU to complete the assessment.

Summary of changes to the protocol

The COVID-19 pandemic resulted in the trial pausing recruitment in March 2020 for 4 months, with
a series of national lockdowns, and led to subsequent reduced rates of new practice and participant
recruitment. Internal pilot objectives were, therefore, difficult to evaluate in the original time frame
and a costed trial extension was required to complete recruitment and follow-up of the trial. Several
substantial amendments were made to the trial protocol, including an approved amendment due to
the impact of COVID-19, to reduce the duration of trial medication and follow-up from 12 months to
6 months and to remove the cost-effectiveness analysis, which is now dependent on further additional
funding. This was done to minimise additional funding required to complete the trial and to prioritise
funds for participant recruitment. Site and hub PIs, hub researchers, and participants were informed of
all protocol amendments following ethical and regulatory approvals. A summary of all protocol changes
can be found in Report Supplementary Material 1.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 13
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Methods

Sample size

We estimated that an evaluable sample size of 414 participants would provide 90% power to detect the
minimum clinically important difference of 35 points between amitriptyline and placebo at 6 months on
the IBS-SSS,23 assuming a maximum standard deviation (SD) of 110 points on the IBS-SSS,39,40 and 5%
two-sided significance level, equating to a small to moderate effect size of 0.32. The 35-point between-
group difference on the IBS-SSS was agreed as a minimum clinically important difference in the ACTIB
trial, which was another treatment trial in IBS in UK primary care.36,37 The evaluable sample size gave
at least 85% power to detect a 15% absolute difference in SGA of relief of IBS symptoms,24 our key
secondary outcome. We planned to recruit 518 participants to allow for 20% loss to follow-up.22

Statistical methods

A detailed statistical analysis plan was written and signed off by the Trial Management Group (TMG) and
Trial Steering Committee (TSC) before analysis was undertaken.

Analyses of data up to 6 months post randomisation were conducted on the intention-to-treat (ITT)
population, defined as all participants randomised, regardless of adherence to the intervention, unless
otherwise indicated. Analyses of treatment-related data beyond 6 months, and secondary outcomes
at 12 months, were conducted on the 12-month ITT population, defined as all participants who were
randomised and who consented to 12-month follow-up, regardless of adherence to the intervention.
Analysis of data beyond 6 months is presented separately in the report as results are applicable only to
a subset of participants and are no longer a fully randomised comparison as participants could choose to
continue treatment or not at 6 months.

An overall two-sided 5% significance level was used for all outcome comparisons. Outcome data
were analysed once only after data lock, at final analysis, and no interim analyses were planned.
Analyses were completed in SAS® (SAS Institute Inc., Cary, NC, USA) version 9.4. Statistical
monitoring of safety data was conducted throughout the trial and reported at agreed 6-monthly
intervals to the DMEC.

Descriptive analysis
Summary statistics, by treatment group (where applicable) and overall, are used to provide a descriptive
analysis of the study conduct, including screening, accrual, protocol violations, withdrawals, treatment
receipt, participant follow-up, and analysis populations informing the study CONSORT diagram.

A flow diagram further summarises the course of participants through the screening and recruitment
process, including total number of patients approached by GP mail-out, as well as the number who
expressed interest, and were screened, eligible, consented, registered, and randomised, along with
reasons for drop-out at each stage. Age and sex of all patients were also summarised at each stage of
the screening and recruitment process and compared with those randomised.

Baseline characteristics and questionnaire scores of recruited participants were summarised overall,
by treatment group, and by availability of 6-month primary outcome data (to inform our missing
data approach).

Treatment delivery and receipt were summarised overall and by treatment group, including details
of treatment received and discontinuation; dosage, titration, and modifications; adherence; kit
replenishment and replacement; ongoing monitoring of suicidal ideation and concomitant medications;
uptake of optional GP review; and participant-reported changes in diet, exercise, other treatments, IBS
symptoms, and any potential contributing factors.

14

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

The success and process of blinding are summarised overall and by arm, including details of an exit
survey of participants, that is which treatment the participant thought they received and why, and end-
of-trial participation unblinding.

Primary outcome analysis

A linear regression model, adjusted for true values of minimisation variables (recruitment hub, stool
type, baseline HADS-depression score) and IBS-SSS score at baseline, was used to test for differences
between the treatment groups on the IBS-SSS at 6 months. Missing data were imputed by treatment
arm via multiple imputation by chained equations with 25 imputations, including recruitment hub,
IBS subtype, sex, age, baseline questionnaire scores (IBS-SSS, PHQ-12, HADS and WSAS), 3-month
IBS-SSS score and 6-month treatment status in the model. The 3-month IBS-SSS score was imputed
within the same model in a preliminary step, incorporating 3-month (rather than 6-month) treatment
status. Sensitivity analyses on a per-protocol population (using multiple imputation), and on participants
with complete data (ITT to data availability), were performed to test robustness of results. Results were
expressed as point estimates of the mean difference, together with 95% confidence intervals (CIs) and
p-values.

Secondary outcome analysis

Continuous outcomes (where applicable at 3, 6 and 12 months), including IBS-SSS, HADS, WSAS and
PHQ-12 scores, were analysed in the same manner as the primary outcome adjusted for the respective
baseline score. Analysis of the PHQ-12 was adjusted additionally for sex due to differences in the
maximum available total score for male and females. Secondary binary outcomes, including SGA of
relief of IBS symptoms at 3, 6 and 12 months, and acceptability of treatment at 6 months were analysed
similarly in logistic regression models with results expressed as odds ratios (ORs) with 95% CIs. Missing
IBS-SSS, HADS, WSAS, PHQ-12, SGA and acceptability data were imputed in the same manner as the
primary outcome, as appropriate to outcome type and incorporating both 3- and 6-month outcomes
where applicable.

We planned to analyse adherence (at week 3, months 3, 6, 9 and 12) using ordinal regression. However,
due to violation of the proportional odds assumption, only descriptive analyses are presented.

Adequate relief of IBS symptoms, measured weekly to 6 months, was analysed using a repeated-
measures model based on available data (all participants in the ITT population with at least one weekly
observation included). A likelihood-based generalised linear mixed model with population-averaged
(marginal) inference and unstructured covariance matrix was used to compare response between
treatment groups at each week and overall, across weeks. The treatment group, true values of
minimisation variables, time (in weeks) and the treatment time interactions were fitted as fixed effects.
Results were expressed as ORs, together with 95% CIs and p-values. Descriptive analysis of aggregated
weekly data included responder status based on the number and proportion of participants reporting
adequate relief in ≥ 50% of weeks (i.e. ≥ 13 of 25 weeks).

Tolerability of trial treatment, based on the participant’s self-reported symptoms on the ASEC at 3, 6 and
12 months, was analysed according to the safety population (see Safety analysis) and using available data
for participants on trial treatment at each time point. A linear regression model, adjusted for true values
of minimisation variables, was used to test for differences between the treatment groups on the total
ASEC score at each time point.

Sensitivity analyses of secondary outcomes were performed to test robustness of results.

These included:

• analysis of complete data (ITT to data availability) where the primary analysis was based on multiple
imputation (for IBS-SSS, HADS, WSAS, PHQ-12, SGA of relief of IBS symptoms, and acceptability)

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 15
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Methods

• analysis of SGA of relief of IBS symptoms:

◦ using an alternative definition of response, with response defined as reporting only considerable

or complete relief of IBS symptoms

◦ as an ordinal outcome using ordinal logistic regression

• analysis of acceptability, with additional multiple imputation for participants who did not start trial
treatment or were lost to follow-up on or before the 6-month call (derived as not acceptable in
primary analysis)

Safety analysis
All participants receiving at least one dose of trial medication were included in the safety population
and analysis; participants who received at least one dose of amitriptyline were included in the
amitriptyline group, regardless of the arm they were allocated to. The number of participants reporting
a SAE, including serious adverse reactions (SARs) and suspected unexpected serious adverse reactions
(SUSARs), and details of all SAEs were reported for each treatment group. The number and details of
emergency unblinding events, pregnancies and deaths were also reported for each treatment group.

Exploratory analysis
Exploratory moderator analyses were conducted to investigate if the 6-month treatment effect on the IBS-
SSS varied by baseline IBS-SSS score, recruitment hub, IBS subtype or mood (baseline HADS-depression
or HADS-anxiety scores), by including an interaction between the treatment arm and each potential
moderator in the primary analysis model. Moderator analyses were also conducted to investigate if the
6-month treatment effect on the key secondary outcome, SGA response, varied by IBS subtype.

Further exploratory analyses of the IBS-SSS at 3 and 6 months were conducted using logistic regression,
adjusted for true values of minimisation variables, to test for differences between the treatment groups
on response rates according to:

• a ≥ 50-point decrease in the total IBS-SSS score

• a ≥ 30% decrease in abdominal pain severity on the IBS-SSS item response score at 3 and 6 months
• a ≥ 30% decrease in abdominal distension severity on the IBS-SSS item response score at 3 and
6 months.

Missing response data, according to the definitions above, were imputed in the same manner as the
primary and secondary outcomes.

Health economics methods

We planned to perform a within-study cost-effectiveness analysis, which adopted the perspective of

the NHS and Personal Social Services and a societal perspective. We planned to use a time horizon of
6 months; hence costs and outcomes were not to be discounted. Our primary outcome was intended to be
cost per QALY, with uncertainty assessed using a within-trial probabilistic sensitivity analysis. This would
be performed using Monte Carlo simulation, with the results presented as incremental cost-effectiveness
ratios and cost-effectiveness acceptability curves. We planned to assume a willingness to pay (lambda)
of £20,000 per QALY. Sensitivity analyses were planned, which would include a 12-month time horizon,
as well as a scenario similar to real NHS practice, with treatment prescribed by a GP with repeated
prescriptions, tests, and necessary appointments. As stated previously, health economic analyses were
removed after a discussion and meeting with the HTA to minimise additional funding required to complete
the trial and to prioritise funds for participant recruitment and are now on hold subject to further funding.

Qualitative study

The nested qualitative study is reported in Nested qualitative study.

16

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Chapter 3 Clinical trial results

Study summary

The numbers of patients identified via GP mail-out, responding and screened for entry into ATLANTIS,
eligible, randomised, followed up at 3, 6 and 12 months, withdrawn and analysed are presented in the
CONSORT diagram in Figure 1.

Mailed out: Researcher call did not take place: 148 (11.8%)
Not interested: 1975 (61.2%) (n = 15,672) patients, • Could not contact patient, n = 51
Reason (not mutually exclusive) 55 general practices • Patient no longer interested, n = 45
• IBS symptoms improved/no help required, • Moved GP practice/out of country, n = 15
n = 968 • Do not have IBS, n = 10
• Do not like the sound of the treatment, n = 313 • Ineligible, n = 6
• Does not have IBS, n = 301 Responded to mail out/ • Unable to take part due to other health issue, n = 3
• Does not wish to take more meds/on other expressed interest: • Trial closed to new recruitment, n = 2
meds, n = 137 (n = 3228) • Currently on a TCA, n = 1
• Do not have time, n = 135 • Unable to take part – phone/technology/transport, n = 1
• Does not wish to take any medications/prefer • Pregnant/plan to get pregnant, n = 1
natural ways, n = 121 • No reason given, n = 13
• Unable to/do not wish to take part – other health
issues, n = 116 Interested Not interested: 15 (1.2%)
• Do not wish to complete questionnaires, n = 109 (n = 1253) (38.8%)
• Taken amitriptyline before/side effects/no Unknown: 10 (0.8%)
benefit, n = 83
Ineligible: 501 (40.0%) (not mutually exclusive)
• Concerned about side effects, n = 73
• Participating in another trial, n = 27 • Inc 1: No IBS diagnosis/not fulfilling Rome IV, n = 269
Eligible
• Life stresses, n = 22 • Inc 3: IBS-SSS score not ≥ 75 despite having tried dietary
579 (46.2%)
• Current on a TCA, n = 17 changes/first-line therapies, n = 87
• Pregnant/plan to get pregnant, n = 12 • Inc 4: Abnormal haemoglobin, WCC or platelets, n = 1
• Moved GP practice/out of country, n = 12 • Inc 5: Abnormal CRP, n = 1
• Do not wish to take part in clinical trials, n = 10 Consented • Inc 6: Coeliac disease, n = 1
• Unable to take part – 528 (91.2%) • Inc 7a: Thoughts of harming/ending their life, n = 3
phone/technology/transport, n = 8 • Inc 7d: Episode of self-harm in the last 12 months, n = 4
• COVID-19, n = 1 • Inc 8: Not post-menopausal, surgically sterile or using
• Unknown/not interested to take part in the trial, contraception, n = 42
n = 184 • Inc 10: No informed consent, n = 1
Registered
525 (99.4%) • Exc 1: Aged 60 or above with no GP review, n = 9
[91.2% of eligible] • Exc 2: Suspected lower gastrointestinal cancer, n = 1
• Exc 3: Diagnosis of IBD or coeliac disease, n = 6
Not randomised: 62 (11.8%)
• Exc 4: Previous diagnosis of colorectal cancer, n = 1
• No longer eligible, n = 47
• Exc 5: Participated or in another CTIMP, n = 2
° Inc 4: Abnormal haemoglobin, WCC, platelets, n = 18 • Exc 6: Pregnant or breastfeeding, n = 6
° Inc 5: Abnormal CRP, n = 27 • Exc 7: Planning to become pregnant, n = 8
° Other: Inc 6, Inc 8, Exc 7, Exc 8, Exc 10, n = 10 Randomised • Exc 8: Using TCAs for another indication, n = 12
• Patient withdrew, personal choice, n = 6 463 (88.2%) • Exc 9: Allergy to TCAs, n = 2
• Moved GP practice/out of country, n = 3 [80.0% of eligible] • Exc 10: Other known contraindications to the use of
• Unable to contact, n = 2 • 6-month follow-up only: 172 TCAs, n = 45
• Unable to/do not wish to take part – other health • 12-month follow-up: 291 • No reason given, n = 11
issues, n = 1
• Baseline questionnaire not completed, n = 1
• Unknown, n = 2

Allocation
Amitriptyline (n = 232) Placebo (n = 231)
Completed 6 months treatment: 173 (74.6%) Completed 6 months treatment: 165 (71.4%)
Completed 12 months treatment: 67 (28.9%) Completed 12 months treatment: 61 (26.4%)

Follow-up

Completed questionnaires Completed questionnaires

3 m: 220 (94.8%) 3 m: 213 (92.2%)
6 m: 204 (87.9%) 6 m: 197 (85.3%)
12 m: 118 (80.3%) 12 m: 107 (74.3%)

Withdrawals Withdrawals
Interviews: 6 (2.6%) Interviews: 13 (5.6%)
Questionnaires: 6 (2.6%) Questionnaires: 13 (5.6%)
Further data collection: 6 (2.6%) Further data collection: 6 (2.6%)

Analysis

Intention to treat (n = 232) Intention to treat (n = 231)

12-month intention to treat (n = 147, 63.5%) 12-month intention to treat (n = 144, 62.3%)

FIGURE 1 Consolidated standards of reporting trial flow diagram.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 17
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Clinical trial results

Screening and recruitment

Screening
A total of 15,672 potentially eligible patients were identified via SnoMed clinical terms searches by 55
general practices and contacted by letter to provide information and invite them to take part in the trial.
Screening subsequently took place for 3228 patients who either responded via reply slip to the general
practice mail-out (n = 3144, 97.4%) or were identified opportunistically following a GP visit, publicity, or
other means and contacted a research nurse directly.

Of the 3228 who responded, 1253 (38.8%) expressed an interest in joining the trial, of whom 1105
(88.2%) had a screening call with a research nurse. Of these, 579 (52.4%) were eligible, of whom 528
(91.2%) consented, 525 (90.7%) were registered and 463 (80.0%) were randomised. Figure 2 and
Appendix 1, Table 42 present screening and recruitment by hub.

The most common reasons why the other 1975 (61.2%) patients were not interested in taking part
were that their IBS symptoms had improved and no further help was required [n = 968 (49.0%) of those
not interested], they did not like the sound of the treatment [n = 313 (15.8%)], they did not have IBS
[n = 301 (15.2%)], they did not wish to take more medications [n = 137 (6.9%)], they did not have time
[n = 135 (6.8%)], they did not wish to take any medications [n = 121 (6.1%)], their other health issues
meant they felt unable to or did not wish to take part [n = 116 (5.9%)], they did not wish to complete
questionnaires [n = 109 (5.5%)], they had taken amitriptyline before and experienced side effects or no
benefit [n = 83 (4.2%)], or they were concerned about side effects [n = 73 (3.7%)].

The most common reasons for ineligibility of 501 (45.3% of those with a screening researcher call)
patients were not having a diagnosis of IBS in the primary care record and fulfilling Rome IV criteria
[n = 269 (53.7%) of those ineligible], not having ongoing symptoms, as defined by an IBS-SSS score ≥ 75
despite having tried dietary changes and first-line therapies [n = 87 (17.4%)], known contraindication to
the use of TCAs [n = 45 (9.0%)], and potential for pregnancy [not post menopausal, surgically sterile or
using effective contraception, n = 42 (8.4%)].

Reasons why 62 (11.8%) registered patients did not go on to be randomised included subsequently not
meeting eligibility criteria [n = 47 (75.8%)], in the majority of cases due to abnormal blood test results,
patient choice [n = 6 (9.7%)] or other reasons [n = 9 (14.5%)].

Screening and recruitment by hub

5592
1387
West of England 612
215
202
184
Mailed out
6218 Responded to mail out
1688 Interested
Southampton 525
256 Eligible
222
192 Registered
Randomised
3862
153
116
West Yorkshire 108
101
87

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500

FIGURE 2 Screening and recruitment by hub.

18

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Recruitment
The first mail-out took place on 18 October 2019 and the first participant was randomised on 5
December 2019 and the last on 11 April 2022, with 232 participants randomised to receive low-dose
amitriptyline and 231 participants randomised to receive placebo, across 55 general practices. Figure 3
shows overall, monthly and cumulative recruitment of participants into the trial. Appendix 1, Figure 8
depicts the timing between practice mail-out and subsequent randomisations.

Recruitment was originally expected to be completed within 18 months. Owing to a delay in trial
opening, a pause in recruitment between March and July 2020, in line with national guidance related to
the COVID-19 pandemic, and subsequent slower than anticipated recruitment rates, again due to the
COVID-19 pandemic, recruitment was extended and took place over 29 months. A number of changes
were made to the trial protocol and study processes to enable recruitment to continue, including a
reduction in follow-up from 12 to 6 months for the final cohort of recruited participants (see Report
Supplementary Material 1). The first 291 (62.9%) recruited participants were therefore consented
to 12-month follow-up, whereas the final 172 (37.1%) could only provide consent to 6 months of
follow-up.

Characteristics of the screened, eligible and randomised participants

The age and sex of the mailed out, screened, interested, eligible and randomised patient populations
were broadly similar (Table 2). Patients who responded to the mail-out or expressed interest were slightly
older (mean age 57.0 vs. 48.4 years) and more likely to be female (75.5% vs. 71.3%). However, the mean
age of those subsequently interested, eligible and randomised was similar to the overall mailed-out
population, and the proportion of males increased through later stages of the screening process.

Protocol violations, withdrawals and follow-up

Protocol violations
Protocol violations were identified for six (1.3%) participants, three in each arm, four of which were
classed as major protocol violations (see Appendix 1, Table 43). One participant was found to be ineligible
6 days after randomisation following receipt of an abnormal blood test result (positive anti-tTG antibody;
major violation). Four participants experienced unplanned treatment errors, including: two participants
(one in each arm) who reported having taken four tablets daily (40 mg), more than the maximum 30 mg
daily allowance, at 3-month follow-up (minor violations). One participant allocated to placebo was sent
the incorrect bottle of medication at randomisation due to a kit number identification error at pharmacy
and received a bottle of amitriptyline; this was identified 14 days post randomisation after which the
participant was asked to return the incorrect bottle and they subsequently discontinued trial medication
(major violation). A further participant allocated to placebo disclosed at 6-month follow-up that they had
taken 30 mg of their friend’s amitriptyline on a single day (major violation). A further protocol violation
was reported for one participant allocated to amitriptyline who was asked to stop and return trial
medication after reporting suicidal ideation at 3 months follow-up. However, they continued to take trial
medication for a further week (major violation).

Research withdrawals
Participants could withdraw from optional interviews, weekly or monthly (3-, 6- or 12-month)
questionnaires, or further data collection. A total of 23 (5.0%) participants withdrew from at least one
study process: 6 (2.6%) in the amitriptyline arm and 17 (7.4%) in the placebo arm (Table 3). Participants
most frequently withdrew from optional interviews and monthly questionnaires: 6 (2.6%) in the
amitriptyline arm and 13 (5.6%) in the placebo arm, all but one of whom also withdrew from the weekly
questionnaire, and 6 participants in each arm withdrew from further data collection. The mean time of
withdrawal was 3.5 months post randomisation. The main reasons for withdrawal included treatment
side effects or lack of benefit, with study withdrawal accompanying treatment discontinuation, personal
choice, difficult personal circumstances, or lack of time.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 19
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 20
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Clinical trial results

60 600

Jan 2021, 518

50 500
48 Apr 2022, 463

45
43

40 39 400

Cumulative recruitment
36 36
Monthly recruitment

34 Monthly registered
32
30
31 31 31 Monthly randomised
30 300 Cumulative target (original)
26 Cumulative randomised
24 24 24
23
20
20 19 19 200
18 18 18
17
16 16 16 16 16 16
15 15 15
14 14 14
12
1111 11
10 10
10 9 100
8
6 6
5 5
4 4
3
2
1 1
0 0
Jul-19

Aug-19

Sep-19

Oct-19

Nov-19

Dec-19

Jan-20

Feb-20

Mar-20

Apr-20

May-20

Jun-20

Jul-20

Aug-20

Sep-20

Oct-20

Nov-20

Dec-20

Jan-21

Feb-21

Mar-21

Apr-21

May-21

Jun-21

Jul-21

Aug-21

Sep-21

Oct-21

Nov-21

Dec-21

Jan-22

Feb-22

Mar-22

Apr-22
Reduced follow-up (6 months)

FIGURE 3 Recruitment graph.

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 2 Demographics of patients mailed-out to, responded, interested, eligible, registered and randomised

Mailed-out Responded Interested Eligible Registered Randomised

(n = 15,672) (n = 3228) (n = 1253) (n = 579) (n = 525) (n = 463)

Age

Mean (SD) 48.4 (16.8) 57.0 (16.9) 51.0 (17.0) 48.4 (16.5) 48.4 (16.3) 48.4 (16.1)

Median (range) 48.0 (18–100) 59.0 (19–98) 51.0 (19–92) 49.0 (19–92) 49.0 (18–92) 49.0 (18–87)

Missing 333 74 11 4 0 0

Sex

Male 4419 (28.8%) 797 (25.5%) 357 (29.3%) 162 (28.7%) 158 (30.1%) 148 (32.0%)

Female 10,918 (71.2%) 2332 (75.5%) 862 (70.7%) 402 (71.3%) 367 (69.9%) 315 (68.0%)

Missing 335 99 34 15 0 0

TABLE 3 Withdrawals

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

Number of participants with a withdrawal 6 (2.6%) 17 (7.4%) 23 (5.0%)

Withdrawn from

 ptional interviews (of participants consented to

O 6 (2.9%) 13 (6.3%) 19 (4.6%)
interview)

Consented to optional interview (of randomised) 204 (87.9%) 206 (89.2%) 410 (88.6%)

Completing questionnaires 6 (2.6%) 13 (5.6%) 19 (4.1%)

Withdrawn from monthly questionnaires only 0 (0.0%) 1 (0.4%) 1 (0.2%)

Withdrawn from weekly and monthly questionnaires 6 (2.6%) 12 (5.2 %) 18 (3.9%)

Further data collection 6 (2.6%) 6 (2.6%) 12 (2.6%)

Time from randomisation to withdrawal (months)

Mean (SD) 3.6 (4.6) 3.5 (3.8) 3.5 (3.9)

Median (range) 2.1 (0.6, 12.7) 1.6 (0.4, 12.3) 1.8 (0.4, 12.7)

IQR 1.0–3.0 0.8–5.8 0.8–5.8

n 6 17 23

Reason for withdrawal

Side effects 1 (16.7%) 3 (17.6%) 4 (17.4%)

Difficult personal circumstances 1 (16.7%) 3 (17.6%) 4 (17.4%)

Lack of benefit 0 (0.0%) 3 (17.6%) 3 (13.0%)

Do not have time 0 (0.0%) 3 (17.6%) 3 (13.0%)

Personal choice 0 (0.0%) 3 (17.6%) 3 (13.0%)

Due to COVID-19 2 (33.3%) 0 (0.0%) 2 (8.7%)

SAE/SUSAR 0 (0.0%) 2 (11.8%) 2 (8.7%)

Unknown 2 (33.3%) 0 (0.0%) 2 (8.7%)

Total 6 (100%) 17 (100%) 23 (100%)

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 21
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Clinical trial results

Questionnaire follow-up
Monthly follow-up questionnaires were completed and returned by 433 (93.5%) of 463 participants at
3 months, 401 (86.6%) of 463 at 6 months and 225 (77.3%) of 291 participants at 12 months (Table 4).
Response rates were slightly higher in the amitriptyline arm compared with the placebo arm, particularly
at 12 months (80.3% vs. 74.3%). Individual questionnaire completion rates (arranged in the order
they appear within each questionnaire pack in Table 4) showed a slight reduction in completion rates
for questionnaires appearing later within the questionnaire packs. All baseline questionnaires were
completed within 1 month prior to randomisation and follow-up questionnaires were largely completed
within a 7-day window either side of the point of follow-up (see Table 4 and Appendix 1, Figure 9). The
majority of the participants completed questionnaires online via REDCap, with online completion for
93.3% at baseline, and 95.2%, 96.0% and 96.9% of responders at 3, 6 and 12 months, respectively and
the remainder completing paper-based questionnaires by post. Participants provided a median of 23
responses to the weekly question ‘Have you had adequate relief of your IBS symptoms?’ up to 6 months
(25 weeks in total) (see Appendix 1, Figure 10). At least one response was provided by all except 10
participants in each arm, and responses were provided for at least 75% of weeks (≥ 19 weeks) for 337
(72.8%) participants, with similar rates between treatment arms.

Comparison of baseline characteristics between participants with and without monthly follow-up
questionnaires (Table 5) indicated that those not completing the primary 6-month follow-up
questionnaire were more likely to have discontinued trial medication before 6 months, to be younger, to
be in the West Yorkshire hub, and to have had more severe scores on the baseline IBS-SSS and WSAS.

Analysis populations

Intention to treat
All 463 randomised participants were included in the ITT population, including 232 allocated to
amitriptyline and 232 to placebo. Prior to the protocol amendment reducing follow-up to 6 months, the
first 291 randomised participants [147 (63.4%) in the amitriptyline arm; 144 (62.3%) in the placebo arm]
were consented to 12-month follow-up and are included in the 12-month ITT population (Figure 1).

Per protocol
The per-protocol population included 376 (81.2%) participants at 3 months and 323 (69.8%) participants
at 6 months, with slightly greater numbers of participants in the amitriptyline arm (Table 6). The majority
of participants excluded from the per-protocol population were excluded because they discontinued trial
medication before either 3 or 6 months. Of those excluded at 3 and 6 months, 69.0% and 72.9% had
discontinued trial medication before 3 and 6 months, respectively, 12.6% and 2.1% had not responded
to the treatment adherence question at the 3- and 6-month researcher follow-up calls, respectively,
and 8.0% and 12.1% were lost to follow-up by 3 and 6 months, respectively. A smaller proportion of
participants had not started treatment, reported inadequate levels of adherence to treatment, breached
eligibility criteria, or had a major protocol violation.

Safety population
The safety population mirrored the ITT and 12-month ITT populations with the exception of three
participants who were excluded as they did not start trial medication, and two participants for
whom treatment cross-over was observed. One participant allocated to placebo received a bottle of
amitriptyline by error within the first 2 weeks of randomisation and is included in the amitriptyline
arm in the 3-, 6- and 12-month safety populations. A further participant allocated to placebo (and
consenting to 6-month follow-up) reported having taken their friend’s amitriptyline on a single day at
their 6-month follow-up call and is included in the amitriptyline arm for the 6-month safety population
(see Table 6).

22

NIHR Journals Library www.journalslibrary.nihr.ac.uk

attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
TABLE 4 Monthly questionnaire completion

Month 3 Month 6 Month 12a

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 147) (n = 144) (n = 291)

Questionnaire completed?

Completedb 220 (94.8%) 213 (92.2%) 433 (93.5%) 204 (87.9%) 197 (85.3%) 401 (86.6%) 118 (80.3%) 107 (74.3%) 225 (77.3%)

 id not
D 12 (5.2%) 18 (7.8%) 30 (6.5%) 28 (12.1%) 34 (14.7%) 62 (13.4%) 29 (19.7%) 37 (25.7%) 66 (22.7%)
complete

Reason not completed

 ithdrawn
W 5 (41.7%) 7 (38.9%) 12 (40.0%) 5 (17.9%) 10 (29.4%) 15 (24.2%) 6 (20.7%) 10 (27.0%) 16 (24.2%)
questionnaires

No response 7 (58.3%) 11 (61.1%) 18 (60.0%) 23 (82.1%) 24 (70.6%) 47 (75.8%) 23 (79.3%) 27 (73.0%) 50 (75.8%)

Total 12 (100%) 18 (100%) 30 (100%) 28 (100%) 34 (100%) 62 (100%) 29 (100%) 37 (100%) 66 (100%)

Health Technology Assessment 2024 Vol. 28 No. 66

Individual questionnaires c

IBS-SSS 219 (94.4%) 213 (92.2%) 432 (93.3%) 204 (87.9%) 197 (85.3%) 401 (86.6%) 118 (80.3%) 107 (74.3%) 225 (77.3%)

SGA 220 (94.8%) 213 (92.2%) 433 (93.5%) 204 (87.9%) 195 (84.4%) 399 (86.2%) 118 (80.3%) 107 (74.3%) 225 (77.3%)

HADS 220 (94.8%) 212 (91.8%) 432 (93.3%) 203 (87.5%) 193 (83.5%) 396 (85.5%) 117 (79.6%) 107 (74.3%) 224 (77.0%)

PHQ-12 N/A N/A N/A 202 (87.1%) 192 (83.1%) 394 (85.1%) N/A N/A N/A

WSAS 219 (94.4%) 212 (91.8%) 431 (93.1%) 202 (87.1%) 193 (83.5%) 395 (85.3%) 117 (79.6%) 107 (74.3%) 224 (77.0%)

ASEC 219 (94.4%) 212 (91.8%) 431 (93.1%) 201 (86.6%) 192 (83.1%) 393 (84.9%) 117 (79.6%) 107 (74.3%) 224 (77.0%)

EQ-5D 218 (94.0%) 210 (90.9%) 428 (92.4%) 200 (86.2%) 192 (83.1%) 392 (84.7%) 117 (79.6%) 107 (74.3%) 224 (77.0%)

 ealth resource
H 217 (93.5%) 210 (90.9%) 427 (92.2%) 200 (86.2%) 192 (83.1%) 392 (84.7%) 117 (79.6%) 107 (74.3%) 224 (77.0%)
use

continued
23
 24
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Clinical trial results

TABLE 4 Monthly questionnaire completion (continued)

Month 3 Month 6 Month 12a

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 147) (n = 144) (n = 291)

Time to completion (days)

Mean (SD) 3.1 (0.3) 3.1 (0.3) 3.1 (0.3) 6.1 (0.2) 6.1 (0.3) 6.1 (0.2) 12.1 (0.2) 12.2 (0.3) 12.1 (0.2)

Median (range) 3.0 (2.7, 5.3) 3.0 (2.7, 4.7) 3.0 (2.7, 5.3) 6.0 (5.7, 7.6) 6.0 (5.9, 7.7) 6.0 (5.7, 7.7) 12.0 (12.0, 13.1) 12.1 (12.0, 13.2) 12.0 (12.0, 13.2)

n 220 213 433 203d

197 400 118 107 225

Timing of completion

≤ ± 7 days 185 (84.1%) 181 (85.0%) 366 (84.5%) 182 (89.2%) 167 (84.8%) 349 (87.0%) 103 (87.3%) 85 (79.4%) 188 (83.6%)

≤ ± 30 days 31 (14.1%) 29 (13.6%) 60 (13.9%) 19 (9.3%) 27 (13.7%) 46 (11.5%) 14 (11.9%) 21 (19.6%) 35 (15.6%)

> ± 30 days 4 (1.8%) 3 (1.4%) 7 (1.6%) 2 (1.0%) 3 (1.5%) 5 (1.2%) 1 (0.8%) 1 (0.9%) 2 (0.9%)

Total completed 220 (100%) 213 (100%) 433 (100%) 203 (100%) 197 (100%) 401 (100%) 118 (100%) 107 (100%) 225 (100%)

a Only participants consented to 12-month follow-up (n = 291, 62.9%) were given the opportunity to provide the 12-month outcome data.
b Participants who completed one or more questionnaire within the pack are categorised as ‘completed’.
c Individual questionnaires were completed at baseline with completion rates: IBS-SSS 100%, HADS 100%, PHQ-12 99.6%, WSAS 98.9%, EQ-5D 97.4% and health resource use 95%.
d Questionnaire completion date not available for one participant.
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 5 Characteristics of participants completing and not completing 6-month questionnaires

Completed (n = 401) Not completed (n = 62) Total (n = 463) p-valuea

Treatment allocation and receipt

Treatment allocation

Amitriptyline 204 (50.9%) 28 (45.2%) 232 (50.1%)

Placebo 197 (49.1%) 34 (54.8%) 231 (49.9%) 0.1883

Did not start or discontinued trial

medication before 6 months

Yes 80 (20.0%) 45 (72.6%) 125 (27.0%)

No 321 (80.0%) 17 (27.4%) 338 (73.0%) < 0.0001

Demographic characteristics

Recruitment hub

West Yorkshire 67 (16.7%) 20 (32.3%) 87 (18.8%)

Wessex 171 (42.6%) 21 (33.9%) 192 (41.5%)

West of England 163 (40.6%) 21 (33.9%) 184 (39.7%) 0.0068

IBS subtype

IBS-C 67 (16.7%) 10 (16.1%) 77 (16.6%)

IBS-D 158 (39.4%) 23 (37.1%) 181 (39.1%)

IBS-M 163 (40.6%) 28 (45.2%) 191 (41.3%)

IBS-U 13 (3.2%) 1 (1.6%) 14 (3.0%) 0.8896

Age at randomisation

Mean (SD) 48.9 (15.8) 45.7 (17.8) 48.5 (16.1)

Median (range) 50.0 (19.0, 86.0) 44.5 (20.0, 87.0) 49.0 (19.0, 87.0) 0.0489

Participant sex

Male 132 (32.9%) 16 (25.8%) 148 (32.0%)

Female 269 (67.1%) 46 (74.2%) 315 (68.0%) 0.4370

Baseline questionnaires

Baseline IBS-SSS score

Mean (SD) 269.3 (88.2) 295.3 (100.7) 272.8 (90.3)

Median (range) 270.0 (10.0, 480.0) 310.0 (60.0, 480.0) 280.0 (10.0, 480.0) 0.0140

IBS-SSS severity

Normal (< 75) 6 (1.5%) 2 (3.2%) 8 (1.7%)

Mild (75–174) 55 (13.7%) 8 (12.9%) 63 (13.6%)

Moderate (175–299) 184 (45.9%) 17 (27.4%) 201 (43.4%)

Severe (300–500) 156 (38.9%) 35 (56.5%) 191 (41.3%)

Baseline HADS-anxiety score

Mean (SD) 7.5 (4.2) 7.7 (4.6) 7.5 (4.3)

Median (range) 7.0 (0.0, 21.0) 7.0 (0.0, 21.0) 7.0 (0.0, 21.0) 0.8471

continued

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 25
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Clinical trial results

TABLE 5 Characteristics of participants completing and not completing 6-month questionnaires (continued)

Completed (n = 401) Not completed (n = 62) Total (n = 463) p-valuea

Baseline HADS-depression score

Mean (SD) 4.2 (3.4) 4.4 (3.4) 4.3 (3.4)

Median (range) 4.0 (0.0, 18.0) 4.0 (0.0, 14.0) 4.0 (0.0, 18.0) 0.8202

Baseline PHQ-12 score

Mean (SD) 6.2 (3.4) 6.8 (4.4) 6.3 (3.5)

Median (range) 6.0 (0.0, 16.4) 5.5 (0.0, 18.0) 6.0 (0.0, 18.0) 0.2396

Missing 3 3 6

Baseline WSAS score

Mean (SD) 10.9 (7.5) 14.3 (9.6) 11.4 (7.9)

Median (range) 10.0 (0.0, 36.0) 13.5 (0.0, 40.0) 10.0 (0.0, 40.0) 0.0024

Missing 18 2 20

a p-value calculated in univariable logistic regression.

Baseline characteristics

Demographics and clinical characteristics

Participants allocated to amitriptyline and placebo were well balanced with respect to randomisation
stratification factors (Table 7), demographic characteristics (Table 8), and clinical characteristics (Table 9).
Wessex randomised 41.5% of participants, West of England 39.7%, and West Yorkshire 18.8% (Table 7).
A small proportion of participants had IBS-U (3.0%), 41.3% had IBS-M, 39.1% IBS-D and 16.6% IBS-C.
The majority of participants (84.2%) had a HADS-D score ˂ 8, indicating the absence of symptoms
of depression.

The mean age at randomisation was 48.5 (SD 16.1) years and over two-thirds (68.0%) of participants
were female (Table 8). Most participants were white (97.6%) with < 1% of each black, Asian, mixed
or other ethnic groups. Over two-thirds of participants were married or living with a partner, and the
majority (> 90%) had received formal education. Approximately a third of participants lived in each of
the least 20% and 20–40% least deprived neighbourhoods in England (IMD quintile 5 and 4), while
5.7% and 13.3% lived in the 20% and 20–40% most deprived neighbourhoods (IMD quintile 1 and 2)
respectively. Further details of IMD by decile and hub are in Appendix 1, Table 44.

All participants had previously tried dietary changes and other first-line therapies and met Rome IV
criteria (Table 9). Over three-quarters had tried antispasmodics (77.5%), 31.3% an anti-diarrhoeal, 22.5%
fibre supplements, 18.4% laxatives and 9.7% reported having tried peppermint oil. Participants were
randomised a mean of 12.5 (SD 11.0) years after their IBS diagnosis (median 10 years, range < 1 week to
67 years) (see Appendix 1, Figure 11).

Baseline questionnaires
Baseline questionnaire scores were largely balanced across trial arms (Table 10). The mean IBS-SSS score
was 272.8 (SD 90.33), and 41.3% of participants reported severe, 43.4% moderate and 13.6% mild IBS
symptoms. The mean HADS-A and HADS-D scores were 7.5 (SD 4.3) and 4.3 (SD 3.4), respectively, with
47.1% and 15.8% reporting at least mild symptoms of anxiety or depression. Over a third of participants
reported having ever received treatment for depression (38.4%) or anxiety (34.3%), with a slightly higher
proportion having received treatment for depression in the placebo arm (42.9%). The mean PHQ-12
score was 6.3 (SD 3.5) and the mean WSAS score was 11.4 (SD 7.4).

26

NIHR Journals Library www.journalslibrary.nihr.ac.uk

 TABLE 6 Per-protocol and safety population
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
Month 3 Month 6 Month 12a

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 147) (n = 144) (n = 291)

In per-protocol population

Yes 193 (83.2%) 183 (79.2%) 376 (81.2%) 172 (74.1%) 151 (65.4%) 323 (69.8%)

No 39 (16.8%) 48 (20.8%) 87 (18.8%) 60 (25.9%) 80 (34.6%) 140 (30.2%)

Total 232 (100%) 231 (100%) 463 (100%) 232 (100%) 231 (100%) 463 (100%)

Reasons for exclusionb

Discontinued trial medication 30 (76.9%) 30 (62.5%) 60 (69.0%) 44 (73.3%) 58 (72.5%) 102 (72.9%)

 reach eligibility and

B 1 (2.6%) 0 (0.0%) 1 (1.1%) 1 (1.7%) 0 (0.0%) 1 (0.7%)
discontinued trial medication

 ajor violation and discon-

M 1 (2.6%) 1 (2.1%) 2 (2.3%) 1 (1.7%) 1 (1.3%) 2 (1.4%)
tinued trial medication

Not started treatment 2 (5.1%) 1 (2.1%) 3 (3.4%) 2 (3.3%) 1 (1.3%) 3 (2.1%)

Lost to follow-up 4 (10.3%) 3 (6.3%) 7 (8.0%) 11 (18.3%) 6 (7.5%) 17 (12.1%)

Health Technology Assessment 2024 Vol. 28 No. 66

Major violation 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (1.3%) 1 (0.7%)

 ot adhered to trial
N 1 (2.6%) 2 (4.2%) 3 (3.4%) 1 (1.7%) 6 (7.5%) 7 (5.0%)
medication

 o response to treatment
N 0 (0.0%) 11 (22.9%) 11 (12.6%) 0 (0.0%) 3 (3.8%) 3 (2.1%)
adherence question

Other non-adherence 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 4 (5.0%) 4 (2.9%)

Total excluded 39 (100%) 48 (100%) 87 (100%) 60 (100%) 80 (100%) 140 (100%)

Safety population

Amitriptyline 230 (100.0%) 1 (0.4%) 231 (50.2%) 230 (100.0%) 2 (0.9%) 232 (50.4%) 145 (100.0%) 1 (0.7%) 146 (50.7%)

Placebo 0 (0.0%) 229 (99.6%) 229 (49.8%) 0 (0.0%) 228 (99.1%) 228 (49.6%) 0 (0.0%) 142 (99.3%) 142 (49.3%)

Excluded 2 1 3 2 1 3 2 1 3

a Per-protocol population not defined for the 12-month population.

b Non-adherence to trial medication based on response to researcher at follow-up call. Other non-adherence where adherence to medication was reported at 6 months. However, the
participant had not been contactable or received treatment replenishment at 3 months.
27
 Clinical trial results

TABLE 7 Randomisation stratification factorsa

Amitriptyline (n = 232) (%) Placebo (n = 231) (%) Total (n = 463) (%)

Recruitment hub

West Yorkshire 43 (18.5) 44 (19.0) 87 (18.8)

Wessex 97 (41.8) 95 (41.1) 192 (41.5)

West of England 92 (39.7) 92 (39.8) 184 (39.7)

IBS subtype

IBS-C 40 (17.2) 37 (16.0) 77 (16.6)

IBS-D 92 (39.7) 89 (38.5) 181 (39.1)

IBS-M 93 (40.1) 98 (42.4) 191 (41.3)

IBS-U 7 (3.0) 7 (3.0) 14 (3.0)

Baseline HADS-D score

0–7 (normal range) 195 (84.1) 195 (84.4) 390 (84.2)

8–21 (mild, moderate, severe depression) 37 (15.9) 36 (15.6) 73 (15.8)

a True values for stratification factors are presented. Incorrect IBS subtype was entered at randomisation for four
participants: two participants allocated to amitriptyline with IBS-D were randomised under IBS-C and IBS-M; two
participants allocated to placebo with IBS-M were randomised under IBS-C and IBS-D. Incorrect HADS-D score
(HADS-D ≥ 8) was entered at randomisation for seven participants (three allocated to amitriptyline, four allocated to
placebo) with HADS-D < 8.

TABLE 8 Demographic characteristics of randomised participantsa

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

Age (years)

Mean (SD) 49.2 (16.2) 47.8 (15.9) 48.5 (16.1)

Median (range) 50 (19, 86) 49 (19, 87) 49 (19, 87)

Participant sex

Female 156 (67.2%) 159 (68.8%) 315 (68.0%)

Male 76 (32.8%) 72 (31.2%) 148 (32.0%)

Ethnic origin

White 226 (97.4%) 225 (97.8%) 451 (97.6%)

Black 0 (0.0%) 1 (0.4%) 1 (0.2%)

Asian 2 (0.9%) 2 (0.9%) 4 (0.9%)

Other ethnic group 1 (0.4%) 1 (0.4%) 2 (0.4%)

Mixed 3 (1.3%) 0 (0.0%) 3 (0.6%)

Prefer not to say 0 (0.0%) 1 (0.4%) 1 (0.2%)

Missing 0 1 1

Marital status

Single 37 (15.9%) 55 (23.8%) 92 (19.9%)

Married 123 (53.0%) 110 (47.6%) 233 (50.3%)

28

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 8 Demographic characteristics of randomised participantsa (continued)

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

Living with partner 43 (18.5%) 36 (15.6%) 79 (17.1%)

Separated 2 (0.9%) 4 (1.7%) 6 (1.3%)

Divorced 20 (8.6%) 19 (8.2%) 39 (8.4%)

Widowed 7 (3.0%) 7 (3.0%) 14 (3.0%)

Highest education level achieved

No formal 13 (5.6%) 18 (7.8%) 31 (6.7%)

GCSE/O level or equivalent 61 (26.4%) 61 (26.5%) 122 (26.5%)

A level or equivalent 54 (23.4%) 54 (23.5%) 108 (23.4%)

Degree 52 (22.5%) 58 (25.2%) 110 (23.9%)

Postgraduate 47 (20.3%) 31 (13.5%) 78 (16.9%)

Diploma 3 (1.3%) 3 (1.3%) 6 (1.3%)

Otherb 1 (0.4%) 5 (2.2%) 6 (1.3%)

Missing 1 1 2

IMD quintilec

1 13 (5.7%) 13 (5.7%) 26 (5.7%)

2 34 (14.8%) 27 (11.7%) 61 (13.3%)

3 38 (16.6%) 33 (14.3%) 71 (15.5%)

4 75 (32.8%) 74 (32.2%) 149 (32.5%)

5 69 (30.1%) 83 (36.1%) 152 (33.1%)

Missing 3 1 4

a Statistics out of all randomised participants except where the missing indicated.
b Other education level: CSE (two reported), professionally qualified, Certificate in Training Practice, RSA/pitman
business/secretarial, city and guilds.
c IMD = Index of Multiple Deprivation: quintile 1 = neighbourhood in the 20% most deprived neighbourhoods in
England, 2 = 20–40%, 3 = 40–60%, 4 = 60–80%, 5 = neighbourhood in the 20% least deprived neighbourhoods in
England.

TABLE 9 Clinical characteristics of randomised participants

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

First-line therapies a

Antispasmodics (e.g. mebeverine) 176 (76.5%) 183 (79.2%) 359 (77.9%)

Anti-diarrhoeals (e.g. loperamide) 70 (30.4%) 75 (32.5%) 145 (31.5%)

Fibre supplements (e.g. ispaghula husk) 52 (22.6%) 52 (22.5%) 104 (22.6%)

Laxatives (e.g. bisacodyl) 51 (22.2%) 34 (14.7%) 85 (18.4%)

Peppermint oil (e.g. Mintec, Colpermin) 18 (7.8%) 27 (11.7%) 45 (9.8%)

Otherb 4 (1.7%) 9 (3.9%) 13 (2.8%)

Missing 2 1 3
continued

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 29
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Clinical trial results

TABLE 9 Clinical characteristics of randomised participants (continued)

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

Years from IBS diagnosis to randomisation

Mean (SD) 13.3 (11.8) 11.8 (10.2) 12.5 (11.0)

Median (range) 10.0 (0.0, 67.0) 9.0 (0.0, 44.0) 10.0 (0.0, 67.0)

IQR 4.0, 21.0 4.0, 18.0 4.0, 20.0

Missing 0 1 1

Years from IBS diagnosis to randomisation

≤ 2 years 31 (13.4%) 36 (15.7%) 67 (14.5%)

≤ 10 years 93 (40.1%) 93 (40.4%) 186 (40.3%)

≤ 20 years 47 (20.3%) 53 (23.0%) 100 (21.6%)

≤ 30 years 42 (18.1%) 31 (13.5%) 73 (15.8%)

> 30 years 19 (8.2%) 17 (7.4%) 36 (7.8%)

Missing 0 1 1

Rome IV criteria

Abdominal pain 232 (100.0%) 231 (100.0%) 463 (100.0%)

Pain relieved or aggravated by defaecation 215 (92.7%) 211 (91.3%) 426 (92.0%)

Pain associated with change in stool frequency 193 (83.2%) 189 (81.8%) 382 (82.5%)

Pain associated with change in stool appearance 212 (91.4%) 204 (88.3%) 416 (89.8%)

Symptom onset 6 months prior to diagnosis 232 (100.0%) 231 (100.0%) 463 (100.0%)

a Not mutually exclusive, percentages calculated out of number of participants randomised.

b Other first-line therapies: amitriptyline, aware of all OTC treatments for IBS, CBD tablet, codeine, colesevalem, colofac,
lansoprazole, omeprazole, prucalopride, turmeric tablets, charcoal tablets, slippery elm.

TABLE 10 Baseline questionnairesa,b

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

Total IBS-SSS scorec

Mean (SD) 273.4 (90.53) 272.1 (90.33) 272.8 (90.33)

Median (range) 280.0 (60, 480) 270.0 (10, 470) 280.0 (10, 480)

IBS-SSS level

< 75 (remission) 3 (1.3%) 5 (2.2%) 8 (1.7%)

75–174 (mild) 37 (15.9%) 26 (11.3%) 63 (13.6%)

175–299 (moderate) 98 (42.2%) 103 (44.6%) 201 (43.4%)

300–500 (severe) 94 (40.5%) 97 (42.0%) 191 (41.3%)

HADS-A scores

Mean (SD) 7.3 (4.3) 7.7 (4.3) 7.5 (4.3)

Median (range) 7.0 (0.0, 21.0) 7.0 (0.0, 20.0) 7.0 (0.0, 21.0)

HADS-A level

30

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 10 Baseline questionnairesa,b (continued)

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

0–7 (normal range) 126 (54.3%) 119 (51.5%) 245 (52.9%)

8–10 (mild anxiety) 55 (23.7%) 58 (25.1%) 113 (24.4%)

11–14 (moderate anxiety) 39 (16.8%) 36 (15.6%) 75 (16.2%)

15–21 (severe anxiety) 12 (5.2%) 18 (7.8%) 30 (6.5%)

HADS-D scores

Mean (SD) 4.4 (3.6) 4.1 (3.2) 4.3 (3.4)

Median (range) 4.0 (0.0, 18.0) 4.0 (0.0, 15.0) 4.0 (0.0, 18.0)

HADS-D level

0–7 (normal range) 195 (84.1%) 195 (84.4%) 390 (84.2%)

8–10 (mild depression) 23 (9.9%) 24 (10.4%) 47 (10.2%)

11–14 (moderate depression) 11 (4.7%) 11 (4.8%) 22 (4.8%)

15–21 (severe depression) 3 (1.3%) 1 (0.4%) 4 (0.9%)

Have you ever been treated for depression?

Yes 79 (34.2%) 99 (42.9%) 178 (38.5%)

No 152 (65.8%) 132 (57.1%) 284 (61.5%)

Missing 1 0 1

Have you ever been treated for anxiety?

Yes 80 (34.6%) 79 (34.2%) 159 (34.4%)

No 151 (65.4%) 152 (65.8%) 303 (65.6%)

Missing 1 0 1

Total PHQ-12 score

Mean (SD) 6.3 (3.5) 6.3 (3.6) 6.3 (3.5)

Median (range) 6.0 (0.0, 17.3) 6.0 (0.0, 18.0) 6.0 (0.0, 18.0)

Missing 4 2 6

WSAS total score

Mean (SD) 11.2 (8.2) 11.5 (7.6) 11.4 (7.9)

Median (range) 9.0 (0.0, 40.0) 11.0 (0.0, 40.0) 10.0 (0.0, 40.0)

Missing 8 12 20

a Statistics out of all randomised participants except where missing indicated.

b IBS-SSS scores range 0–500, higher scores indicate more severe IBS symptoms. HADS scores range 0–21, higher
scores indicate more severe levels of anxiety and depression. PHQ-12 scores range 0–24 in women and 0–22 in
men, higher scores indicate more severe somatic symptoms. WSAS scores range 0–40, higher scores indicate greater
impairment in functioning.
c IBS-SSS score also collected prior to baseline during the initial eligibility screening call with mean score 290 (SD 79.8),
range 80–470. Eight participants had an IBS-SSS score ≥ 75 during the initial eligibility screening, but < 75 points on
the subsequent baseline questionnaire; they were considered eligible based on the initial eligibility screening call.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 31
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Clinical trial results

Six-month treatment delivery and receipt

Treatment receipt and completion

A total of 338 (73.0%) participants completed 6 months treatment, 173 (74.6%) allocated to
amitriptyline, and 165 (71.4%) allocated to placebo (Table 11), and 105 (22.7%) discontinued trial
medication before 6 months, 46 (19.8%) allocated to amitriptyline and 59 (25.5%) allocated to
placebo. Another 17 (3.7%) participants were lost to follow-up and three participants (< 1%) did not
start treatment.

TABLE 11 Six-month treatment receipt

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

Completed 6 months treatment

Yes 173 (74.6%) 165 (71.4%) 338 (73.0%)

Discontinued treatment before 6 months 46 (19.8%) 59 (25.5%) 105 (22.7%)

Lost to follow-up before 6 months 11 (4.7%) 6 (2.6%) 17 (3.7%)

Did not start treatmenta 2 (0.9%) 1 (0.4%) 3 (0.6%)

Time from randomisation to treatment start (days)

Mean (SD) 7.5 (5.66) 7.4 (6.05) 7.5 (5.86)

Median (range) 6.0 (0.0, 46.0) 6.0 (1.0, 43.0) 6.0 (0.0, 46.0)

IQR 4.0–9.0 4.0–9.5 4.0–9.0

Missing 7 3 10

n 225 228 453

Time from randomisation to treatment start

≤ 7 days 145 (64.4%) 152 (66.7%) 297 (65.6%)

≤ 14 days 60 (26.7%) 53 (23.2%) 113 (24.9%)

≤ 21 days 13 (5.8%) 16 (7.0%) 29 (6.4%)

> 21 days 7 (3.1%) 7 (3.1%) 14 (3.1%)

Missing 5 2 7

Optional GP review taken place?

Yes 4 (1.7%) 10 (4.3%) 14 (3.0%)

In person 2 3 5

Over the telephone 2 7 9

Time to treatment discontinuation (months)

Mean (SD) 2.5 (1.62) 2.7 (1.76) 2.6 (1.70)

Median (range) 2.3 (0.2, 5.7) 2.8 (0.2, 5.9) 2.5 (0.2, 5.9)

IQR 1.1–4.2 0.8–4.4 1.1–4.2

n 46 59 105

Reason for discontinuation

Side effect 30 (12.9%) 20 (8.7%) 50 (10.8%)

Lack of benefit 7 (3.0%) 18 (7.8%) 25 (5.4%)

32

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 11 Six-month treatment receipt (continued)

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

Non-specific or personal choice 5 (2.2%) 15 (6.5%) 20 (4.3%)

SAR b
1 (0.4%) 3 (1.3%) 3 (0.6%)

Safety (including allergic reactions to IMP) 1 (0.4%) 2 (0.9%) 3 (0.6%)

Stopped in error on advice of GP 1 (0.4%) 0 (0.0%) 1 (0.2%)

Randomised in error 1 (0.4%) 0 (0.0%) 1 (0.2%)

Administrative error 0 (0.0%) 1 (0.4%) 1 (0.2%)

Total 232 (100%) 231 (100%) 463 (100%)

a One participant started a new medication that contraindicated amitriptyline and was unable to start trial medication,
one participant was not contactable and their posted trial medications were returned to the post office undelivered
and unopened, one participant felt too unwell and chose not to start taking trial medication.
b The SAR in the amitriptyline arm, and one in the placebo arm related to suicidal ideation.

An optional GP review at 1-month post randomisation was requested by only 18 (3.9%) participants and
was reported to have taken place either in person or over the telephone for 14 (3.0%) participants, 4
(1.7%) allocated to amitriptyline and 10 (4.3%) allocated to placebo.

Participants started treatment a median of 6 days post randomisation, and the majority (88.6%) started
treatment within 2 weeks. In participants who discontinued treatment before 6 months, the median time
to discontinuation was 2.5 months, with a slightly earlier time to discontinuation in the amitriptyline arm
(median 2.3 vs. 2.8 months). The most common reason for treatment discontinuation was side effects in
30 (12.9%) and 20 (8.7%) participants allocated to amitriptyline and placebo respectively, followed by
lack of benefit in 7 (3.0%) amitriptyline and 18 (7.8%) placebo participants, and non-specific or personal
choice in 5 (2.2%) amitriptyline and 15 (6.5%) placebo participants. The most common side effects
leading to treatment discontinuation were drowsiness (in 13 participants allocated to amitriptyline
and 6 to placebo), deterioration of mood (9 amitriptyline, 5 placebo), constipation (4 amitriptyline, 3
placebo) and headache (2 amitriptyline, 4 placebo) (see Appendix 1, Figure 12). Note that side effects in
all participants according to the ASEC are reported in detail in Tolerability at 3 and 6 months.

Across all randomised participants (ITT population), the median time from randomisation to the end-of-
trial treatment was 5.8 months, with similar duration observed across trial arms (Table 12). Appendix 1,
Figure 13 further presents the distribution of time from randomisation to treatment end date for all
participants and by follow-up duration, and Further 12-month treatment delivery and secondary end points
provides further details of treatment beyond 6 months for participants in the 12-month ITT population.

TABLE 12 Time from randomisation to treatment end date

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

Months to last dose (from randomisation)

Mean (SD) 7.3 (3.74) 6.8 (3.72) 7.0 (3.73)

Median (range) 5.9 (0.2, 14.0) 5.8 (0.2, 12.2) 5.8 (0.2, 14.0)

IQR 5.7–11.9 4.9–11.9 5.5–11.9

Missing 21 16 37

n 211 215 426

continued

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 33
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Clinical trial results

TABLE 12 Time from randomisation to treatment end date (continued)

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

Months to last dose (from randomisation)

≤ 1 month 10 (4.7%) 16 (7.4%) 26 (6.1%)

≤ 2 months 11 (5.2%) 9 (4.2%) 20 (4.7%)

≤ 3 months 11 (5.2%) 6 (2.8%) 17 (4.0%)

≤ 4 months 2 (0.9%) 12 (5.6%) 14 (3.3%)

≤ 5 months 9 (4.3%) 11 (5.1%) 20 (4.7%)

≤ 6 months 79 (37.4%) 85 (39.5%) 164 (38.5%)

≤ 7 months 9 (4.3%) 8 (3.7%) 17 (4.0%)

≤ 8 months 2 (0.9%) 2 (0.9%) 4 (0.9%)

≤ 9 months 6 (2.8%) 2 (0.9%) 8 (1.9%)

≤ 10 months 2 (0.9%) 1 (0.5%) 3 (0.7%)

≤ 11 months 2 (0.9%) 1 (0.5%) 3 (0.7%)

≤ 12 months 37 (17.5%) 43 (20.0%) 80 (18.8%)

> 12 months 31 (14.7%) 19 (8.8%) 50 (11.7%)

Total 211 (100%) 215 (100%) 426 (100%)

Treatment adherence, dosage titration and modification

Over 90% of participants on trial medication at each of week 3, month 3 and month 6, reported the
highest level of treatment adherence, taking at least one tablet ‘every day or nearly every day’, with
similar rates across trial arms at each time point (see Table 13 and Appendix 1, Figure 14). At 3 weeks
post randomisation, most participants were taking a dose of 10 mg (44.9% amitriptyline vs. 35.4%
placebo) or 20 mg (37.0% amitriptyline vs. 39.2% placebo) once daily, with a slightly higher proportion
of participants in the placebo arm taking 30 mg once-daily (see Table 13 and Appendix 1, Figure 15). By
3 months, similar proportions of participants randomised to amitriptyline were taking 20 mg (35.2%)
or 30 mg (37.8%) once daily, although by 6 months this increased to 42.8% taking 30 mg once daily.
However, in the placebo arm, 57.0% of participants titrated their dose to 30 mg once daily by 3 months
and this remained similar at 6 months.

Between 3 weeks and 3 months post randomisation, just over half of participants still on treatment had
modified their dose (50.5% amitriptyline, 55.2% placebo) of whom the majority had an increase in dose
(82.3% amitriptyline, 88.0% placebo). Dose modifications were less frequent between 3 and 6 months,
reported in 33 (19.2%) and 17 (11.2%) participants still receiving amitriptyline and placebo respectively,
with half the modifications to a higher and half to a lower dose, and with similar proportions across
trial arms.

Dosage and adherence details reported in participants who had discontinued treatment are presented in
Appendix 1, Tables 45 and 46.

Further treatment summaries

Off-trial amitriptyline
One participant (0.2% of all participants, 1.0% of participants who discontinued trial medication before
6 months) reported taking amitriptyline off-trial after treatment discontinuation within 6 months
of randomisation.

34

NIHR Journals Library www.journalslibrary.nihr.ac.uk

attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
TABLE 13 Treatment adherence and dose for participants on treatment: at 3 weeks, 3 and 6 months

Week 3 Month 3 Month 6

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 222) (n = 223) (n = 445) (n = 194) (n = 196) (n = 390) (n = 173) (n = 165) (n = 338)

Participant taken at least one tablet daily

 very/nearly every
E 212 (98.1%) 204 (95.8%) 416 (97.0%) 185 (95.4%) 179 (96.8%) 364 (96.0%) 163 (94.2%) 146 (90.1%) 309 (92.2%)
day

≥ half the days 4 (1.9%) 6 (2.8%) 10 (2.3%) 8 (4.1%) 4 (2.2%) 12 (3.2%) 9 (5.2%) 9 (5.6%) 18 (5.4%)

< half of the days 0 (0.0%) 2 (0.9%) 2 (0.5%) 0 (0.0%) 2 (1.1%) 2 (0.5%) 1 (0.6%) 4 (2.5%) 5 (1.5%)

 one/nearly no
N 0 (0.0%) 1 (0.5%) 1 (0.2%) 1 (0.5%) 0 (0.0%) 1 (0.3%) 0 (0.0%) 3 (1.9%) 3 (0.9%)
days

Missing 6 10 16 0 11 11 0 3 3

Current dose of trial medication

 × 10 mg every
1 1 (0.5%) 4 (1.9%) 5 (1.2%) 3 (1.6%) 4 (2.2%) 7 (1.8%) 6 (3.5%) 4 (2.5%) 10 (3.0%)
other day

Health Technology Assessment 2024 Vol. 28 No. 66

1 × 10 mg daily 97 (44.9%) 75 (35.4%) 172 (40.2%) 49 (25.4%) 23 (12.4%) 72 (19.0%) 40 (23.1%) 22 (13.9%) 62 (18.7%)

2 × 10 mg daily 80 (37.0%) 83 (39.2%) 163 (38.1%) 68 (35.2%) 53 (28.5%) 121 (31.9%) 53 (30.6%) 43 (27.2%) 96 (29.0%)

3 × 10 mg daily 38 (17.6%) 50 (23.6%) 88 (20.6%) 73 (37.8%) 106 (57.0%) 179 (47.2%) 74 (42.8%) 89 (56.3%) 163 (49.2%)

Missing 6 11 17 1 10 11 0 7 7

Dose modification since last follow-up call

Yes 96 (50.5%) 100 (55.2%) 196 (52.8%) 33 (19.2%) 17 (11.2%) 50 (15.4%)

Higher dose 79 (82.3%) 88 (88.0%) 167 (85.2%) 17 (51.5%) 8 (47.1%) 25 (50.0%)

Lower dose 17 (17.7%) 12 (12.0%) 29 (14.8%) 16 (48.5%) 9 (52.9%) 25 (50.0%)

No 94 (49.5%) 81 (44.8%) 175 (47.2%) 139 (80.8%) 135 (88.8%) 274 (84.6%)

Missing 4 15 19 1 13 14
35
 Clinical trial results

Suicidal ideation and drugs leading to discontinuation

Two participants (one in the placebo arm and one in the amitriptyline arm) reported experiencing
thoughts of self-harm, one at the 3-week and one at the 3-month follow-up call. Both events were
reported as a SAR (see Safety) and participants discontinued trial medication subsequently. No
participants reported taking, or discontinued trial medication due to taking, monoamine oxidase
inhibitors or drugs prolonging the QT interval during the study during researcher concomitant
medication reviews at the 1-week, 3-week, 3-month or 6-month follow-up call.

Treatment replenishment and replacement

Treatment replenishment at 3 weeks and 3 months is presented in Appendix 1, Table 47. A total of 20
treatment kit replacements were conducted for 18 participants (10 amitriptyline, 8 placebo) throughout
the trial. The reasons for the replacement requests were damaged or lost bottles, participants running
out of medication before their next scheduled follow-up call, or administrative errors, where a
replacement was performed instead of replenishment, or a pharmacy error occurred.

Changes in diet, exercise, other IBS treatments, and IBS symptoms at 6 months
Of 338 participants completing 6 months treatment and researcher follow-up, 67 (19.9%) reported
having tried at least one new diet for IBS during the study, 66 (19.6%) reported increasing the amount of
exercise they did to help their IBS symptoms, whereas 11 (3.3%) reported reducing their exercise, and 34
(10.1%) tried one or more other treatments for their IBS symptoms, with similar rates across trial arms
(Table 14). Over two-thirds of participants on amitriptyline [118 (68.2%)] reported having experienced
improved IBS symptoms, compared with just over half of participants on placebo [89 (54.6%)]. The
majority of participants reporting improvement attributed this to the ATLANTIS trial medication [103
(87.3%) on amitriptyline; 75 (84.3%) on placebo]. Among participants who had discontinued treatment
before 6 months, a lower proportion reported experiencing improved IBS symptoms [13 (28.3%) on
amitriptyline; 10 (29.1%) on placebo]. Appendix 1, Table 48 further details the types of new diets, other
treatments and attributed reasons for improvement in IBS symptoms.

Primary end point: 6-month global symptoms of irritable bowel syndrome (irritable
bowel syndrome Severity Scoring System)

Summary statistics of available data for the IBS-SSS up to and including the 6-month follow-up
are presented in Table 15 and Appendix 1, Figure 16. The IBS-SSS item level scores can be found in
Appendix 1, Table 49 and Appendix 1, Figure 17. The total IBS-SSS score was available at 6 months for
401 (86.6%) participants [204 (87.9%) on amitriptyline; 197 (85.3%) on placebo]. Although IBS-SSS
scores in each arm were similar at baseline, the mean 6-month scores were 170.4 (SD 107.7) and 200.1
(SD 114.4) in participants allocated to amitriptyline and placebo, respectively. A higher proportion of
participants had remission of IBS symptoms according to the IBS-SSS (score < 75 points) at 6 months in
the amitriptyline arm compared with placebo (23.5% vs. 15.7%), and a lower proportion of participants
had severe IBS symptoms on the IBS-SSS (score ≥ 300 points) in the amitriptyline arm compared with
placebo (15.2% vs. 23.4%).

Amitriptyline was superior to placebo at 6 months in ITT analysis, using linear regression adjusted for
covariates with multiple imputation, with strong evidence (p < 0.05) of a reduced total IBS-SSS score
at 6 months with amitriptyline compared with placebo and an estimated adjusted mean difference of
−27.01 (95% CI −46.91 to −7.10; p = 0.008). Results of sensitivity analyses were consistent, albeit with
larger reductions in the 6-month IBS-SSS score, with amitriptyline compared with placebo (Table 16).

Baseline IBS-SSS score was strongly associated with outcome such that higher baseline scores were
associated with higher 6-month scores, with a 0.51 (95% CI 0.40 to 0.62; p < 0.001) increase in 6-month

36

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 14 Number of participants trying a new diet, changing their amount of exercise, or trying other irritable bowel
syndrome treatments during the study, and experienced improvement in irritable bowel syndrome symptoms

Month 6 Discontinued treatment before month 6

Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 173) (n = 165) (n = 338) (n = 46) (n = 59) (n = 105)

Tried a new diet for IBS 33 (19.1%) 34 (20.9%) 67 (19.9%) 0 (0.0%) 6 (12.8%) 6 (7.6%)

Changed amount of exercise

Increased the amount 38 (22.0%) 28 (17.2%) 66 (19.6%) 3 (9.3%) 5 (10.6%) 8 (10.1%)

of exercise

 educed the amount

R 4 (2.3%) 7 (4.3%) 11 (3.3%) 1 (3.1%) 0 (0.0%) 1 (1.3%)
of exercise

Tried other treatments 18 (10.4%) 16 (9.8%) 34 (10.1%) 5 (15.6%) 2 (4.3%) 7 (8.9%)

for IBS symptoms

Experienced improved 118 (68.2%) 89 (54.6%) 207 (61.6%) 13 (28.3%) 10 (21.3%) 23 (29.1%)
IBS symptoms

Missing 0 2 2 14 12 26

IBS-SSS score for every unit increase in baseline score in the primary analysis (Table 16). Similar effects
were observed in sensitivity analysis. There were no other statistically significant covariate effects.

There was weak evidence (p < 0.1) that participants with IBS-C had higher 6-month IBS-SSS scores
compared with participants with IBS-M or IBS-U, with a mean adjusted difference of 24.7 (95% CI
−3.5 to 52.9; p = 0.086) in primary analysis, and an extenuated effect in per-protocol analysis and a
reduced effect in complete case analysis. Participants recruited in West Yorkshire tended to have lower
scores than participants recruited from Wessex or West of England. However, only weak evidence
of a statistically significant effect was observed in complete case analysis and not in the primary or
per-protocol analysis.

Missing data
Overall, across trial arms, univariable analysis identified recruitment hub as predictive (p < 0.05) of
missing data status, baseline HADS-D, baseline HADS-A and baseline PHQ-12 scores as predictive
of outcome, and age, baseline IBS-SSS, and baseline WSAS scores, and 6-month treatment status as
key characteristics predictive of both missing data status and outcome (see Table 5 and Appendix 1,
Table 50).

Primary analysis, therefore, imputed missing 6-month IBS-SSS scores using multiple imputation
(stratified by treatment group) and incorporated IBS-SSS score at baseline and 3 months, the planned
covariates of recruitment hub, IBS subtype, and HADS-D score, as well as additional variables found to
be predictive of missingness and/or outcomes including 6-month treatment status, age, baseline WSAS,
HADS-A and PHQ-12 score. Sex was also included in the imputation model to allow for consistency
across imputation models for all outcomes (as a covariate in PHQ-12 analysis).

Model checking
Graphical and statistical tests of the adequacy of the linear regression model for treatment and
covariates were generally satisfactory and are presented in Appendix 1, Figure 18.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 37
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 38
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Clinical trial results

TABLE 15 Total IBS-SSS score, IBS-SSS severity and change in IBS-SSS score from baseline

Baseline Month 3 Month 6

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463)

Total IBS-SSS score

Mean (SD) 273.4 (90.53) 272.1 272.8 173.0 (106.63) 194.6 (107.53) 183.7 (107.50) 170.4 (107.73) 200.1 (114.46) 185.0 (111.94)
(90.33) (90.33)

Median (range) 280.0 (60, 480) 270.0 (10, 280.0 (10, 180.0 (0, 460) 190.0 (0, 430) 180.0 (0, 460) 170.0 (0, 500) 190.0 (0, 450) 180.0 (0, 500)
470) 480)

IQR 210.0–330.0 200.0–330.0 210.0–330.0 80.0–250.0 110.0–270.0 100.0–260.0 80.0–250.0 110.0–290.0 90.0–270.0

Missing 0 0 0 13 18 31 28 34 62

 ean difference
M −23.30 (−41.96 to −27.01 (−46.91 to
(95% CI), p-value −4.64), 0.014 −7.10), 0.008

IBS-SSS severity

< 75 (remission) 3 (1.3%) 5 (2.2%) 8 (1.7%) 49 (22.4%) 34 (16.0%) 83 (19.2%) 48 (23.5%) 31 (15.7%) 79 (19.7%)

75–174 (mild) 37 (15.9%) 26 (11.3%) 63 (13.6%) 59 (26.9%) 64 (30.0%) 123 (28.5%) 58 (28.4%) 58 (29.4%) 116 (28.9%)

175–299 98 (42.2%) 103 (44.6%) 201 (43.4%) 84 (38.4%) 70 (32.9%) 154 (35.6%) 67 (32.8%) 62 (31.5%) 129 (32.2%)
(moderate)

300–500 (severe) 94 (40.5%) 97 (42.0%) 191 (41.3%) 27 (12.3%) 45 (21.1%) 72 (16.7%) 31 (15.2%) 46 (23.4%) 77 (19.2%)

Missing 13 18 31 28 34 62

Difference in IBS-SSS score from baseline

Mean (SD) −99.8 (107.67) −76.1 (107.09) −88.1 (107.92) −99.2 (112.88) −68.9 (109.26) −84.3 (112.01)

Median (range) −80.0 (−370.0 to −70.0 (−430.0 −80.0 (−430.0 to −90.0 (−360.0 to −50.0 (−380.0 to −80.0 (−380.0 to
170.0) to 250.0) 250.0) 160.0) 200.0) 200.0)

IQR −170.0 to −20.0 −140.0 to 0.0 −160.0 to −20.0 −170.0 to −20.0 −130.0 to 0.0 −160.0 to −10.0
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
Baseline Month 3 Month 6

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463)

Missing 13 18 31 28 34 62

≥ 50-point reduction in total IBS-SSS score from baseline

Yes 149 (68.0%) 126 (59.2%) 275 (63.7%) 131 (64.2%) 106 (53.8%) 237 (59.1%)

No 70 (32.0%) 87 (40.8%) 157 (36.3%) 73 (35.8%) 91 (46.2%) 164 (40.9%)

Missing 13 18 31 28 34 62

≥ 30% reduction in abdominal pain severity from baseline (Item 1b)

Yes 114 (52.3%) 100 (46.9%) 214 (49.7%) 113 (55.7%) 84 (42.6%) 197 (49.3%)

Health Technology Assessment 2024 Vol. 28 No. 66

No 104 (47.7%) 113 (53.1%) 217 (50.3%) 90 (44.3%) 113 (57.4%) 203 (50.8%)

Missing 14 18 32 29 34 63

≥ 30% reduction in abdominal distention severity from baseline (Item 3b)

Yes 100 (45.7%) 86 (40.4%) 186 (43.1%) 95 (46.6%) 74 (37.6%) 169 (42.1%)

No 119 (54.3%) 127 (59.6%) 246 (56.9%) 109 (53.4%) 123 (62.4%) 232 (57.9%)

Missing 13 18 31 28 34 62
39
 40
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Clinical trial results

TABLE 16 Six-month total IBS-SSS score: linear regression model – parameter estimates in primary, complete case and per-protocol analysisa

Primary analysis (n = 463, multiple imputation) Complete case (n = 401) Per-protocol (n = 323, multiple imputationsb)

Parameter estimate Parameter estimate Parameter estimate

(95% CI) SE p-value (95% CI) SE p-value (95% CI) SE p-value

Intercept 57.71 (21.03 to 94.40) 18.71 0.002** 58.90 (21.05 to 96.74) 19.25 0.002** 48.38 (7.28 to 89.49) 20.97 0.021*

Treatment: amitriptyline (vs. placebo) −27.01 (−46.91 to −7.10) 10.15 0.008** −30.87 (−50.88 to −10.86) 10.18 0.003** −31.76 (−54.18 to −9.33) 11.44 0.006**

Baseline IBS-SSS score 0.51 (0.40 to 0.62) 0.06 < 0.001** 0.50 (0.39 to 0.62) 0.06 < 0.001** 0.52 (0.39 to 0.65) 0.07 < 0.001**

IBS subtype (vs. IBS-M or IBS-U) 0.240

IBS-C 24.68 (−3.50 to 52.87) 14.37 0.086 23.81 (−4.95 to 52.57) 14.63 0.104 31.33 (−0.92 to 63.58) 16.46 0.057

IBS-D 0.36 (−21.41 to 22.12) 11.10 0.974 1.75 (−20.23 to 23.73) 11.18 0.876 15.40 (−9.02 to 39.83) 12.46 0.216

Baseline HADS-D score 0.66 (−2.36 to 3.68) 1.54 0.669 0.71 (−2.28 to 3.71) 1.52 0.640 −0.44 (−3.91 to 3.04) 1.77 0.805

Recruitment hub (vs. Wessex) 0.081

West of England 7.87 (−13.52 to 29.25) 10.91 0.471 6.49 (−15.44 to 28.43) 11.16 0.561 2.52 (−21.46 to 26.50) 12.23 0.837

West Yorkshire −17.93 (−46.39 to 10.52) 14.50 0.217 −26.58 (−55.60 to 2.44) 14.76 0.072 −23.37 (−57.13 to 10.38) 17.22 0.175

* Indicates parameters significant at the 5% level.

** Indicates significance at the 1% level.
a Parameter estimates indicate the difference in total IBS-SSS score for each covariate compared with the reference value (listed after ‘vs.’ in the table for categorical covariates, and for
a unit increase for continuous covariates).
b n = 14 participants in the per-protocol population were missing data, six in the amitriptyline arm and six in the placebo arm. Multiple imputation was used as per the primary analysis.
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Key secondary end point: subjective global assessment of relief of irritable bowel
syndrome symptoms at 6 months

A total of 399 (86.2%) participants [204 (87.9%) on amitriptyline; 195(84.4%) on placebo] provided a
response to the SGA of relief of IBS symptoms during the past week at 6-month follow-up. Table 17
and Appendix 1, Figure 19 show a higher proportion of participants reported IBS symptoms as being
completely, considerably, or somewhat relieved in the amitriptyline arm compared with placebo (61.3%
vs. 45.1%, primary responder definition), with response rates of 35.8% versus 22.6% for complete or
considerable relief (sensitivity responder definition).

Amitriptyline was superior to placebo at 6 months in ITT analysis (Table 18), using logistic regression
adjusted for covariates with multiple imputation, with strong evidence of an increased odds of
response (IBS symptoms completely, considerably or somewhat relieved) on the SGA at 6 months with
amitriptyline compared with placebo and an OR of 1.78 (95% CI 1.19 to 2.66; p = 0.005). Results of
sensitivity analyses were consistent, albeit with larger estimated treatment effects, in both complete
case and sensitivity analysis using the alternative response (complete or considerable relief) definition
(Table 18).

There were no statistically significant covariate effects in primary analysis. As seen in the primary
outcome, there was weak evidence (p < 0.1) that participants with IBS-C had reduced odds of
responding compared with participants with IBS-M or IBS-U (OR 0.58, 95% CI 0.32 to 1.04; p = 0.066).
A similar effect was observed in complete case analysis but not in sensitivity analysis using the
alternative response (complete or considerable relief) definition. Although there was no evidence of a
difference across hub in primary analysis, there was good evidence of an increased odds of response
in participants in West Yorkshire compared with Wessex in sensitivity analysis using the alternative
response definition (OR 2.02, 95% CI 1.13 to 3.61; p = 0.017) and weak evidence in complete
case analysis.

Secondary analysis
The treatment effect estimated from secondary analysis using ordinal logistic regression, rather than
dichotomising responses, was consistent with the primary analysis (Table 18). The estimated odds of a
better response (ordinal regression models the odds of a better response, assuming proportional odds
between the cumulative odds of: complete relief vs. less than complete relief; at least considerable relief
vs. less than considerable relief; at least some relief vs. less than some relief; and at least unchanged
vs. worse symptoms) in the amitriptyline arm compared with the placebo arm was 1.72 (95% CI 1.20 to
2.46; p = 0.003).

Model checking
Graphical and statistical tests of the adequacy of the logistic regression and ordinal logistic regression
models were satisfactory (see Appendix 1, Figure 20 and Appendix 1, Key secondary end point).

Secondary end points

Global symptoms of irritable bowel syndrome (irritable bowel syndrome Severity

Scoring System) at 3 months
The total IBS-SSS score was available at 3 months for 432 (93.3%) participants [219 (94.4%) on
amitriptyline; 213 (92.2%) on placebo]. The mean 3-month total IBS-SSS scores were 173.0 (SD
106.6) and 194.6 (SD 107.5) in participants allocated to amitriptyline and placebo, respectively
(Table 15). A higher proportion of participants had remission of IBS symptoms on the IBS-SSS (IBS-SSS
score < 75 points) at 3 months in the amitriptyline arm compared with placebo (22.4% vs. 16.0%), and
a lower proportion of participants had severe IBS on the IBS-SSS (IBS-SSS score ≥ 300 points) in the
amitriptyline arm compared with placebo (12.3% vs. 21.1%).

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 41
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 42
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Clinical trial results

TABLE 17 Subjective global assessment of relief of IBS symptoms at 3 and 6 months

Month 3 Month 6

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463) Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

Relief of IBS symptoms

a

1-Completely relieved 7 (3.2%) 5 (2.3%) 12 (2.8%) 8 (3.9%) 4 (2.1%) 12 (3.0%)

2-Considerably relieved 67 (30.5%) 42 (19.7%) 109 (25.2%) 65 (31.9%) 40 (20.5%) 105 (26.3%)

3-Somewhat relieved 65 (29.5%) 58 (27.2%) 123 (28.4%) 52 (25.5%) 44 (22.6%) 96 (24.1%)

4-Unchanged 78 (35.5%) 97 (45.5%) 175 (40.4%) 73 (35.8%) 99 (50.8%) 172 (43.1%)

5-Worse 3 (1.4%) 11 (5.2%) 14 (3.2%) 6 (2.9%) 8 (4.1%) 14 (3.5%)

Missing 12 18 30 28 36 64

Responder (response 1–3)

Yes 139 (63.2%) 105 (49.3%) 244 (56.4%) 125 (61.3%) 88 (45.1%) 213 (53.4%)

Odds ratio (95% CI), p-valueb 1.70 (1.15 to 2.53), 1.78 (1.19 to 2.66), 0.005
0.008

Responder (response 1–2)

Yes 74 (33.6%) 47 (22.1%) 121 (27.9%) 73 (35.8%) 44 (22.6%) 117 (29.3%)

a Relief of IBS symptoms during the past week, in particular overall well-being and symptoms of abdominal discomfort and altered bowel habit.
b Odds ratio (amitriptyline vs. placebo) estimated using logistic regression adjusted for covariates and using multiple imputation of missing data.
 TABLE 18 Six-month SGA of relief of IBS symptoms: logistic and ordinal regression – parameter estimates in primary, sensitivity and secondary analysis
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
Sensitivity analysis Secondary analysis

Primary analysis (Responder 1–3 vs. Complete case (Responder 1–3 vs. Alternative responder definition
4–5), (n = 463) 4–5), (n = 399) (Responder 1–2 vs. 3–5), (n = 463) Ordinal regressiona, (n = 463)

Odds ratio Odds ratio Odds ratio Odds ratio

P. est. SE p-value (95% CI) P. est. SE p-value (95% CI) P. est. SE p-value (95% CI) P. est. SE p-value (95% CI)

Intercept −0.08 0.25 0.751 −0.02 0.25 0.945 −1.10 0.28 < 0.0001

1
 -Completely −3.74 0.36 <0.001
relieved

2
 -Considerably −1.10 0.24 <0.001
relieved

3
 -Somewhat −0.07 0.23 0.780
relieved

4-Unchanged 3.10 0.34 <0.001

Treatment: amitripty- 0.58 0.20 0.005** 1.78 (1.19 to 0.69 0.21 0.001** 2.00 (1.33 to 0.63 0.23 0.006** 1.88 (1.20 to 0.54 0.18 0.003** 1.72 (1.20
line (vs. placebo) 2.66) 3.00) 2.95) to 2.46)b

IBS subtype (vs. 0.188

IBS-M or IBS-U)

Health Technology Assessment 2024 Vol. 28 No. 66

IBS-C −0.54 0.30 0.066 0.58 (0.32 to −0.54 0.30 0.068 0.58 (0.33 to −0.26 0.33 0.421 0.77 (0.40 to −0.40 0.26 0.124 0.67 (0.40
1.04) 1.04) 1.46) to 1.12)

IBS-D −0.11 0.22 0.625 0.90 (0.58 to −0.13 0.23 0.569 0.88 (0.56 to −0.13 0.24 0.588 0.88 (0.54 to −0.13 0.20 0.501 0.88 (0.60
1.38) 1.37) 1.41) to 1.29)

Baseline HADS-D −0.02 0.03 0.518 0.98 (0.92 to −0.03 0.03 0.373 0.97 (0.92 to −0.06 0.03 0.102 0.95 (0.89 to −0.03 0.03 0.321 0.97 (0.92
score 1.04) 1.03) 1.01) to 1.03)

Recruitment hub 0.140

(vs. Wessex)

West of England −0.11 0.22 0.631 0.90 (0.59 to −0.07 0.22 0.767 0.94 (0.60 to −0.11 0.25 0.671 0.90 (0.55 to −0.09 0.20 0.665 0.92 (0.62
1.38) 1.45) 1.47) to 1.36)

West Yorkshire 0.44 0.28 0.115 1.55 (0.90 to 0.52 0.30 0.084 1.69 (0.93 to 0.70 0.30 0.017* 2.02 (1.13 to 0.49 0.26 0.057 1.63 (0.98
2.69) 3.07) 3.61) to 2.69)

* Indicates parameters significant at the 5% level.

** Indicates significance at the 1% level.
a Intercept (vs. 5-worse) in ordinal regression.
b Complete case analysis gives treatment effect of OR 1.93 [95% CI (1.33 to 2.79), p < 0.001].
43
 Clinical trial results

As per the 6-month outcome, amitriptyline was superior to placebo, with strong evidence (p < 0.05) of
a reduced total IBS-SSS score at 3 months with amitriptyline compared with placebo and an estimated
adjusted mean difference of −23.30 (95% CI −41.96 to −4.64; p = 0.014) (Tables 15 and 19). There were
no statistically significant covariate effects (see Appendix 1, Table 51). Similar results were obtained in the
complete case and per-protocol analysis, with increased treatment effects.

Subjective global assessment of relief of irritable bowel syndrome symptoms at 3

months
A total of 433 (93.5%) participants [220 (94.8%) on amitriptyline; 213 (92.2%) on placebo] provided
a response to SGA of relief of IBS symptoms during the past week at 3-month follow-up. A higher
proportion of participants reported IBS symptoms as being completely, considerably, or somewhat
relieved in the amitriptyline arm compared with placebo (63.2% vs. 49.3%), with response rates of 33.6%
vs. 22.1% for complete or considerable relief (see Table 17 and Appendix 1, Figure 19).

As per the 6-month outcome, amitriptyline was superior to placebo, with strong evidence (p < 0.05) of
an increased odds of response for SGA of relief of IBS symptoms at 3 months in the amitriptyline arm
compared with the placebo arm and an OR of 1.70 (95% CI 1.15 to 2.53; p = 0.008) (Tables 17 and 19).
Similar results were obtained in sensitivity and secondary analysis, with increased odds of response with
amitriptyline (Table 19).

There were no statistically significant covariate effects (see Appendix 1, Table 52). There was weak
evidence (p < 0.1) that participants with IBS-C had reduced odds of responding compared with
participants with IBS-M or IBS-U (OR 0.60, 95% CI 0.34 to 1.05; p = 0.071) in primary analysis and
complete case analysis but not in secondary analysis using ordinal regression or in sensitivity analysis
using the alternative responder definition.

Hospital Anxiety and Depression Scale-A scores at 3 and 6 months

An improvement in participants’ HADS-A scores was seen in both groups over time and compared with
baseline, with a mean score of 7.5 (SD 4.3) at baseline, 6.6 (SD 4.2) at 3 months, and 6.8 (SD 4.2) at
6 months (Table 20). There was no evidence of a treatment effect on HADS-A scores at 3 or 6 months in
primary or sensitivity analysis (see Tables 20 and 21, Appendix 1, Table 53 and Figure 21). Adjusted mean
differences indicated higher scores, but not statistically significantly different, in the amitriptyline arm

TABLE 19 Treatment effect estimates of 3-month total IBS-SSS score and SGA of relief of IBS symptoms secondary
outcomes: primary, secondary and sensitivity analysis

IBS-SSS Mean differencea (amitriptyline – placebo) p-value N

Primary analysis −23.30 (−41.96, −4.64) 0.014 463
Complete case −23.95 (−42.35, −5.56) 0.011 433
Per protocol −27.70 (−47.23, −8.17) 0.005 373
SGA of relief Odds ratio (amitriptyline vs. placebo)
b
p-value N
Primary analysis (Responder 1–3 vs. 4–5) 1.70 (1.15, 2.53) 0.008 463
Sensitivity analysis
Complete case (Responder 1–3 vs. 4–5) 1.81 (1.23, 2.67) 0.003 433
 lternative responder definition
A 1.81 (1.17, 2.79) 0.008 463
(Responder 1–2 vs. 3–5)
Secondary analysis
Ordinal regression 1.80 (1.26, 2.58) 0.001 463

a Mean difference (amitriptyline vs. placebo) estimated using linear regression adjusted for covariates and using multiple
imputation of missing data in primary analysis, and no imputation in complete case sensitivity analysis.
b Odds ratio (amitriptyline vs. placebo) estimated using logistic and ordinal regression (as appropriate) adjusted for
covariates and using multiple imputation of missing data in primary analysis, alternative responder sensitivity analysis
and ordinal secondary analysis, and no imputation in complete case sensitivity analysis.

44

NIHR Journals Library www.journalslibrary.nihr.ac.uk

 TABLE 20 Hospital Anxiety and Depression Scale-A, HADS-D, WSAS and PHQ-12 scores at baseline and 3 and 6 monthsa
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
Baseline Month 3 Month 6

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463)

HADS-A score

Mean (SD) 7.3 (4.3) 7.7 (4.3) 7.5 (4.3) 6.5 (4.4) 6.6 (4.0) 6.6 (4.2) 6.7 (4.4) 6.9 (4.0) 6.8 (4.2)

Median (range) 7.0 (0.0, 21.0) 7.0 (0.0, 20.0) 7.0 (0.0, 21.0) 6.0 (0.0, 21.0) 7.0 (0.0, 17.0) 6.0 (0.0, 21.0) 6.0 (0.0, 20.0) 7.0 (0.0, 16.0) 7.0 (0.0, 20.0)

Missing 0 0 0 12 19 31 29 38 67

Mean difference (95% 0.05 (−0.53 to 0.63), 0.08 (−0.49 to

CI), p-value 0.861 0.65), 0.775

Anxiety level

0–7 (Normal) 126 (54.3%) 119 (51.5%) 245 (52.9%) 138 (62.7%) 131 (61.8%) 269 (62.3%) 127 (62.6%) 112 (58.0%) 239 (60.4%)

8–10 (Mild anxiety) 55 (23.7%) 58 (25.1%) 113 (24.4%) 46 (20.9%) 46 (21.7%) 92 (21.3%) 41 (20.2%) 44 (22.8%) 85 (21.5%)

11–14 (moderate 39 (16.8%) 36 (15.6%) 75 (16.2%) 24 (10.9%) 26 (12.3%) 50 (11.6%) 22 (10.8%) 31 (16.1%) 53 (13.4%)
anxiety)

15–21 (severe anxiety) 12 (5.2%) 18 (7.8%) 30 (6.5%) 12 (5.5%) 9 (4.2%) 21 (4.9%) 13 (6.4%) 6 (3.1%) 19 (4.8%)

Health Technology Assessment 2024 Vol. 28 No. 66

HADS-D score

Mean (SD) 4.4 (3.6) 4.1 (3.2) 4.3 (3.4) 3.5 (3.3) 3.6 (3.2) 3.5 (3.3) 3.9 (3.6) 4.0 (3.5) 4.0 (3.6)

Median (range) 4.0 (0.0, 18.0) 4.0 (0.0, 15.0) 4.0 (0.0, 18.0) 3.0 (0.0, 15.0) 3.0 (0.0, 14.0) 3.0 (0.0, 15.0) 3.0 (0.0, 18.0) 3.0 (0.0, 16.0) 3.0 (0.0, 18.0)

Missing 0 0 0 12 19 31 30 38 68

Mean difference (95% −0.22 (−0.71 to 0.26), −0.20 (−0.75 to

CI), p-value 0.369 0.34), 0.462

Depression level

0–7 (normal) 195 (84.1%) 195 (84.4%) 390 (84.2%) 196 (89.1%) 187 (88.2%) 383 (88.7%) 175 (86.6%) 158 (81.9%) 333 (84.3%)

8–10 (mild depression) 23 (9.9%) 24 (10.4%) 47 (10.2%) 12 (5.5%) 15 (7.1%) 27 (6.3%) 14 (6.9%) 29 (15.0%) 43 (10.9%)

11–14 (moderate 11 (4.7%) 11 (4.8%) 22 (4.8%) 8 (3.6%) 10 (4.7%) 18 (4.2%) 10 (5.0%) 4 (2.1%) 14 (3.5%)
depression)

15–21 (severe 3 (1.3%) 1 (0.4%) 4 (0.9%) 4 (1.8%) 0 (0.0%) 4 (0.9%) 3 (1.5%) 2 (1.0%) 5 (1.3%)
depression)

continued
45
 46
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Clinical trial results

TABLE 20 HADS-A, HADS-D, WSAS and PHQ-12 scores at baseline and 3 and 6 Monthsa (continued)

Baseline Month 3 Month 6

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463)

WSAS scoreb

Mean (SD) 11.2 (8.2) 11.5 (7.6) 11.4 (7.9) 9.3 (7.6) 9.5 (6.3) 9.4 (7.0) 8.1 (7.6) 9.4 (7.8) 8.7 (7.7)

Median (range) 9.0 (0.0, 40.0) 11.0 (0.0, 40.0) 10.0 (0.0, 40.0) 8.0 (0.0, 39.0) 9.0 (0.0, 28.0) 8.0 (0.0, 39.0) 6.0 (0.0, 40.0) 8.0 (0.0, 39.0) 7.0 (0.0, 40.0)

Missing 8 12 20 22 33 55 37 47 84

Mean difference (95% −0.27 (−1.36 to 0.83), −1.04 (−2.30 to

CI), p-value 0.633 0.23), 0.108

PHQ-12 scorec

Mean (SD) 6.3 (3.5) 6.3 (3.6) 6.3 (3.5) 5.7 (3.4) 5.9 (3.2) 5.8 (3.3)

Median (range) 6.0 (0.0, 17.3) 6.0 (0.0, 18.0) 6.0 (0.0, 18.0) 5.2 (0.0, 19.0) 6.0 (0.0, 16.0) 5.5 (0.0, 19.0)

Missing 4 2 6 30 39 69

Mean difference (95% −0.04 (−0.58 to

CI), p-value 0.49), 0.877

a Mean difference (amitriptyline vs. placebo) estimated using linear regression adjusted for covariates and using multiple imputation of missing data.
b 92/458 (20.1%) at baseline, 98/431 (22.7%) at 3 months, 86/395 (21.8%) at 6 months and 48/224 (21.4%) at 12 months reported they were retired or chose not to have a job for
reasons unrelated to their IBS resulting in lower WSAS scores.
c Mean baseline PHQ-12 score was 5.3 (SD 3.3) in males and 6.8 (SD 3.5) in females with similar scores across arms. Mean 6-month score was 5.2 (SD 3.4) in males and 6.1 (SD 3.2) in
females, with slightly lower scores in the amitriptyline arm as per the total sample.
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 21 Treatment effect estimates of 3- and 6-month HADS, WSAS and PHQ-12 secondary outcomes: primary and
sensitivity analysisa

3 months 6 months

Mean difference Mean difference

(amitriptyline – placebo) p-value N (amitriptyline – placebo) p-value N

HADS-A

Primary analysis 0.05 (−0.53, 0.63) 0.861 463 0.08 (−0.49, 0.65) 0.775 463

Complete case 0.07 (−0.50, 0.64) 0.815 432 −0.07 (−0.66, 0.51) 0.808 396

HADS-D

Primary analysis −0.22 (−0.71, 0.26) 0.369 463 −0.20 (−0.75, 0.34) 0.462 463

Complete case −0.27 (−0.75, 0.21) 0.264 432 −0.37 (−0.89, 0.15) 0.161 395

WSAS

Primary analysis −0.27 (−1.36, 0.83) 0.633 463 −1.04 (−2.30, 0.23) 0.108 463

Complete case −0.38 (−1.48, 0.72) 0.499 398 −1.29 (−2.61, 0.02) 0.054 367

PHQ-12

Primary analysis −0.04 (−0.58, 0.49) 0.877 463

Complete case −0.22 (−0.73, 0.29) 0.400 392

a Mean difference (amitriptyline vs. placebo) estimated using linear regression adjusted for covariates and using multiple
imputation of missing data in primary analysis, and no imputation in complete case sensitivity analysis.

by 0.05 (95% CI −0.53 to 0.63; p = 0.861) at 3 months and 0.08 (95% CI −0.49 to 0.63; p = 0.775) at
6 months in primary analysis. However, the direction of effect was reversed in 6-month complete case
analysis, which indicated lower scores, again not statistically significantly different, in the amitriptyline
arm by −0.07 (95% CI −0.66 to 0.51; p = 0.808).

Hospital Anxiety and Depression Scale-D scores at 3 and 6 months

An improvement in participants’ HADS-D scores was seen in both groups over time and compared with
baseline, with a mean score of 4.3 (SD 3.4) at baseline, 3.5 (SD 3.3) at 3 months, and 4.0 (SD 3.6) at
6 months (Table 20). There was no evidence of a treatment effect on HADS-D scores at 3 or 6 months
in primary or sensitivity analysis (see Tables 20 and 21, Appendix 1, Table 54 and Figure 21). Scores
were lower, but not statistically significantly different, in the amitriptyline arm by −0.22 (95% CI −0.71
to 0.26; p = 0.369) at 3 months and −0.20 (95% CI −0.75 to 0.34; p = 0.462) at 6 months in primary
analysis, with similar but extenuated non-significant effects in complete case analysis.

Ability to work and participate in other activities (work and social adjustment scale
scores) at 3 and 6 months
An improvement in participants’ WSAS scores was seen in both groups compared with baseline, with
a mean score of 11.4 (SD 7.9) at baseline, 9.4 (SD 7.0) at 3 months, and 8.7 (SD 7.7) at 6 months
(Table 20). There was no evidence of a treatment effect on WSAS scores at either 3 or 6 months in
primary or sensitivity analysis (see Tables 20 and 21, Appendix 1, Table 55 and Figure 22). Scores were
lower, but not statistically significantly different, in the amitriptyline arm by −0.27 (95% CI −1.36 to
0.83; p = 0.369) at 3 months and −1.04 (95% CI −2.30 to 0.23; p = 0.108) at 6 months in primary
analysis, with extenuated non-significant effects in complete case analysis.

Irritable bowel syndrome-associated somatic symptom-reporting (patient health

questionnaire-12 scores) at 6 months
Again, an improvement in participants’ PHQ-12 scores was observed in both groups compared with
baseline, with a mean score of 6.3 (SD 3.5) at baseline and 5.8 (SD 3.3) at 6 months (Table 20). There was
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 47
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Clinical trial results

no evidence of a treatment effect on PHQ-12 scores at 6 months in primary or sensitivity analysis (see
Tables 20 and 21, Appendix 1, Table 56). Scores were lower, but not statistically significantly different, in
the amitriptyline arm by −0.04 (95% CI −0.58 to 0.49; p = 0.877) at 6 months in primary analysis, with
an extenuated non-significant effect in complete case analysis.

Participant-reported weekly adequate relief of irritable bowel syndrome symptoms

Summary statistics of available data for weekly adequate relief of IBS symptoms up to 26 weeks post
randomisation are presented in Table 22 and Appendix 1, Figure 23. In 443 (96%) participants reporting
data for at least 1 week, the mean number and proportion of weeks with adequate relief was 10.2 (SD
7.77) weeks and 48.3% (SD 33.6) of weeks in the amitriptyline arm, and 8.0 (7.76) weeks and 38.0% (SD
34.2%) of weeks in the placebo arm.

TABLE 22 Overall weekly relief summary

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

Number of weekly questions completed

Mean (SD) 19.9 (6.59) 19.5 (6.91) 19.7 (6.74)

Median (range) 23.0 (0.0, 25.0) 23.0 (0.0, 25.0) 23.0 (0.0, 25.0)

IQR 18.0–24.0 17.0–24.0 17.0–24.0

Number of weekly questions completed N (%)

0 week 10 (4.3%) 10 (4.3%) 20 (4.3%)

1–12 weeks 24 (10.3%) 32 (13.9%) 56 (12.1%)

13–18 weeks 28 (12.1%) 22 (9.5%) 50 (10.8%)

19–24 weeks 134 (57.8%) 135 (58.4%) 269 (58.1%)

25 weeks 36 (15.5%) 32 (13.9%) 68 (14.7%)

Number of weeks with adequate relief

Mean (SD) 10.2 (7.77) 8.0 (7.76) 9.1 (7.84)

Median (range) 11.0 (0.0, 25.0) 5.0 (0.0, 25.0) 8.0 (0.0, 25.0)

IQR 3.0–17.0 0.0–14.0 1.0–16.0

Missinga 10 10 20

% of weeks with adequate relief (of 25 weeks) b

Mean (SD) 41.0 (31.1) 32.0 (31.1) 36.5 (31.4)

Median (range) 44.0 (0.0, 100) 20.0 (0.0, 100) 32.0 (0.0, 100)

IQR 12.0–68.0 0.0–56.0 4.0–64.0

Relief for at least ≥ 13/25 weeksa

Yes 90 (40.5%) 67 (30.3%) 157 (35.4%)

No 132 (59.5%) 154 (69.7%) 286 (64.6%)

a Missing where no weekly relief responses provided. Not missing where at least 1 week provided.
b Missing weeks classed as no relief. As a proportion of completed weeks (rather than the total 25 weeks) gave a mean
% weeks with adequate relief of 48.3% (SD 33.6%) with amitriptyline and 38.0% (SD 34.2%) with placebo, with 123
(55.4%) participants in the amitriptyline arm and 93 (42.1%) participants in the placebo arm reporting relief for at least
≥ 50 of completed weeks.

48

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Amitriptyline was superior to placebo, with strong evidence of an increased likelihood of adequate relief
with amitriptyline compared with placebo and an overall OR of 1.56 (95% CI 1.20 to 2.03; p < 0.001)
across all weeks (see Figure 4, Appendix 1 and Table 57) in repeated-measures analysis. The odds of relief
were increased (OR > 1) with amitriptyline compared with placebo at all weeks, with good evidence
(p < 0.05) for 12 of 25 weeks (weeks 3, 6, 8–9, 11–16, 23, 24), weak evidence (p < 0.1) for 5 weeks
(weeks 10, 17–18, 22, 24), and the effect was not statistically significant for 8 weeks (weeks 1–2, 4–5,
7, 19–21).

(a) Estimated odds ratio of relief (amitriptyline vs. placebo)

4
3.75
3.5
3.25
3
2.75
2.5
Odds ratio

2.25 Band
Odds ratio
2
Odds ratio
1.75
1.5
1.25
1
0.75
0.5
0.25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Week
Overall odds ratio over time is 1.565 (95% CI 1.205 to 2.031)

(b) Predicted proportion of participants with relief each week

1.0

0.9
Proportion of participants with relief

0.8

0.7

0.6 Amitriptyline
Placebo
Amitriptyline
0.5
Placebo
0.4

0.3

0.2

0.1

0.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Week
Estimated for a ‘reference’ patient with mixed or unclassified stool type, HADS-D score of 4 from Southampton

FIGURE 4 Estimated treatment effect on weekly relief: logistic regression: 443 participants with at least one weekly
response. The band in panel (a) represents the estimated OR and 95% CI overall across all weeks.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 49
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Clinical trial results

Acceptability of trial medication at 6 months

Acceptability of trial medication, based on participants’ 6-month response and prior treatment
discontinuation, was available for a total of 424 (91.6%) participants (with similar rates by arm). A higher
proportion of participants reported acceptability in the amitriptyline arm compared with placebo (57.8%
vs. 46.9%) (Table 23). Conversely a lower proportion of participants reported they did not find the
medication acceptable (14.2% vs. 21.6%) and had discontinued treatment before 6 months (21.8% vs.
27.7%) in the amitriptyline arm compared with placebo. Three participants did not start treatment and
18 participants were lost to follow-up before 6 months; these participants were classed as not finding
the medication acceptable in the primary analysis.

Data were missing at random for 39 participants due to an administrative error in one site where
acceptability was not asked of participants recruited to the reduced 6-month follow-up. As all
participants’ missing data were from the same hub and still on treatment at 6 months, multiple
imputation was performed by allocation within participants on treatment from the missing hub only.

Amitriptyline was superior to placebo in the ITT analysis, with good evidence (p < 0.05) of an increased
odds of acceptability with amitriptyline compared with placebo and an adjusted OR of 1.60 (95% CI 1.08
to 2.35; p = 0.018) (Table 24). Similar results were obtained in sensitivity analysis.

TABLE 23 Acceptability of trial medication at 6 months

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463)

Does the participant find the medication acceptable at 6 months?

Yes 122 (57.8%) 100 (46.9%) 222 (52.4%)

Answered ‘Yes’ to the acceptability at 6 months 122 (57.8%) 99 (46.5%) 221 (52.1%)

 cceptability missing but remained on

A 0 (0.0%) 1 (0.5%) 1 (0.2%)
treatment > 6 months

No 89 (42.2%) 113 (53.1%) 202 (47.6%)

Answered ‘No’ to the acceptability at 6 months 30 (14.2%) 46 (21.6%) 76 (17.9%)

Discontinued treatment < 6 months 46 (21.8%) 59 (27.7%) 105 (24.8%)

Did not start trial medication 2 (0.9%) 1 (0.5%) 3 (0.7%)

Lost to follow-up ≤ 6 months 11 (5.2%) 6 (2.7%) 17 (4.0%)

Othera 0 (0.0%) 1(0.5%) 1 (0.2%)

Missing 21 18 39

a Participant still on treatment but not contactable at 6-month call.

TABLE 24 Six-month acceptability: logistic regression models – adjusted OR of acceptability in primary and
sensitivity analysis

Adjusted OR (amitriptyline vs. placebo) p-value N

Primary analysis 1.60 (1.08, 2.35) 0.018 463

Sensitivity analysis

Complete case 1.59 (1.08, 2.35) 0.020 424

Complete case (excluding n = 21 lost to 1.74 (1.16, 2.60) 0.007 403

follow-up or did not start treatment)

50

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Adherence
The number and proportion of participants with the highest level of adherence to therapy (taking
one tablet every day or nearly every day) decreased over time and was reported by 416/447 (93.1%)
participants at week 3, 364/452 (80.5%) at 3 months and 309/460 (67.2%) at 6 months (Table 25,
excluding participants on treatment with missing data). The majority of the remaining participants were
classified according to the lowest level of adherence, having discontinued or not started trial medication,
or having been lost to follow-up. Rates were similar across trial arms at 3 weeks and 3 months. However,
by 6 months a higher proportion of participants in the amitriptyline arm reported the highest level of
adherence compared with placebo, 163 (70.3%) versus 146 (64.0%) participants respectively; and a slightly
lower proportion of participants had the lowest level of adherence (25.4% vs. 28.9%, respectively).

Due to the large proportion of participants reporting the highest and lowest levels of adherence (every
day or nearly every day through to discontinued, did not start treatment, or lost to follow-up), and
small numbers of participants reporting adherence within these extremes, the proportional hazards
assumption was not met, and only descriptive analyses are presented.

Tolerability at 3 and 6 months

An overall summary of 21 possible treatment-emergent AEs self-reported by participants at 3 and
6 months, as captured by the ASEC for participants in the safety population and still on treatment, are
reported in Table 26.

Of 385 (83.7%) and 318 (69.1%) participants on treatment (or treatment status unknown) and
completing the ASEC at 3 and 6 months respectively, most participants reported at least one mild to
severe side effect (> 95%), and 310 (80.5%) participants at 3 months and 440 (75.5%) participants at
6 months reported at least one moderate to severe side effect, with similar rates across trial arms. A
slightly greater proportion of participants reported at least one severe side effect with amitriptyline
compared with placebo at both 3 [58 (30.1%) amitriptyline; 46 (24.0%) placebo] and 6 months
[45 (27.1%) amitriptyline; 37 (24.3%) placebo]. The total ASEC score (ranging from 0 to 63), which
quantifies both the number and severity of symptoms reported, was slightly higher in the amitriptyline
arm compared with placebo, with an overall mean of 9.2 (SD 5.88) at 3 months and 9.0 (SD 6.13)
at 6 months. In adjusted complete case analysis, there was a statistically significant increase in the
total ASEC score with amitriptyline compared with placebo at 3 months (1.39, 95% CI 0.29 to 2.50;
p = 0.013) but not at 6 months (0.26, 95% CI −0.98 to 1.51; p = 0.681). Further details of the types of
AEs can be found in Figures 5 and 6, Appendix 1, Table 58. Adverse events occurred more frequently
in the amitriptyline arm compared with placebo, related mainly to its known anticholinergic effects,
including dry mouth, drowsiness, blurred vision, and problems with urination. However, rates fell
generally between 3 and 6 months and few (< 5%) were severe, except for constipation and diarrhoea
(< 10%).

Further 12-month treatment delivery and secondary end points

Treatment receipt, adherence, dosage and titration

Of the 291 participants in the 12-month ITT population, 128 (44%) completed 12 months treatment;
67 (45.6%) of 147 allocated to amitriptyline, and 61 (42.4%) of 144 allocated to placebo (Table 27).
The median time from randomisation to the end-of-trial treatment was 10.4 months, with a longer
duration observed in the amitriptyline arm compared with placebo (median 11.5 vs. 8.3 months).
Similar to all randomised participants, just under a quarter (24.4%) discontinued trial medication before
6 months; 33 (22.4%) of 147 allocated to amitriptyline, and 38 (26.4%) of 144 allocated to placebo.
At the 6-month follow-up time point, participants were given the choice to continue treatment or
not; of 208 participants who were still on treatment at 6 months, 91 (85.8%) of 106 and 81 (79.4%) of
102 participants chose to continue treatment in the amitriptyline and placebo arms, respectively, with
15 (14.2%) participants in the amitriptyline arm and 21 (20.6%) in the placebo arm choosing to stop

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 51
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 52
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Clinical trial results

TABLE 25 Adherence to therapy end-point summary – up to 6 monthsa

Week 3 Month 3 Month 6

Has the participant taken at Amitriptyline Placebo Amitriptyline Placebo Total Amitriptyline Placebo Total
least one tablet daily (n = 232) (n = 231) Total (n = 463) (n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463)

Every or nearly every day 212 (93.8%) 204 (92.3%) 416 (93.1%) 185 (79.7%) 179 (81.4%) 364 (80.5%) 163 (70.3%) 146 (64.0%) 309 (67.2%)

≥ half the days 4 (1.8%) 6 (2.7%) 10 (2.2%) 8 (3.4%) 4 (1.8%) 12 (2.7%) 9 (3.9%) 9 (3.9%) 18 (3.9%)

< half of the days 0 (0.0%) 2 (0.9%) 2 (0.4%) 0 (0.0%) 2 (0.9%) 2 (0.4%) 1 (0.4%) 4 (1.8%) 5 (1.1%)

None or nearly none of the days 0 (0.0%) 1 (0.5%) 1 (0.2%) 1 (0.4%) 0 (0.0%) 1 (0.2%) 0 (0.0%) 3 (1.3%) 3 (0.7%)

Discontinued, did not start 10 (4.4%) 8 (3.6%) 18 (4.0%) 38 (16.4%) 35 (15.9%) 73 (16.2%) 59 (25.4%) 66 (28.9%) 125 (27.2%)
treatment, or lost to follow-up

Missing 6 10 16 0 11 11 0 3 3

a This table differs from Table 13 as participants who had discontinued trial medication, did not start trial medication, or were lost-to follow-up are included as the lowest level of
adherence to trial medication.
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
TABLE 26 Participant-reported tolerability on the ASEC at 3 and 6 monthsa

Month 3 Month 6

Amitriptyline (n = 231) Placebo (n = 229) Total (n = 460) Amitriptyline (n = 232) Placebo (n = 228) Total (n = 460)

N on treatment/unknownb 194 (84.0%) 196 (85.6%) 390 (84.8%) 174 (75.0%) 164 (71.9%) 338 (73.5%)

N on treatment and ASEC completed 193 (83.5%) 192 (83.8%) 385 (83.7%) 166 (71.6%) 152 (66.7%) 318 (69.1%)

Total score

Mean (SD) 9.9 (6.00) 8.4 (5.67) 9.2 (5.88) 9.3 (6.07) 8.7 (6.19) 9.0 (6.13)

Median (range) 9.0 (0.0, 32.0) 8.0 (0.0, 33.0) 8.0 (0.0, 33.0) 8.0 (0.0, 30.0) 7.0 (0.0, 28.0) 7.0 (0.0, 30.0)

IQR 6.0–13.0 4.0–12.0 5.0–12.0 5.0–12.0 4.0–12.0 4.0–12.0

N 193 192 385 166 152 318

Mean differencec (95% CI) 1.39 (0.29 to 2.50) 0.26 (−0.98 to

1.51)

p-value 0.013 0.681

ASEC symptoms

Health Technology Assessment 2024 Vol. 28 No. 66

No symptoms 5 (2.6%) 7 (3.6%) 12 (3.1%) 2 (1.2%) 3 (2.0%) 5 (1.6%)

≥ 1 mild-severe symptom 188 (97.4%) 185 (96.4%) 373 (96.9%) 164 (98.8%) 149 (98.0%) 313 (98.4%)

≥ 1 moderate – severe symptom 156 (80.8%) 154 (80.2%) 310 (80.5%) 127 (76.5%) 113 (74.3%) 240 (75.5%)

≥ 1 severe symptom 58 (30.1%) 46 (24.0%) 104 (27.0%) 45 (27.1%) 37 (24.3%) 82 (25.8%)

N mild-severe symptoms

Mean (SD) 6.8 (3.74) 5.8 (3.56) 6.3 (3.68) 6.5 (3.57) 6.0 (3.65) 6.3 (3.61)

Median (range) 6.0 (0.0, 19.0) 5.0 (0.0, 19.0) 6.0 (0.0, 19.0) 6.0 (0.0, 17.0) 5.0 (0.0, 17.0) 6.0 (0.0, 17.0)

IQR 4.0–8.0 3.0–8.0 4.0–8.0 4.0–9.0 3.0–8.0 3.0–9.0

N 193 192 385 166 152 318

a ASEC = Antidepressant Side-Effect Checklist (scores range 0–63, lower scores are better). Data presented according to the 3- and 6-month safety population.
b Includes participants lost to researcher telephone follow-up where treatment status could not be determined.
c Mean difference (amitriptyline vs. placebo) estimated using linear regression adjusted for covariates (complete case, missing data not imputed).
53
 54
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Clinical trial results

1. Dry mouth 122 (63.2%) 12 52 58 52 29 6 87 (45.3%)

2. Drowsiness 128 (66.3%) 6 55 67 45 21 67 (34.9%)

3. Insomnia (difficulty sleeping) 78 (40.4%) 3 26 49 53 46 6 108 (56.3%)

4. Blurred vision 29 (15.0%) 6 22 20 4 24 (12.5%)

5. Headache 74 (38.3%) 3 24 47 58 23 4 85 (44.3%)

6. Constipation 110 (57.0%) 18 42 50 37 37 15 89 (46.4%)

7. Diarrhoea 117 (60.6%) 19 38 60 50 63 13 126 (65.6%)

8. Increased appetite 54 (28.0%) 3 24 27 23 18 3 44 (22.9%)

9. Decreased appetite 34 (17.6%) 3 6 25 21 7 28 (14.6%) ASEC score

10. Nausea or vomiting 35 (18.1%) 4 5 26 21 3 2 26 (13.5%) 1 = mild
Symptom

2 = moderate
11. Problems with urination 31 (16.1%) 4 8 19 12 3 2 23 (12.0%)
3 = severe
12. Problems with sexual function 29 (15.0%) 5 6 18 12 3 2 23 (12.0%)

13. Palpitations 56 (29.0%) 11 44 31 6 37 (19.3%)

14. Feeling light-headed on standing 73 (37.8%) 14 58 49 14 63 (32.8%)

15. Feeling like the room is spinning 29 (15.0%) 3 25 16 8 24 (12.5%)

16. Sweating 71 (36.8%) 2 25 44 36 19 5 60 (31.3%)

17. Increased body temperature 56 (29.0%) 20 36 30 17 48 (25.0%)

18. Tremor 17 (8.8%) 16 10 2 13 (6.8%)

19. Disorientation 24 (12.4%) 3 20 8 8 (4.2%)

20. Yawning 67 (34.7%) 3 19 45 53 15 68 (35.4%)

21. Weight gain Amitriptyline arm (N = 193) 72 (37.3%) 6 27 39 38 16 5 59 (30.7%) Placebo arm (N = 192)

150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
Number of participants experienced mild/moderate/severe symptoms

N is number of participants who were on treatment and completed the ASEC questionnaire in each arm and percentages calculated out of N

FIGURE 5 Participant-reported tolerability on the ASEC at 3 months (safety population for participants on treatment).
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
1. Dry mouth 90 (54.2%) 5 30 55 35 18 3 56 (36.8%)

2. Drowsiness 88 (53.0%) 8 28 52 34 17 52 (34.2%)

3. Insomnia (difficulty sleeping) 77 (46.4%) 4 32 41 47 40 9 96 (63.2%)

4. Blurred vision 28 (16.9%) 5 22 11 2 14 (9.2%)

5. Headache 78 (47.0%) 6 20 52 54 23 3 80 (52.6%)

6. Constipation 93 (56.0%) 11 30 52 40 31 7 78 (51.3%)

7. Diarrhoea 98 (59.0%) 9 40 49 52 42 9 103 (67.8%)

8. Increased appetite 45 (27.1%) 2 18 25 24 8 2 34 (22.4%)

9. Decreased appetite 17 (10.2%) 6 10 12 10 22 (14.5%)

10. Nausea or vomiting 26 (15.7%) 6 19 23 2 26 (17.1%) ASEC score

Symptom

1 = mild
11. Problems with urination 36 (21.7%) 3 10 23 15 5 20 (13.2%)
2 = moderate
12. Problems with sexual function 24 (14.5%) 5 10 9 6 6 4 16 (10.5%) 3 = severe

13. Palpitations 41 (24.7%) 3 7 31 28 7 3 38 (25.0%)

14. Feeling light-headed on standing 69 (41.6%) 5 63 42 10 2 54 (35.5%)

Health Technology Assessment 2024 Vol. 28 No. 66

15. Feeling like the room is spinning 20 (12.0%) 19 14 4 19 (12.5%)

16. Sweating 54 (32.5%) 3 21 30 28 16 5 49 (32.2%)

17. Increased body temperature 35 (21.1%) 2 7 26 26 9 36 (23.7%)

18. Tremor 13 (7.8%) 11 8 2 11 (7.2%)

19. Disorientation 13 (7.8%) 2 10 9 10 (6.6%)

20. Yawning 63 (38.0%) 3 17 43 38 10 2 50 (32.9%)

21. Weight gain Amitriptyline arm (N = 166) 73 (44.0%) 6 22 45 23 22 4 49 (32.2%) Placebo arm (N = 152)

120 110 100 90 80 70 60 50 40 30 20 10 0 10 20 30 40 50 60 70 80 90 100 110 120

Number of participants experienced mild/moderate/severe symptoms

N is number of participants who were on treatment and completed the ASEC questionnaire in each arm and percentages calculated out of N

FIGURE 6 Participant-reported tolerability on the ASEC at 6 months (safety population for participants on treatment).
55
 Clinical trial results

treatment. Of the 172 participants who continued treatment beyond 6 months, a further 16 (16.5%)
of 91 in the amitriptyline arm, and 11 (13.6%) of 81 in the placebo arm discontinued treatment before
the full 12-month period. The most common reason for treatment discontinuation after 6 months was
lack of benefit in 3 (2.0%) of 147 and 6 (4.2%) of 144 participants allocated to amitriptyline and placebo
respectively, followed by non-specific or personal choice in 5 (3.4%) amitriptyline and 3 (2.1%) placebo
participants, and side effects in 4 (2.7%) amitriptyline and 1 (< 1%) placebo participant.

TABLE 27 Twelve-month treatment receipt

Amitriptyline (n = 147) Placebo (n = 144) Total (n = 291)

Participant continued trial medication beyond 6 months

Yes 91 (61.9%) 81 (56.3%) 172 (59.1%)

No 15 (10.2%) 21 (14.6%) 36 (12.4%)

Did not start/discontinued/lost to follow-up < 6 months 41 (27.9%) 42 (29.2%) 83 (28.5%)

Completed 12 months’ treatment

Yes 67 (45.6%) 61 (42.4%) 128 (44.0%)

Discontinued trial medication 64 (43.5%) 70 (48.6%) 134 (46.0%)

Before 6 months 33 (22.4%) 38 (26.4%) 71 (24.4%)

Chose not to continue treatment beyond 6 months 15 (10.2%) 21 (14.6%) 36 (12.4%)

After 6 months 16 (10.9%) 11 (7.6%) 27 (9.3%)

Lost to follow-up 14 (9.5%) 12 (8.3%) 26 (8.9%)

Before 6 months 6 (4.1%) 3 (2.1%) 9 (3.1%)

After 6 months 8 (5.4%) 9 (6.3%) 17 (5.8%)

Not started treatment 2 (1.4%) 1 (0.7%) 3 (1.0%)

Reason for discontinuation > 6 months

Lack of benefit 3 (2.0%) 6 (4.2%) 9 (3.1%)

Non-specific or personal choice 5 (3.4%) 3 (2.1%) 8 (2.7%)

Side effecta 4 (2.7%) 1 (0.7%) 5 (1.7%)

Administrative error 1 (0.7%) 1 (0.7%) 2 (0.7%)

SAE or SAR 1 (0.7%) 0 (0.0%) 1 (0.3%)

Safety (including allergic reactions to IMP) 1 (0.7%) 0 (0.0%) 1 (0.3%)

Other reasonb 1 (0.7%) 0 (0.0%) 1 (0.3%)

Months to last dose (from randomisation)

Mean (SD) 8.5 (4.16) 8.0 (4.14) 8.3 (4.15)

Median (range) 11.5 (0.2, 14.0) 8.3 (0.2, 12.2) 10.4 (0.2, 14.0)

Missing 16 13 29

n 131 131 262

Months to last dose (from randomisation)

≤ 1 month 9 (6.9%) 8 (6.1%) 17 (6.5%)

≤ 2 months 5 (3.8%) 6 (4.6%) 11 (4.2%)

56

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 27 Twelve-month treatment receipt (continued)

Amitriptyline (n = 147) Placebo (n = 144) Total (n = 291)

≤ 3 months 8 (6.1%) 6 (4.6%) 14 (5.3%)

≤ 4 months 1 (0.8%) 7 (5.3%) 8 (3.1%)

≤ 5 months 7 (5.3%) 6 (4.6%) 13 (5.0%)

≤ 6 months 15 (11.5%) 25 (19.1%) 40 (15.3%)

≤ 7 months 6 (4.6%) 5 (3.8%) 11 (4.2%)

≤ 8 months 2 (1.5%) 2 (1.5%) 4 (1.5%)

≤ 9 months 6 (4.6%) 2 (1.5%) 8 (3.1%)

≤ 10 months 2 (1.5%) 1 (0.8%) 3 (1.1%)

≤ 11 months 2 (1.5%) 1 (0.8%) 3 (1.1%)

≤ 12 months 37 (28.2%) 43 (32.8%) 80 (30.5%)

> 12 months 31 (23.7%) 19 (14.5%) 50 (19.1%)

Total 131 (100%) 131 (100%) 262 (100%)

a Side effects included constipation for one participant in each arm, drowsiness, micturition difficulties, and night sweats
for one participant each in the amitriptyline arm.
b Other reason for discontinuation: moved out of area and changed GP practice.

Over 90% of participants on trial medication continued to report the highest level of treatment
adherence beyond 6 months at month 9 and 12, taking at least one tablet daily ‘every day or nearly
every day’ with similar rates across trial arms (Table 28 and Appendix 1, Figure 14). In the amitriptyline
arm, the proportion of participants on each dose remained similar at 6, 9 and 12 months, with just under
half of participants on the highest dose of 30 mg (see Table 28 and Appendix 1, Figure 15). In contrast,
in the placebo arm, 55 (57.9%) participants were on 30 mg at 6 months, 47 (69.1%) at 9 months and
40 (65.6%) at 12 months. Relatively few participants changed their dose after 6 months. A total of 17
(12.4%) changed dose between 6 and 9 months; rates were similar across arms. However, participants
in the amitriptyline arm were more likely to lower their dose, whereas those on placebo were more likely
to increase their dose. Only 12 (9.5%) participants remaining on treatment changed dose between and 9
and 12 months post randomisation, with a higher number in the amitriptyline arm.

No participants in the 12-month ITT population who discontinued treatment after 6 months reported
taking amitriptyline off-trial after treatment discontinuation.

Secondary end points

Global symptoms of irritable bowel syndrome (irritable bowel syndrome Severity

Scoring System) at 12 months
The total IBS-SSS score was available at 12 months for 225 (77.3%) participants [118 (80.3%) on
amitriptyline; 107 (74.3%) on placebo]. The mean 12-month total IBS-SSS scores were 160.7 (SD 113.7)
and 176.7 (SD 107.2) in participants allocated to amitriptyline and placebo, respectively (Table 29). A
higher proportion of participants had remission of IBS symptoms on the IBS-SSS (IBS-SSS score < 75
points) at 12 months in the amitriptyline arm compared with placebo (28.8% vs. 20.6%). There was weak
evidence (p < 0.10) of a reduced total IBS-SSS score with amitriptyline compared with placebo, with
an estimated adjusted mean difference of −22.59 (95% CI −49.35 to 4.16; p = 0.098) in 12-month ITT
analysis. Similar results were obtained in sensitivity analysis (Table 30 and Appendix 1, Table 59).

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 57
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 58
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Clinical trial results

TABLE 28 Treatment adherence, dose and replenishment for participants on treatment: at 6, 9 and 12 months

Month 6 Month 9 Month 12

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 106) (n = 102) (n = 208)a (n = 74) (n = 69) (n = 143) (n = 67) (n = 61) (n = 128)

Has the participant taken at least one tablet daily?

Every or nearly 99 (93.4%) 89 (89.9%) 188 (91.7%) 65 (92.9%) 64 (94.1%) 129 (93.5%) 61 (91.0%) 57 (93.4%) 118 (92.2%)
every day

≥ half the days 6 (5.7%) 5 (5.1%) 11 (5.4%) 5 (7.1%) 4 (5.9%) 9 (6.5%) 5 (7.5%) 3 (4.9%) 8 (6.3%)

 half of the
< 1 (0.9%) 3 (3.0%) 4 (2.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (1.6%) 1 (0.8%)
days

 one or nearly
N 0 (0.0%) 2 (2.0%) 2 (1.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (1.5%) 0 (0.0%) 1 (0.8%)
none of the
days

Missing 0 (0.0%) 3 3 4 1 5

Current dose of trial medication

1 × 10 mg 4 (3.8%) 4 (4.2%) 8 (4.0%) 2 (2.8%) 1 (1.5%) 3 (2.2%) 2 (3.0%) 1 (1.6%) 3 (2.3%)

every other day

1 × 10 mg daily 26 (24.5%) 10 (10.5%) 36 (17.9%) 17 (23.9%) 5 (7.4%) 22 (15.8%) 15 (22.4%) 5 (8.2%) 20 (15.6%)

2 × 10 mg daily 30 (28.3%) 26 (27.4%) 56 (27.9%) 20 (28.2%) 15 (22.1%) 35 (25.2%) 20 (29.9%) 15 (24.6%) 35 (27.3%)

3 × 10 mg daily 46 (43.4%) 55 (57.9%) 101 (50.2%) 32 (45.1%) 47 (69.1%) 79 (56.8%) 30 (44.8%) 40 (65.6%) 70 (54.7%)

Missing 0 7 7 3 1 4

Dose modification since last follow-up call

Yes 24 (22.9%) 13 (14.1%) 37 (18.8%) 9 (12.7%) 8 (12.1%) 17 (12.4%) 8 (12.3%) 4 (6.6%) 12 (9.5%)

Higher dose 13 (54.2%) 7 (53.8%) 20 (54.1%) 3 (33.3%) 7 (87.5%) 10 (58.8%) 5 (62.5%) 2 (50.0%) 7 (58.3%)

Lower dose 11 (45.8%) 6 (46.2%) 17 (45.9%) 6 (66.7%) 1 (12.5%) 7 (41.2%) 3 (37.5%) 2 (50.0%) 5 (41.7%)

No 81 (77.1%) 79 (85.9%) 160 (81.2%) 62 (87.3%) 58 (87.9%) 120 (87.6%) 57 (87.7%) 57 (93.4%) 114 (90.5%)

Missing 1 10 11 4 15 19 1 13 14

a Number of participants who consented to 12 months follow-up and still on treatment at the time of 6 months follow-up call.
 TABLE 29 Baseline and 12-month IBS-SSS and SGA outcomes
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
Baseline Month 12

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463) Amitriptyline (n = 147) Placebo (n = 144) Total (n = 291)

Total IBS-SSS score

Mean (SD) 273.4 (90.53) 273.4 (90.53) 273.4 (90.53) 160.7 (113.69) 176.7 (107.23) 168.3 (110.71)

Median (range) 280.0 (60, 480) 280.0 (60, 480) 280.0 (60, 480) 145.0 (0, 440) 170.0 (0, 390) 160.0 (0, 440)

IQR 210.0–330.0 210.0–330.0 210.0–330.0 70.0–240.0 90.0–270.0 80.0–250.0

Missing 0 0 0 29 37 66

Mean difference (95% CI), p-value a

−22.59 (−49.35 to 4.16), 0.098

IBS-SSS level

< 75 (remission) 3 (1.3%) 3 (1.3%) 3 (1.3%) 34 (28.8%) 22 (20.6%) 56 (24.9%)

75–174 (mild) 37 (15.9%) 37 (15.9%) 37 (15.9%) 36 (30.5%) 34 (31.8%) 70 (31.1%)

175–299 (moderate) 98 (42.2%) 98 (42.2%) 98 (42.2%) 34 (28.8%) 33 (30.8%) 67 (29.8%)

300–500 (severe) 94 (40.5%) 94 (40.5%) 94 (40.5%) 14 (11.9%) 18 (16.8%) 32 (14.2%)

Health Technology Assessment 2024 Vol. 28 No. 66

Difference in IBS-SSS score from baseline

Mean (SD) −109.2 (117.58) −80.4 (98.86) −95.5 (109.80)

Median (range) −110.0 (−370.0, 280.0) −70.0 (−410.0, 110.0) −90.0 (−410.0, 280.0)

IQR −200.0 to −30.0 −140.0 to 0.0 −170.0 to −20.0

≥ 50 point reduction in total IBS-SSS score from baseline

Yes 84 (71.2%) 65 (60.7%) 149 (66.2%)

No 34 (28.8%) 42 (39.3%) 76 (33.8%)

≥ 30% reduction in abdominal pain severity from baseline (item 1b)

Yes 65 (55.1%) 49 (45.8%) 114 (50.7%)

No 53 (44.9%) 58 (54.2%) 111 (49.3%)

continued
59
 60
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Clinical trial results

TABLE 29 Baseline and 12-month IBS-SSS and SGA outcomes (continued)

Baseline Month 12

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463) Amitriptyline (n = 147) Placebo (n = 144) Total (n = 291)

≥ 30% reduction in abdominal distention severity from baseline (item 3b)

Yes 58 (49.2%) 45 (42.1%) 103 (45.8%)

No 60 (50.8%) 62 (57.9%) 122 (54.2%)

SGA of relief of IBS symptomsb

1 – Completely relieved 6 (5.1%) 4 (3.7%) 10 (4.4%)

2 – Considerably relieved 36 (30.5%) 26 (24.3%) 62 (27.6%)

3 – Somewhat relieved 28 (23.7%) 20 (18.7%) 48 (21.3%)

4 – Unchanged 47 (39.8%) 53 (49.5%) 100 (44.4%)

5 – Worse 1 (0.8%) 4 (3.7%) 5 (2.2%)

Missing 29 37 66

Responder (response 1–3)

Yes 70 (59.3%) 50 (46.7%) 120 (53.3%)

Odds ratio (95% CI), p-valuec 1.58 (0.94 to 2.64), 0.083

a Mean difference (amitriptyline vs. placebo) estimated using linear regression adjusted for covariates and using multiple imputation of missing data.
b Relief of IBS symptoms during the past week, in particular overall well-being and symptoms of abdominal discomfort and altered bowel habit.
c Odds ratio (amitriptyline vs. placebo) estimated using logistic regression adjusted for covariates and using multiple imputation of missing data.
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 30 Twelve-month treatment effect estimates in primary and sensitivity analysis of irritable bowel syndrome
Severity Scoring System, subjective global assessment, Hospital Anxiety and Depression Scale and Work and Social
Adjustment Scale secondary outcomes

Mean difference (amitriptyline – placebo) p-value N

IBS-SSS
Primary analysis −22.59 (−49.35, 4.16) 0.098 291
Complete case −24.34 (−50.49, 1.81) 0.068 225
Anxiety (HADS-A)
Primary analysis −0.38 (−1.22, 0.47) 0.385 291
Complete case −0.18 (−1.02, 0.65) 0.662 224
Depression (HADS-D)
Primary analysis −0.88 (−1.71, −0.06) 0.036 291
Complete case −0.85 (−1.63, −0.07) 0.032 224
Ability to work and participate in activities (WSAS)
Primary analysis −2.14 (−3.80, −0.49) 0.011 291
Complete case −1.70 (−3.24, −0.15) 0.031 202
Odds ratio (amitriptyline – placebo) p-value N
SGA
Primary analysis 1.58 (0.94, 2.64) 0.083 291
Complete case 1.73 (1.01, 2.95) 0.046 225

Subjective global assessment of relief of irritable bowel syndrome symptoms at

12 months
A higher proportion of participants reported IBS symptoms as being completely, considerably, or
somewhat relieved in the amitriptyline arm compared with placebo, with response rates of 59.3% versus
46.7% at 12 months (Table 29). There was weak evidence of an increased odds of response for SGA of
relief of IBS symptoms with amitriptyline compared with placebo, with an OR of 1.58 (95% CI 0.94 to
2.64; p = 0.083, n = 291) (Table 30) in 12-month ITT analysis and good evidence in sensitivity analysis
(OR 1.73, 95% CI 1.01 to 2.95; p = 0.046, n = 225) (Table 30 and Appendix 1, Table 60).

Hospital Anxiety and Depression Scale-A scores at 12 months

As observed at 3 and 6 months, an improvement in participants’ HADS-A scores was seen in both
groups at 12 months compared with baseline. However, although scores were lower in the amitriptyline
compared with the placebo arm there was no evidence of an effect between treatment arms in
12-month ITT analysis (−0.38, 95% CI −1.22 to 0.47; p = 0.385) or sensitivity analysis (Tables 30 and 31).

Hospital Anxiety and Depression Scale-D scores at 12 months

Despite no evidence of a treatment effect on participants’ HADS-D at 3 months and 6 months,
amitriptyline was found to be superior to placebo at 12 months, with a significant difference in mean
HADS-D scores between arms in 12-month ITT analysis (−0.88, 95% CI −1.71 to −0.06; p = 0.036)
(Table 31). Similar results were obtained in sensitivity analysis (Table 30).

Ability to work and participate in other activities (Work and Social Adjustment
Scale scores) at 12 months
Amitriptyline was also superior to placebo in terms of ability to work or participate in other activities
at 12 months according to the WSAS, with a significant difference in mean WSAS score between arms
in 12-month ITT analysis (−2.14, 95% CI; −3.80 to −0.49; p = 0.011) (Table 31). Similar results were
obtained in sensitivity analysis (Table 30).
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 61
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 62
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Clinical trial results

TABLE 31 Baseline and 12-month HADS and WSAS outcomesa

Baseline Month 12

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463) Amitriptyline (n = 147) Placebo (n = 144) Total (n = 291)

HADS-A score

Mean (SD) 7.3 (4.3) 7.7 (4.3) 7.5 (4.3) 6.9 (4.6) 7.3 (4.1) 7.1 (4.4)

Median (range) 7.0 (0.0, 21.0) 7.0 (0.0, 20.0) 7.0 (0.0, 21.0) 6.0 (0.0, 21.0) 7.0 (0.0, 18.0) 7.0 (0.0, 21.0)

Missing 0 0 0 30 37 67

Mean difference (95% CI), p-value −0.38 (−1.22 to 0.47), 0.385

Anxiety level

0–7 (normal) 126 (54.3%) 119 (51.5%) 245 (52.9%) 72 (61.5%) 56 (52.3%) 128 (57.1%)

8–10 (mild anxiety) 55 (23.7%) 58 (25.1%) 113 (24.4%) 21 (17.9%) 28 (26.2%) 49 (21.9%)

11–14 (moderate anxiety) 39 (16.8%) 36 (15.6%) 75 (16.2%) 14 (12.0%) 19 (17.8%) 33 (14.7%)

15–21 (severe anxiety) 12 (5.2%) 18 (7.8%) 30 (6.5%) 10 (8.5%) 4 (3.7%) 14 (6.3%)

HADS-D score

Mean (SD) 4.4 (3.6) 4.1 (3.2) 4.3 (3.4) 3.8 (3.5) 4.6 (3.8) 4.2 (3.7)

Median (range) 4.0 (0.0, 18.0) 4.0 (0.0, 15.0) 4.0 (0.0, 18.0) 3.0 (0.0, 18.0) 4.0 (0.0, 17.0) 3.0 (0.0, 18.0)

Missing 0 0 0 30 37 67

Mean difference (95% CI), p-value −0.88 (−1.71 to −0.06), 0.036

Depression level

0–7 (Normal) 195 (84.1%) 195 (84.4%) 390 (84.2%) 103 (88.0%) 87 (81.3%) 190 (84.8%)

8–10 (Mild depression) 23 (9.9%) 24 (10.4%) 47 (10.2%) 6 (5.1%) 13 (12.1%) 19 (8.5%)

11–14 (Moderate depression) 11 (4.7%) 11 (4.8%) 22 (4.8%) 6 (5.1%) 5 (4.7%) 11 (4.9%)

15–21 (Severe depression) 3 (1.3%) 1 (0.4%) 4 (0.9%) 2 (1.7%) 2 (1.9%) 4 (1.8%)

attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
Baseline Month 12

Amitriptyline (n = 232) Placebo (n = 231) Total (n = 463) Amitriptyline (n = 147) Placebo (n = 144) Total (n = 291)

WSAS scoreb

Mean (SD) 11.2 (8.2) 11.5 (7.6) 11.4 (7.9) 7.9 (7.0) 10.1 (7.2) 8.9 (7.2)

Median (range) 9.0 (0.0, 40.0) 11.0 (0.0, 40.0) 10.0 (0.0, 40.0) 6.0 (0.0, 34.0) 10.0 (0.0, 35.0) 7.0 (0.0, 35.0)

Missing 8 12 20 37 41 78

Health Technology Assessment 2024 Vol. 28 No. 66

Mean difference (95% CI), p-value −2.14 (−3.80 to −0.49), 0.011

a Mean difference (amitriptyline vs. placebo) estimated using linear regression adjusted for covariates and using multiple imputation of missing data.
b 92/458 (20.1%) at baseline, 98/431 (22.7%) at 3 months, 86/395 (21.8%) at 6 months and 48/224 (21.4%) at 12 months reported they were retired or chose not to have a job for
reasons unrelated to their IBS, resulting in lower WSAS scores.
63
 Clinical trial results

Tolerability at 12 months
Of 128 (44.4%) participants on treatment (or treatment status unknown) and completing the ASEC at
12 months, most participants reported at least one mild to severe side effect (95.7%) at 12 months, with
similar rates across trial arms (Table 32). Although the total ASEC score was higher with amitriptyline
compared with placebo, there was no evidence of an effect between treatment arms in 12-month safety
analysis (−1.04, 95% CI −3.32 to 1.24; p = 0.368, n = 117). Further details of the types of AEs can be
found in Figure 7 and Appendix 1, Table 58.

Exploratory analysis

50-point reduction in irritable bowel syndrome Severity Scoring System at 3 and 6 months
Compared with baseline, there was a mean reduction in the total IBS-SSS score of 99.8 (SD 107.7) and
76.1 (SD 107.1) in participants allocated to amitriptyline and placebo, respectively at 3 months, and a
mean reduction of 99.2 (SD 112.9) and 68.9 (SD 109.3) at 6 months. A higher proportion of participants
had a reduction of ≥ 50 points in the amitriptyline arm compared with placebo at both 3 (68.0%
vs. 59.2%) and 6 months (64.2% vs. 53.8%) (Table 15). Adjusted analysis found weak evidence of an
association with treatment, such that participants in the amitriptyline arm were 1.49 times (95% CI 0.97
to 2.28; p = 0.068) more likely to have a ≥ 50-point reduction than those in the placebo arm at 3 months
and 1.48 times (95% CI 0.97 to 2.2; p = 0.068) more likely at 6 months (see Appendix 1, Table 61).

Thirty per cent reduction in 6-month irritable bowel syndrome Severity Scoring
System abdominal pain and distention
A higher proportion of participants reported a ≥ 30% reduction on individual abdominal pain and
abdominal distention severity items (items 1b and 3b) in the amitriptyline arm compared with placebo at
both 3 and 6 months (Table 15). With amitriptyline versus placebo, 52.3% versus 46.9% at 3 months, and
55.7% versus 42.6% at 6 months reported a ≥ 30% reduction in abdominal pain, and 45.7% versus 40.4%
at 3 months and 46.6% versus 37.6% at 6 months reported a ≥ 30% reduction in abdominal distention.

Adjusted analysis found good evidence of an increased likelihood of a ≥ 30% reduction in abdominal
pain severity with amitriptyline compared with placebo (OR 1.66, 95% CI 1.12 to 2.46; p = 0.012) at
6 months, but no evidence of a statistically significant effect at 3 months. There was no evidence of an
effect on abdominal distention at 3 or 6 months. Full analysis model effects are presented in Appendix 1,
Tables 62 and 63.

Moderator analysis of 6-month irritable bowel syndrome Severity Scoring System score
Moderator analysis found no evidence of a moderating effect of recruitment hub, IBS subtype, baseline
IBS-SSS, HADS-A scores or HADS-D scores on the primary 6-month treatment effect on the primary
outcome (Table 33 and Appendix 1, Table 64). Although no statistically significant interaction effects were
observed, larger treatment effects, in favour of amitriptyline, were observed in participants:

• from West Yorkshire compared with Wessex and West of England (see Appendix 1, Figures 24 and 25)
• with IBS-C or IBS-D compared with those with IBS-M or IBS-U (see Appendix 1, Figures 26 and 27)
• with higher baseline IBS-SSS scores (see Appendix 1, Figure 28)
• with lower HADS-A scores (see Appendix 1, Figure 29)
• with higher baseline HADS-D scores (see Appendix 1, Figure 30).

Moderator analysis of 6-month subjective global assessment of relief of irritable bowel

syndrome symptoms
Moderator analysis found no evidence of a moderating effect of IBS subtype on the 6-month treatment
effect on the key secondary outcome SGA of relief of IBS symptoms (Tables 34 and 35). Although no
statistically significant interaction was observed, larger treatment effects, in favour of amitriptyline,
were observed in participants with IBS-D compared with those with IBS-C and IBS-M or IBS-U (see
Appendix 1, Figures 31 and 32).

64

NIHR Journals Library www.journalslibrary.nihr.ac.uk

attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
1. Dry mouth 33 (54.1%) 1 12 20 16 6 2 24 (42.9%)

2. Drowsiness 28 (45.9%) 2 9 17 17 7 1 25 (44.6%)

3. Insomnia (difficulty sleeping) 25 (41.0%) 1 11 13 17 14 4 35 (62.5%)

4. Blurred vision 9 (14.8%) 1 8 10 10 (17.9%)

5. Headache 21 (34.4%) 9 12 15 5 3 23 (41.1%)

6. Constipation 30 (49.2%) 4 10 16 18 10 3 31 (55.4%)

7. Diarrhoea 37 (60.7%) 5 8 24 18 18 4 40 (71.4%)

8. Increased appetite 21 (34.4%) 2 8 11 7 7 1 15 (26.8%)

9. Decreased appetite 5 (8.2%) 1 4 6 6 (10.7%)

ASEC score
10. Nausea or vomiting 9 (14.8%) 11 7 5 1 6 (10.7%) 1 = mild
Symptom

2 = moderate
11. Problems with urination 11 (18.0%) 1 6 4 7 3 10 (17.9%) 3 = severe

12. Problems with sexual function 9 (14.8%) 4 5 4 3 1 8 (14.3%)

13. Palpitations 14 (23.0%) 2 12 12 6 18 (32.1%)

Health Technology Assessment 2024 Vol. 28 No. 66

14. Feeling light-headed on standing 22 (36.1%) 5 17 14 4 4 22 (39.3%)

15. Feeling like the room is spinning 6 (9.8%) 6 3 21 6 (10.7%)

16. Sweating 19 (31.1%) 1 4 14 16 6 1 23 (41.1%)

17. Increased body temperature 13 (21.3%) 4 9 6 4 10 (17.9%)

18. Tremor 8 (13.1%) 1 7 2 2 (3.6%)

19. Disorientation 2 (3.3%) 2 3 1 4 (7.1%)

20. Yawning 18 (29.5%) 4 14 18 4 3 25 (44.6%)

21. Weight gain Amitriptyline arm (N = 61) 28 (45.9%) 5 6 17 18 5 2 25 (44.6%) Placebo arm (N = 56)

60 50 40 30 20 10 0 10 20 30 40 50 60
Number of participants experienced mild/moderate/severe symptoms

FIGURE 7 Participant-reported tolerability on the ASEC at 12 months (safety population for participants on treatment).
65
 Clinical trial results

TABLE 32 Participant-reported tolerability on the ASEC at 12 monthsa

Amitriptyline (n = 146) Placebo (n = 142) Total (n = 288)

Number on treatmentb 67 (45.9%) 61 (43.0%) 128 (44.4%)

Number on treatment and 61 (41.8%) 56 (39.4%) 117 (40.6%)

completed the ASEC

Total score

Mean (SD) 8.6 (5.65) 9.6 (7.41) 9.0 (6.54)

Median (range) 9.0 (0.0, 24.0) 8.0 (0.0, 36.0) 8.0 (0.0, 36.0)

IQR 4.0–12.0 4.0–13.0 4.0–13.0

N 61 56 117

 ean difference (95% CI),

M −1.04 (−3.32 to 1.24), 0.368
p-valuec

ASEC symptoms

No symptoms 2 (3.3%) 3 (5.4%) 5 (4.3%)

≥ 1 mild-severe symptom 59 (96.7%) 53 (94.6%) 112 (95.7%)

 1 moderate – severe
≥ 41 (67.2%) 40 (71.4%) 81 (69.2%)
symptom

≥ 1 severe symptom 16 (26.2%) 15 (26.8%) 31 (26.5%)

Total number of mild-severe symptoms

Mean (SD) 6.0 (3.56) 6.6 (4.15) 6.3 (3.85)

Median (range) 6.0 (0.0, 15.0) 6.5 (0.0, 19.0) 6.0 (0.0, 19.0)

N 61 56 117

a ASEC = Antidepressant Side-Effect Checklist (scores range 0–63, lower scores are better). Data presented according to
the 12-month safety population.
b Includes participants lost to researcher telephone follow-up where treatment status could not be determined.
c Mean difference (amitriptyline vs. placebo) estimated using linear regression adjusted for covariates (complete case,
missing data not imputed).

Safety

Serious adverse events and reactions

A total of 10 SAEs and SARs were reported for 10 (2.2%) of 460 participants in the safety population;
6 (2.6%) and 4 (1.7%) participants in the amitriptyline and placebo safety populations, respectively
(Table 36). SAEs were reported for five participants, four in the amitriptyline arm and one in the placebo
arm. Five SARs (expected and suspected to be related to the trial medication) were reported for a
further five participants, two in the amitriptyline arm and three in the placebo arm.

Other safety events

No SUSARs, deaths, or pregnancies occurred during the trial.

66

NIHR Journals Library www.journalslibrary.nihr.ac.uk

attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
TABLE 33 Moderator analyses of the treatment effect on the 6-month total IBS-SSS score

Primary analysis Complete case

Mean difference (amitriptyline Interaction estimate Interaction, p- Mean difference (amitriptyline Interaction estimate Interaction,
Moderator vs. placebo) (95% CI) value vs. placebo) (95% CI) p-value

Recruitment hub 0.457

Wessex −21.95 (−52.77, 8.86) −26.89 (−57.69, 3.90)

West Yorkshire −48.13 (−96.39, 0.13) −26.17 (−84.71 to 32.36) 0.380 −59.13 (−108.14, −10.11) −42.4 (−83.3 to −1.62)

West of England −22.28 (−53.52, 8.97) −0.32 (−44.29 to 43.64) 0.989 −23.40 (−54.88, 8.08) 8.08 (−22.8 to 39.0)

IBS subtype 0.246

IBS-M or IBS-U −7.98 (−38.13, 22.16) −12.07 (−42.31, 18.17)

IBS-C −50.24 (−99.14, −1.34) −42.26 (−99.56 to 15.04) 0.148 −54.20 (−103.53, −4.88) −42.13 (−100.14 to 15.88) 0.154

IBS-D −38.66 (−69.70, −7.61) −30.67 (−74.15 to 12.81) 0.167 −41.89 (−73.73, −10.04) 29.82 (−73.72 to 14.09) 0.183

Baseline IBS-SSS score −0.14 (−0.37 to 0.09) 0.236 −0.14 (−0.369 to 0.091) 0.235

Lower quartile = 210 −18.23 (−41.77, 5.31)

Health Technology Assessment 2024 Vol. 28 No. 66

Median = 280 −28.01 (−48.11, −7.91)

Upper quartile = 330 −35.00 (−59.87, −10.13)

Baseline HADS-A score 1.86 (−2.91 to 6.64) 0.444 2.43 (−2.30 to 7.16) 0.313

Lower quartile = 4.0 −31.64 (−57.06, −6.22)

Median = 7.0 −26.05 (−45.89, −6.21)

Upper −20.45 (−43.95, 3.04)

quartile = 10.0

Baseline HADS-D score −1.69 (−7.68 to 4.31) 0.581 −0.86 (−6.84 to 5.11) 0.776

Lower quartile = 1.0 −21.51 (−48.74, 5.71)

Median = 4.0 −26.57 (−46.47, −6.68)

Upper quartile = 7.0 −31.63 (−58.06, −5.21)

67
 Clinical trial results

TABLE 34 Moderator analyses of the treatment effect on the 6-month SGA of relief of IBS symptoms by IBS subtype

Primary analysis Complete case analysis

Subgroup effect, OR Subgroup effect, OR

(95% CI) (amitriptyline Interaction, Interaction, (95% CI) (amitriptyline vs. Interaction,
vs. placebo) OR (95% CI) p-value placebo) p-value

IBS subtype (amitriptyline vs. placebo)

IBS-M or IBS-U 1.50 (0.83 to 2.73) 1.71 (0.93 to 3.14) 0.604

IBS-C 1.56 (0.58 to 4.20) 1.04 (0.33 0.947 1.65 (0.62 to 4.43) 0.959
to 3.28)

IBS-D 2.27 (1.21 to 4.26) 1.51 (0.64 0.347 2.59 (1.35 to 4.95) 0.360
to 3.60)

Blinding

Emergency unblinding
No emergency unblinding requests were made throughout the trial.

Treatment allocation exit survey

A treatment allocation exit survey was undertaken by the research nurse or clinical study officer, asking
participants which treatment they thought they had received, at the participants’ 6-month treatment
call or at the point of treatment discontinuation for participants who discontinued treatment before
6 months. Of the participants who completed 6 months treatment, 129 (74.6%) in the amitriptyline
arm and 91 (56.5%) in the placebo arm guessed their allocation correctly (Table 37). A slightly higher
proportion of participants who discontinued treatment before 6 months guessed their allocation
correctly [27 (78.1%) amitriptyline; 30 (63.8%) placebo]. On a scale of 0 (not at all certain) to 10
(completely sure), participant certainty about their choice was a mean of 7.3 (SD 1.79) at 6 months and
7.5 (SD 2.22) at treatment discontinuation.

Eighty participants (46.2%) in the amitriptyline arm made their report of which treatment they thought
they had received at 6 months because the treatment worked, whereas 33 participants (19.1%) made
their choice because the treatment did not work and 61 (35.3%) because they had a side effect. Of
the participants in the placebo arm at 6 months, 58 (36.0%) made their report of which treatment they
thought they had received because it worked, whereas 70 (43.5%) made their report because it did
not work and 22 (13.7%) because they had a side effect. Of participants who discontinued treatment
before 6 months, a higher proportion of participants made their report because they had a side effect
[20 (62.5%) amitriptyline; 10 (21.3%) placebo] and a lower proportion of participants in both arms made
their report because the treatment worked [7 (21.9%) amitriptyline; 4 (8.5%) placebo].

End-of-trial participation unblinding

A total of 346 (74.7%) participants were provided with their treatment allocation following their trial
participation, 176 (75.9%) participants in the amitriptyline arm and 170 (73.6%) in the placebo arm
(Table 38). Most [312 (90.2%)] participants requested that their GP be provided with their treatment
allocation directly via the research team, and the remainder [34 (9.8%)] requested their GP not be
provided with their allocation. The primary method by which participants could request their treatment
allocation was via completion of, and response within, the final follow-up questionnaire, the method
used by 314 (90.8%) of participants, while the remainder [32 (9.2%)] requested their allocation via
correspondence with the trial researcher at their final 12-month researcher treatment follow-up call
prior to implementation of the allocation request within participants’ final questionnaire pack. Only 10

68

NIHR Journals Library www.journalslibrary.nihr.ac.uk

attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
TABLE 35 Six-month SGA of relief of IBS symptoms score by IBS subtype

IBS-C IBS-D IBS-M or IBS-U

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 40) (n = 37) (n = 77) (n = 92) (n = 89) (n = 181) (n = 100) (n = 105) (n = 205)

SGA of relief of IBS symptoms

 – Completely
1 1 (2.8%) 3 (9.7%) 4 (6.0%) 3 (3.7%) 0 (0.0%) 3 (1.9%) 4 (4.6%) 1 (1.1%) 5 (2.9%)
relieved

 –
2 11 (30.6%) 3 (9.7%) 14 (20.9%) 27 (33.3%) 14 (18.4%) 41 (26.1%) 27 (31.0%) 23 (26.1%) 50 (28.6%)
Considerably
relieved

 – Somewhat
3 6 (16.7%) 6 (19.4%) 12 (17.9%) 22 (27.2%) 18 (23.7%) 40 (25.5%) 24 (27.6%) 20 (22.7%) 44 (25.1%)
relieved

Health Technology Assessment 2024 Vol. 28 No. 66

4 – Unchanged 17 (47.2%) 17 (54.8%) 34 (50.7%) 28 (34.6%) 40 (52.6%) 68 (43.3%) 28 (32.2%) 42 (47.7%) 70 (40.0%)

5 – Worse 1 (2.8%) 2 (6.5%) 3 (4.5%) 1 (1.2%) 4 (5.3%) 5 (3.2%) 4 (4.6%) 2 (2.3%) 6 (3.4%)

Missing 4 6 10 11 13 24 13 17 30

Responder (score 1–3)

Yes 18 (50.0%) 12 (38.7%) 30 (44.8%) 52 (64.2%) 32 (42.1%) 84 (53.5%) 55 (63.2%) 44 (50.0%) 99 (56.6%)

No 18 (50.0%) 19 (61.3%) 37 (55.2%) 29 (35.8%) 44 (57.9%) 73 (46.5%) 32 (36.8%) 44 (50.0%) 76 (43.4%)

Missing 4 6 10 11 13 24 13 17 30
69
 Clinical trial results

TABLE 36 Serious AE summary

Amitriptyline (n = 231) Placebo (n = 229) Total (n = 460)

Total participants with a SAE or SAR 6 (2.6%) 4 (1.7%) 10 (2.2%)

Total number of SAE/SARS 6 4 10

Total participants with a SAE 4 (1.7%) 1 (0.4 %) 5 (1.1 %)

Total number of SAEs 4 1 5

Total participants with a SAR 2 (0.9%) 3 (1.3%) 5 (1.1 %)

Total number of SARs 2 3 5

Out of all SAEs and SARs 6 (100%) 4 (100%) 10 (100%)

MedDRA/body system codea

Cardiac disorders 0 (0.0%) 2 (50.0%) 2 (20.0%)

Gastrointestinal disorders 1 (16.7%) 0 (0.0%) 1 (10.0%)

Injury, poisoning and procedural complications 1 (16.7%) 0 (0.0%) 1 (10.0%)

Psychiatric disorders 2 (33.3%) 1 (25.0%) 3 (30.0%)

Renal and urinary disorders 2 (33.3%) 1 (25.0%) 3 (30.0%)

Seriousness criteria (not mutually exclusive)b

Life-threatening 1 (16.7%) 1 (20.0%) 2 (18.2%)

Required or prolonged hospitalisation 3 (50.0%) 3 (60.0%) 6 (54.5%)

Other important medical event 2 (33.3%) 1 (20.0%) 3 (27.3%)

Outcome

Recovered 6 (100.0%) 3 (75.0%) 9 (90.0%)

Recovered with sequelae 0 (0.0%) 1 (25.0%) 1 (10.0%)

Days from randomisation to onset

Mean (SD) 142.3 (63.31) 94.3 (89.96) 123.1 (74.44)

Median (range) 124.5 (83.0, 231.0) 76.0 (20.0, 205.0) 118.0 (20.0, 231.0)

IQR 85.0–206.0 21.0–167.5 83.0–205.0

Days from onset to recovery

Mean (SD) 52.2 (75.72) 12.5 (17.75) 36.3 (60.91)

Median (range) 15.0 (3.0, 194.0) 4.5 (2.0, 39.0) 10.0 (2.0, 194.0)

IQR 3.0–83.0 2.5–22.5 3.0–39.0

Action taken

None 3 (50.0%) 1 (25.0%) 4 (40.0%)

Treatment delayed 2 (33.3%) 0 (0.0%) 2 (20.0%)

Treatment stopped 1 (16.7%) 3 (75.0%) 4 (40.0%)

Most recent dose

10 mg 1 (16.7%) 2 (50.0%) 3 (30.0%)

20 mg 2 (33.3%) 0 (0.0%) 2 (20.0%)

30 mg 3 (50.0%) 2 (50.0%) 5 (50.0%)

70

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 36 Serious adverse event summary (continued)

Amitriptyline (n = 231) Placebo (n = 229) Total (n = 460)

Other causes

Yesc 3 (50.0%) 2 (50.0%) 5 (50.0%)

No 3 (50.0%) 2 (50.0%) 5 (50.0%)

a SAR: two cardiac disorders, one gastrointestinal disorder, two psychiatric disorders (one placebo and one amitriptyline).
SAE: one injury, poisoning and procedural complications, one psychiatric disorder (amitriptyline), three renal and
urinary disorders.
b Seriousness criteria are not mutually exclusive; one participant who experienced a cardiac disorder had required/
prolonged hospitalisation and it was life-threatening.
c Other causes (amitriptyline): one participant thought IBS had played a role (reported as a SAE), two missing other
causes (both SARs). Other causes (placebo, both reported as a SAR): poor social support and work stress for one
participant, concomitant medications, losartan (an antihypertensive) for another participant.

TABLE 37 Exit survey of which treatment participants thought they received

Month 6 Discontinued treatment before month 6

Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 173) (n = 165) (n = 338) (n = 46) (n = 59) (n = 105)

Treatment participant thought they received

Active drug 129 (74.6%) 70 (43.5%) 199 (59.6%) 25 (78.1%) 17 (36.2%) 42 (53.2%)

Placebo 44 (25.4%) 91 (56.5%) 135 (40.4%) 7 (21.9%) 30 (63.8%) 37 (46.8%)

Missing 0 4 4 14 12 26

How certain was the participant (0 = Not at all, to 10 = Completely sure)

Mean (SD) 7.4 (1.81) 7.2 (1.77) 7.3 (1.79) 7.6 (2.39) 7.4 (2.11) 7.5 (2.22)

Median (range) 8.0 (1.0, 10.0) 7.0 (0.0, 10.0) 7.0 (0.0, 10.0) 8.0 (0.0, 10.0) 8.0 (2.0, 10.0) 8.0 (0.0,
10.0)

Missing 0 3 3 14 13 27

Reason for treatment allocation prediction a

Treatment 80 (46.2%) 58 (36.0%) 138 (41.3%) 7 (21.9%) 4 (8.5%) 11 (13.9%)

worked

 reatment didn’t
T 33 (19.1%) 70 (43.5%) 103 (30.8%) 7 (21.9%) 25 (53.2%) 32 (40.5%)
work

 articipant had a
P 61 (35.3%) 22 (13.7%) 83 (24.9%) 20 (62.5%) 10 (21.3%) 30 (38.0%)
side effect

Just a guess 17 (9.8%) 19 (11.8%) 36 (10.8%) 2 (6.3%) 7 (14.9%) 9 (11.4%)

 articipant had
P 9 (5.2%) 16 (9.9%) 25 (7.5%) 0 (0.0%) 3 (6.4%) 3 (3.8%)
no side effects

 ppearance or
A 5 (2.9%) 1 (0.6%) 6 (1.8%) 1 (3.1%) 1 (2.1%) 2 (2.5%)
taste of the tablet

 aken amitripty-
T 2 (1.2%) 4 (2.5%) 6 (1.8%) 2 (6.3%) 5 (10.6%) 7 (8.9%)
line previously

Otherb 1 (0.6%) 0 (0.0%) 1 (0.3%) 0 0 0

a Not mutually exclusive.

b Other reason: noticed a change in symptoms but not an improvement.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 71
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Clinical trial results

TABLE 38 End-of-trial participation unblinding summary

Amitriptyline (n = 232) (%) Placebo (n = 231) (%) Total (n = 463) (%)

Participant unblinded following trial participation

Yes 176 (75.9) 170 (73.6) 346 (74.7)

No 56 (24.1) 61 (26.4) 117 (25.3)

Of unblinded participants

Did the participant want to unblind their GP?

Yes 158 (89.8) 154 (90.6) 312 (90.2)

No 18 (10.2) 16 (9.4) 34 (9.8)

Unblinding requested via

Final questionnaire 163 (92.6) 151 (88.8) 314 (90.8)

Researcher correspondence/e-mail 13 (7.4) 19 (11.2) 32 (9.2)

Participant contacted the qualitative team to discuss their allocation?

Yesa 1 (0.6) 3 (1.8) 4 (1.2)

No 175 (99.4) 167 (98.2) 342 (98.8)

Total unblinded 176 (100) 170 (100) 346 (100)

Reason participants not unblinded

Did not complete final questionnaire 34 (60.7) 50 (82.0) 84 (71.8)

Did not respond to unblinding question in 8 (14.3) 1 (1.6) 9 (7.7)

final questionnaire

Ended trial before final questionnaire included 12 (21.4) 4 (6.6) 16 (13.7)

unblinding request/lost to follow-upb

Requested NOT to be unblinded in final 2 (3.6) 6 (9.8) 8 (6.8)

questionnaire

Total not unblinded 56 (100) 61 (100) 117 (100)

a Qualitative team resolved queries for all participants where contact was made.
b Before the final questionnaire included the unblinding request, participants were able to request their treatment
allocation at their final 12-month researcher follow-up call. For these 16 participants, the 12-month call did not take
place as they had all discontinued trial medication before 12 months.

(2.9%) participants were provided with their allocation via post, where participants’ preference was post
or where a valid e-mail address was not provided, with the remainder via e-mail.

Of the 117 (25.3%) participants who were not provided their treatment allocation following trial
participation, in most cases this was because participants had not completed their final questionnaire
or had not responded to the allocation request within their completed final questionnaire. Only 8
(6.8%) participants requested not to be provided with their allocation, and the remaining 16 (13.7%)
participants had completed their final questionnaire prior to implementation of the allocation request
within the final questionnaire pack and had also discontinued trial medication before the final 12-month
researcher treatment follow-up call.

Research nurse/clinical study officer unblinding

The research nurse or clinical study officer was inadvertently unblinded for four participants after their
final follow-up and treatment allocation had been revealed: in three cases by the participant, and in one
case by a GP research nurse to clarify participant details.

72

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Chapter 4 Nested qualitative study

Aims

The overarching aim was to explore participants’ and GPs’ experiences of treatments and participating
in the ATLANTIS trial. The objectives were to identify factors that facilitate or impede prescribing
and uptake of low-dose amitriptyline in IBS, to identify participants’ and GPs’ perspectives on the
broader impact of the trial, and to explore psychosocial and contextual factors that might shape
wider use of amitriptyline for IBS. The purpose was to use these in-depth findings to support the
interpretation of trial outcomes and to inform future efforts to promote wider use of amitriptyline for
IBS where appropriate.

Methods

Design
The qualitative study was nested within the main ATLANTIS trial. The qualitative team comprised
a female research fellow with experience of conducting qualitative research about IBS, a female
health psychologist, two female GPs, a male psychology student and a female psychology student.
Semistructured telephone interviews were conducted with a sub-sample of trial participants and GPs
involved in the trial. Reflexive thematic analysis,41,42 incorporating techniques from grounded theory,43
was used to analyse the qualitative data. Data collection and initial analyses proceeded iteratively: that
is, coding started after the first few interviews were conducted and informed subsequent interviews.
Although analysis was primarily inductive, that is, driven by the data, the common-sense model of illness
perception44 and normalisation process theory45 informed the development of the interview topic guides
and were consulted during the analysis to aid our interpretation of data around experiences of IBS and
treatments and wider implementation of amitriptyline for IBS in primary care. Qualitative findings were
related to the findings of the main trial by comparing themes across participant groups and triangulating
the themes against key quantitative findings.

Ethical considerations
As part of the main trial consent procedures, all participants could consent to be contacted about taking
part in optional semi-structured telephone interviews at two time points: 6 months and 12 months
post randomisation. All trial participants who had consented to be contacted about the qualitative
study and had reached their 6-month post-randomisation time point were sent a qualitative study
invitation pack by the ATLANTIS trial team (CTRU). The qualitative study invitation pack comprised
a covering letter, participant information sheet and qualitative interview consent form (see Report
Supplementary Material 1). Trial participants were invited to express their interest in taking part in an
interview by e-mailing a completed written consent form to the qualitative researcher (EJT). All GPs from
participating practices (excluding those GPs also on the trial management team) were invited directly
by the qualitative researcher to take part in one telephone interview. A list of general practices and PI
contact details (name and e-mail addresses) was sent to the qualitative researcher by the ATLANTIS trial
team (CTRU) via a secure messaging system. GPs were contacted about the qualitative study following
their participant recruitment period (approximately 5 months after they completed their mail-out). Each
PI was e-mailed a copy of the GP participant information and consent form and asked to forward this
information to all GPs involved in participant recruitment for ATLANTIS at their practice (see Report
Supplementary Material 1). GPs were asked to complete and return the consent form if they were
interested in taking part in an interview. Written informed consent was obtained by the qualitative team
prior to carrying out all interviews.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 73
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Nested qualitative study

In addition to obtaining written consent, interviewers reiterated the purpose of the interviews and
obtained verbal consent prior to starting each interview. At the end of each interview, participants were
given another opportunity to ask any questions and were thanked for their contribution. They were also
offered a copy of their transcript and a copy of study findings when available.

All interviews were pseudonymised on transcription and participant ID numbers were assigned to
ensure confidentiality and minimise the risk of participant identification in sharing/reporting findings.

The qualitative study was included in the main ethics application approved by Yorkshire and the Humber
(Sheffield) Research Ethics Committee (19/YH/0150).

Sampling and recruitment

The aim was to interview a diverse sample of trial participants to include approximately 20 interviewees
from each arm of the trial (amitriptyline/placebo). The lead qualitative researcher (EJT) was blinded
to participant allocation for most but not all qualitative interviews: she was blinded for all 6-month
interviews; however, for participants who had been told their treatment allocation beforehand, this
was discussed during their 12-month interview (this happened for seven participants); EJT also knew
the treatment allocation of four trial participants through her role supporting unblinding to treatment
allocation (one of whom was interviewed as part of the qualitative study). The rest of the qualitative
team remained blinded to participant allocation throughout recruitment, data collection, and data
analysis until preliminary analysis of the main trial data was complete (1 March 2023). The final sample
size was dependent on saturation, and when we determined we had achieved a rigorous, credible
analysis in relation to our aims. Interviewing participants from the amitriptyline arm allowed us to
identify factors related to acceptability, uptake, and psychosocial context. Interviewing participants from
the placebo arm enabled us to explore between-group qualitative comparisons to provide insight into
the quantitative results. Interviewing the same participants at 6 and 12 months allowed us greater depth
to explore changes over time and the potential to better understand any differences in the quantitative
results between 6 and 12 months.

Our aim was to purposively sample trial participants to incorporate variety in sex, age, recruitment hub
(West of England, West Yorkshire, Wessex), baseline symptom severity scores (IBS-SSS), and those who
decided to continue or stop treatment at 6 months. Sampling for variety on key characteristics helps
ensure that the qualitative findings encapsulate the breadth of participants’ experiences and are not
overshadowed by any one subgroup. However, such sampling was not possible due to the impact of
the COVID-19 pandemic delaying qualitative study recruitment. Instead, we employed convenience
sampling and interviewed all trial participants who expressed an interest in the qualitative study and
responded to the qualitative researcher’s request to arrange an interview. We were still able to recruit
and explore experiences from a variety of trial participants, generating a comprehensive account of
participants’ perspectives based on interviews with 20 participants from each trial arm.

Between October 2019 and April 2022, the ATLANTIS trial recruited a total of 463 participants.
Between April 2020 and March 2022, 140 qualitative study invitations were sent to all trial participants
who had consented to be contacted about the qualitative study. Forty-two of these participants were
interviewed, a further three participants contacted the qualitative researcher to decline to take part
due to ill health or going away and the rest did not respond. Sixteen interviews were conducted in the
pilot phase of the trial (between April and June 2020) with participants who had reached their 2-month
time point. The timing of this first interview for participants in the pilot phase was brought forward
(to 2 instead of 6 months) to fit within the time frame for the internal pilot. Twenty-six interviews
were conducted with participants who had reached their 6-month time point between April 2021 and
March 2022. Participants were invited in batches to permit iterative interviewing and data analysis.
Each month the ATLANTIS trial team sent out batches of qualitative study invitations to participants
who had consented to be contacted about the qualitative study and had reached the required time

74

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

point. Recruitment to interviews ended when no new themes emerged, and existing themes were well
developed within a diverse sample.

Of the 42 participants interviewed at 2 or 6 months, only 29 were eligible for follow-up interviews at
12 months. This was because following national guidance trial recruitment paused from March to July
2020 due to COVID-19, which resulted in 13 participants reaching their 12-month time point beyond
the end of the qualitative study interviewing period. Of the 29 trial participants invited to take part in
a follow-up interview, 19 individuals were interviewed, 2 declined due to lack of availability and 8 did
not respond.

In total, 55 general practices were involved in the ATLANTIS trial, 42 of which were available to be
contacted during the qualitative study time frame. Forty-two qualitative study invitations were sent to
general practices inviting any GP involved in the ATLANTIS trial to take part in an interview about their
experiences. Ten GPs declined to take part due to lack of capacity and 16 did not respond. Sixteen GP
interviews with a diverse sample (full and part-time, gender and years as a GP) were conducted between
October 2020 and March 2022.

Qualitative interviews
Semistructured telephone interviews were used to elicit trial participants’ and GPs’ experiences of
the trial treatments and trial participation. Three separate topic guides were developed: (1) 6-month
participant interview, (2) 12-month participant interview, (3) GP interview (see Report Supplementary
Material 1). Each topic guide was developed collaboratively by the qualitative research team by drawing
on existing literature, relevant theories (as detailed below), and input from patient collaborators.
The topic guides consisted of open-ended questions and prompts used by the interviewer to elicit
participants’ views and experiences of the trial and their thoughts and feelings about the trial treatments
and managing IBS. Topic guides were used flexibly to ensure that interviewers explored all required
topics while remaining open to exploring participants’ individual experiences and perspectives in-depth,
to allow novel and unanticipated insights to emerge. All interviews (GP and trial participant) were
audio-recorded using a digital audio-recorder (except one interview which was audio-recorded via MS
Teams). Field notes were taken to capture the interviewer’s impressions and reflections, and any aspects
not captured by the audio-recorder. These notes were used to aid initial coding of each transcript
and were reflected upon during the analysis. At the end of the interview, interviewees were thanked
and debriefed.

Participant interviews
Participant topic guides were informed by the common-sense model of illness perception,44 which
provides a conceptual framework for understanding how participants experience treatments and make
treatment decisions within the context of chronic illness; this has proved relevant in previous qualitative
work on IBS.46

The extended common-sense model (CSM) including the necessity-concerns framework is relevant
to theorising how the beliefs of patients around symptoms and treatment may relate to adherence to
medication. It suggests that patients construct cognitive and emotional representations around the
identity, cause, duration, consequences and controllability of a health condition, and that treatment
adherence is related to their understanding of the condition and its treatments, as well as the perceived
need for treatment and concerns about any negative effects.47

The 6-month telephone interviews explored trial participants’ experiences of taking part in the trial,
their trial treatment (while still blinded), their experiences of other treatments for IBS, their thoughts
about the COVID-19 pandemic, their thoughts about the future and the process of unblinding to
treatment allocation. Most of the interviews were conducted by a female qualitative research fellow
(n = 36) and the rest (n = 6) were conducted by a male doctoral student. The 6-month interviews lasted

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 75
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Nested qualitative study

from 17 to 65 minutes (mean 40 minutes). They took place over a 23-month period (between April 2020
and March 2022).

The 12-month telephone interviews explored trial participants’ reflections on taking part in the trial and
the treatment they were allocated to (where possible), their thoughts about the unblinding process, any
other treatments they had tried since the 6-month interviews, their thoughts about the pandemic and
about the future. Most of the 12-month interviews were conducted by a female qualitative research
assistant (n = 15) and the rest (n = 4) were conducted by a male doctoral student. These interviews
lasted from 15 to 60 minutes (mean 29 minutes). They took place over a 16-month period (from October
2020 to February 2022).

General practitioner interviews

The GP interview topic guide was informed by key domains from normalisation process theory (NPT),45
which provides a conceptual framework that specifies the factors and processes likely to hinder or
enable widespread implementation of new practices. The four constructs of NPT are coherence (people
making sense of the processes), cognitive participation (engaging with the process), collective action
(the work that is required to operationalise the process), and reflexive monitoring (reflecting on and
appraising new working practices). Drawing on this theory helps to explain the factors and processes
encountered during implementation and how these can facilitate the embedding of an intervention
(such as amitriptyline for IBS in primary care) into everyday practice.

General practitioner telephone interviews explored GPs’ experiences of prescribing amitriptyline within
the trial (and within the broader contexts of primary care and IBS management), and potential barriers
and facilitators to widespread post-trial implementation in primary care. Most of the GP interviews
were conducted by a female qualitative research fellow (n = 15) and one was conducted by a female
master’s student. GP interviews lasted from 18 to 45 minutes (mean 27 minutes). They took place over a
17-month period (between October 2020 and March 2022).

Qualitative data analysis

All interviews were transcribed verbatim by a professional transcription company and any identifiable
data (e.g. person names, place names) removed. Analysis started after the first few interviews and
proceeded iteratively. This enabled early insights to be explored more fully in later interviews and the
topic guides to be revised accordingly, for example, adding questions about unblinding to treatment
allocation. The six phases of reflexive thematic analysis42 were implemented alongside coding
techniques from grounded theory43 (e.g. open coding, line-by-line coding, constant comparison) to
explore the interview data. In phase 1 (data familiarisation), one author (EJT) repeatedly read through
all the transcripts and listened back to the audio-recordings. In phase 2 (generating initial codes), EJT
conducted line-by-line coding. Trial participant 6-month and 12-month interviews and GP interviews
were all coded separately. Codes were derived inductively from the data and grouped together to
produce separate initial coding frames for the three interview data sets. After coding all the transcripts,
in phase 3 (searching for themes) the codes were then sorted into potential themes by recognising
meaningful repeated patterns and identifying key concepts in the data. To enhance the quality and
credibility of the analysis, detailed coding manuals were produced containing names and descriptions
of potential themes and subthemes along with examples quotes. At this stage participants’ 6- and
12-month data were compared and combined into a participant coding manual. A separate coding
manual was produced for the GP interview data.

The initial three phases were conducted by one author (EJT), a qualitative research fellow. In phase 4
(reviewing themes), potential themes/subthemes detailed in the coding manuals were discussed with,
and iteratively developed by, members of the research team (EJT, FB, HE, SA) and PPI members (MC,
EJW) to offer diverse inferences and interpretation of the data and to avoid idiosyncratic interpretations.

76

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Reviewing themes for fit with coded extracts and entire data set involved reviewing the original data
for relevant incidents for each potential theme and expanding, merging, or generating new themes.
A negative case analysis was carried out, which explicitly searched for ideas in the data that were
potentially inconsistent with our interpretations. This allowed us to identify and reflect upon important,
but rare, views and the limits of the analysis.

In phase 5 (defining and naming themes), matrices and diagramming were used to compare the themes’
similarities and differences between the 6-month and 12-month trial participant interviews and between
the trial participant and GP interviews. Themes were reviewed and refined to form broader cross-cutting
themes producing an overarching narrative that drew on the participant 6-month, participant 12-month,
and GP interviews. In phase 6 (reporting), quotes were reviewed and discussed by the qualitative team
to provide compelling examples for theme and subthemes and to write up our main report.

Although primarily inductive, the common-sense model of illness perception and normalisation process
theory were reviewed against the coding manuals to help in interpreting initial findings related to (1)
participants’ experiences of IBS and treatments and (2) wider implementation of amitriptyline for IBS
in primary care. NVivo (version 12) (QSR International, Warrington, UK)48 was used to manage data,
implement and record coding and to perform thematic comparisons as described above.

After drafting the main report and when the quantitative trial results became available, the qualitative
team then returned to the data to undertake further analysis of the qualitative themes in relation to the
quantitative trial results. Two key analyses were undertaken.

1. NVivo’s48 Matrix Queries function was used to cross-tabulate talk about key themes by partici-
pants’ treatment allocation (amitriptyline vs. placebo). This enabled an exploration of similarities
and differences between how people in each trial arm talked about their experiences of IBS and the
ATLANTIS trial; findings from this analysis are integrated into the presentation of themes in Barriers
and facilitators to uptake of low-dose, Experiences of taking part in the ATLANTIS trial, Reflections on
managing IBS during the COVID-19 pandemic and Reflections on treatment allocation.
2. Quantitative and qualitative findings were cross-tabulated, considering points of convergence,
divergence, and explanation. This enabled the qualitative findings to be related to the quantitative
findings in a systematic manner. This analysis is presented in Relating the qualitative and quantitative
findings.

The findings are presented in five main sections, which discuss: participants’ and GPs’ barriers and
facilitators to low-dose amitriptyline for IBS, and how these relate to the context of treatment-seeking;
trial participants’ experiences of taking part in the ATLANTIS trial; trial participants’ experiences of being
blinded to low-dose amitriptyline or placebo; trial participants’ reflections on managing IBS during the
COVID-19 pandemic in the first lockdown in 2020 (during which 16 interviews were conducted) and
during subsequent lockdowns and ongoing social restrictions in 2021–2.

Findings

Participants
Forty-two trial participants took part in a 6-month telephone interview. Fifty-five per cent (n = 23) were
allocated to the amitriptyline treatment arm and 45% (n = 19) were allocated to the placebo treatment
arm. Of these 42 participants, 19 took part in a 12-month telephone interview. Nine of these 19
participants (47%) were allocated to the amitriptyline treatment arm and 10 were allocated to receive
placebo (53%). Baseline characteristics of trial participants who took part in the 6-month and 12-month
interviews are shown in Table 39. Sixteen GPs took part in a telephone interview; their characteristics
are shown in Table 40.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 77
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Nested qualitative study

TABLE 39 Baseline characteristics of interviewees at 6 and 12 months

6 months (n = 42) 12 months (n = 19)

Baseline characteristics n % n %

Sex

Female 30 71 15 79

Male 12 29 4 21

Age

Median age (years) 54 54

Range (years) 21–83 25–83

Recruitment hub

Wessex 16 38 6 32

West Yorkshire 9 22 6 32

West of England 17 40 7 37

Educational level

No formal 1 2 1 5

GCSE 7 17 3 16

A level 11 26 5 26

Degree 9 21 4 21

Postgraduate 13 31 5 26

Professionally qualified 1 2 1 5

Ethnicity

White 41 98 18 95

Asian 1 2 1 5

IBS-SSS at baseline

Mild 9 22 4 21

Moderate 19 45 9 47

Severe 14 33 6 32

Barriers and facilitators to uptake of low-dose

Amitriptyline for irritable bowel syndrome

Thematic analysis of trial participant and GP interviews highlighted barriers and facilitators to low-dose
amitriptyline for IBS, as well as an overarching theme that explained why participants and GPs in this
trial were willing to try it despite some concerns. Key barriers, that is, factors likely to hinder prescribing
and uptake of low-dose amitriptyline for IBS, include concerns about antidepressant use, medicalising
IBS, and side effects. Key facilitators include the familiarity of amitriptyline, the low and flexible
recommended dosage, its potential to offer benefits beyond IBS symptom relief and perceived ease of
treatment. The barriers and facilitators to low-dose amitriptyline for IBS were expressed in the context
of desire for a novel approach to IBS: GPs were keen to offer more options for IBS and patients sought

78

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 40 Demographic data for GP interviewees (n = 16)

N %

Sex

Female 8 50

Male 8 50

Age

Median age (years) 45

Range (years) 34–60

Recruitment hub

Wessex 7 44

West Yorkshire 0 0

West of England 9 56

Ethnicity

White 12 75

Asian 3 19

Mixed 1 6

Employment status

Part-time 11 69

Full-time 5 31

Years worked as GP

Median (years) 18

Range (years) 3–30

a cure for their symptoms. The next three sections explore and illustrate these barriers, facilitators, and
the desire for a cure.

Potential barriers to uptake of low-dose amitriptyline for irritable bowel syndrome

Concern about amitriptyline being an antidepressant was raised by both trial participants and
GPs as a potential barrier to patient uptake. A dominant worry for participants was the potential
psychological effects of taking an antidepressant such as amitriptyline in that it might ‘alter their mind’ or
‘zombify them’.

My only concern was that it would alter my mind in some way because it’s used to treat depression. I don’t
know how it works treating depression, but that’s not anything I know anything about.
P2, female aged 45–54 years with severe IBS, 6-month interview

Patient reluctance towards taking an antidepressant for a ‘functional physical health problem’ was also
highlighted among GP interviewees. Typically, GPs expressed doubt about prescribing amitriptyline for
people with IBS, especially with mild symptoms, as they felt such patients would be reluctant to take an
antidepressant for IBS.

It would only be for the ones who are really struggling … I think at that level of symptoms when people
are feeling a bit desperate about their symptoms. There may be some reticence at the slightly lower-end

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 79
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Nested qualitative study

spectrum of IBS for people to start taking an antidepressant … in patients potentially with mild
symptoms, they may be reluctant to use a medication that’s labelled an antidepressant for their physical
health problem.
GP2, male GP aged 35–44 years with 15–20 years’ experience

Perceived social stigma of taking an antidepressant was highlighted by both GPs and participants as a
barrier to uptake due to concern that amitriptyline may be perceived negatively by others and not easily
understood to be a treatment for other (non-mental health) conditions such as IBS.

The worst thing about it is that it can be used as an antidepressant, and it’s almost like, if somebody saw
that on your prescription, they wouldn’t think, oh, she’s taking that because she has IBS; they would think,
oh, she must have stress or depression, and there is still a stigma about that. I think that’s probably the
biggest negative, is that kind of stigma … a lot of people are aware of the name of amitriptyline, and then
it’s got that connotation, so I think that’s a negative. If you called it something completely different, and
said, ‘It’s a treatment for IBS’, you probably wouldn’t worry about somebody knowing you were having it.
P22, female aged 55–64 years with mild IBS, 12-month interview

General practitioners were concerned about the wider consequences of prescribing a long-term ‘drug’
for people with IBS and suggested that this may be a barrier to prescribing amitriptyline for IBS patients.
They predominantly worried that prescribing amitriptyline for IBS could ‘medicalise’ IBS and highlighted
potential contraindications to its prescription and polypharmacy issues as potential barriers.

I think that we’re very cautious or wary of the fact that there are within a stage where we’re thinking
about polypharmacy and about multiple medications and is the prescribing of medicines the right thing
to do for everybody? Obviously, it’s not necessarily … I think that there is both caution in general about
prescribing – particularly in elderly patients, older patients – and caution about polypharmacy … I think
there is an issue to explore about then the potential for medicalising, introducing a medicine for instance
to somebody who therefore may not require a medicine and may just require lifestyle options.
GP11, male GP aged 45–54 years with over 20 years’ experience

Furthermore, GP interviewees felt that prescribing low-dose amitriptyline for IBS could lead to increased
administrative burden from the resultant repeat prescriptions and follow-up review appointments.

More drugs to sign off! I keep prescribing, my list is already too long.
GP1, female GP aged 55–64 years with over 20 years’ experience

Amitriptyline has side effects and it has to be prescription only, whereas Buscopan and Colofac are over
the counter. It’s medicalising the situation and increasing the workload in terms of monitoring a drug and
repeat prescription system, etc. That is an additional workload compared to somebody who the diagnosis
is made and they’re sending them off to get their own meds.
GP7, female GP aged 55–64 years with over 20 years’ experience

General practitioner interviewees also expressed concerns about patient tolerability of side effects,
potential of overdose and wider impacts of prescribing amitriptyline, including anticholinergic burden in
older patients. Careful consideration of the suitability of low-dose amitriptyline for people with IBS was
thus evident in the data.

Often these patients are quite young, but as they get older, then it’s an anticholinergic and it adds to
the anticholinergic burden and we know that people fall over and get confused and having long-term
anticholinergic burden is not a great idea in terms of overall well-being.
GP1, female GP aged 55–64 years with over 20 years’ experience

80

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Reluctance to ‘medicalise’ IBS was also evident in the trial participant interview data. Concerns about
drug side effects/dependency were expressed in the participant interviews but this seemed to have
been tempered by the sense that if they experienced any adverse effects, they would likely be known
side effects and as such manageable. This may be reflective of the fact that they are participants
who have all signed up for a drug trial and were provided with detailed information about possible
side effects.

I suppose I was a bit concerned that amitriptyline was a drug that I’d heard of, and I was worried that I
might get addicted to it in some way, because it’s antidepressant I understood.
P6, female aged over 65 years with mild IBS, 6-month interview

I suppose with any low-dose medication that can be used for depression, anxiety, and whatnot, I did think
weight gain might be part of it, but again I thought if it helps with my stomach I’m prepared to deal with
any sort of mild side effects. If it were to help, I’d be up for a bit of weight gain if that’s what it took.
P42, female aged 18–25 years with severe IBS, 6-month interview

Potential facilitators of uptake of low-dose amitriptyline for irritable bowel

syndrome
The familiarity of amitriptyline appeared to be an important factor helping participants and GPs feel
reassured about using amitriptyline for IBS. Trial participants seemed to take comfort in amitriptyline
being a well-established drug rather than a new or experimental treatment. Prior personal experience
or knowing others already taking amitriptyline for non-mental health conditions appeared to help
normalise a drug treatment, and specifically amitriptyline, for IBS.

From what I understand, it’s been used for years so it’s quite a well-known drug, so I didn’t feel uneasy
using it. It’s been tried and tested, so I think if we can use existing drugs to treat other things, then it’s
worth giving them a go.
P18, female aged 25–34 years with moderate IBS, 6-month interview

Well, I’ve got a couple of friends that are on them, but not for their tummies, they’ve got them because
they’ve got a bit of stress in their lives, and have lost their husbands and stuff, and it helps them sleep.
I was quite happy to go on them, because I knew that I’d got friends on it; it wasn’t something that I
didn’t recognise.
P9, female aged over 65 years with moderate IBS, 6-month interview

The familiarity of amitriptyline was also highlighted in the GP interviews as a potential facilitator to
prescribing. Amitriptyline was commonly viewed as a widely available and inexpensive treatment with
well-established knowledge of side effects, that anecdotally it ‘seems logical’ that it could be successful
is helping to resolve IBS symptoms.

It’s freely available and, as I said, very cheap, so there isn’t going to be that barrier in terms of an
expensive, new medication.
GP2, male GP aged 35–45 years with 15–20 years’ experience

My initial thoughts were that it was actually a very good idea in that anecdotally there’s certainly been
a suggestion about amitriptyline helping people with multiple issues regarding pain and possibly pain,
certainly neuropathic-type pain.
GP11, male GP aged 45–54 years with over 20 years’ experience

General practitioners demonstrated existing knowledge of amitriptyline and seemed confident in making
prescribing decisions about people with IBS. A common view was that it is more appropriate to prescribe

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 81
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Nested qualitative study

low-dose amitriptyline to people experiencing pain and diarrhoea rather than constipation due to not
wanting to ‘give them something that’s going to also constipate them’.

I think I’d more likely use it for somebody who’s got cramps and bloating and maybe loose motions, than
constipation really. If they’ve got those sorts of symptoms, particularly pain I think, then I’m more likely to
use it for that I’m more inclined to use it for the pain aspect of IBS.
GP8, female GP aged 45–54 years with over 20 years’ experience

Distinguishing low-dose use for IBS from traditional antidepressant use was highlighted as a potential
facilitator of uptake. Trial participants felt assured that amitriptyline for IBS is prescribed at a lower dose
that it would be for treating depression, feeling that it would be safer and that there was more distance/
separation from being on a treatment for mental health conditions.

I think it’s an antidepressant, isn’t it, or it’s used as an antidepressant obviously in larger doses. I wasn’t
really concerned because the dosages are pretty low to be fair and it is used for other things, so you’d like
to think it was relatively safe to be used for IBS as well.
P7, female, aged 45–54 years with moderate IBS, 6-month interview

Similarly, GP interviewees reflected on the importance of addressing patients’ concerns about

amitriptyline being an antidepressant. A common belief was that patients could be reassured by
explaining that, although traditionally an antidepressant, amitriptyline for IBS is prescribed at a much
lower dose and is also commonly used for a variety of other conditions. Some GP interviewees also
reflected that explaining that amitriptyline is being prescribed for ‘physical symptoms’ can help ensure
that people with IBS are not left feeling that the GP thinks IBS is ‘all in their head’.

Not to blow my own trumpet, but I’m always really careful with, I mean, amitriptyline can be used for a
variety of conditions, can’t it? I’m always very careful that I, because the common thing is, and usually
patients I haven’t spoken to who come back and tend to want to talk about the amitriptyline they’ve been
prescribed and usually the case of, ‘You’ve prescribed me an antidepressant, I’m not depressed’. I’m very
careful to say or remind patients that if you’re taking, amitriptyline is used for a number of reasons and it
traditionally was used as an antidepressant, but we’re using it at much lower doses to what you want as
an antidepressant.
GP4, male GP aged 35–44 years, with 10–15 years’ experience

I’m always quite careful when I prescribe amitriptyline for anything to explain to people that, ‘Look, this is
an antidepressant, but we’re not looking at using it as an antidepressant’. I put that in as one of the first
things I say because otherwise, people go away, read the leaflet, ring up and say, ‘Why have you given me
this?’ I always tend to couch it like that and say, ‘Look, it’s been around for a long time. That was originally
its use, but it’s been found to helpful in lots of other conditions’. Then I’ll often say to patients, ‘It’s used
for migraine prophylaxis, sciatica, chronic pains, etc.’, because otherwise, they think you just think it’s all
in their head and you’re giving them an antidepressant. So I definitely couch it like that. I find that when I
put it like that, people seem to be reasonably receptive to it as an idea.
GP14, male GP aged 45–54 years with 15–20 years’ experience

Recognising potential benefits beyond IBS symptom relief was also highlighted as a potential facilitator
to prescribing and uptake. Taking amitriptyline was viewed by trial participants as potentially having
other benefits beyond IBS symptom relief, including improving emotional health (e.g. elevating mood/
reducing stress), relieving symptoms in other pain-related conditions and improving sleep.

I think because it’s to do with, yes, it was a very, very small amount of, like an antidepressant, but maybe
that would help control any emotional feelings that I was going through, maybe that would then help with
my IBS.
P18, female aged 25–34 years with moderate IBS, 6-month interview

82

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

It was amitriptyline, and I have taken amitriptyline in the past for migraines and also, I was recommended
to take them for my back. I’ve got lower back problems, so I thought, oh, that’d be quite good. I could kill
all the birds with one stone! IBS, migraines, back problems!
P3, female aged over 65 years with moderate IBS, 6-month interview

Some participants described experiencing side effect of drowsiness but welcomed the opportunity of a
good night’s sleep, which had been disrupted by their IBS symptoms and/or stress.

I just wait for the Channel 4 News to end, and then take it at eight o’clock because I’ve worked out
clockwise, that if I take it at bedtime, then I wake up a bit groggy. Whereas if I take it at eight o’clock, it
sort of kicks in about the time I go to bed. I’m only doing it for the sleep. No, I’m not, but it is helpful. So
the spin-off from this is if anybody wants a sleeping tablet, then amitriptyline’s a really good one.
P40, male aged 45–54 years with moderate IBS, 6-month interview

I definitely found an improved sleep, because I was drowsy, going to bed. When I have stressful periods,
that is something that suffers, is my sleep; getting off to sleep – because I have all sorts going through my
head. The tablets actually really helped with that.
P25, female aged 25–34 years with moderate IBS, 12-month interview

Similarly, in GP interviews amitriptyline was viewed as an advantageous choice for some people. As well
as providing reassurance about the low and flexible dose range of amitriptyline for IBS, it appeared to
be common practice among GP interviewees to stress to patients that they ‘could use some of the side
effects to their advantage’ and reassure patients of other potential benefits taking amitriptyline such as
addressing sleep problems.

I generally sell it to them like if they struggle to sleep, then the amitriptyline can help them sleep.
GP5, male GP aged 25–34 years with 5–10 years’ experience

I think the people who I’m thinking of who would probably take this amitriptyline are the people where
they’re quite debilitated by their symptoms. They’re probably affecting their sleep; it’s probably affecting
their mood, so the fact that it was previously used as an antidepressant, even though that’s not what
we’re using it for in this context, may actually also be a benefit. I think it’s really looking at both arguments
and saying, ‘Well, it may help improve your mood even though we don’t use it in that context and the
doses are much lower’. Being able to explain the benefits, the sleep – if you’re struggling with your sleep, it
may help with that.
GP10, female GP aged 45–54 years with 15–20 years’ experience

Emphasising ease of treatment was viewed as another important facilitator of patient uptake of
amitriptyline. Trial participants seemed to appreciate the small size of the tablets making them easier to
swallow and only having to take tablets once a day making it easy to fit into their daily routine.

Well, they’re nice and small, I suppose, physically! They’re easy to swallow and all that sort of thing.
P6, female aged over 65 years with mild IBS, 6-month interview

I don’t like taking pills per se. I’m not a pill-popper, but I have been taking them, and I would be quite
happy to continue to take them. Easy. Easy to swallow. Yes, not a problem.
P32, female aged 45–54 years with mild IBS, 6-month interview

I take them at seven o’clock now, on the dot. Taking the tablets wasn’t a problem. I take other tablets as
well. I take a, the smallest dose of statin in the evenings as well, so it goes hand in hand with that.
P5, male aged over 65 years with moderate IBS, 6-month interview

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 83
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Nested qualitative study

Another important factor was having the flexibility of adjusting the dose from one 10 mg tablet
(10 mg) to 2 × 10 mg (20 mg) or 3 × 10 mg tablets (30 mg). Although some participants were uneasy
about adjusting the dose themselves and felt they would prefer any dose adjustment to be a medical
judgement and ‘not just willy-nilly by yourself’, most trial participants reported a sense of empowerment
and appreciation at having the flexibility to adjust the dose according to their needs and increasing
dosage at their own pace.

I felt like I could reduce them or increase – do you know what I mean? There was a sort of sense of being
able to say, ‘No, I don’t need three every day to make them work’. Do you know what I mean? It was like
there was that sense of, okay, I can see that by reducing them it got worse, and by increasing them it got
better. There was that sort of flexibility within the study to try that out a bit.
P23, female aged 45–54 years with severe IBS, 12-month interview

I have more control with this. Whereas the previous medical trials, it was you have to take this at certain times.
This, I could take one tablet or two tablets, or three tablets, whatever I felt was suitable for me. That was quite
nice to have the flexibility, definitely … having control over it and going with what my body felt like if you see
what I mean. The maximum tablets I did go up to was two. I just felt three was a bit too much. I just didn’t feel
comfortable with it. Whereas the two was the good in between and I was happy enough with that.
P17, female aged 25–34 years with severe IBS, 6-month interview

Participants described using the written instructions and flow chart provided by the trial team and
talking to the researchers to guide their dose titration. Most found the written information helpful and
straightforward. A few participants felt the amount of written information was potentially overwhelming
(‘I was absolutely deluged with stuff’). A few mentioned they would have liked further guidance on how
to manage the process of stopping their tablets.

It [dose titration] was fine. There was a flowchart to follow. I followed the flowchart and I took the
extra medicine.
P37, male aged 35–44 years with severe IBS, 6-month interview

Interviewer: How has it been taking the tablets, then, thinking about the process of adjusting the dose,
and just taking the tablets?
P23: Fine. I think, because again, there was quite a lot of information sent about that, so there was the
sort of suggestion that you could start at one a day, but then, if you felt you had any side effects go to
every other day, or increase to two a day, or two every other day. So there was that sort of information
about, okay, you’ve got to trial and error this a bit to see what you feel is right for you.
P23, female aged 45–54 years with severe IBS, 6-month interview

I actually had some quite bad, kind of, what I considered potentially withdrawal symptoms from the study,
as well. […] It might be kind of attributed to a stressful time, but I feel like if there was more information on
weaning off the tablets, that might have been useful to kind of prevent that potentially from happening.

P25, female aged 25–34 years with moderate IBS, 6-month interview

Similarly, GP interviewees also reflected on how prescribing a ‘dose range rather than straight dose’ of
amitriptyline was a ‘good idea’ and is ‘now common practice’. They also expressed that explaining to
patients that they can self-titrate the dose (increasing or reducing as required) helps GPs to promote
the benefits of taking amitriptyline for IBS to patients, potentially increasing adherence and reducing
appointment time.

I think you need to empower patients to be able to increase the dose themselves otherwise that’s three
appointments just to get them to 30 milligrams. … I think the idea that they can titrate themselves is a
good one.
GP3, male GP aged 45–54 years with 15–20 years’ experience

84

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

I think it’s a good idea to be able to wax and wane it. People know their own body and particularly with
IBS things change so much even for one person … so I think it’s a good idea.
GP16, female GP aged 35–44 years with ˂ 5 years’ experience

Desire for a cure

Participants’ desire for a cure and GPs’ desire to offer more patient choice around IBS treatments
seemed to explain how patients contribute to an overarching theme that the potential positives of
trying amitriptyline for IBS outweighed any concerns. Participants expressed frustration at having
unresolved symptoms and feeling they tried everything else (‘you name it; I think I’ve tried it; nothing
seems to work’) and as such were highly motivated to try something new, especially if it offered the
hope of sustained symptom relief. Hopes for a cure seemed to be a key motivator for taking part in the
ATLANTIS trial.

That’s one of the reasons I was quite keen to come on this trial because it offered a ray of hope. A miracle
cure! There are things you can try but none of them seem to be very effective. So really, if they could find
those little blue pills that cure it, it would be brilliant.
P1, male aged over 65 years with severe IBS, 6-month interview

If you can get some sort of cure for it. That’s what the appealing thing was, if they can come up with
some cure.
P28, male aged 45–64 years with moderate IBS, 6-month interview

While GP interviewees did not view amitriptyline as a potential cure for IBS, they recognised the
potential benefit of providing another IBS treatment option and being able to offer increased patient
choice. GP interviewees commonly appreciated having something else to offer people with IBS and
described how prescribing amitriptyline for IBS would allow them to add ‘another string to your bow’ or
have ‘another tool in the box’.

Well, I think it’s always useful to have another tool in your box, isn’t it? If you’ve tried other things that
haven’t worked, it’s useful to have something else you can give a patient and it’s cheap.
GP3, male GP aged 45–54 years with 15–20 years’ experience

Well, it’s another string to your bow, so it’s another thing to be able to offer. Yes, I think it’s just another
offer, really … and these patients have – not necessarily something else.
GP14, male GP aged 45–54 years with 15–20 years’ experience

Being able to provide reliable patient information about amitriptyline for IBS was viewed as an important
component in offering more choice and encouraging future prescribing of amitriptyline for IBS. GPs also
reflected on the need to update national guidelines and the patient information leaflets that accompany
amitriptyline tablets if ATLANTIS demonstrated successful trial outcomes.

If it is effective, then getting that into the patient literature so that when we send them something to
have a think about … patients can have a think about it and the pros and cons … having that resource to
hand. I mean, if it’s integrated into our online systems round the country to describe the physician support
information and integrated into the computer system would be amazing.
GP1, female GP aged 55–65 years, over 20 years’ experience

Comparison between participants allocated to placebo and low-dose

amitriptyline
Talk about barriers and facilitators to low-dose amitriptyline for IBS was very similar among
trial participants who had been allocated to placebo and those who had been allocated to
low-dose amitriptyline.
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 85
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Nested qualitative study

Experiences of taking part in the ATLANTIS trial

Participants described volunteering for ATLANTIS in the hope that an effective treatment could be
found for them and/or others with IBS, often in the context of support from significant others. They
described experiencing fluctuating IBS symptoms and described symptom changes over time ranging
from improved symptoms through no clear change to possibly worsening symptoms. The predominant
view among trial participants was that ATLANTIS was a straightforward and convenient trial to be
involved with. Two main aspects of the trial led to positive evaluations of trial participation: the
convenience of the tablets and questionnaires; and receiving support from the trial researchers (three
research nurses and one clinical study officer). A few small delays were frustrating, but glitches were
experienced as well-managed by the trial team and frustrations with specific questionnaires were also
noted but did not detract from overall positive evaluations of experiences in the trial.

Volunteering to find something to relieve irritating bowel syndrome symptoms

Trial participants described joining the ATLANTIS trial within the context of ongoing symptoms, to find
a way to resolve or to relieve IBS symptoms or to ‘make a difference’. Some participants emphasised
the desire for personal benefit while others also spoke about the potential for the trial results to
benefit others.

As I’ve had quite extreme problems with IBS, I was quite happy of any chance it might improve things. […]
that’s one of the reasons I was quite keen to come on this trial because it offered a ray of hope.
P1, male aged over 65 years with severe IBS, 6-month interview

Yes, so I was contact by my doctors just to ask if I would take part as I’ve been suffering with IBS for quite
some time now, so it was really just, I’ve tried different things, I thought I’d give this a try and see if it
worked for me. Or if it doesn’t, if it’s helpful for other people who suffer with it as well.
P18, female aged 25–34 years with moderate IBS, 6-month interview

Participants described talking with family members and/or friends about the ATLANTIS trial and being
supported to take part.

Well, I’ve spoken to a couple of my good friends, and said this is what I have gone on to, and they’re
saying, ‘Well, we wish you luck, that you’ll hopefully find something to sort it’, because I probably show
in my face when things are really, really bad because everything’s knotting and grinding, and all the rest
of it. I’m sure I’m frowning, and close friends tend to pick up on stuff like that anyway. They were pleased
when I said that I’d been selected and was going to give it a go, it was like, ‘Good luck, I hope it works out
for you’.
P4, male aged over 65 years with mild IBS, 6-month interview

Symptom fluctuations and changes

Trial participants described experiencing fluctuating IBS symptoms and changes over time which ranged
from greatly improved symptoms through no clear change to potentially slightly worsening symptoms.

I think it’s virtually disappeared. There’s still a few things I can’t eat, but not many. Before, I couldn’t eat
anything with wheat in; it would make it worse, whereas now I can. I’ve just got a few things left, like
lentils I don’t seem to do well with. They’re a very tiny number, so now I can just go out to eat; there’s
always something to eat. The same as friends, I don’t have to ask them to do special meals they wouldn’t
otherwise do. So yes, I don’t have any problems sleeping; I’m not woken up by cramps and pains. It’s made
a huge difference.
P26, female aged over 65 years with severe IBS, 6-month interview

86

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

It’s come back a bit, but not as bad as it used to be. I used to get a bout maybe every two or three weeks,
whereas now it’s maybe once a month. So, it is an improvement, but it’s not as good as it was at the
beginning of the trial.
P8, female aged 55–64 years with mild IBS, 12-month interview

Some weeks it’s worse than others, but yes, I’d say most weeks I experience some form of IBS. Sometimes
it’s less painful. Sometimes it’s more. Sometimes it’s, yes, it’s always very up and down. I think it’s to do
with my stress levels and stuff. It depends on how I’m feeling.
P18, female aged 25–34 years with moderate IBS, 12-month interview

Yes, when I started the study, I started off taking one tablet a day. Then the instructions were that if I’d
had no ill effects, I should start to increase it to two after a certain period and then again if there were no
ill effects, I should increase it to three. So, I’ve been taking three a day for quite a long time and it doesn’t
seem to make any difference at all.
P2, female aged 45–54 years with severe IBS, 6-month interview

It’s probably worse than it was before. Yes, the last couple of months it’s not been great. Well, as it
is sometimes, it goes better and sometimes it goes worse, and it just seems to be slightly worse at
the moment.
P10, male aged 55–64 years with mild IBS, 12-month interview

The convenience of trial treatments and questionnaires

Having to take tablets just once a day, ‘Just taking tablets, not a strain’, completing short online
questionnaires (at a time convenient to participants) and receiving links and reminders via text and
e-mail all seemed to fuel a common perception that the ATLANTIS trial was straightforward and not a
burden on participants’ time.

It’s been very non-intrusive. They’ve been very efficient sending the medication and I’ve filled in the weekly
surveys. It’s been easy enough. It’s not been arduous. They’re quite short, generally. The survey was just
one question. It hasn’t involved any pain or disruption to my life, really. It’s been quite easy to fit in.
P26, female aged over 65 years with severe IBS, 6-month interview

It was all fairly straightforward. It’s not a particular hassle to take the tablets. I think I always felt as
though I didn’t really have to think about it very much, because if I needed new tablets, or if there was a
questionnaire to fill in or something I’d get a text, so I was always prompted, wasn’t in any way difficult.
No, it’s just all fairly straightforward.
P24, female aged 45–54 years with mild IBS, 12-month interview

Receiving support from the researchers

Trial participants reported highly valuing the researcher support throughout the trial. Researchers were
seen as friendly, helpful, supportive, and informative, ‘always available to answer any participant queries
or concerns’ without being bothersome.

In terms of communication and everything, that’s been really, really positive. The main contact that I have,
she’s great [the Researcher]. She’s really friendly, and she’s always made me feel really at ease. I actually
feel like when I’m talking to her she’s listening to me, which maybe doesn’t sound a lot, but when you’ve
been to the doctor so many times and just been passed off, for somebody to actually listen to you is nice in
itself, really. I feel like she is genuinely interested in how the study is going.
P31, female aged 35–44 years with moderate IBS, 6-month interview

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 87
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Nested qualitative study

They were always contactable. As I say, when I had a problem entering the survey that one week, I just
texted [Researcher’s Name] and they took it from there, so I’ve got no problems with the way it’s been
dealt with.
P3, female aged over 65 years with moderate IBS, 6-month interview

Delays and glitches

Trial participants reported common glitches in trial processes around medicine supplies and receiving
online questionnaires and reminders. These were generally viewed as managed well (i.e. quickly
resolved) by the trial team. Early in the trial, some participants described a long start-up period. This
seemed to be due to complications with blood tests and was limited to the first few participants
recruited into the trial.

We had one minor issue where there was some confusion and I needed more tablets, and there was a
cock-up on your end and they didn’t get sent, but it was soon resolved and they got it to me well in time so
I didn’t have any days without. It was fine … Even when there was an error, it was resolved. So fantastic.
P37, male aged 35–44 years with severe IBS, 6-month interview

I thought it was very slow at first. I thought, God, I thought I was supposed to be going on this trial and
it seemed to take months. It did take months, I think. I just seemed to think am I ever going to start this,
you know?
P1, male aged over 65 years with severe IBS, 6-month interview

Frustrations of converting complexities of symptom fluctuations into a binary

response option
One common concern about the trial processes related to the perceived inadequacy of binary response
options in some of the questionnaires, in particular the weekly question that asked, ‘Have you had
adequate relief of your IBS symptoms?’, with response options Yes or No. Trial participants commonly
expressed frustration and concerns about whether such response options accurately captured their
experiences. For example, participants described finding it difficult to give a global judgement on
whether relief had been ‘adequate’ over the course of a whole week, when the meaning of ‘adequate’
was open to interpretation, and when symptoms fluctuate considerably within a short space of time.

I thought the weekly one, the kind of just yes or no response, maybe it should have been more detailed to
catch any different changes.
P21, male aged 25–34 years with moderate IBS, 6-month interview

I do think that the questionnaire that we get every week, which I think the question is, ‘Have you had
adequate relief of your IBS symptoms this week?’ I find that really, really hard to answer yes or no. I do
answer it yes, but it’s subjective, isn’t it, that ‘adequate’, and also in a whole week. Some days, not other
days. So that I don’t know if it’s been the most helpful question.
P23, female aged 45–54 years with severe IBS, 6-month interview

Comparison between participants allocated to placebo and low-dose

amitriptyline
Experiences of taking part in the ATLANTIS trial were broadly similar in many ways across participants
allocated to placebo and participants allocated to low-dose amitriptyline. The exception to this pattern
was in the description of experiences of IBS symptoms during the trial. While participants in both trial
arms reported symptomatic improvements during the trial, these were more commonly expressed

88

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

by participants in the low-dose amitriptyline arm. While participants in both trial arms reported little
change or possibly worsening symptoms during the trial, these were more commonly expressed by
participants in the placebo arm. Overall, participants in both arms of the trial volunteered for similar
reasons and were generally positive about their experiences in the trial, they found the tablets and
questionnaires convenient and not unduly burdensome, they experienced a few glitches, they found it
difficult to answer binary questions about IBS symptoms, and they valued the support received from
the researchers.

Reflections on managing IBS during the coronavirus disease discovered in 2019

pandemic
Trial participants predominantly felt that the COVID-19 pandemic had had minimal impact on their
IBS as their symptoms had been easier to manage with social restrictions imposed by lockdowns and
measures to reduce social contact.

I suppose to some extent managing my IBS in general, maybe, was a little bit easier, because I didn’t have
the stress of going out to dinner or meeting people, because you couldn’t do any of those things, so being
sat at home and feeling bloated was easier than being out.
P31, female aged 35–44 years with moderate IBS, 6-month interview

I think it’s much improved because I don’t have the pressure of being in an alien environment, or travelling,
or having to eat food that I have no choice but to eat if I go to a meeting or out to lunch with a client.
You’re kind of restricted, whereas when you’re in lockdown and at home, you can eat what you want to
eat, and you’re not embarrassed by the consequences because it’s only you. So, it’s made it much easier to
cope with the symptoms.
P32, female aged 55–54 years with mild IBS, 6-month interview

Despite often experiencing increased stress/anxiety due to the pandemic, for example, financial worries,
general anxiety about pandemic, participants typically reported that their symptoms were easier to
manage due to reduced worries about going out (home working, not eating out, not having to find public
toilets) and eating better (more home-cooked food).

I suppose with the pandemic it has caused things to make my IBS worse because of the stress, but then
I can’t go out, so in a way it’s easier being at home near the toilet anyway. So, I haven’t got the normal
stresses of planning days out or anything like that, but I’ve got a lot of other new stresses about money
and things like that. So, I suppose it’s been a good test case really, because I’ve had some less stresses, but
some more stresses as well.
P2, female aged 45–54 years with severe IBS, 6-month interview

Comparison between participants allocated to placebo and low-dose

amitriptyline
Talk about the impact of the COVID-19 pandemic on IBS was very similar among trial participants who
had been allocated to placebo and those who had been allocated to low-dose amitriptyline.

Reflections on treatment allocation

Trial participants appeared to construct narratives around their treatment-arm allocation based on
their early and ongoing experiences of their treatment, around their experience (or lack of experience)
of symptom improvement and side effects. Four themes captured participants’ talk about treatment
allocation – clues to low-dose amitriptyline allocation; clues to placebo allocation; curiosity and
concerns about finding out one’s treatment allocation; and disappointment about one’s treatment
allocation. Comparisons between participants across the two trial arms are integrated within the
presentation of each of these themes.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 89
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Nested qualitative study

Clues to low-dose amitriptyline allocation

Trial participants thought they had been allocated to low-dose amitriptyline if they noticed changes
in their health. Some participants expressed the thought that they were taking low-dose amitriptyline
because they had noticed improvements in their IBS symptoms, such as pain and bowel function.

I think I’ve received real pills. I’m seeing less symptoms. Not getting the stomach cramps quite as often.
When I do go to the bathroom, it’s more of a solid consistency, rather than runnier.
P13, female aged 35–44 years with moderate IBS, 6-month interview

Well, initially I must admit I thought, because it cleared up, I didn’t have any problems after about a
fortnight into the study, I felt that it was getting better every day, I didn’t have any and then it come to
the point where I didn’t have anything for weeks and weeks and I thought I must be on the amitriptyline
because why otherwise have my symptoms all stopped?
P29, female aged over 65 years with mild IBS, 6-month interview

Some trial participants thought they were taking low-dose amitriptyline because they experienced side
effects associated with amitriptyline such as dry mouth and sleepiness.

Well, you know, if I was a betting man, I would bet on the fact that I’m on an actual drug. I’m going to feel
really cheated if I find out I am not. It will show that placebo produces symptoms that you imagine. I really
have got a very dry throat. I’ve had that since straight after the first week, and that’s one of the symptoms
that they mention. Also, I found that they were making me slightly constipated at first, that’s why I didn’t
double up immediately to two, but there we go. No, I mean, yes, I would say that I’m taking amitriptyline.
P5, male aged over 65 years with moderate IBS, 6-month interview

Well, I think I’m taking the antidepressants or whatever they are because I sleep like a log which I really,
really like! I’ve never slept as well in years as I do at the moment, right from the start. So that’s the only
thing I can put it down to, is the tablets.
P8, female aged 55–64 years with mild IBS, 12-month interview

Some trial participants seemed to be less confident when asked to guess their treatment allocation
and talked about hoping that the changes they had noticed meant that they were taking low-dose
amitriptyline. This more tentative hope seemed to be related to the challenges of interpreting symptom
changes given the fluctuating nature of IBS symptoms and the multiple possible influences on them.

I hoped I was because I genuinely believed that there had been a difference. I did wonder whether the
improvement had been because I hadn’t been at work, and I was less stressed. So that could have been
part of the improvement, but I’ve not regressed backwards, so that makes me think, well, perhaps I am on
the amitriptyline, because I’ve definitely been stressed since I’ve gone back.
P13, female aged 35–44 years with moderate IBS, 12-month interview

When discussing marked benefits or side effects that they attributed to low-dose amitriptyline, trial
participants drew comparisons with placebos that suggested placebos were perceived as unlikely to
generate such effects.

It’s been great. I found the amitriptyline really helped. If this is a placebo effect then my mind is way
too powerful and men should fear me! Since starting to take it, I would say that my bowel function is
overwhelmingly just normal […]. The IBS symptoms are not wholly gone, but overwhelmingly, they’re so
much more improved than they were.
P33, female aged 25–34 years with moderate IBS, 6-month interview

90

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Straightaway it did what I was warned, one of the side effects, which was the dry mouth. I do have a dry
mouth. […]. The only good thing, really good thing that I think that I’ve noticed, is that – and again, I’ll feel
very silly if I’ve just been on the placebo, is I don’t sleep well, and I can sleep for England now!
P9, female aged over 65 years with moderate IBS, 6-month interview

Talk about experiencing symptom improvements and/or side effects and interpreting this to mean one
was taking low-dose amitriptyline was slightly more common among participants who had actually been
allocated to low-dose amitriptyline compared with those allocated to placebo.

Clues to placebo allocation

Consistent with interpreting symptom changes as meaning one was taking low-dose amitriptyline, trial
participants interpreted a lack of symptom change to mean they were taking placebo. If they were still
experiencing IBS symptoms (‘it’s no worse, no better’) and/or they felt that they had not experienced any
of the common side effects associated with amitriptyline, participants typically reported with conviction
that they must be on placebo tablets.

I’m on, convinced that I’m on placebo, so, because it’s made, and I’m on three a day and it makes no
difference whatsoever. I wouldn’t mind putting a few quid on that I’m on the placebo! Either that or the
drug doesn’t work, at all. There’s no change.
P1, male aged over 65 years with severe IBS, 6-month interview

I think I’m on the placebo, well I’m 99 […]. Unless you tell me – you probably don’t know anyway, but I’ve
decided I’m on placebo because there’s been absolutely no change at all, no side effects and I have had no
beneficial effects whatsoever unfortunately.
P3, female aged over 65 years with moderate IBS, 6-month interview

Talk about experiencing no noticeable change in symptoms and interpreting this to mean one was taking
placebo was slightly more common among participants who had actually been allocated to placebo. Only
participants who had actually been allocated to placebo described thinking they were in the placebo arm
because they had not experienced any side effects.

Curiosity and concerns about finding out one’s treatment allocation

Typically, regardless of actual treatment allocation, trial participants were very keen to know their
actual treatment allocation. Nevertheless, concerns were expressed around actual trial treatments
not matching expectations/constructed narratives. Some trial participants who thought they were
taking low-dose amitriptyline because of changes in their IBS and/or side effects expressed concerns
that they would feel foolish if told they had actually been taking placebo. This seemed to be based
on understandings of health that drew sharp distinctions between physical and mental processes,
and assumptions that placebos can’t trigger any physical symptom changes, which in turn implied for
participants that any perceived benefit would mean their IBS symptoms had been generated mentally
and were somehow less ‘real’ than previously thought.

I shall feel really stupid if it’s the placebo! I’d just feel really stupid if I’ve been taking a placebo and I’ve
suddenly felt much better. I would start to question myself then as to why I went through years of, you
know, what caused it. It is because then you think, well, it must have been in my head then. I must have
been willing it or something, I must be imagining it, but I know I wasn’t imagining it, definitely I wasn’t. I
know how much, so much better I do feel now.
P29, female aged over 65 years with mild IBS, 6-month interview

You think if I do get relief from the IBS, then find out, if we ever do find out whether it’s a placebo, then I’ll
know it’s just in my head. Then I’ve got to rethink what have I been thinking the cause was over the years?

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 91
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Nested qualitative study

Maybe actually something completely different. Then I’d think that I’d been a fake for all these years!
Again, it’s something out of my control and if it happens, it happens. I’ll come to that when it does.
P10, male aged 55–64 years with mild IBS, 6-month interview

Some participants also explicitly expressed concern about being embarrassed for having wrongly
thought they were taking low-dose amitriptyline and having shared that with trial staff and others.

I was a bit nervous. I didn’t like the thought that I would react to something, and it would turn out to be
a placebo, because I thought oh it’s a bit weird. Then it makes you feel a bit stupid. So, I thought, what if I
say, ‘Oh it’s amazing’ and then I find out I’ve had the placebo? I’d feel a bit of an idiot.
P2, female aged 45–54 years with severe IBS, 6-month interview

If it’s not the drug I’m going to feel a bit stupid because I’ve told you [interviewer] and [the research nurse]
without reservation that I’m convinced it is the drug and that the drug has worked. Obviously, if that was
not to be the case, well, the other line is that in some way my mind must think that that placebo is doing
me good. As I say, I’m convinced that won’t be the case, but I might be proved wrong … If it’s negative, if
it’s not the drug I shall feel very silly talking to the nurse having told her that I’m taking something! I shall
just hang my head in shame.
P5, male aged over 65 years with moderate IBS, 12-month interview

However, not all participants were concerned about possibly having experienced benefit from a placebo.

Well, I suppose regardless of whether I’m on the placebo and it’s had a psychological impact or whether
I’m on the real thing that’s had a physical impact, I do think it has made some improvement to my
symptoms, so I would say that I have a positive view of it!
P7, female, aged 45–54 years with moderate IBS, 12-month interview

Fewer concerns were expressed about the other possible mismatch between perceived and actual
treatment allocations, that is thinking one had been receiving placebo and then finding out it had
been low-dose amitriptyline. The main concern in this scenario was about the implications for the
effectiveness of low-dose amitriptyline for IBS. Given that participants interpreted a lack of symptom
change as indicating they were taking placebo, if it turned out they were actually taking low-dose
amitriptyline then that would suggest this was not an effective treatment for them and their search for
IBS relief would continue.

If they’re not [amitriptyline], it means that I can stop taking the damn things, and if they are the real
things, well, you’re back to square one. They clearly don’t work!
P1, male aged over 65 years with severe IBS, 12-month interview

Disappointment about one’s treatment allocation

A few participants expressed slight disappointment after finding out their treatment allocation.
Participants who were disappointed to find out that they had been taking low-dose amitriptyline were
disappointed not to have experienced benefit from the medicine and/or disappointed that this was
another treatment that had not helped them.

Yes, interesting because I thought I was taking the placebo. Yes, it was interesting that I was actually
taking the real thing. I mean, it was disappointing for me that it didn’t work for me, but I mean, that’s not
to say it didn’t work for other people. I’m still glad that I took part and did it. I think just because it wasn’t
working for me, I assumed that it was the placebo, I think.
P18, female aged 25–34 years with moderate IBS, 12-month interview

92

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

No, because you know it’s a new trial, you know you can be on the placebo, so there were no expectations
there, except that you were hoping. You were hoping that this was going to make it all go away, or at least
alleviate an awful lot of it, but it didn’t. It didn’t.
P9, female aged over 65 years with moderate IBS, 12-month interview

I can’t see it getting any better. I just can’t see how – I feel like I’ve tried a lot of things, so apart from
perhaps pursuing more like a medical route, pushing my GP to maybe see a gastro doctor, I don’t really
know. I think it’s hard because you feel a bit pathetic going to your doctor saying, ‘I’m bloated’, because
they’ll be like, ‘So?’ [Both laugh] ‘You’re not going to die, you’re fine’. So I don’t feel like I really want to push
for anything extra because – it’s annoying, but it’s not like I’ve got diabetes or anything really serious. It’s
just one of those things. I can’t really see it changing. I can’t really see it getting any better. I’ll just have to
live like this forever.
P42, female aged 18–24 years with severe IBS, 6-month interview

Participants who were disappointed to find out that they had been taking placebo were mainly
disappointed that they had not had the opportunity to try low-dose amitriptyline as part of the main
ATLANTIS trial. However, they understood that their treatment had been allocated at random.

Fine. It’s part of the study. I understand these things. That’s fine, no problem. It would have been nice to
have known if I’d had the real ones whether it had any effect or not.
P27, female aged over 65 years with moderate IBS, 12-month interview

Interestingly, one participant was disappointed not to have benefitted from taking placebo suggesting
they appreciated the potential for placebo effects more than those participants who talked about
placebos in more negative ways.

I’m a very positive person and I expected it to work, and that’s why I’m jolly disappointed nothing
is happening.
P14, female aged over 65 years with severe IBS, 6-month interview

Relating the qualitative and quantitative findings

Table 41 presents a mapping of selected key quantitative findings against the qualitative findings.
Quantitatively, both arms showed some improvements in IBS symptoms over time. The qualitative
findings position this within the context of participants’ ongoing desire for symptom relief. The
qualitative findings further suggest that the increased ease of managing IBS during the pandemic
lockdowns and the valued support received from researchers in the trial might explain some of the
symptom improvements in both arms.

Quantitatively, low-dose amitriptyline improved IBS symptoms significantly more than placebo. This
pattern was also observed in participants’ qualitative talk about symptom changes during the trial.
Participants’ reflections on treatment allocation suggest some, but not all, participants may have
accurately guessed their treatment allocation; these qualitative data are insufficient to suggest whether
or how much such guessing may have contributed to between-arm differences.

Quantitatively, rates of adherence to trial medications were high. This maps to the qualitative data on
the acceptability of amitriptyline as a familiar, known, drug and the convenience of taking small tablets
on a once-daily basis.

Quantitatively, rates of questionnaire completion were high. Qualitatively, participants typically found
the questionnaires convenient and straightforward and accepted the need for these as part of the trial.
However, they also found it frustrating to have to give a binary response in the context of fluctuating
symptoms, and support from researchers may have been important in overcoming this.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 93
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 94
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Nested qualitative study

TABLE 41 Mapping themes to key quantitative findings

Quantitative finding

High rates of
Both arms showed improved IBS Low-dose amitriptyline improved IBS adherence to trial
symptoms over time symptoms significantly more than placebo medications High rates of questionnaire completion

Qualitative Barriers and Participants in both arms entered Amitriptyline was

theme/topic facilitators the trial with some concerns; these seen as a familiar
were outweighed by facilitators drug with familiar
and/or wanting relief from IBS side effects.
symptoms.

Experiences Participants in both arms valued the Participants receiving low-dose amitriptyline Participants in both Participants in both arms found the trial
in the trial accessible support received from the were more likely to report symptom arms found the questionnaires convenient. Participants in both
researchers. improvements and less likely to report no trial medications arms found binary questions frustrating and
change compared with those receiving convenient to take. difficult to answer due to the fluctuating nature
placebo. of IBS symptoms.

Reflections Some participants in both arms felt

on the the pandemic eased the impact of
pandemic IBS.

Reflections Participants attributed symptomatic

on treatment improvements or side effects to low-dose
allocation amitriptyline; participants attributed no
changes to placebo.
Not all participants who received low-dose
amitriptyline felt their symptoms had
improved.
Participants who received placebo were
disappointed not to have tried low-dose
amitriptyline.
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Discussion

Overview
This qualitative study conducted and thematically analysed 77 semistructured interviews with 42
participants and 16 GPs taking part in the ATLANTIS trial. A multidisciplinary team including patient
collaborators has explored multiple aspects of participants’ and GPs’ experiences of treatments and
participating in the ATLANTIS trial: barriers and facilitators to uptake of low-dose amitriptyline for IBS;
experiences of taking part in the ATLANTIS trial; reflections on managing IBS during the COVID-19
pandemic; reflections on treatment allocation. Each set of findings is summarised and discussed in
turn below, drawing out implications for theory, research, and/or future efforts to promote wider use
of amitriptyline for IBS where appropriate. The qualitative findings are related to the trial findings;
strengths and limitations are considered.

Barriers and facilitators

Among participants and GPs who took part in the ATLANTIS trial, potential barriers that could hinder
prescribing and uptake of low-dose amitriptyline for IBS included: participants’ and GPs’ concerns
about possible stigma associated with a medication commonly known as antidepressant; medicalising
IBS and the associated administrative burdens for GPs of increased prescribing; and side effects
including possible mental health effects (expressed by patients and linked to seeing amitriptyline as
an antidepressant) and potentially contributing to anticholinergic burden in some (older) patients
(expressed by GPs).

Stigma associated with mental illness has decreased in England in the past decade,49 but stigma
related to the use of antidepressants may be distinct from stigma related to depression.50 Notably,
participant interviewees’ concerns about the stigma of taking antidepressants did not deter them from
volunteering for the ATLANTIS trial, although of course other patients with stronger concerns may have
been deterred.

Participants and GPs were encouraged to try/prescribe amitriptyline for IBS by the low and flexible
recommended dosage, its potential to offer benefits beyond IBS symptom relief including for example,
its effects on sleep, and perceived ease of treatment (once daily dosing and small tablets) including the
participant self-titration process, which most participants found acceptable and empowering. Simple
treatment regimens may facilitate adherence to medication,51 where adherence to complex dietary
regimens such as FODMAP can be poor52 and difficult for patients to manage within the context of
daily life.46,53 Being able to reframe minor ‘adverse’ effects as potential benefits (such as effects on sleep)
could help reduce concerns and thus might increase adherence.47,54 Empowering patients to self-titrate
their dose, with the support of a dose titration document carefully developed with PPI and clinician
input, is consistent with increasing patient engagement55 and participation within patient-centred care.56

Consistent with the common-sense model of illness perception44 and previous work in IBS,46,57
participant-perceived ongoing need for symptom relief facilitated their uptake of a novel treatment,
in this case low-dose amitriptyline for IBS. Some participants expressed this in terms of wanting a
‘cure’, suggesting that they may perceive IBS as an acute condition to be cured instead of a long-term
condition to be managed.58 Our findings were also broadly consistent with the necessity-concerns
framework, according to which patients adhere to a specific medication for a specific condition when
their perceived need for the treatment is greater than their concerns about it.47 While participant
interviewees expressed concerns about low-dose amitriptyline for IBS, these were outweighed by the
perceived need and desire for symptom relief and the benefits that participants hoped for (at the start of
the trial) and experienced (during the trial).

Our findings map to three key concepts from normalisation process theory.59 The intervention
was ‘coherent’ to GPs, in that low-dose amitriptyline for IBS made sense to GP interviewees. GPs
demonstrated ‘cognitive participation’ in the intervention, in that they understood the potential benefits

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 95
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Nested qualitative study

of low-dose amitriptyline for people with IBS, appreciated its ease of use and therefore committed to it.
In order to facilitate ‘collective action’, GPs would value additional patient-facing resources to support
prescribing low-dose amitriptyline for IBS.

Barriers and facilitators to low-dose amitriptyline for IBS were identified in the context of participants’
desire for a novel approach to IBS: GPs were keen to offer more options for IBS and patients sought a
cure for their symptoms. This is consistent with previous work examining the wider impact of IBS on
patients’ lives and highlighting the challenges of treatment-seeking.46,60,61 It suggests GPs and primary
care patients with IBS may consider low-dose amitriptyline for IBS despite concerns and/or after their
concerns are appropriately addressed.

Overall, the analysis of barriers and facilitators suggests that low-dose amitriptyline for IBS is likely
to be acceptable to and, in some cases, welcomed by GPs and participants as an additional treatment
option. The familiarity of amitriptyline may both hinder uptake (given its association with depression)
and facilitate it (given that it is known and taken in a low dose, distinct from the antidepressant
dose, by others for a range of conditions and has comparatively mild and in some cases potentially
beneficial side effects such as on sleep). GPs’ and participants’ desire for another treatment option
for IBS does not obviate the need to address concerns about amitriptyline. Addressing concerns and
promoting facilitators could in turn promote wider use of low-dose amitriptyline for IBS and might be
achieved through:

• Clear communication to clinicians, for example in clinical guidelines, that distinguishes low-dose
amitriptyline for IBS from amitriptyline for other conditions (especially depression).
• Resources to support GP–patient communication to distinguish low-dose amitriptyline for IBS from
amitriptyline for other conditions (especially depression). This might include, for example, tips for GPs
discussing amitriptyline for IBS with patients, online materials to support or reinforce messages given
during consultations, tailored packaging and patient inserts, and education for pharmacists.
• Clear guidance about low-dose amitriptyline for IBS and anticholinergic burden. This should highlight
that low-dose is lower risk and that, currently, anticholinergic burden risk scores do not account for
this, thus they can overinterpret risk with low-dose amitriptyline.
• Guidance and resources for GPs and patients to support patients managing their own dose titration.

Experiences of the trial

Similar to other trial participants, participants described volunteering for ATLANTIS in the hope that
an effective treatment could be found for them and/or others with IBS, often in the context of support
from significant others.62 Similar to previous studies of patients’ experiences of IBS, participants
described experiencing and trying to monitor their fluctuating IBS symptoms.60,63 Participants found it
difficult to answer binary outcome measures that did not permit them to communicate these symptom
fluctuations; qualitative studies nested in other IBS trials have also observed the challenges for
participants of providing confident global evaluations of their condition.63 Future studies in IBS should
move away from binary outcome measures given the difficulties faced by participants completing them
and the potential for bias and/or measurement error that such difficulties may introduce. Participants
particularly valued the convenience of the tablets and brief, online, questionnaires, and the accessibility
and support received from the researchers. This valued support from the researchers, both in relation to
completing outcome measures and providing reassurance, may have contributed to the improvements
seen in the placebo arm64 and reinforces the importance of good support for patients in clinical settings
when initiating amitriptyline.

Irritable bowel syndrome during the coronavirus disease discovered in 2019 pandemic
Several quantitative survey studies have examined the impact of the COVID-19 pandemic on IBS
symptoms. Findings have highlighted negative impacts of the COVID-19 pandemic on IBS. For example,
compared with those seen in the previous 12 months, people seen in one tertiary care centre in England
for refractory IBS during the pandemic had higher IBS symptom severity and other symptoms.65 In

96

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

comparison, an international survey of 305 people self-reporting IBS found that the majority reported
no change in their IBS symptoms, while 27% reported symptom improvements and 12% reported
symptom deterioration.66 Our findings suggest a nuanced picture in which some people with IBS found
it easier to cope with their symptoms and experienced less stress and concern about their symptoms
during COVID-19 lockdowns (in the absence of eating outside the home) compared with before
COVID-19.

Treatment allocation
Trial participants appeared to construct narratives around their treatment-arm allocation based on their
early and ongoing experiences of their treatment, around their experience (or lack of experience) of
symptom improvement and side effects. Participants thought, or hoped, that they had been allocated to
low-dose amitriptyline when they perceived symptom benefit (though found this challenging to interpret
in the context of typical fluctuations) and/or side effects such as dry mouth or sleepiness. Conversely,
participants thought they had been allocated to placebo when they perceived ongoing IBS symptoms
with no noticeable improvements and/or they felt that they had not experienced any of the common
side effects associated with amitriptyline. This is consistent with evidence from other trials in IBS,
suggesting common-sense attributions of symptom changes based at least in part on patients’ beliefs
about the (beneficial and adverse) effects of the active drug compared with placebo in the context of
their illness and symptom perceptions.67–69

Only participants allocated to the placebo arm discussed experiencing a lack of side effects and
interpreting this to mean they were in the placebo arm. This is consistent with evidence suggesting
medication side effects may contribute to patient unblinding in IBS trials;70 patient unblinding may in
turn bias effect size estimates in favour of the active drug, although effective blinding of investigators is
also very important.71

As in previous trials, participants expressed curiosity about finding out their own personal treatment
allocation in the trial and sometimes expressed concerns that their perceived treatment allocation may
not match their actual treatment allocation.63,72,73 Some participants were disappointed on finding out
they were taking low-dose amitriptyline because they had not experienced benefit from the medicine
and so their search for symptom relief had to resume. Some participants were disappointed to find
out that they had been taking placebo because they had not had the opportunity to try low-dose
amitriptyline as part of the ATLANTIS trial. It is unsurprising that some participants felt disappointed in
these ways, given the refractory nature of ATLANTIS participants’ IBS and the desire for symptom relief
that underpinned decisions to volunteer for ATLANTIS. Participants’ curiosity, concerns, and reactions
to being unblinded to treatment allocation support our development and use in ATLANTIS of a novel
information leaflet to support sensitive participant unblinding to treatment allocation that addresses
common concerns (to be reported separately).

Strengths and limitations

Despite using convenience sampling instead of the planned purposive sampling, a diverse sample of
participants from each trial arm were included in the qualitative study, ensuing that our findings are
not based on a narrow subset of the people who took part in the trial. Although we achieved a diverse
range of views, participants willing to undertake the qualitative interviews may not hold the same views
as all the participants in the study. Unfortunately, no GPs from the West Yorkshire hub agreed to be
interviewed: one declined the invitation without giving a reason and one was part of the qualitative
team; no other responses were received. There were fewer participating practices and GPs in West
Yorkshire compared with the other two hubs, and West Yorkshire practices were more severely impacted
by longer periods in lockdown, limiting capacity for research activity.

Conducting interviews with the same participants repeatedly over time permitted the development of
rapport and the elicitation of experiences over a longer time period than is often achieved. Future trials
might benefit from including a very early interview time point (e.g. at 2–4 weeks) to explore participants’

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 97
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Nested qualitative study

initial impressions and experiences of trial interventions; in the case of ATLANTIS, this would have
enabled us to capture participants’ perspectives on titrating closer in time to when they were engaged in
this aspect of the trial.

Interviewing participants and GPs involved in ATLANTIS permitted a more comprehensive analysis of
barriers and facilitators to low-dose amitriptyline for IBS than would have been possible by including
only one of these groups. This was essential for developing insights and recommendations for future
widespread implementation of low-dose amitriptyline for IBS.

The multidisciplinary nature of the qualitative team including input from patient collaborators meant
that we approached the data from diverse perspectives, achieving richer insights than would otherwise
have been possible.

Including stand-alone objectives for the qualitative component, as well as relating the qualitative
findings to the quantitative findings, maximised the value of adding a qualitative component to a major
RCT. Remaining blinded to treatment allocation and not knowing the main trial results while conducting
the primary qualitative data analysis ensured interpretations were not shaped by researchers’
perspectives on treatment allocation and efficacy.74 Subsequent unblinding of researchers facilitated
deeper qualitative analysis and mapping of qualitative to quantitative findings, thus enhancing the value
of the nested qualitative work and increasing integration during the interpretation phase.75

Conclusion
This qualitative study has explored participants’ and GPs’ experiences of treatments and participating
in the ATLANTIS trial. Findings build on previous work examining participants’ experiences of IBS in
the community and in other clinical trials. Facilitators and barriers to prescribing and uptake of low-
dose amitriptyline in IBS have been identified and explored within the broader context of participants’
and GPs’ experiences of living with and seeking treatment for IBS. This analysis has generated
recommendations to support dissemination activities to enable wider use of low-dose amitriptyline for
IBS which in turn should provide more management options and patient benefit. Analysis of participants’
experiences of the trial including outcome reporting, researcher support, and treatment allocation
has also generated insights that can inform future research. Future work should develop resources to
implement recommended actions to support wider use of low-dose amitriptyline for IBS.

98

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Chapter 5 Discussion
ATLANTIS was a pragmatic, double-blind randomised trial to assess the effectiveness and cost-
effectiveness of low-dose amitriptyline as a second-line treatment for IBS in primary care. To our
knowledge, it is the largest trial of a TCA in IBS ever undertaken and the first to be based entirely in
primary care. The trial was designed to address a key research priority identified by NICE guidance
for management of IBS in primary care in 2015.16 The trial recruited participants who had ongoing
troublesome symptoms despite trying first-line therapies, with very similar characteristics to those
with a diagnosis of IBS on GP records who were invited by letter to participate, meaning the results are
relevant and generalisable to usual clinical practice in primary care. The median duration of symptoms
was 10 years, and more than 80% had moderate to severe symptoms.

In this population, low-dose amitriptyline was shown to be superior for the trial’s primary outcome, with
a significant difference in mean IBS-SSS score between arms of 27.0, and a mean decrease in IBS-SSS
of almost 100 points at both 3 and 6 months, compared with baseline. Low-dose amitriptyline was
also superior to placebo for the key secondary outcome for effectiveness, with an OR for SGA of relief
of global IBS symptoms of 1.78. Amitriptyline also improved other IBS symptom outcomes, including
adequate relief of IBS symptoms and a ≥ 30% decrease in abdominal pain on the IBS-SSS at 6 months.
However, there was no significant effect of low-dose amitriptyline on abdominal distention, somatoform
symptom-reporting scores, or anxiety or depression scores, nor was there any impact on work and social
activities during 6-month follow-up.

In the subset of participants recruited to 12 months of follow-up, and with the choice to continue
treatment beyond 6 months, 44% of participants completed 12 months treatment. Despite the mixed
sample, weak evidence of a significant effect in favour of low-dose amitriptyline remained on the
IBS-SSS, with a mean difference of 22.6 and on the SGA of relief of global IBS symptoms with an OR of
1.58. In contrast to the 6-month results, there was a statistically significant effect in favour of low-dose
amitriptyline on HADS-depression scores, with a mean difference of −0.88, and on WSAS scores, with a
mean difference of −2.14.

Significantly more participants randomised to low-dose amitriptyline found it acceptable to take than
placebo and almost three-quarters adhered to the drug during the 6-month trial. Although AEs were
more frequent with low-dose amitriptyline, the drug was generally safe and well-tolerated. The AEs
reported in participants receiving amitriptyline in excess of those reported by the placebo arm mainly
related to its known anticholinergic effects, including drowsiness and dry mouth, but most participants
reported these as mild, and SAEs and SARs were rare. However, withdrawals due to AEs were slightly
more frequent with low-dose amitriptyline, occurring in 12.9% assigned to amitriptyline compared with
8.7% in the placebo arm.

The 6-month duration of treatment in ATLANTIS is in line with European Medicines Agency (EMA)
recommendations for IBS treatment trials76 and is longer than most drug trials in IBS, where efficacy is
usually assessed over 12 weeks. The results of the trial are, therefore, likely to be more representative
of the effectiveness of low-dose amitriptyline in IBS, a condition that, for many people, is a relapsing
and remitting disorder.15 The outcomes studied are ones that are accepted widely in pragmatic trials
conducted in IBS, including a mean change in the total IBS-SSS and adequate relief of symptoms of IBS.
Follow-up rates for participant-reported outcomes at 6 months were high (over 85%). As the analyses
were conducted on all participants, irrespective of adherence, and with imputation of missing data, it is
unlikely that the effectiveness of low-dose amitriptyline for IBS in primary care has been overestimated.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 99
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Discussion

Limitations

The primary outcomes in the ATLANTIS trial differed from those recommended for drug trials in IBS
by the Food and Drug Administration (FDA) in the USA and the EMA in Europe.76,77 It would have been
impractical to adhere to these in a pragmatic 6-month trial recruiting participants in primary care for
the following reasons. First, these are IBS subtype-specific, which would have meant using different
outcome measures in different groups of participants. Second, we recruited participants with IBS of
all subtypes, including IBS-M or IBS-U, but there is no clear consensus on recommended end points
for these two subtypes. Third, completion of a daily diary, which would be required to assess FDA or
EMA end points, would have been too burdensome for participants for a 6-month period. However, the
secondary outcome of adequate relief of IBS symptoms, which allows relief for 50% of weeks of the trial
to be assessed, and the exploratory outcome of a ≥ 30% improvement in abdominal pain on the IBS-SSS,
were both significantly higher with low-dose amitriptyline. These approximate to the FDA and EMA
recommended end points and are more stringent without requiring completion of a daily diary outcome
specific to IBS subtype.

Over 80% of recruited participants had either IBS-D or IBS-M. This means it may be harder to judge the
effectiveness of low-dose amitriptyline definitively in those with IBS-C or IBS-U. However, although
exploratory moderator analysis found no statistical evidence of a moderating effect, larger treatment
effects, in favour of amitriptyline, were observed in participants with IBS-C or IBS-D compared with
those with IBS-M or IBS-U on the IBS-SSS, and in participants with IBS-D compared with those with
IBS-C, IBS-M or IBS-U on the SGA of relief of IBS symptoms. As we were not powered for the moderator
analysis, the trial results cannot be used to make inferences about effectiveness of amitriptyline
according to IBS subtype. Similarly, we are also likely to be underpowered for some of our secondary
end points, including effect on anxiety and depression scores. These may be expected to improve as
symptoms of IBS improve but, perhaps, to a lesser degree, and therefore the differences we detected
were not statistically significant.

Although the primary outcome was at 6 months, we had planned to offer all participants the option to
continue blinded trial medication until 12 months and follow up all participants until 12 months post
randomisation. However, the COVID-19 pandemic interrupted the delivery of the trial, and a funded
extension was required to complete recruitment. This meant that follow-up after 7 October 2021 was
curtailed to 6 months for new participants to minimise additional funding required to complete the trial
and to prioritise funds for participant recruitment. Thus, interpretation of the 12-month outcomes is
difficult, as there were fewer participants able to choose to continue trial medication than anticipated.
For similar reasons, the health economic analyses were removed from the trial. This means that we could
not perform a cost-effectiveness analysis as part of the trial. However, given that amitriptyline is cheap,
we do not feel this should be a barrier to the implementation and uptake of the findings of the trial. In
addition, we cannot exclude that the COVID-19 pandemic itself may have had an impact on trial results,
and that participants in both arms experienced a larger improvement in IBS symptoms than expected
due to the reduced ability to work and socialise for long periods of time.

Another limitation is that participants entering the trial were not ethnically diverse, which may limit
generalisability of the findings to these populations – see Equality, diversity and inclusion for more detail.
Finally, because we used the Rome IV criteria to define IBS, the generalisability of the results to patients
with a GP’s diagnosis of IBS, who may not meet these criteria, is uncertain.

Generalisability

The definition of IBS used consisted of the current recommended symptom-based criteria, the Rome
IV criteria,34 together with limited diagnostic testing to exclude known organic ‘mimics’ of IBS in all
participants. This is in line with current UK guidance for the diagnosis of IBS.16,78 Participants were

100

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

recruited irrespective of IBS subtype, with symptoms ranging from mild to severe, and these came from
a broad range of general practices in three different geographical regions in the UK. Amitriptyline was
provided in addition to usual care and current first-line IBS medications, and participants were able to
self-titrate their dose between 10 mg on alternate days and 30 mg every day according to symptom
response and side effects. This is how amitriptyline is commonly managed for other conditions in
primary care and is pragmatic, reflecting usual clinical practice outside a trial setting.

The results of the trial are, therefore, likely to be generalisable to many patients with IBS in UK primary
care who have not experienced adequate improvement in their symptoms with first-line therapies. The
qualitative findings on guidance and resources needed for improved clinician-patient communication
to distinguish low-dose amitriptyline for IBS from use as an antidepressant and to support patients
managing their own dose titration are also likely to be generalisable to clinical interactions in secondary
care settings.

Interpretation

The Rome IV criteria are known to select a group of patients with higher symptom severity than
previous iterations.79 This is borne out by the mean IBS-SSS scores seen at baseline, which were in the
moderate to severe range in more than 80% of participants, and the median duration of IBS among
participants was 10 years. Given this, and the relatively long treatment duration of 6 months, the
placebo response rates observed in ATLANTIS may appear relatively high, with 36% of those assigned
to the placebo arm reporting that they thought the trial medication had worked. However, the placebo
response in trials in IBS is known to be high,80,81 and there is evidence that patients with IBS are
more likely to respond to a placebo than to a no-treatment control intervention.82,83 Other possible
explanations for this within the trial design include the fact that all participants were provided with
the NICE-approved BDA dietary advice sheet, that regular follow-up and usual GP care were allowed
throughout the trial, and the regular supportive telephone calls from a research nurse to assist with dose
titration and re-supply of trial medication. In addition, as demonstrated by the nested qualitative study
results, participants may have felt a sense of empowerment and being more in control of their symptoms
because they were able to self-titrate their medication dose during the trial in response to symptoms
and side effects. Finally, a regression towards the mean effect during follow-up is well-recognised in
clinical trials. However, all of this makes the highly statistically significant benefit seen across almost all
IBS outcome measures particularly noteworthy.

Several prior meta-analyses of TCAs in IBS have demonstrated that these drugs, as a class, are superior
to placebo.7,9,84 However, there are limitations of the trials that have been included in these meta-
analyses. Most trials, to date, have been relatively small. None of the trials have exceeded a maximum
treatment duration of 3 months, nor have any previous trials been conducted entirely in primary care.
Finally, many have used definitions of IBS that are no longer in active use and have assessed efficacy
using outdated or non-rigorous outcomes. The largest RCT conducted to date recruited 216 female
patients with various functional bowel disorders in the USA, 172 of whom had IBS.85 This trial used the
TCA desipramine, commencing at a dose of 50 mg o.d. for 1 week, titrated up to 150 mg o.d. by week 3.
As in ATLANTIS, the commonest side effects related to the anticholinergic effects of the drug, including
dry mouth in 48% and drowsiness in 20% of participants. Rates of discontinuation of desipramine
due to AEs were higher than observed in ATLANTIS, at 20%. This may relate to the higher dosage of
desipramine used, compared with the low dose of amitriptyline studied here. In this trial, the primary
outcome, which was a composite of patient satisfaction, improvement in symptoms, and increased
engagement in social activities, was not met. There was a 60% response rate with desipramine versus
47% with placebo (p = 0.128). However, in subgroup analysis desipramine was superior to placebo in
those with moderate, rather than severe, symptoms and in those with IBS-D. Interestingly, despite the
high dose of desipramine used, the presence of abnormal depression scores at baseline had no impact
on treatment response. In another 8-week trial recruiting 54 patients with IBS-D in secondary care in

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 101
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Discussion

Iran,20 the response rate with amitriptyline 10 mg o.d. was 70%, compared with 41% with placebo, but
this difference was not statistically significant at the 5% level (p = 0.054). Based on our required sample
size, it is likely the trial was underpowered. AE rates in this trial were similar between treatment arms.
Treatment effects in ATLANTIS on our primary outcome, the IBS-SSS, were generally larger in those
with IBS-C or IBS-D, those with lower baseline HADS-anxiety scores, and those with higher baseline
IBS-SSS scores.

The magnitude of the observed difference in treatment effect on the IBS-SSS increased between the
3- and 6-month follow-up points from 23 points to 27 points. In addition, the difference in rates of
a ≥ 30% improvement in abdominal pain on the IBS-SSS between amitriptyline and placebo was only
statistically significant at 6 months, as the placebo response in terms of abdominal pain fell from 46.9%
to 42.6% and increased in the amitriptyline arm from 52.3% to 55.7%. These observations highlight
the importance of allowing adequate time for low-dose amitriptyline to have a beneficial effect on
symptoms in IBS and are compatible with reports of a decrease in placebo response rates as trial
duration increases.80

There was no significant effect of low-dose amitriptyline on somatoform symptom-reporting, anxiety, or

depression scores during the 6 months of the trial. This supports previous hypotheses that the benefit
of low-dose amitriptyline in IBS is arising from a combination of its peripheral actions on gastrointestinal
motility and pain sensation,86,87 rather than via a reduction in extra-intestinal symptom-reporting, or
an improvement in symptoms of anxiety or depression, all of which are well-recognised to associate
strongly with a diagnosis of IBS.88,89 Additionally, there was no significant impact on either ability to work
or social functioning, according to WSAS scores. However, the reduction in scores was generally greater
among participants randomised to low-dose amitriptyline. Although IBS can have substantial impacts on
these measures of activities of daily living,90,91 the treatment duration may have been too short to detect
any meaningful improvements, as supported by superior effects detected on the WSAS and HADS-D
with amitriptyline at 12, but not 3 or 6, months. Additionally, we cannot exclude the possibility that the
COVID-19 pandemic had an impact on this measure.

The flexible dosing schedule allowed participants to increase or reduce their dose according to symptom
response and side effects. It is interesting to note that by 3 months 57.0% of participants in the placebo
arm had already titrated to 30 mg o.d., whereas only 37.8% of those randomised to amitriptyline had
reached this. Although 42.8% were taking 30 mg o.d. of amitriptyline by 6 months, almost one-in-four
participants were only taking 10 mg o.d. This flexible dosing may also have contributed to tolerability.
Amitriptyline at low dose was safe and well-tolerated by participants. There were few SAEs or SARs,
no SUSARs or deaths, no pregnancies, and no emergency unblinding events. Although treatment-
emergent AEs were generally higher among those receiving amitriptyline most of these were mild, and
the overall rates of AEs fell between 3 and 6 months. However, to some extent this is not unexpected
as participants reporting AEs may have been more likely to stop treatment before 6 months and
ASEC data were only collected for those still on treatment. Some of the side effects reported, such as
constipation and diarrhoea, are known symptoms of IBS, and many of the other possible side effects on
the ASEC were reported at a similar, or higher, frequency by the placebo arm. In any case, participants
were significantly more likely to report that they found amitriptyline acceptable to take, compared with
placebo, at 6 months. Finally, adherence rates were high in the amitriptyline arm: 83.2% at 3 months
and 74.1% at 6 months. This underlines that if the rationale for the use of a TCA for IBS is explained
clearly, as in the information materials provided to participants in this trial, with appropriate support, the
majority of people are willing and able to take the drug.

Overall evidence

The ATLANTIS trial provides strong evidence for the effectiveness of low-dose amitriptyline, at a dose
of between 10 mg o.d. and 30 mg o.d., as a second-line treatment for IBS in primary care, when first-line

102

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

treatment has not led to an adequate improvement in symptoms. Amitriptyline was more effective
than placebo across a range of IBS symptom measures, and was safe and well-tolerated, when titrated
according to symptom response and side effects. Therefore, GPs should offer low-dose amitriptyline
to patients who have ongoing troublesome symptoms despite trying first-line therapies. Management
guidelines for IBS in primary care need to change to reflect the findings of this definitive trial.
Additionally, as mentioned in the qualitative chapter, guidance and resources are needed to support
GP–patient communication to distinguish low-dose amitriptyline for IBS from use as an antidepressant
and to support patients managing their own dose titration.

Recommendations for future research

It is important to assess the health economic benefits of low-dose amitriptyline for IBS, especially in
light of the positive trial results. Although health economic data were collected during the ATLANTIS
trial, unfortunately the analysis was unable to be conducted due to the impact of COVID-19 on the trial,
with funding being redirected to participant recruitment. Further funding is needed to make use of these
valuable data.

ATLANTIS has demonstrated that low-dose amitriptyline is superior to placebo as a second-line

treatment for IBS in secondary care. However, it is unclear whether there are particular groups
of patients who are more likely to respond to the drug. It is also unclear whether other low-dose
antidepressant drugs, such as selective serotonin reuptake inhibitors or serotonin norepinephrine
reuptake inhibitors, are also effective for IBS in primary care. In addition, the question remains whether
using low-dose amitriptyline as a first-line treatment, potentially earlier in the course of IBS symptoms,
is an effective strategy in primary care.

Equality, diversity and inclusion

The participants recruited were not ethnically diverse, despite considerable efforts to reach out
to ethnic minorities with IBS during the trial. Given that a recent global survey demonstrated that
the prevalence of IBS is similar across multiple countries,2 this likely represents under-sampling of
this population due to inequities in, or barriers to, accessing health care and research, rather than
differences in the epidemiology of IBS. However, unlike many treatment trials in IBS, where often
80–90% of participants are female, more than 30% of recruited participants were male, there were a
wide range of ages and educational backgrounds among participants, and different deprivation indices
were well-represented. In addition, the trial was conducted in three distinct geographical regions
and those recruited were very similar in age and gender to those invited after searching for an IBS
diagnosis on GP lists.

Patient and public involvement

Patient and public involvement representatives were involved prior to the grant application stage
and throughout the trial, analysis, and interpretation of the results. They will continue to play a very
important role in dissemination of the results.

The trial team includes a public co-investigator who leads ‘Let’sCureIBS’, a local patient support group.
He provided valuable, personal experience of IBS and facilitated wider engagement with group members
at key points throughout the study. The CTRU PPI lead developed and implemented PPI plans and
provided support to public contributors. Examples of PPI input and impact include:

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 103
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Discussion

• Study design – a workshop was held with patients at the grant application development stage, which
informed key design and operational decisions: for example, participants being given the choice to
cease or continue treatment at 6 months, the use of electronic questionnaires, additional participant
support from GPs and researchers, and medication being sent to people’s homes via post.
• Project management – our public co-investigator joined the TMG and qualitative subgroup, enabling
us to seek a patient perspective on emerging issues throughout the study.
• Project oversight – an additional public contributor joined the TSC, ensuring that the patient
perspective was included in high-level decisions about the study.
• Site set-up – one hub conducted a workshop with research staff who would be recruiting to
the study. The workshop was attended by a public contributor and included role-play scenarios
co-developed by patients, which enabled staff to rehearse recruitment conversations.
• Participant communication – PPI was an important part of shaping all participant communication,
including invitation and thank you letters, dose titration guidance, and unblinding information. PPI
feedback was particularly valuable during development of the participant information sheets. For
example, people highlighted the need to explain clearly why amitriptyline might be used for IBS:
that is, due to its impact on pain and bowel function rather than mood. This is important due to the
potential stigma associated with a drug also used as an antidepressant.
• Data collection – the online system used for PROM completion was user-tested and PPI helped
shape the instructions given to participants.
• Nested qualitative study – PPI input guided the development of GP and patient interview topic
guides. Public contributors also provided general advice to interviewers; for example, how
to approach the topic of IBS in a sensitive manner, and the importance of being clear about
the boundaries of their role and the fact they are not able to provide medical advice. A public
co-investigator was also involved in qualitative data analysis, contributing to the development of a
coding framework, themes, and interpretation.
• Interpretation and dissemination of results – an interpretation workshop was held, where preliminary
results were presented to public contributors. They were asked to consider what the results mean for
patients and the NHS and how they should be shared. That workshop has influenced the conclusions
presented in this monograph, as well as our ongoing dissemination plans. The group highlighted some
important dissemination messages for patients and GPs:
◦ Titration is useful and needs to be supported by GPs and good quality information.

◦ Patients need to be made aware that it may take some time to get the right dose for them and feel

the benefit of amitriptyline.

◦ GP contact and support are an important part of this intervention.

◦ The fact that HADS scores did not improve is very relevant to patients, as it supports the fact that

this is not a psychological intervention.

Conclusions

This rigorously conducted, pragmatic trial is the largest trial of amitriptyline for IBS undertaken to date,
worldwide and the first in primary care. It provides definitive results indicating that GPs should offer
low-dose amitriptyline to patients with IBS whose symptoms do not improve with first-line therapies.
This recommendation should be widely disseminated to clinical settings in primary and secondary care
and incorporated into guidelines for IBS management. We will publicise the results to participants, and
other people with IBS, via the ATLANTIS trial website (https://ctru.leeds.ac.uk/atlantis/) and X (formerly
Twitter) account (@ATLANTISTrial).

Guidance and resources are needed to support GP–patient communication to distinguish low-dose
amitriptyline for IBS from its use as an antidepressant and to support patients managing their own dose
titration. The dose titration document successfully used by participants in this trial is included in Report
Supplementary Material 1.

104

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Additional information
CRediT contribution statement

Alexandra Wright-Hughes (https://orcid.org/0000-0001-8839-6756): Data curation (lead), Formal

analysis, Methodology, Resources, Software, Validation, Visualisation (lead), Writing – original draft,
Writing – editing and reviewing.

Alexander C Ford (https://orcid.org/0000-0001-6371-4359): Conceptualisation (lead), Funding

acquisition (lead), Investigation, Methodology, Resources, Supervision, Visualisation, Writing – original
draft, Writing – editing and reviewing.

Sarah L Alderson (https://orcid.org/0000-0002-5418-0495): Conceptualisation, Funding acquisition,

Investigation, Methodology, Resources, Supervision, Writing – editing and reviewing.

Pei Loo Ow (https://orcid.org/0000-0001-6025-6372): Data curation, Formal analysis, Methodology,

Resources, Software, Visualisation, Writing – editing and reviewing.

Matthew J Ridd (https://orcid.org/0000-0002-7954-8823): Funding Acquisition, Investigation,

Methodology, Resources, Supervision, Writing – editing and reviewing.

Robbie Foy (https://orcid.org/0000-0003-0605-7713): Conceptualisation, Funding acquisition,

Investigation, Methodology, Resources, Supervision, Writing – editing and reviewing.

Felicity L Bishop (https://orcid.org/0000-0002-8737-6662): Conceptualisation, Funding acquisition,

Investigation, Methodology, Resources, Supervision, Visualisation, Writing – original draft, Writing –
editing and reviewing.

Matthew Chaddock: Funding acquisition, Methodology, Writing – editing and reviewing; PPI activities.

Heather Cook (https://orcid.org/0000-0003-0223-417X): Investigation, Methodology, Project

administration, Resources, Writing – editing and reviewing.

Deborah Cooper: Investigation, Resources, Writing – editing and reviewing.

Catherine Fernandez (https://orcid.org/0000-0001-8561-5642): Investigation, Methodology, Project

administration, Resources, Writing – editing and reviewing.

Elspeth A Guthrie (https://orcid.org/0000-0002-5834-6616): Conceptualisation, Funding acquisition,

Methodology, Writing – editing and reviewing.

Suzanne Hartley (https://orcid.org/0000-0003-2346-9461): Funding acquisition, Methodology, Project

administration, Supervision, Writing – editing and reviewing.

Amy Herbert (https://orcid.org/0009-0008-6109-6006): Investigation, Resources, Writing – editing

and reviewing.

Daniel Howdon (https://orcid.org/0000-0001-8052-2893): Methodology, Resources, Writing – editing

and reviewing.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 105
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Additional information

Delia P Muir (https://orcid.org/0000-0003-1136-3416): Funding acquisition, Methodology, Writing –

editing and reviewing, PPI activities.

Sonia Newman (https://orcid.org/0009-0007-6827-2861): Investigation, Resources.

Christopher A Taylor (https://orcid.org/0009-0009-8220-4923): Methodology, Project administration,

Resources, Software, Writing – editing and reviewing.

Emma J Teasdale (https://orcid.org/0000-0001-9147-193X): Formal Analysis, Investigation, Resources,

Visualisation, Writing – original draft, Writing – editing and reviewing.

Ruth Thornton (https://orcid.org/0009-0006-4041-0233): Investigation, Resources, Writing – editing

and reviewing.

Hazel A Everitt (https://orcid.org/0000-0001-7362-8403): Conceptualisation (lead), Funding acquisition

(lead), Investigation, Methodology, Resources, Supervision, Visualisation, Writing – original draft, Writing
– editing and reviewing.

Amanda J Farrin (https://orcid.org/0000-0002-2876-0584): Conceptualisation (lead), Data curation,

Formal analysis (lead), Funding acquisition (lead), Methodology (lead), Resources, Supervision, Validation,
Visualisation, Writing – original draft, Writing – editing and reviewing.

Sandy Tubeuf: Funding acquisition, Methodology, Investigation.

Gina Bianco: Investigation.

Taposhi Nath: Investigation.

Amy West: Investigation.

Sarah T Brown: Methodology.

Acknowledgements

Trial Steering Committee

Dr Tim Holt, TSC Chair (University of Oxford), Professor David Sanders (Sheffield Teachings Hospitals
NHS Foundation Trust), Professor Kerry Hood (Cardiff University), Dr Maureen Twiddy (University of
Hull), Dr Peter Paine (Salford Royal NHS Foundation Trust), Asociate Professor Mara Violato (University
of Oxford), Dr Iryna Schlackow (University of Oxford), Ms Jill Durnell (PPI Representative).

Data monitoring ethics committee

Professor Peter Whorwell DMEC Chair (Wythenshawe Hospital), Miss Natalie Rowland (University of
Birmingham), Professor Christopher Burton (University of Sheffield), Professor Peter Bower (University
of Manchester).

Leeds Institute of Clinical Trials Research

Florence Day, Catriona Parker, Sarah Brown, Tom Morris, Richard Brindle, Alasdair Fellows, Jake
Emmerson, Taposhi Nath, Thomas Smith, Amy West, Aaron Dowse, Aimee Christodoulou, Sandra Lopes
Goncalves Graca, Gina Bianco, Maggie Barratt, Aminah Malik, and Damien Hindmarch.

University of Leeds
Sandy Tubeuf and Roberta Longo.

106

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Clinical Research Networks

With thanks to the staff at the Clinical Research Networks who supported the hub research staff with
the collection of trial data across the three locations:

Wessex Clinical Research Network;

West of England Clinical Research Network;

Yorkshire and Humber Clinical Research Network.

Recruiting general practitioners

With thanks to all the research staff at the participating centres who provided and cared for trial
participants and collected trial data:

Wessex
Friarsgate Practice, Dr Stephen Fowler and Dr Stephanie Hughes

Swanage Medical Practice, Dr Claire Hombersley

Highcliffe Medical Centre, Dr Zelda Cheng

Gratton Surgery, Dr Laila Nicholson

Park and St Francis Surgery, Dr Boon Yong

Chawton Park Surgery, Dr Emma Bowen-Simpkins

Liphook and Liss surgery, Dr Anna Lalonde

Oaks Healthcare, Dr Nicola Millen

Highlands Practice, Dr Rem Lee

Westlands Medical Practice, Dr Helen Pandya

The Adam Practice, Dr Patrick Moore

Trafalgar Medical Group Practice, Dr Shruti Singh

The Bosmere Medical Practice, Dr Dirk Konig

Emsworth Surgery, Dr Andrew Slingsby

Abbeywell Surgery, Dr Nicola Lester

Solent GP Surgery, Dr Tara Watson

Wareham Surgery, Dr Nathan Francis

Raymond Road surgery, Dr Stuart Robinson

Brook House Surgery, Dr Stuart Robinson

Three Chequers Medical Practice, Dr Craig Kyte

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 107
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Additional information

West of England
Portishead Medical Group, Dr Matthew Ridd

Tyntesfield Medical Group, Dr Edward Mann

Greenway Community Practice, Dr Liz Grimshaw

Mendip Vale Medical Practice, Dr Richard Reed

Westbury on Trym Primary Care Centre, Dr Jenny Eachus

Nightingale Valley Practice, Dr Katharine Alsop

Fishponds Family Practice, Dr Simon Atkins

Whiteladies Medical Group, Dr Stephen Granier

The Family Practice, Dr James Baker

Beechwood Medical Practice, Dr Justine de Mink

Fallodon Way Medical Centre, Dr Katia Chapman

Clevedon Medical Centre, Dr Ann-Marie Streeton

Pembroke Road Surgery, Dr Rohan Perera

Horfield Health Centre, Dr Farida Ahmad

Monks Park (PIC to Mendip Vale Medical Practice), Dr Richard Reed

Frome Valley Medical Centre, Dr Jane Goram

Student Health Services, Dr Bushra Shahid

Trowbridge Health Centre, Dr Tobias Cookson

Phoenix Health Group, Dr Naomi Vernon

Hathaway Surgery, Dr Phillip Grimmer

Close Farm Surgery, Dr Ashutosh Singh

CONCORD Medical Centre, Dr Alastair Hay

West Yorkshire
Craven Road Medical Practice and Hollybank Surgery, Dr Robbie Foy

Blackburn Road Medical Centre, Dr Nitish Singh

Croft Medical Centre (formerly Bradford Road Medical Centre), Dr MB Ahmed

Oaklands Health Centre, Dr Sarah Alderson

108

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Meltham Road Surgery, Dr Naim Hasanie

Slaithwaite Health Centre (accruals to go through Oaklands Health Centre), Dr Sarah Alderson

Haworth Medical Practice, Dr Tessa Mayo

Affinity Care Bradford – Shipley Medical Practice, Dr Ikechukwuka Nwachukwu

Affinity Care Bradford – Willows Medical Centre, Dr Ikechukwuka Nwachukwu

Affinity Care Bradford – Cowgill Surgery, Dr Ikechukwuka Nwachukwu

Affinity Care Bradford – Haigh Hall Medical Centre, Dr Ikechukwuka Nwachukwu

Affinity Care Bradford – Thornton Medical Practice, Dr Ikechukwuka Nwachukwu

Affinity Care Bradford – Sunnybank Medical Centre, Dr Ikechukwuka Nwachukwu

Patient and public involvement

Mr Matthew Chaddock, Ms Jill Durnell, and Ms Emma-Jane Williamson.

Participants
We are grateful to all the trial participants for their essential contribution to the trial.

Data-sharing statement

All data requests should be submitted to the corresponding author for consideration and would be
subject to review by a subgroup of the trial team, which will include the data guarantor. Access to
anonymised data may be granted following this review. All data-sharing activities would require a
data-sharing agreement.

Ethics statement

Ethical approval for the study was given by the Yorkshire and The Humber Sheffield Research Ethics
Committee on 5 August 2019 (reference number 19/YH/0150). Confirmation of capacity and
capability was obtained from the recruiting centres as well as the PICs in primary care. The trial was
registered with the International Standard Randomised Controlled Trial Register under the reference
number ISRCTN48075063.

Information governance statement

The University of Leeds organisation/institution is committed to handling all personal information in

line with the UK Data Protection Act (2018) and the General Data Protection Regulation (EU GDPR)
2016/679. Under the Data Protection legislation, the University of Leeds is the Data Controller, and you
can find out more about how we handle personal data, including how to exercise your individual rights
and the contact details for our Data Protection Officer here: https://dataprotection.leeds.ac.uk/.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 109
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Additional information

Disclosure of interests

Full disclosure of interests: Completed ICMJE forms for all authors, including all related interests, are
available in the toolkit on the NIHR Journals Library report publication page at https://doi.org/10.3310/
BFCR7986.

Primary conflicts of interest: Alexandra Wright-Hughes: NIHR grant funding paid to her institution,
Data Monitoring and Ethics Committee and Trial Steering Committee member of NIHR and MRC funded
projects, travel reimbursement for expert Committee membership of the Yorkshire and Northeast
Regional Advisory Committee for NIHR Research for Patient Benefit, and payment to her institution for
role as protocol editor for Trials journal. Alexander C Ford: NIHR grant funding paid to his institution.
Sarah L Alderson: NIHR, YCR, and Health Data Research UK grant funding paid to her institution,
consulting fees from West Yorkshire Integrated Care Board paid to her institution, speaker’s payments
from Xytal, member of an HS&DR Funding Committee, and Data Monitoring and Ethics Committee
member for an NIHR-funded study. Pei Loo Ow: none. Matthew J Ridd: NIHR grant funding paid to
his institution, Co-Chair for SAPC and NIHR SPCR skin/allergy research groups, committee member
for NIHR In Practice Fellowships, member of an HTA General Committee, an ESP – Evidence Synthesis
Programme Advisory Group, an ESP – Evidence Synthesis Programme Grants Committee, and an ESP –
NIHR Incentive Awards Committee. Robbie Foy: NIHR (HTA, PGfAR and HSDR) and YCR grant funding
paid to his institution, Chair of NICE Implementation Strategy Group, member of an NIHR Dissemination
Centre Advisory Group, member of UK Harkness Fellowship selection Committee, and Chair of
Independent Steering Groups and Data Monitoring and Ethics Committee member for NIHR-funded
studies. Felicity L Bishop: none. Matthew Chaddock: none. Heather Cook: none. Deborah Cooper: none.
Catherine Fernandez: none. Elspeth A Guthrie: NIHR and Leeds Hospitals Charity grant funding paid to
her institution. Suzanne Hartley: none. Amy Herbert: none. Daniel Howdon: NIHR and ESRC funding
paid to his institution, consulting fees from Organisation for Economic Co-operation and Development,
and United Nations Asia Pacific Region, speaker’s payment paid to institution from University of
Lucerne. Delia P. Muir: none. Sonia Newman: member of an HTA MNCH Methods Group. Christopher
M Taylor: none. Emma J Teasdale: none. Ruth Thornton: none. Hazel A Everitt: NIHR grant funding paid
to her institution, personal/institutional income received as a result of a licence agreement with Mahana
Therapeutics, and consulting fees/share options paid from Mahana Therapeutics. Amanda J Farrin: NIHR
grant funding (HTA, EME, PGfAR, HS&DR, NIHR/MRC) paid to her institution, Data Monitoring and
Ethics Committee and Trial Steering Committee member of NIHR and BHF funded projects, NIHR senior
investigator, and member of an NIHR CTU Standing Advisory Committee, an HTA Funding Committee
Policy Group (formerly CSG), an HTA Clinical Evaluation and Trials Committee, and a Prophylaxis
Platform Study Funding Committee.

Publications

Alderson SL, Wright‑Hughes A, Ford AC, Farrin A, Hartley S, Fernandez C, et al. Amitriptyline at low-
dose and titrated for irritable bowel syndrome as second-line treatment (the ATLANTIS trial): protocol
for a randomised double-blind placebo-controlled trial in primary care. Trials 2022;23:552. https://doi.
org/10.1186/s13063-022-06492-6

Ford AC, Wright-Hughes A, Alderson SL, Ow P-L, Ridd MJ, Foy R, et al.; ATLANTIS trialists. Amitriptyline
at low-dose and titrated for irritable bowel syndrome as second-line treatment in primary care
(ATLANTIS): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet 2023;402:1773–85.
https://doi.org/10.1016/S0140-6736(23)01523-4.

110

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Conferences

• 6th International Clinical Trials Methodology Conference – ‘Methodological considerations in the

unblinding of participants in a double-blinded randomised controlled trial: why, when, and how’.
Poster presented, 3–6 October 2022.
• British Society of Gastroenterology – Amitriptyline at Low-dose and Titrated for Irritable Bowel
Syndrome as Second-line Treatment in primary care (ATLANTIS): A Double-blind Placebo-controlled
Trial (main results). Oral presentation, 21 June 2023, Liverpool, UK. Awarded prize for ‘Best Neuro-
gastroenterology Oral Presentation’.
• Society of Academic Primary Care Conference – Amitriptyline at Low-dose and Titrated for Irritable
Bowel Syndrome as Second-line Treatment in Primary Care (The ATLANTIS Study): A Double-blind
Placebo-controlled Trial (main results). Oral presentation, 18–20 July 2023, Brighton, UK.
• Society of Academic Primary Care Conference – Low-dose amitriptyline for irritable bowel
syndrome (IBS): patients’ and GPs’ views on barriers and facilitators of prescribing and uptake. Oral
presentation, 18–20 July 2023, Brighton, UK.
• Society of Academic Primary Care Conference – Unblinding to treatment allocation in randomised
placebo-controlled trials: a new process and analysis of patient perspectives from a trial of low-dose
amitriptyline for irritable bowel syndrome (IBS) in primary care. Oral presentation, 18–20 July 2023,
Brighton, UK.
• United European Gastroenterology conference – Amitriptyline at low-dose and titrated for irritable
bowel syndrome as second-line treatment (ATLANTIS): a randomised double-blind placebo-
controlled trial in primary care. Oral presentation, 14–17 October 2023, Copenhagen.
• Royal College of General Practitioners Conference – Amitriptyline at Low-dose and Titrated for
Irritable Bowel Syndrome as Second-line Treatment in primary care (ATLANTIS): A Double-blind
Placebo-controlled Trial. Oral and e-poster presentation, 19–20 October 2023, Glasgow, UK.
Awarded the winning ePoster in the ‘Clinical’ category.
• North American Primary Care Research Group (NPCRG) – Amitriptyline at Low-dose and Titrated for
Irritable Bowel Syndrome in Primary Care: ATLANTIS: Randomised Controlled Trial, Oral Presentation,
30 October to 3 November 2023, San Francisco, USA. Awarded Distinguished Paper Award.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 111
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

References
1. Ford AC, Sperber AD, Corsetti M, Camilleri M. Irritable bowel syndrome. Lancet
2020;396:1675–88.
2. Sperber AD, Bangdiwala SI, Drossman DA, Ghoshal UC, Simren M, Tack J, et al. Worldwide
prevalence and burden of functional gastrointestinal disorders, results of Rome Foundation
global study. Gastroenterology 2021;160:99–114.e3.
3. Thompson WG, Heaton KW, Smyth GT, Smyth C. Irritable bowel syndrome in general practice:
prevalence, characteristics, and referral. Gut 2000;46:78–82.
4. Goodoory VC, Ng CE, Black CJ, Ford AC. Direct healthcare costs of Rome IV or Rome
III – defined irritable bowel syndrome in the United Kingdom. Aliment Pharmacol Ther
2022;56:110–20.
5. Pace F, Molteni P, Bollani S, Sarzi-Puttini P, Stockbrügger R, Bianchi Porro G, Drossman DA.
Inflammatory bowel disease versus irritable bowel syndrome: a hospital-based, case-control
study of disease impact on quality of life. Scand J Gastroenterol 2003;38:1031–8.
6. Shivaji UN, Ford AC. Prevalence of functional gastrointestinal disorders among consecutive new
patient referrals to a gastroenterology clinic. Frontline Gastroenterology 2014;5:266–71.
7. Ruepert L, Quartero AO, de Wit NJ, van der Heijden GJ, Rubin G, Muris JW. Bulking agents,
antispasmodics and antidepressants for the treatment of irritable bowel syndrome. Cochr Datab
Syst Rev 2011;2011:CD003460.
8. Ford AC, Quigley EM, Lacy BE, Lembo AJ, Saito YA, Schiller LR, et al. Effect of antidepressants
and psychological therapies, including hypnotherapy, in irritable bowel syndrome: systematic
review and meta-analysis. Am J Gastroenterol 2014;109:1350–65; quiz 1366.
9. Ford AC, Lacy BE, Harris LA, Quigley EM, Moayyedi P. Effect of antidepressants and psycholog-
ical therapies in irritable bowel syndrome: an updated systematic review and meta-analysis. Am
J Gastroenterol 2019;114:21–39.
10. Sindrup SH, Otto M, Finnerup NB, Jensen TS. Antidepressants in the treatment of neuropathic
pain. Basic Clin Pharmacol Toxicol 2005;96:399–409.
11. Talley NJ, Locke GR, Saito YA, Almazar AE, Bouras EP, Howden CW, et al. Effect of amitriptyline
and escitalopram on functional dyspepsia: a multi-center, randomized, controlled study.
Gastroenterology 2015;149:340–9.e2.
12. Saarto T, Wiffen PJ. Antidepressants for neuropathic pain. Cochr Datab Syst Rev 2007;2007:
CD005454.
13. Wong J, Motulsky A, Abrahamowicz M, Eguale T, Buckeridge DL, Tamblyn R. Off-label indica-
tions for antidepressants in primary care: descriptive study of prescriptions from an indication
based electronic prescribing system. BMJ 2017;356:j603.
14. Gorard DA, Libby GW, Farthing MJ. Influence of antidepressants on whole gut orocaecal transit
times in health and irritable bowel syndrome. Aliment Pharmacol Ther 1994;8:159–66.
15. Ford AC, Forman D, Bailey AG, Axon ATR, Moayyedi P. Irritable bowel syndrome: a 10-year nat-
ural history of symptoms, and factors that influence consultation behavior. Am J Gastroenterol
2008;103:1229–39; quiz 1240.
16. Hookway C, Buckner S, Crosland P, Longson D. Irritable bowel syndrome in adults in primary
care: summary of updated NICE guidance. BMJ 2015;350:h701.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 113
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 References

17. Shivaji UN, Ford AC. Beliefs about management of irritable bowel syndrome in primary care:
cross-sectional survey in one locality. Prim Health Care Res Dev 2014;16:263–9.
18. Everitt H, McDermott L, Leydon G, Yules H, Baldwin D, Little P. GPs’ management strat-
egies for patients with insomnia: a survey and qualitative interview study. Br J Gen Pract
2014;64:e112–19.
19. Nigam P, Kapoor KK, Rastog CK, Kumar A, Gupta AK. Different therapeutic regimens in irritable
bowel syndrome. J Assoc Physicians India 1984;32:1041–4.
20. Vahedi H, Merat S, Momtahen S, Kazzazi AS, Ghaffari N, Olfati G, Malekzadeh R. Clinical trial:
the effect of amitriptyline in patients with diarrhea-predominant irritable bowel syndrome.
Aliment Pharmacol Ther 2008;27:678–84.
21. Thoua NM, Murray CD, Winchester WJ, Roy AJ, Pitcher MCL, Kamm MA, Emmanuel AV.
Amitriptyline modifies the visceral hypersensitivity response to acute stress in the irritable
bowel syndrome. Aliment Pharmacol Ther 2009;29:552–60.
22. Alderson SL, Wright-Hughes A, Ford AC, Farrin A, Hartley S, Fernandez C, et al. Amitriptyline
at low-dose and titrated for irritable bowel syndrome as second-line treatment (The ATLANTIS
trial): protocol for a randomised double-blind placebo-controlled trial in primary care. Trials
2022;23:552.
23. Francis CY, Morris J, Whorwell PJ. The irritable bowel severity scoring system: a simple
method of monitoring irritable bowel syndrome and its progress. Aliment Pharmacol Ther
1997;11:395–402.
24. Muller-Lissner S, Koch G, Talley NJ, Drossman D, Rueegg P, Dunger-Baldauf C, Lefkowitz M.
Subject’s global assessment of relief: an appropriate method to assess the impact of treatment
on irritable bowel syndrome-related symptoms in clinical trials. J Clin Epidemiol 2003;56:310–6.
25. Kroenke K, Spitzer RL, Williams JBW. The PHQ-15: validity of a new measure for evaluating the
severity of somatic symptoms. Psychosom Med 2002;64:258–66.
26. Spiller RC, Humes DJ, Campbell E, Hastings M, Neal KR, Dukes GE, Whorwell PJ. The Patient
Health Questionnaire 12 Somatic Symptom scale as a predictor of symptom severity and
consulting behaviour in patients with irritable bowel syndrome and symptomatic diverticular
disease. Aliment Pharmacol Ther 2010;32:811–20.
27. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand
1983;67:361–70.
28. Mundt JC, Marks IM, Shear MK, Greist JH. The Work and Social Adjustment Scale: a simple
measure of impairment in functioning. Br J Psychiatry 2002;180:461–4.
29. Kennedy T, Jones R, Darnley S, Seed P, Wessely S, Chalder T. Cognitive behaviour therapy in
addition to antispasmodic treatment for irritable bowel syndrome in primary care: randomised
controlled trial. BMJ 2005;331:435–7.
30. Moss-Morris R, McAlpine L, Didsbury LP, Spence MJ. A randomized controlled trial of a cogni-
tive behavioural therapy-based self-management intervention for irritable bowel syndrome in
primary care. Psychol Med 2010;40:85–94.
31. Uher R, Farmer A, Henigsberg N, Rietschel M, Mors O, Maier W, et al. Adverse reactions to
antidepressants. Br J Psychiatry 2009;195:202–10.
32. EuroQol G. EuroQol – a new facility for the measurement of health-related quality of life. Health
Pol 1990;16:199–208.
33. Bushnell DM, Martin ML, Ricci JF, Bracco A. Performance of the EQ-5D in patients with irritable
bowel syndrome. Value Health 2006;9:90–7.

114

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

34. Lacy BE, Mearin F, Chang L, Chey WD, Lembo AJ, Simren M, Spiller R. Bowel disorders.
Gastroenterology 2016;150:1393–407.
35. NICE. Suspected Cancer: Recognition and Referral. 2015. URL: www.nice.org.uk/guidance/NG12/
chapter/1-Recommendations-organised-by-site-of-cancer#lower-gastrointestinal-tract-cancers
(accessed 29 June 2023).
36. Everitt H, Landau S, Little P, Bishop FL, McCrone P, O’Reilly G, et al.; ACTIB Trial Team.
Assessing Cognitive behavioural Therapy in Irritable Bowel (ACTIB): protocol for a randomised
controlled trial of clinical-effectiveness and cost-effectiveness of therapist delivered cognitive
behavioural therapy and web-based self-management in irritable bowel syndrome in adults.
BMJ Open 2015;5:e008622.
37. Everitt HA, Landau S, O’Reilly G, Sibelli A, Hughes S, Windgassen S, et al.; ACTIB Trial Group.
Assessing telephone-delivered cognitive-behavioural therapy (CBT) and web-delivered CBT
versus treatment as usual in irritable bowel syndrome (ACTIB): a multicentre randomised trial.
Gut 2019;68:1613–23.
38. British Dietetic Association. Irritable Bowel Syndrome and Diet. URL: www.bda.uk.com/resource/
irritable-bowel-syndrome-diet.html (accessed 29 June 2023).
39. Sisson G, Ayis S, Sherwood RA, Bjarnason I. Randomised clinical trial: a liquid multi-strain
probiotic vs. placebo in the irritable bowel syndrome – a 12 week double-blind study. Aliment
Pharmacol Ther 2014;40:51–62.
40. Peckham EJ, Relton C, Raw J, Walters C, Thomas K, Smith C, et al. Interim results of a ran-
domised controlled trial of homeopathic treatment for irritable bowel syndrome. Homeopathy
2014;103:172–7.
41. Braun V, Clarke V. One size fits all? What counts as quality practice in (reflexive) thematic
analysis? Qual Res Psychol 2021;18:328–52.
42. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol 2006;3:77–101.
43. Strauss A, Corbin J. Basics of Qualitative Research. Thousand Oaks, CA: SAGE; 1998.
44. Leventhal H, Phillips LA, Burns E. The Common-Sense Model of Self-Regulation (CSM): a
dynamic framework for understanding illness self-management. J Behav Med 2016;39:935–46.
45. May C, Finch T, Mair F, Ballini L, Dowrick C, Eccles M, et al. Understanding the implementation
of complex interventions in health care: the normalization process model. BMC Health Serv Res
2007;7:1–7.
46. Harvey JM, Sibelli A, Chalder T, Everitt H, Moss-Morris R, Bishop FL. Desperately seeking
a cure: treatment seeking and appraisal in irritable bowel syndrome. Br J Health Psychol
2018;23:561–79.
47. Horne R, Chapman SCE, Parham R, Freemantle N, Forbes A, Cooper V. Understanding patients’
adherence-related beliefs about medicines prescribed for long-term conditions: a meta-analytic
review of the necessity-concerns framework. PLOS ONE 2013;8:e80633.
48. QSR International Pty Ltd. NVivo. Version 12 ed. Warrington: QSR International Pty Ltd; 2018.
49. Gagné T, Henderson C, McMunn A. Is the self-reporting of mental health problems sensitive
to public stigma towards mental illness? A comparison of time trends across English regions
(2009–19). Soc Psychiatry Psychiatr Epidemiol 2023;58:671–80.
50. Castaldelli-Maia JM, Scomparini LB, Andrade AG, Bhugra D, de Toledo Ferraz Alves TC, D’Elia
G. Perceptions of and attitudes toward antidepressants stigma attached to their use: a review. J
Nerv Ment Dis 2011;199:866–71.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 115
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 References

51. Ingersoll K, Cohen J. The impact of medication regimen factors on adherence to chronic treat-
ment: a review of literature. J Behav Med 2008;31:213–24.
52. Mari A, Hosadurg D, Martin L, Zarate-Lopez N, Passananti V, Emmanuel A. Adherence with
a low-FODMAP diet in irritable bowel syndrome: are eating disorders the missing link? Eur J
Gastroenterol Hepatol 2019;31:178–82.
53. Tuck CJ, Reed DE, Muir JG, Vanner SJ. Implementation of the low FODMAP diet in functional
gastrointestinal symptoms: a real-world experience. Neurogastroenterol Motil 2020;32:e13730.
54. Stewart S-JF, Moon Z, Horne R. Medication nonadherence: health impact, prevalence, corre-
lates and interventions. Psychol Health 2023;38:726–65.
55. Timmermans S. The engaged patient: the relevance of patient–physician communication for
twenty-first-century health. J Health Soc Behav 2020;61:259–73.
56. NHS England. The NHS Long Term Plan. 2019. URL: www.longtermplan.nhs.uk/publication/nhs-
long-term-plan/ (accessed 29 June 2023).
57. Schwille-Kiuntke J, Rüdlin SL, Junne F, Enck P, Brenk-Franz K, Zipfel S, Rieger MA. Illness
perception and health care use in individuals with irritable bowel syndrome: results from an
online survey. BMC Fam Pract 2021;22:154.
58. Tonkin-Crine S, Bishop FL, Ellis M, Moss-Morris R, Everitt H. Exploring patients’ views of a
cognitive behavioral therapy-based website for the self-management of irritable bowel syn-
drome symptoms. J Med Internet Res 2013;15:e190.
59. Murray E, Treweek S, Pope C, MacFarlane A, Ballini L, Dowrick C, et al. Normalisation process
theory: a framework for developing, evaluating and implementing complex interventions. BMC
Med 2010;8:63.
60. Shorey S, Demutska A, Chan V, Siah KTH. Adults living with irritable bowel syndrome (IBS): a
qualitative systematic review. J Psychosom Res 2021;140:110289.
61. Sibelli A, Moss-Morris R, Chalder T, Bishop FL, Windgassen S, Everitt H. Patients’ perspectives
on GP interactions after cognitive behavioural therapy for refractory IBS: a qualitative study in
UK primary and secondary care. Br J Gen Pract 2018;68:e654–62.
62. Sheridan R, Martin-Kerry J, Hudson J, Parker A, Bower P, Knapp P. Why do patients take part
in research? An overview of systematic reviews of psychosocial barriers and facilitators. Trials
2020;21:259.
63. Kaptchuk TJ, Shaw J, Kerr CE, Conboy LA, Kelley JM, Csordas TJ, et al. ‘Maybe I made up the
whole thing’: placebos and patients’ experiences in a randomized controlled trial. Cult Med
Psychiatry 2009;33:382–411.
64. Kaptchuk TJ, Kelley JM, Conboy LA, Davis RB, Kerr CE, Jacobson EE, et al. Components of
placebo effect: randomised controlled trial in patients with irritable bowel syndrome. BMJ (Clin
Res Ed) 2008;336:999–1003.
65. Noble H, Hasan SS, Whorwell PJ, Vasant DH. The symptom burden of irritable bowel syndrome
in tertiary care during the COVID-19 pandemic. Neurogastroenterol Motil 2022;34:e14347.
66. Quek SXZ, Loo EXL, Demutska A, Chua CE, Kew GS, Wong S, et al. Impact of the coronavirus
disease 2019 pandemic on irritable bowel syndrome. J Gastroenterol Hepatol 2021;36:2187–97.
67. Stone DA, Kerr CE, Jacobson E, Conboy LA, Kaptchuk TJ. Patient expectations in placebo-­
controlled randomized clinical trials. J Eval Clin Pract 2005;11:77–84.
68. Bishop FL, Jacobson EE, Shaw JR, Kaptchuk TJ. Debriefing to placebo allocation: a phenom-
enological study of participants’ experiences in a randomized clinical trial. Qual Health Res
2012;22:1138–49.

116

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

69. Bishop FL, Jacobson EE, Shaw JR, Kaptchuk TJ. Scientific tools, fake treatments, or triggers
for psychological healing: how clinical trial participants conceptualise placebos. Soc Sci Med
2012;74:767–74.
70. Shah E, Triantafyllou K, Hana AA, Pimentel M. Adverse events appear to unblind clinical trials in
irritable bowel syndrome. Neurogastroenterol Motil 2014;26:482–8.
71. Schulz KF, Grimes DA. Blinding in randomised trials: hiding who got what. Lancet
2002;359:696–700.
72. Shalowitz DI, Miller FG. Disclosing individual results of clinical research: implications of respect
for participants. JAMA 2005;294:737–40.
73. Shalowitz DI, Miller FG. Communicating the results of clinical research to participants: attitudes,
practices, and future directions. PLOS Med 2008;5:e91.
74. Moore GF, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, et al. Process evaluation of
complex interventions: Medical Research Council guidance. BMJ (Clin Res Ed) 2015;350:h1258.
75. Fetters MD, Curry LA, Creswell JW. Achieving integration in mixed methods designs – principles
and practices. Health Serv Res 2013;48:2134–56.
76. European Medicines Agency. Guideline on the Evaluation of Medicinal Products for the
Treatment of Irritable Bowel Syndrome. 2014. URL: www.ema.europa.eu/en/documents/
scientific-guideline/guideline-evaluation-medicinal-products-treatment-irritable-bowel-syn-
drome-revision-1_en.pdf (accessed 29 June 2023).
77. Food and Drug Administration. Guidance for Industry: Irritable Bowel Syndrome – Clinical
Evaluation of Drugs for Treatment. 2012. URL: www.fda.gov/media/78622/download (accessed
29 June 2023).
78. Vasant DH, Paine PA, Black CJ, Houghton LA, Everitt HA, Corsetti M, et al. British Society
of Gastroenterology guidelines on the management of irritable bowel syndrome. Gut
2021;70:1214–40.
79. Black CJ, Yiannakou Y, Houghton LA, Ford AC. Epidemiological, clinical, and psychological
characteristics of individuals with self-reported irritable bowel syndrome based on the Rome IV
vs Rome III criteria. Clin Gastroenterol Hepatol 2020;18:392–398.e2.
80. Ford AC, Moayyedi PM. Factors affecting placebo response rate in irritable bowel syndrome.
Aliment Pharmacol Ther 2010;32:144–58.
81. Barberio B, Savarino EV, Black CJ, Ford AC. Placebo response rates in trials of licensed drugs for
irritable bowel syndrome with constipation or diarrhea: meta-analysis. Clin Gastroenterol Hepatol
2022;20:e923–44.
82. Lembo A, Kelley JM, Nee J, Ballou S, Iturrino J, Cheng V, et al. Open-label placebo vs
double-blind placebo for irritable bowel syndrome: a randomized clinical trial. Pain
2021;162:2428–35.
83. Kaptchuk TJ, Friedlander E, Kelley JM, Sanchez MN, Kokkotou E, Singer JP, et al. Placebos
without deception: a randomized controlled trial in irritable bowel syndrome. PLOS ONE 2010;
5:e15591.
84. Black CJ, Yuan Y, Selinger CP, Camilleri M, Quigley EMM, Moayyedi P, Ford AC. Efficacy of
soluble fibre, antispasmodic drugs, and gut-brain neuromodulators in irritable bowel syndrome:
a systematic review and network meta-analysis. Lancet Gastroenterol Hepatol 2020;5:117–31.
85. Drossman DA, Toner BB, Whitehead WE, Diamant NE, Dalton CB, Duncan S, et al. Cognitive-
behavioral therapy versus education and desipramine versus placebo for moderate to severe
functional bowel disorders. Gastroenterology 2003;125:19–31.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 117
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 References

86. Gorard DA, Libby GW, Farthing MJ. Effect of a tricyclic antidepressant on small intestinal motil-
ity in health and diarrhea-predominant irritable bowel syndrome. Dig Dis Sci 1995;40:86–95.
87. Morgan V, Pickens D, Gautam S, Kessler R, Mertz H. Amitriptyline reduces rectal pain related
activation of the anterior cingulate cortex in patients with irritable bowel syndrome. Gut
2005;54:601–7.
88. Patel P, Bercik P, Morgan DG, Bolino C, Pintos-Sanchez MI, Moayyedi P, Ford AC. Irritable
bowel syndrome is significantly associated with somatisation in 840 patients, which may drive
bloating. Aliment Pharmacol Ther 2015;41:449–58.
89. Zamani M, Alizadeh-Tabari S, Zamani V. Systematic review with meta-analysis: the prevalence
of anxiety and depression in patients with irritable bowel syndrome. Aliment Pharmacol Ther
2019;50:132–43.
90. Goodoory VC, Ng CE, Black CJ, Ford AC. Impact of Rome IV irritable bowel syndrome on work
and activities of daily living. Aliment Pharmacol Ther 2022;56:844–56.
91. Frandemark A, Tornblom H, Jakobsson S, Simren M. Work productivity and activity impairment
in irritable bowel syndrome (IBS): a multifaceted problem. Am J Gastroenterol 2018;113:1540–9.

118

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Appendix 1 Additional clinical results tables

and figures
Study summary and baseline characteristics

TABLE 42 Patient screening and recruitment by hub

West Yorkshire Wessex West of England Total

Mailed out

Patients 3862 (24.6%) 6218 (39.7%) 5592 (35.7%) 15,672 (100%)

GP practices 13 20 22 55

Responded to mail-out 153 (4.7%) 1688 (52.3%) 1387 (43.0%) 3228 (100%)

Interested 116 (9.3%) 525 (41.9%) 612 (48.8%) 1253 (100%)

Screening call 113 (10.2%) 501 (45.3%) 491 (44.4%) 1105 (100%)

Eligible 108 (18.7%) 256 (44.2%) 215 (37.1%) 579 (100%)

Consented 103 (19.5%) 223 (42.2%) 202 (38.3%) 528 (100%)

Registered 101 (19.2%) 222 (42.3%) 202 (38.5%) 525 (100%)

Randomised

Patients 87 (18.8%) 192 (41.5%) 184 (39.7%) 463 (100%)

GP practicesa 11 20 21 52

a Three practices conducted mailouts but did not recruit any participants as: two practices were PIC sites for other
practices, and one was a very small practice (n = 44 mailed out).

TABLE 43 Number and reasons for protocol violations

Amitriptyline (n = 232) (%) Placebo (n = 231) (%) Total (n = 463) (%)

Protocol violation identified? 3 (1.3) 3 (1.3) 6 (1.3)

Reason for protocol violation

Breached eligibility criteria 1 (33.3) 0 (0.0) 1 (16.7)

Unplanned treatment error 1 (33.3) 3 (100) 4 (66.7)

Other protocol violation 1 (33.3) 0 (0.0) 1 (16.7)

Type of protocol violation

Major 2 (66.7) 2 (66.7) 4 (66.7)

Minor 1 (33.3) 1 (33.3) 2 (33.3)

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 119
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Appendix 1

Practice52 (N = 2) West of England

Practice51 (N = 10)
Practice50 (N = 3)
Practice49 (N = 18)
Practice48 (N = 4)
Practice47 (N = 13)
Practice46 (N = 7)
Practice45 (N = 5)
Practice44 (N = 2)
Practice43 (N = 8)
Practice42 (N = 27)
Practice41 (N = 3)
Practice40 (N = 5)
Practice39 (N = 32)
Practice38 (N = 8)
Practice37 (N = 10)
Practice36 (N = 7)
Practice35 (N = 6)
Practices that have randomised patients

Practice34 (N = 2)
Practice33 (N = 4)
Practice32 (N = 8)
Practice31 (N = 3)
Wessex
Practice30 (N = 2)
Practice29 (N = 16)
Practice28 (N = 3)
Practice27 (N = 6) Mail out
Practice26 (N = 20)
Practice25 (N = 11) Randomisation
Practice24 (N = 17)
Practice23 (N = 8)
Practice22 (N = 17)
Practice21 (N = 9)
Practice20 (N = 17)
Practice19 (N = 5)
Practice18 (N = 11)
Practice17 (N = 3)
Practice16 (N = 9)
Practice15 (N = 6)
Practice14 (N = 16)
Practice13 (N = 5)
Practice12 (N = 8) West Yorkshire
Practice11 (N = 8)
Practice10 (N = 2)
Practice9 (N = 7)
Practice8 (N = 7)
Practice7 (N = 3)
Practice6 (N = 7)
Practice5 (N = 7)
Practice4 (N = 21)
Practice3 (N = 1)
Practice2 (N = 16)
Practice1 (N = 8)
Oct Dec Feb Apr Jun Aug Oct Dec Feb Apr Jun Aug Oct Dec Feb Apr
2019 2020 2021 2022
Months post practice mail out

FIGURE 8 Time from general practice mail-out to randomisation. N refers to the number of randomised participants in
each practice.

Randomisation Month 3 Month 6 Month 12

480
460
440
420
400
380 Placebo
360
340 Baseline questionnaires
320 completed
300 Month 3 questionnaires
completed
280
Participant index

Month 6 questionnaires
260 completed
240 Month 12 questionnaires
220 completed
200 Not completed
180
160
140
120 Amitriptyline
100
80
60
40
20
0
–1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Time in months

FIGURE 9 Timing of questionnaire completion.

120

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Histogram of number of weekly relief questions completed

60

50

40
Number of participants

Amitriptyline
30 Placebo

20

10

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Number of weekly questions completed

Histogram of number of participants completing each weekly relief questions

230
220
210
200
190
180
170
160
Number of participants

150
140
130
120 Amitriptyline
110 Placebo
100
90
80
70
60
50
40
30
20
10
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Week

FIGURE 10 Weekly adequate relief question completion.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 121
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Appendix 1

TABLE 44 Index of multiple deprivation decile by treatment allocation and recruitment hub

Treatment allocation Recruitment hub

Amitriptyline Placebo West Yorkshire Wessex West of England Total

Decilea (n = 232) (n = 231) (n = 87) (n = 192) (n = 184) (n = 463)

1 3 (1.3%) 3 (1.3%) 1 (1.1%) 2 (1.0%) 3 (1.6%) 6 (1.3%)

2 10 (4.4%) 10 (4.3%) 9 (10.6%) 4 (2.1%) 7 (3.8%) 20 (4.4%)

3 13 (5.7%) 14 (6.1%) 8 (9.4%) 12 (6.3%) 7 (3.8%) 27 (5.9%)

4 21 (9.2%) 13 (5.7%) 13 (15.3%) 13 (6.8%) 8 (4.4%) 34 (7.4%)

5 19 (8.3%) 19 (8.3%) 8 (9.4%) 15 (7.9%) 15 (8.2%) 38 (8.3%)

6 19 (8.3%) 14 (6.1%) 10 (11.8%) 17 (8.9%) 6 (3.3%) 33 (7.2%)

7 38 (16.6%) 38 (16.5%) 14 (16.5%) 35 (18.3%) 27 (14.8%) 76 (16.6%)

8 37 (16.2%) 36 (15.7%) 12 (14.1%) 30 (15.7%) 31 (16.9%) 73 (15.9%)

9 36 (15.7%) 39 (17.0%) 9 (10.6%) 29 (15.2%) 37 (20.2%) 75 (16.3%)

10 33 (14.4%) 44 (19.1%) 1 (1.2%) 34 (17.8%) 42 (23.0%) 77 (16.8%)

Missing 3 1 2 1 1 4

a Decile 1 = neighbourhood in the 10% most deprived neighbourhoods in England, 10 = neighbourhood in the 10% least
deprived neighbourhoods in England.

122

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Years from IBS diagnosis to trial randomisation

25

20

15
Number of participants

alloc
Amitriptyline
Placebo

10

0
0 20 40 60
Years

Years from IBS diagnosis to trial randomisation by baseline IBS-SSS level

IBS-SSS Level = 75–174 (Mild IBS) IBS-SSS Level = 175–299 (Moderate IBS) IBS-SSS Level = 300–500 (Severe IBS)
40

35

30

25
Number of participants

alloc
Amitriptyline
20 Placebo

15

10

0
0 20 40 60 0 20 40 60 0 20 40 60
Years from IBS diagnosis to trial randomisation

FIGURE 11 Years from IBS diagnosis to trial randomisation. (a) Overall, (b) by IBS-SSS severity.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 123
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Appendix 1

Treatment delivery and receipt

Reduced appetite
Confusion
Diarrhoea
Bloating
Altered taste
Micturition difficulties
Hair loss
Abdominal pain
Difficulty passing urine Amitriptyline
Placebo
Palpitations
Skin rash
Dry mouth
Dizzy
Non specified
Nausea
Heartburn
Headaches
Constipated
Deterioration of mood
Drowsy
0 2 4 6 8 10 12 14
Number of participants

FIGURE 12 Reported side effects as a reason for treatment discontinuation.

All participants 6 months follow-up 12 months follow-up

80
Number of participants

60

alloc
Amitriptyline
40 Placebo

20

0
0 5 10 15 0 5 10 15 0 5 10 15
Time from randomisation to last dose date (months)

FIGURE 13 Distribution of time from randomisation to treatment end date.

124

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Week 3 Month 3 Month 6 Month 9 Month 12

200

150 Has the patient taken at

least one tablet daily
Every day or nearly
every day
Frequency

Half the days or more

than half the days
Less than half of the days
100 None or nearly none of
the days
Missing

50

0
Amitriptyline Placebo Amitriptyline Placebo Amitriptyline Placebo Amitriptyline Placebo Amitriptyline Placebo

FIGURE 14 Treatment adherence reported at each follow-up time point (for participants on trial medication).

Week 1 Week 3 Month 3 Month 6 Month 9 Month 12

200

150 Current dose of

trial treatment
1 x 10 mg every
Frequency

other day
1 x 10 mg daily
100 2 x 10 mg daily
3 x 10 mg daily
Missing

50

0
Am

Pl

Am

Pl

Am

Pl

Am

Pl

Am

Pl

Am

Pl
a

ac

ac

ac
ce

ce

ce
itr

itr

itr

itr

itr

itr
eb ne

eb ne

eb ne
bo e

bo

bo
ip

ip

ip

ip

ip

ip
o

o
ty

ty

ty

ty

ty

ty
lin

lin

li

lin

li

li
e

FIGURE 15 Treatment dose reported at each follow-up time point (for participants on trial medication).

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 125
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Appendix 1

TABLE 45 Treatment adherence and dose reported at last follow-up for participants discontinued trial medication
before month 6

Before month 3 Between month 3 and month 6

Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 32) (n = 31) (n = 63) (n = 14) (n = 28) (n = 42)

Has the patient taken at least one tablet daily?

Every/nearly every 22 (91.7%) 19 (86.4%) 41 (89.1%) 13 (92.9%) 23 (85.2%) 36

day (87.8%)

≥ half the days 2 (8.3%) 2 (9.1%) 4 (8.7%) 1 (7.1%) 3 (11.1%) 4 (9.8%)

< half of the days 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (3.7%) 1 (2.4%)

None/nearly none 0 (0.0%) 1 (4.5%) 1 (2.2%) 0 (0.0%) 0 (0.0%) 0 (0.0%)

of the days

Missing 8 9 17 0 1 1

Current dose of trial treatment

1 × 10 mg every 0 (0.0%) 1 (3.2%) 1 (1.7%) 0 (0.0%) 1 (3.6%) 1 (2.4%)

other day

1 × 10 mg daily 10 (37.0%) 18 (58.1%) 28 (48.3%) 6 (42.9%) 6 (21.4%) 12

(28.6%)

2 × 10 mg daily 11 (40.7%) 7 (22.6%) 18 (31.0%) 6 (42.9%) 10 (35.7%) 16

(38.1%)

3 × 10 mg daily 6 (22.2%) 5 (16.1%) 11 (19.0%) 2 (14.3%) 11 (39.3%) 13

(31.0%)

Missing 5 0 5 0 0 0

TABLE 46 Treatment adherence and dose at trial medication discontinuation: after month 6, all discontinuations

After month 6 At point of discontinuation

Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 16) (n = 11) (n = 27) (n = 62) (n = 70) (n = 132)

Has the patient taken at least one tablet daily?

Every/nearly every 14 (87.5%) 11 (100.0%) 25 (92.6%) 38 (79.2%) 40 (70.2%) 78 (74.3%)

day

≥ half the days 2 (12.5%) 0 (0.0%) 2 (7.4%) 6 (12.5%) 12 (21.1%) 18 (17.1%)

< half of the days 0 (0.0%) 0 (0.0%) 0 (0.0%) 3 (6.3%) 3 (5.3%) 6 (5.7%)

None/nearly none 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (2.1%) 2 (3.5%) 3 (2.9%)

of the days

Missing 0 0 0 14 13 27

Current dose of trial treatment

1 × 10 mg every 1 (6.3%) 2 (18.2%) 3 (11.1%) 2 (4.3%) 3 (5.2%) 5 (4.8%)

other day

1 × 10 mg daily 5 (31.3%) 0 (0.0%) 5 (18.5%) 28 (59.6%) 13 (22.4%) 41 (39.0%)

2 × 10 mg daily 2 (12.5%) 5 (45.5%) 7 (25.9%) 7 (14.9%) 18 (31.0%) 25 (23.8%)

3 × 10 mg daily 8 (50.0%) 4 (36.4%) 12 (44.4%) 10 (21.3%) 24 (41.4%) 34 (32.4%)

Missing 0 0 0 15 12 27

126

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 47 Treatment replenishment for participants on trial medication

Week 3 Month 3

Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 222) (n = 223) (n = 445) (n = 194) (n = 196) (n = 390)

Bottles requested

0 9 (4.1%) 19 (8.5%) 28 (6.3%) 20 (10.3%) 26 46 (11.8%)

(13.3%)

1 14 (6.3%) 14 (6.3%) 28 (6.3%) 33 (17.0%) 19 (9.7%) 52 (13.3%)

2 127 (57.2%) 95 222 37 (19.1%) 28 65 (16.7%)

(42.6%) (49.9%) (14.3%)

3 70 (31.5%) 91 161 48 (24.7%) 49 97 (24.9%)

(40.8%) (36.2%) (25.0%)

4 2 (0.9%) 4 (1.8%) 6 (1.3%) 47 (24.2%) 60 107 (27.4%)

(30.6%)

5 0 0 0 9 (4.6%) 14 (7.1%) 23 (5.9%)

Month 6 Month 9

Amitriptyline Placebo Total Amitriptyline Placebo Total

(n = 173) (n = 165) (n = 338) (n = 74) (n = 69) (n = 143)

Bottles requested

0 89 (51.4%) 89 178 8 (10.8%) 5 (7.2%) 13 (9.1%)

(53.9%) (52.7%)

1 14 (8.1%) 8 (4.8%) 22 (6.5%) 7 (9.5%) 1 (1.4%) 8 (5.6%)

2 12 (6.9%) 15 (9.1%) 27 (8.0%) 14 (18.9%) 11 25 (17.5%)

(15.9%)

3 23 (13.3%) 24 47 19 (25.7%) 24 43 (30.1%)

(14.5%) (13.9%) (34.8%)

4 27 (15.6%) 22 49 19 (25.7%) 23 42 (29.4%)

(13.3%) (14.5%) (33.3%)

5 8 (4.6%) 7 (4.2%) 15 (4.4%) 7 (9.5%) 5 (7.2%) 12 (8.4%)

Note
Each bottle contained 65 tablets.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 127
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Appendix 1

TABLE 48 Details of new diets, other IBS treatments, and attributed reasons for any improvement in IBS symptoms

Discontinued treatment before

Month 6 month 6

Amitriptyline Amitriptyline
(%) Placebo (%) Total (%) (%) Placebo (%) Total (%)

New diet during the studya

Low carbohydrate 6 (18.2) 5 (14.7) 11 (16.4) 0 (0.0) 0 (0.0) 0 (0.0)

Gluten-free or low gluten 5 (15.2) 3 (8.8) 8 (11.9) 0 (0.0) 0 (0.0) 0 (0.0)

Weight loss programme 1 (3.0) 5 (14.7) 6 (9.0) 0 (0.0) 0 (0.0) 0 (0.0)

Low FODMAP 0 (0.0) 5 (14.7) 5 (7.5) 0 (0.0) 1 (16.7) 1 (16.7)

Increase vegetables or fruit 5 (15.2) 0 (0.0) 5 (7.5) 0 (0.0) 1 (16.7) 1 (16.7)

intake

Reduce meat intake 5 (15.2) 0 (0.0) 5 (7.5) 0 (0.0) 1 (16.7) 1 (16.7)

Reduce portion sizes 3 (9.1) 1 (2.9) 4 (6.0) 0 (0.0) 0 (0.0) 0 (0.0)

Dairy-free or low diary 2 (6.1) 1 (2.9) 3 (4.5) 0 (0.0) 2 (33.3) 2 (33.3)

Fermented food or drinks 2 (6.1) 1 (2.9) 3 (4.5) 0 (0.0) 0 (0.0) 0 (0.0)

Low fibre 1 (3.0) 2 (5.9) 3 (4.5) 0 (0.0) 0 (0.0) 0 (0.0)

Low lactose or lactose-free 1 (3.0) 1 (2.9) 2 (3.0) 0 (0.0) 0 (0.0) 0 (0.0)

Paleo 0 (0.0) 1 (2.9) 1 (1.5) 0 (0.0) 0 (0.0) 0 (0.0)

Mediterranean 0 (0.0) 1 (2.9) 1 (1.5) 0 (0.0) 0 (0.0) 0 (0.0)

Vegetarian 1 (3.0) 0 (0.0) 1 (1.5) 0 (0.0) 1 (16.7) 1 (16.7)

Other 7 (21.2)b 14 (41.2)c 21 (31.3) 0 (0.0) 1 (16.7) 1 (16.7)d

Total 33 (100) 34 (100) 67 (100) 0 6 (100) 6 (100)

Other treatments for IBS symptoms during the studye

Antispasmodics 6 (33.3) 2 (12.5) 8 (23.5) 1 (20.0) 0 (0.0 1 (14.3)

Probiotics 3 (16.7) 3 (18.8) 6 (17.6) 1 (20.0) 1 (50.0) 2 (28.6)

Peppermint oil/tablets/tea 2 (11.1) 2 (12.5) 4 (11.8) 0 (0.0) 1 (50.0) 1 (14.3)

Prebiotics 2 (11.1) 1 (6.3) 3 (8.8) 0 (0.0) 0 (0.0) 0 (0.0)

Anti-diarrhoeal 2 (11.1) 1 (6.3) 3 (8.8) 0 (0.0) 0 (0.0) 0 (0.0)

Laxatives 1 (5.6) 1 (6.3) 2 (5.9) 0 (0.0) 0 (0.0) 0 (0.0)

Homeopathy 0 (0.0) 1 (6.3) 1 (2.9) 1 (20.0) 0 (0.0) 1 (14.3)

Other 6 (33.3)f 6 (37.5)g 12 (35.3) 2 (40.0)h 0 (0.0) 2 (28.6)

Missing 0 (0.0) 1 (6.3) 1 (2.9) 0 0 0

Total 18 (100) 16 (100) 34 (100) 5 (100) 2 (100) 7 (100)

What attributed to improved IBS symptomsi

ATLANTIS medication 103 (87.3) 75 (84.3) 178 (86.0) 11 (84.6) 7 (70.0) 18 (78.3)

Changes in diet 15 (12.7) 11 (12.4) 26 (12.6) 0 (0.0) 2 (20.0) 2 (8.7)

Changes in exercise 5 (4.2) 3 (3.4) 8 (3.9) 0 (0.0) 0 (0.0) 0 (0.0)

Uncertain 4 (3.4) 4 (4.5) 8 (3.9) 2 (15.4) 0 (0.0) 2 (8.7)

Less stress 4 (3.4) 2 (2.2) 6 (2.9) 0 (0.0) 0 (0.0) 0 (0.0)

128

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 48 Details of new diets, other IBS treatments, and attributed reasons for any improvement in
IBS symptoms (continued)

Discontinued treatment before

Month 6 month 6

Amitriptyline Amitriptyline
(%) Placebo (%) Total (%) (%) Placebo (%) Total (%)

 hanges in work/personal
C 2 (1.7) 3 (3.4) 5 (2.4) 0 (0.0) 1 (10.0) 1 (4.3)
environments

Changes in other treatments 2 (1.7) 0 (0.0) 2 (1.0) 0 (0.0) 1 (10.0) 1 (4.3)

Other treatment/therapies 0 (0.0) 2 (2.2) 2 (1.0) 0 (0.0) 0 (0.0) 0 (0.0)

Other 2j (1.7) 3k (3.4) 5 (2.4) 0 (0.0) 0 (0.0) 0 (0.0)

Total 118 (100) 89 (100) 207 (100) 13 (100) 10 (100) 23 (100)

a Non-mutually exclusive.
b Other diets (amitriptyline): avoided chilli, daily walnuts, eat individual food in certain order, low-acid diet, swapped bran
breakfast for oats, cut out sweetener, cut out food that aggravate IBS.
c Other diets (placebo): cook from fresh, cut out fatty food, cut out sugar and milk in coffee, drink more water, doesn’t
eat after certain time, food supplement, more ‘healthy’ eating, real food supplement, reduce caffeine, reduce tomato
and onion, reduce triggering food such as bread, reducing diet, stopped cow milk, wheat limited, being cautious with
triggering food.
d Other diets (discontinuation): reduced sulphites.
e Non-mutually exclusive.
f Other treatments (amitriptyline): OTC, yoga posses, esomeprazole, psyllium husk, dulcolax, fibre gel.
g Other treatments (placebo): herbal teas, only eating between 10 a.m. and 8 p.m., Andrews Liver Salts, goats milk,
loperamide as needed, esomeprazole.
h Other treatments (discontinuation): bulk powder for fluid absorption, CBD supplements.
i Non-mutually exclusive.
j Other reasons attributed to improved symptoms (amitriptyline): mental outlook, drink more fluid.
k Other reasons attributed to improved symptoms (placebo): gradual improvement in symptoms over many years, being
busy and distracted by work, more of a routine during lockdown.

Primary end point

Baseline (N = 463) Month 3 (N = 463) Month 6 (N = 463) Month 12 (N = 291)
240

220

200

180

160
IBS-SSS Level
140 <75 (normal bowel
function)
Frequency

120 75–174 (mild IBS)

175–299 (moderate IBS)
300–500 (severe IBS)
100
Missing

80

60

40

20

0
Amitriptyline Placebo Amitriptyline Placebo Amitriptyline Placebo Amitriptyline Placebo

FIGURE 16 Total IBS-SSS score severity at each time point.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 129
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Appendix 1

IBS-SSS score with unadjusted 95% CI – by population

1 – ITT population (N = 463) 2 – ITT 12m population (N = 291)

300

250
Total IBS-SSS score

alloc
Amitriptyline
+ Placebo

200

150

Baseline Month 3 Month 6 Month 12 Baseline Month 3 Month 6 Month 12

IBS-SSS item scores with unadjusted 95% CI – ITT population

Q1 Pain severity Q2 Days pain Q3 Distention severity
70

60

50

40

30

20
Score

Q4 Dissatisfaction Q5 Interference alloc

70 Amitriptyline
+ Placebo
60

50

40

30

20
Baseline Month 3 Month 6 Baseline Month 3 Month 6 Baseline Month 3 Month 6

FIGURE 17 Unadjusted total IBS-SSS score and item level scores with 95% CIs based on available data.

130

NIHR Journals Library www.journalslibrary.nihr.ac.uk

 TABLE 49 Irritable bowel syndrome Severity Scoring System items
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
Baseline Month 3 Month 6 Month 12

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total
(n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 147) (n = 144) (n = 291)

1a. Current (past 10 days) abdominal pain?

No 13 (5.6%) 21 (9.1%) 34 (7.3%) 73 (33.3%) 63 (29.6%) 136 (31.5%) 76 (37.3%) 59 (29.9%) 135 (33.7%) 50 (42.4%) 40 (37.4%) 90 (40.0%)

Yes 219 (94.4%) 210 (90.9%) 429 (92.7%) 146 (66.7%) 150 (70.4%) 296 (68.5%) 128 (62.7%) 138 (70.1%) 266 (66.3%) 68 (57.6%) 67 (62.6%) 135 (60.0%)

Missing 13 18 31 28 34 62 29 37 66

1b. Severity of abdominal pain

0 – No pain 13 (5.6%) 21 (9.1%) 34 (7.4%) 73 (33.3%) 64 (30.0%) 137 (31.7%) 76 (37.3%) 59 (29.9%) 135 (33.7%) 51 (43.2%) 40 (37.4%) 91 (40.4%)

10 4 (1.7%) 1 (0.4%) 5 (1.1%) 5 (2.3%) 7 (3.3%) 12 (2.8%) 3 (1.5%) 5 (2.5%) 8 (2.0%) 3 (2.5%) 2 (1.9%) 5 (2.2%)

20 10 (4.3%) 17 (7.4%) 27 (5.8%) 12 (5.5%) 12 (5.6%) 24 (5.6%) 14 (6.9%) 12 (6.1%) 26 (6.5%) 8 (6.8%) 8 (7.5%) 16 (7.1%)

30 28 (12.1%) 29 (12.6%) 57 (12.3%) 34 (15.5%) 25 (11.7%) 59 (13.7%) 28 (13.7%) 21 (10.7%) 49 (12.2%) 13 (11.0%) 8 (7.5%) 21 (9.3%)

40 27 (11.7%) 26 (11.3%) 53 (11.5%) 30 (13.7%) 34 (16.0%) 64 (14.8%) 27 (13.2%) 26 (13.2%) 53 (13.2%) 10 (8.5%) 10 (9.3%) 20 (8.9%)

50 37 (16.0%) 31 (13.4%) 68 (14.7%) 17 (7.8%) 19 (8.9%) 36 (8.3%) 14 (6.9%) 23 (11.7%) 37 (9.2%) 9 (7.6%) 15 (14.0%) 24 (10.7%)

Health Technology Assessment 2024 Vol. 28 No. 66

60 42 (18.2%) 34 (14.7%) 76 (16.5%) 18 (8.2%) 22 (10.3%) 40 (9.3%) 20 (9.8%) 21 (10.7%) 41 (10.2%) 5 (4.2%) 8 (7.5%) 13 (5.8%)

70 35 (15.2%) 41 (17.7%) 76 (16.5%) 22 (10.0%) 19 (8.9%) 41 (9.5%) 14 (6.9%) 15 (7.6%) 29 (7.2%) 8 (6.8%) 11 (10.3%) 19 (8.4%)

80 24 (10.4%) 22 (9.5%) 46 (10.0%) 7 (3.2%) 10 (4.7%) 17 (3.9%) 4 (2.0%) 13 (6.6%) 17 (4.2%) 10 (8.5%) 4 (3.7%) 14 (6.2%)

90 7 (3.0%) 7 (3.0%) 14 (3.0%) 1 (0.5%) 1 (0.5%) 2 (0.5%) 2 (1.0%) 0 (0.0%) 2 (0.5%) 0 (0.0%) 1 (0.9%) 1 (0.4%)

 00 – Very severe
1 4 (1.7%) 2 (0.9%) 6 (1.3%) 2 (1.0%) 2 (1.0%) 4 (1.0%) 1 (0.8%) 0 (0.0%) 1 (0.4%)
pain

Missing 1 0 1 13 18 31 28 34 62 29 37 66

2. N days abdominal pain

0 days 13 (5.6%) 21 (9.1%) 34 (7.4%) 73 (33.3%) 64 (30.0%) 137 (31.7%) 76 (37.3%) 59 (29.9%) 135 (33.7%) 50 (42.4%) 40 (37.4%) 90 (40.0%)

1 13 (5.6%) 8 (3.5%) 21 (4.5%) 20 (9.1%) 11 (5.2%) 31 (7.2%) 11 (5.4%) 4 (2.0%) 15 (3.7%) 9 (7.6%) 6 (5.6%) 15 (6.7%)

2 32 (13.9%) 25 (10.8%) 57 (12.3%) 37 (16.9%) 26 (12.2%) 63 (14.6%) 32 (15.7%) 29 (14.7%) 61 (15.2%) 16 (13.6%) 14 (13.1%) 30 (13.3%)

continued
131
 132
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Appendix 1
TABLE 49 IBS-SSS items (continued)

Baseline Month 3 Month 6 Month 12

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total
(n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 147) (n = 144) (n = 291)

3 37 (16.0%) 43 (18.6%) 80 (17.3%) 27 (12.3%) 33 (15.5%) 60 (13.9%) 25 (12.3%) 26 (13.2%) 51 (12.7%) 13 (11.0%) 12 (11.2%) 25 (11.1%)

4 25 (10.8%) 26 (11.3%) 51 (11.0%) 18 (8.2%) 19 (8.9%) 37 (8.6%) 15 (7.4%) 26 (13.2%) 41 (10.2%) 6 (5.1%) 10 (9.3%) 16 (7.1%)

5 34 (14.7%) 30 (13.0%) 64 (13.9%) 9 (4.1%) 16 (7.5%) 25 (5.8%) 14 (6.9%) 14 (7.1%) 28 (7.0%) 8 (6.8%) 8 (7.5%) 16 (7.1%)

6 23 (10.0%) 17 (7.4%) 40 (8.7%) 10 (4.6%) 13 (6.1%) 23 (5.3%) 5 (2.5%) 14 (7.1%) 19 (4.7%) 5 (4.2%) 7 (6.5%) 12 (5.3%)

7 17 (7.4%) 14 (6.1%) 31 (6.7%) 7 (3.2%) 9 (4.2%) 16 (3.7%) 7 (3.4%) 6 (3.0%) 13 (3.2%) 2 (1.7%) 3 (2.8%) 5 (2.2%)

8 7 (3.0%) 18 (7.8%) 25 (5.4%) 6 (2.7%) 8 (3.8%) 14 (3.2%) 5 (2.5%) 6 (3.0%) 11 (2.7%) 2 (1.7%) 2 (1.9%) 4 (1.8%)

9 5 (2.2%) 8 (3.5%) 13 (2.8%) 1 (0.5%) 0 (0.0%) 1 (0.2%) 4 (2.0%) 2 (1.0%) 6 (1.5%) 0 (0.0%) 1 (0.9%) 1 (0.4%)

 0 days = pain
1 25 (10.8%) 21 (9.1%) 46 (10.0%) 11 (5.0%) 14 (6.6%) 25 (5.8%) 10 (4.9%) 11 (5.6%) 21 (5.2%) 7 (5.9%) 4 (3.7%) 11 (4.9%)
every day

Missing 1 0 (0.0%) 1 13 18 31 28 34 62 29 37 66

3a. Current (past 10 days) abdominal distention?

No 25 (10.8%) 18 (7.8%) 43 (9.3%) 77 (35.2%) 51 (23.9%) 128 (29.6%) 67 (32.8%) 51 (25.9%) 118 (29.4%) 39 (33.1%) 32 (29.9%) 71 (31.6%)

Yes 207 (89.2%) 213 (92.2%) 420 (90.7%) 142 (64.8%) 162 (76.1%) 304 (70.4%) 137 (67.2%) 146 (74.1%) 283 (70.6%) 79 (66.9%) 75 (70.1%) 154 (68.4%)

Missing 13 18 31 28 34 62 29 37 66

3b. Severity of abdominal distention

0 – No distention 25 (10.8%) 18 (7.8%) 43 (9.3%) 77 (35.2%) 51 (23.9%) 128 (29.6%) 68 (33.3%) 51 (25.9%) 119 (29.7%) 39 (33.1%) 32 (29.9%) 71 (31.6%)

10 3 (1.3%) 3 (1.3%) 6 (1.3%) 2 (0.9%) 6 (2.8%) 8 (1.9%) 5 (2.5%) 3 (1.5%) 8 (2.0%) 4 (3.4%) 0 (0.0%) 4 (1.8%)

20 11 (4.7%) 4 (1.7%) 15 (3.2%) 11 (5.0%) 11 (5.2%) 22 (5.1%) 18 (8.8%) 11 (5.6%) 29 (7.2%) 6 (5.1%) 5 (4.7%) 11 (4.9%)

30 17 (7.3%) 23 (10.0%) 40 (8.6%) 18 (8.2%) 30 (14.1%) 48 (11.1%) 19 (9.3%) 18 (9.1%) 37 (9.2%) 22 (18.6%) 16 (15.0%) 38 (16.9%)

40 29 (12.5%) 30 (13.0%) 59 (12.7%) 30 (13.7%) 22 (10.3%) 52 (12.0%) 24 (11.8%) 22 (11.2%) 46 (11.5%) 10 (8.5%) 13 (12.1%) 23 (10.2%)

50 24 (10.3%) 33 (14.3%) 57 (12.3%) 31 (14.2%) 23 (10.8%) 54 (12.5%) 25 (12.3%) 31 (15.7%) 56 (14.0%) 11 (9.3%) 10 (9.3%) 21 (9.3%)

60 39 (16.8%) 39 (16.9%) 78 (16.8%) 16 (7.3%) 25 (11.7%) 41 (9.5%) 20 (9.8%) 19 (9.6%) 39 (9.7%) 8 (6.8%) 11 (10.3%) 19 (8.4%)
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
Baseline Month 3 Month 6 Month 12

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total
(n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 147) (n = 144) (n = 291)

70 35 (15.1%) 35 (15.2%) 70 (15.1%) 15 (6.8%) 17 (8.0%) 32 (7.4%) 14 (6.9%) 20 (10.2%) 34 (8.5%) 8 (6.8%) 10 (9.3%) 18 (8.0%)

80 32 (13.8%) 22 (9.5%) 54 (11.7%) 14 (6.4%) 18 (8.5%) 32 (7.4%) 6 (2.9%) 11 (5.6%) 17 (4.2%) 5 (4.2%) 6 (5.6%) 11 (4.9%)

90 10 (4.3%) 10 (4.3%) 20 (4.3%) 2 (0.9%) 6 (2.8%) 8 (1.9%) 3 (1.5%) 7 (3.6%) 10 (2.5%) 1 (0.8%) 2 (1.9%) 3 (1.3%)

 00 – Very severe
1 7 (3.0%) 14 (6.1%) 21 (4.5%) 3 (1.4%) 4 (1.9%) 7 (1.6%) 2 (1.0%) 4 (2.0%) 6 (1.5%) 4 (3.4%) 2 (1.9%) 6 (2.7%)
distention

Missing 13 18 31 28 34 62 29 37 66

4. Dissatisfaction with bowel functioning in past 10 days?

 – Not
0 3 (1.3%) 2 (0.9%) 5 (1.1%) 14 (6.4%) 7 (3.3%) 21 (4.9%) 9 (4.4%) 8 (4.1%) 17 (4.2%) 7 (5.9%) 3 (2.8%) 10 (4.4%)
dissatisfied

10 2 (0.9%) 6 (2.6%) 8 (1.7%) 15 (6.8%) 11 (5.2%) 26 (6.0%) 15 (7.4%) 9 (4.6%) 24 (6.0%) 8 (6.8%) 12 (11.2%) 20 (8.9%)

20 14 (6.0%) 4 (1.7%) 18 (3.9%) 24 (11.0%) 28 (13.1%) 52 (12.0%) 25 (12.3%) 18 (9.1%) 43 (10.7%) 18 (15.3%) 10 (9.3%) 28 (12.4%)

30 12 (5.2%) 10 (4.3%) 22 (4.8%) 25 (11.4%) 19 (8.9%) 44 (10.2%) 19 (9.3%) 18 (9.1%) 37 (9.2%) 9 (7.6%) 11 (10.3%) 20 (8.9%)

40 15 (6.5%) 13 (5.7%) 28 (6.1%) 19 (8.7%) 25 (11.7%) 44 (10.2%) 20 (9.8%) 15 (7.6%) 35 (8.7%) 13 (11.0%) 10 (9.3%) 23 (10.2%)

Health Technology Assessment 2024 Vol. 28 No. 66

50 26 (11.2%) 25 (10.9%) 51 (11.0%) 28 (12.8%) 25 (11.7%) 53 (12.3%) 32 (15.7%) 26 (13.2%) 58 (14.5%) 10 (8.5%) 10 (9.3%) 20 (8.9%)

60 28 (12.1%) 40 (17.4%) 68 (14.7%) 24 (11.0%) 17 (8.0%) 41 (9.5%) 23 (11.3%) 26 (13.2%) 49 (12.2%) 18 (15.3%) 11 (10.3%) 29 (12.9%)

70 35 (15.1%) 39 (17.0%) 74 (16.0%) 28 (12.8%) 37 (17.4%) 65 (15.0%) 21 (10.3%) 28 (14.2%) 49 (12.2%) 16 (13.6%) 18 (16.8%) 34 (15.1%)

80 37 (15.9%) 52 (22.6%) 89 (19.3%) 21 (9.6%) 24 (11.3%) 45 (10.4%) 22 (10.8%) 26 (13.2%) 48 (12.0%) 11 (9.3%) 15 (14.0%) 26 (11.6%)

90 20 (8.6%) 20 (8.7%) 40 (8.7%) 7 (3.2%) 8 (3.8%) 15 (3.5%) 8 (3.9%) 12 (6.1%) 20 (5.0%) 3 (2.5%) 4 (3.7%) 7 (3.1%)

 00 – Very
1 40 (17.2%) 19 (8.3%) 59 (12.8%) 14 (6.4%) 12 (5.6%) 26 (6.0%) 10 (4.9%) 11 (5.6%) 21 (5.2%) 5 (4.2%) 3 (2.8%) 8 (3.6%)
dissatisfied

Missing 0 (0.0%) 1 1 13 18 31 28 34 62 29 37 66

continued
133
 134
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Appendix 1
TABLE 49 IBS-SSS items (continued)

Baseline Month 3 Month 6 Month 12

Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total Amitriptyline Placebo Total
(n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 232) (n = 231) (n = 463) (n = 147) (n = 144) (n = 291)

5. Interference of abdominal pain, discomfort, altered bowel function

0 – Not at all 4 (1.7%) 3 (1.3%) 7 (1.5%) 24 (11.0%) 14 (6.6%) 38 (8.8%) 22 (10.8%) 16 (8.1%) 38 (9.5%) 14 (11.9%) 13 (12.1%) 27 (12.0%)

10 7 (3.0%) 8 (3.5%) 15 (3.2%) 32 (14.6%) 29 (13.6%) 61 (14.1%) 39 (19.1%) 28 (14.2%) 67 (16.7%) 22 (18.6%) 20 (18.7%) 42 (18.7%)

20 17 (7.3%) 17 (7.4%) 34 (7.4%) 32 (14.6%) 26 (12.2%) 58 (13.4%) 25 (12.3%) 23 (11.7%) 48 (12.0%) 21 (17.8%) 10 (9.3%) 31 (13.8%)

30 28 (12.1%) 22 (9.6%) 50 (10.8%) 28 (12.8%) 28 (13.1%) 56 (13.0%) 25 (12.3%) 22 (11.2%) 47 (11.7%) 16 (13.6%) 20 (18.7%) 36 (16.0%)

40 20 (8.6%) 18 (7.8%) 38 (8.2%) 37 (16.9%) 21 (9.9%) 58 (13.4%) 21 (10.3%) 17 (8.6%) 38 (9.5%) 10 (8.5%) 9 (8.4%) 19 (8.4%)

50 30 (12.9%) 24 (10.4%) 54 (11.7%) 20 (9.1%) 29 (13.6%) 49 (11.3%) 25 (12.3%) 20 (10.2%) 45 (11.2%) 12 (10.2%) 4 (3.7%) 16 (7.1%)

60 34 (14.7%) 48 (20.9%) 82 (17.7%) 13 (5.9%) 17 (8.0%) 30 (6.9%) 11 (5.4%) 22 (11.2%) 33 (8.2%) 10 (8.5%) 9 (8.4%) 19 (8.4%)

70 31 (13.4%) 36 (15.7%) 67 (14.5%) 11 (5.0%) 29 (13.6%) 40 (9.3%) 12 (5.9%) 21 (10.7%) 33 (8.2%) 4 (3.4%) 12 (11.2%) 16 (7.1%)

80 28 (12.1%) 23 (10.0%) 51 (11.0%) 11 (5.0%) 12 (5.6%) 23 (5.3%) 13 (6.4%) 16 (8.1%) 29 (7.2%) 7 (5.9%) 9 (8.4%) 16 (7.1%)

90 12 (5.2%) 19 (8.3%) 31 (6.7%) 3 (1.4%) 5 (2.3%) 8 (1.9%) 4 (2.0%) 9 (4.6%) 13 (3.2%) 1 (0.8%) 1 (0.9%) 2 (0.9%)

100 – Completely 21 (9.1%) 12 (5.2%) 33 (7.1%) 8 (3.7%) 3 (1.4%) 11 (2.5%) 7 (3.4%) 3 (1.5%) 10 (2.5%) 1 (0.8%) 0 (0.0%) 1 (0.4%)

Missing 0 (0.0%) 1 1 13 18 31 28 34 62 29 37 66
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Pearson residuals for total score 6 m

15
2

1
10
Residual

Percent
0

5
–1

–2
0
100 200 300 –3 –2.2 –1.4 –0.6 0.2 1 1.8 2.6
Predicted mean Residual

3 Residual statistics
Minimum –2.351
2 Observations 401
Mean –6E-16
1 Maximum 2.5929
Std dev 0.9912
Residual

0 Fit statistics
Objective 4798.9
–1 AIC 4800.9
AICC 4800.9
BIC 4804.9
–2

–3
–3 –2 –1 0 1 2 3
Quantile

FIGURE 18 Plots of residuals for 6-month total IBS-SSS score.

Missing data exploration

We prespecified imputation by treatment arm including planned analysis covariates (data informing
stratification: recruitment hub, stool pattern, HADS-D score; and baseline score). Missing data
exploration was performed to identify additional auxiliary variables to include in the multiple imputation
models based on variables found to be predictive of missingness (see Table 7) or outcome for the primary
6-month total IBS-SSS score. Auxiliary variables explored were: baseline HADS-A, baseline WSAS, and
baseline PHQ-12 scores, age, sex and treatment indicators for whether patients were still on treatment
at 6 months.

Other potential baseline variables were not considered due to sparsity of the cells (marital status,
ethnicity, education level) or due to their value of information compared with other included auxiliary
variables (previous treatment for anxiety or depression, time since IBS diagnosis).

Univariable logistic and linear regression were used to identify auxiliary variables predictive of
missingness and outcome respectively (see Table 49). Overall, across trial arms, univariable analysis
found all auxiliary variables, except for sex, to be predictive (p < 0.05) of missing data status, outcome,
or both. Recruitment hub was predictive of missing data status; baseline HADS-D, baseline HADS-A,
baseline and PHQ-12 scores were predictive of outcome; and age, baseline IBS-SSS, and baseline
WSAS scores, and 6-month treatment status were predictive of both missing data status and outcome.
Although sex was not found to be predictive of missing data status or outcome for the 6-month total
IBS-SSS score, as sex was a covariate in the PHQ-12 analysis model (due to differences in available total
scores for male and females), we included sex in the multiple imputation models for all outcomes to
ensure consistency.
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 135
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Appendix 1

Missing data were, therefore, imputed by treatment arm via multiple imputation by chained equations
with 25 imputations, including recruitment hub, IBS subtype, sex, age, baseline questionnaire scores
(IBS-SSS, PHQ-12, HADS and WSAS), 3-month IBS-SSS score, and 6-month treatment status in
the model. The 3-month IBS-SSS score was imputed within the same model in a preliminary step,
incorporating 3-month (rather than 6-month) treatment status. The 12-month IBS-SSS score was
imputed in a further separate imputation model including the 12-month intention-to-treat population,
incorporating 12-month treatment status and 3- and 6-month IBS-SSS scores (also imputed in
preliminary steps based on 3- and 6-month treatment status, respectively).

The same imputation variables were incorporated into multiple imputations models for SGA or relief,
HADS-A, HADS-D, WSAS, PHQ-12 and acceptability outcomes.

Key secondary end point

Primary analysis
Residual plots (see Figure 20) show that assumptions for logistic regression hold; residuals fell within −2
to 2 and no extreme outliers were identified.

Secondary analysis
The score test for the proportional odds assumption [p = 0.903 (from complete case ordinal logistic
regression of SGA of relief of IBS symptoms)] indicated that the assumption holds and the ORs for the
treatment effect can be interpreted as constant across all possible cut points of the outcome.

Secondary end points: 3-month irritable bowel syndrome Severity Scoring System
and subjective global assessment

TABLE 50 Primary outcome missing data exploration – included in multiple imputation model

6-month total IBS-SSS score

Predictive of missingness, p-value Predictive of outcome, p-value

Covariates

Recruitment hub 0.0068 0.2626

Baseline HADS-D score 0.8202 0.0462

IBS subtype 0.8896 0.5023

Baseline IBS-SSS score 0.0140 < 0.0001

Potential auxiliary variables

Baseline HADS-A score 0.8471 0.0004

Baseline PHQ-12 score 0.2396 0.0003

Baseline WSAS score 0.0024 < 0.0001

Age 0.0489 0.0012

Sex 0.4370 0.4654

On treatment at 6 months < 0.0001 0.0003

136

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Month 3 (N = 463) Month 6 (N = 463) Month 12 (N = 291)

240

220

200

180 SGA of relief IBS

symptoms
160 1 – Completely
relieved
Frequency

140 2 – Considerably
relieved
120 3 – Somewhat
relieved
100 4 – Unchanged
5 – Worse
80 Missing

60

40

20

0
Amitriptyline Placebo Amitriptyline Placebo Amitriptyline Placebo

FIGURE 19 Subjective global assessment of relief of IBS symptoms at 3 months, 6 months and 12 months.

Influence diagnostics

1 1
Deviance residual
Pearson residual

0 0

–1 –1

–2 –2 Pricat
Non-responder
(score 4–5)
Responder
Standardised deviance residual
Standardised Pearson residual

1 1 (score 1–3)

0 0

–1 –1

–2 –2
0 100 200 300 400 0 100 200 300 400
Case number Case number

FIGURE 20 Plots of residuals for logistic regression of the 6-month SGA. From complete case logistic regression of SGA of
relief of IBS symptoms (symptoms completely, considerably, or somewhat relieved vs. unchanged or worse).

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 137
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 138
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Appendix 1
TABLE 51 Three-month total IBS-SSS score: linear regression – primary (ITT) and sensitivity (complete case, per protocol) analysis

Primary analysis (n = 463) Complete case (n = 433) Per protocol (n = 373)

Parameter estimates Std Parameter estimates Std Parameter estimates Std

(95% CI) Error p-value (95% CI) Error p-value (95% CI) Error p-value

Intercept 53.91 (19.01 to 88.82) 17.80 0.002 52.24 (17.67 to 86.81) 17.59 0.003 45.33 (8.24 to 82.42) 18.92 0.017

Treatment: amitriptyline (vs. −23.30 (−41.96 to −4.64) 9.52 0.014 −23.95 (−42.35 to −5.56) 9.36 0.011 −27.70 (−47.23 to −8.17) 9.96 0.005
placebo)

Baseline IBS-SSS score 0.49 (0.38 to 0.59) 0.05 < 0.49 (0.38 to 0.59) 0.05 < 0.50 (0.39 to 0.61) 0.06 < 0.001
0.001 0.001

IBS subtype (vs. IBS-M or IBS-U) 0.278

IBS-C 21.92 (−4.53 to 48.38) 13.50 0.104 21.06 (−5.50 to 47.63) 13.52 0.120 25.09 (−3.35 to 53.53) 14.51 0.084

IBS-D 2.35 (−17.85 to 22.55) 10.31 0.820 2.17 (−18.03 to 22.36) 10.27 0.833 13.85 (−7.54 to 35.25) 10.92 0.204

Baseline HADS-D score 2.14 (−0.60 to 4.87) 1.40 0.126 1.86 (−0.90 to 4.63) 1.41 0.185 1.39 (−1.50 to 4.28) 1.48 0.346

Recruitment hub (vs. Wessex) 0.119

West of England 4.94 (−15.29 to 25.18) 10.32 0.632 5.49 (−14.82 to 25.81) 10.34 0.595 −0.65 (−22.07 to 20.77) 10.93 0.953

West Yorkshire −20.83 (−47.31 to 5.66) 13.50 0.123 −21.86 (−48.03 to 4.32) 13.32 0.102 −24.57 (−53.02 to 3.89) 14.52 0.091
 TABLE 52 Three-month SGA of relief of IBS symptoms: logistic and ordinal regression – primary, sensitivity and secondary analysis
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
Sensitivity analysis Secondary analysis

Primary analysis (responder 1–3 vs. Complete case (responder 1–3 vs. Alternative responder definition Ordinal regressiona
4–5) (n = 463) 4–5) (n = 433) (responder 1–2 vs. 3–5) (n = 463) (n = 463)

P. p- Odds ratio p- Odds ratio Odds ratio Odds ratio

est. SE value (95% CI) P. est. SE value (95% CI) P. est. SE p-value (95% CI) P. est. SE p-value (95% CI)

Intercept 0.16 0.24 0.517 0.19 0.24 0.442 −1.14 0.28 < 0.0001

1 – Completely −3.69 0.36 < 0.0001

relieved

2 – Considerably −1.07 0.23 < 0.0001

relieved

3 – Somewhat 0.14 0.22 0.546

relieved

4 – Unchanged 3.28 0.34 < 0.0001

Treatment: amitripty- 0.53 0.20 0.008 1.70 (1.15 to 0.59 0.20 0.003 1.81 (1.23 to 0.59 0.22 0.008 1.81 (1.17 to 0.59 0.18 0.001 1.80 (1.26
line (vs. placebo) 2.53) 2.67) 2.79) to 2.58)b

IBS subtype (vs. 0.178

Health Technology Assessment 2024 Vol. 28 No. 66

IBS-M or IBS-U)

IBS-C −0.51 0.28 0.071 0.60 (0.34 to −0.53 0.28 0.064 0.59 (0.34 to 0.07 0.31 0.829 1.07 (0.58 to −0.34 0.26 0.188 0.71 (0.43
1.05) 1.03) 1.97) to 1.18)

IBS-D −0.14 0.21 0.503 0.87 (0.57 to −0.16 0.22 0.461 0.85 (0.56 to 0.02 0.24 0.936 1.02 (0.64 to −0.11 0.19 0.549 0.89 (0.61
1.32) 1.30) 1.63) to 1.30)

Baseline HADS-D −0.02 0.03 0.432 0.98 (0.92 to −0.02 0.03 0.451 0.98 (0.92 to −0.05 0.03 0.152 0.95 (0.89 to −0.04 0.03 0.139 0.96 (0.91
score 1.03) 1.04) 1.02) to 1.01)

Recruitment hub (vs. 0.643

Wessex)

West of England −0.05 0.21 0.833 0.96 (0.63 to −0.05 0.22 0.810 0.95 (0.62 to −0.11 0.24 0.650 0.90 (0.56 to −0.04 0.19 0.851 0.96 (0.66
1.46) 1.45) 1.44) to 1.41)

West Yorkshire 0.13 0.27 0.628 1.14 (0.67 to 0.21 0.28 0.452 1.24 (0.71 to 0.28 0.29 0.341 1.32 (0.74 to 0.20 0.25 0.426 1.22 (0.75
1.95) 2.15) 2.34) to 1.98)

a Intercept (vs. 5-worse) in ordinal regression.

b Complete case analysis gives treatment effect of: OR 1.87 (95% CI 1.32 to 2.66), p < 0.001.
139
 Appendix 1

Secondary end points: HADS-A, HADS-D, WSAS and PHQ-12

HADS-A score with unadjusted 95% CI – by population

1 – ITT population (N = 463) 2 – ITT 12 m population (N = 291)
8.5

8.0
HADS-A total scores

7.5

alloc
Amitriptyline
Placebo
7.0

6.5

6.0

Baseline Month 3 Month 6 Month 12 Baseline Month 3 Month 6 Month 12

HADS-D score with unadjusted 95% CI – by population

1 – ITT population (N = 463) 2 – ITT 12 m population (N = 291)

5.0
HADS-D total scores

4.5

alloc
Amitriptyline
4.0 Placebo

3.5

3.0
Baseline Month 3 Month 6 Month 12 Baseline Month 3 Month 6 Month 12

FIGURE 21 Unadjusted HADS-A and HADS-D scores with 95% CIs based on available data.

140

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

WSAS score with unadjusted 95% CI – by population

1 – ITT population (N = 463) 2 – ITT 12 m population (N = 291)

12
WSAS total score

10
alloc
Amitriptyline
Placebo

6
Baseline Month 3 Month 6 Month 12 Baseline Month 3 Month 6 Month 12

FIGURE 22 Unadjusted WSAS scores with 95% CIs based on available data.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 141
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 142

TABLE 53 Hospital Anxiety and Depression Scale-A (3, 6 and 12 months): linear regression – primary (ITT, 12-month ITT) and sensitivity (complete case) analysis
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Appendix 1
3 months 6 months 12 months

Parameter estimates Parameter estimates Parameter estimates

(95% CI) SE p-value (95% CI) SE p-value (95% CI) SE p-value

Primary analysis

Intercept 1.31 (0.49 to 2.14) 0.42 0.002 1.33 (0.52 to 2.14) 0.41 0.001 2.45 (1.15 to 3.74) 0.66 < 0.001

Treatment: amitriptyline (vs. placebo) 0.05 (−0.53 to 0.63) 0.30 0.861 0.08 (−0.49 to 0.65) 0.29 0.775 −0.38 (−1.22 to 0.47) 0.43 0.385

Baseline HADS-A score 0.71 (0.62 to 0.79) 0.04 < 0.001 0.69 (0.61 to 0.78) 0.04 <0.001 0.71 (0.59 to 0.83) 0.06 < 0.001

IBS subtype (vs. IBS-M or IBS-U)

IBS-C 0.02 (−0.82 to 0.86) 0.43 0.970 −0.17 (−1.00 to 0.67) 0.43 0.698 −0.90 (−2.08 to 0.29) 0.60 0.139

IBS-D −0.61 (−1.24 to 0.01) 0.32 0.055 −0.60 (−1.24 to 0.04) 0.33 0.068 −1.04 (−1.94 to −0.15) 0.46 0.022

Baseline HADS-D score −0.03 (−0.13 to 0.08) 0.05 0.597 0.04 (−0.07 to 0.14) 0.05 0.485 −0.00 (−0.17 to 0.16) 0.08 0.956

Recruitment hub (vs. Wessex)

West of England 0.48 (−0.15 to 1.12) 0.32 0.135 0.61 (−0.02 to 1.25) 0.32 0.059 0.45 (−0.48 to 1.38) 0.47 0.345

West Yorkshire 1.10 (0.27 to 1.94) 0.43 0.010 0.86 (0.03 to 1.69) 0.42 0.041 0.84 (−0.35 to 2.03) 0.60 0.165

Complete case

Intercept 1.22 (0.42 to 2.03) 0.41 0.003 1.32 (0.50 to 2.15) 0.42 0.002 2.07 (0.82 to 3.31) 0.63 0.001

Treatment: amitriptyline (vs. placebo) 0.07 (−0.50 to 0.64) 0.29 0.815 −0.07 (−0.66 to 0.51) 0.30 0.808 −0.18 (−1.02 to 0.65) 0.42 0.662

Baseline HADS-A score 0.71 (0.63 to 0.80) 0.04 < 0.001 0.68 (0.60 to 0.77) 0.04 < 0.001 0.74 (0.61 to 0.86) 0.06 < 0.001

IBS subtype (vs. IBS-M or IBS-U) 0.130 0.167 0.023

IBS-C −0.01 (−0.83 to 0.81) 0.42 0.974 −0.11 (−0.94 to 0.73) 0.42 0.800 −1.01 (−2.20 to 0.18) 0.60 0.097

IBS-D −0.60 (−1.23 to 0.02) 0.32 0.058 −0.60 (−1.24 to 0.04) 0.33 0.066 −1.23 (−2.15 to −0.32) 0.46 0.008

Baseline HADS-D score −0.03 (−0.14 to 0.07) 0.05 0.522 0.05 (−0.05 to 0.16) 0.05 0.322 −0.02 (−0.18 to 0.15) 0.08 0.847

Recruitment hub (vs. Wessex) 0.034 0.063 0.421

West of England 0.48 (−0.15 to 1.11) 0.32 0.131 0.59 (−0.05 to 1.23) 0.33 0.073 0.54 (−0.39 to 1.46) 0.47 0.252

West Yorkshire 1.05 (0.24 to 1.85) 0.41 0.011 0.89 (0.05 to 1.73) 0.43 0.038 0.67 (−0.56 to 1.89) 0.62 0.285
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
TABLE 54 Hospital Anxiety and Depression Scale-D (3, 6 and 12 months): linear regression – primary (ITT, 12-month ITT) and sensitivity (complete case) analysis.

3 months 6 months 12 months

Parameter estimates Parameter estimates Parameter estimates

(95% CI) SE p-value (95% CI) SE p-value (95% CI) SE p-value

Primary analysis

Intercept 1.13 (0.53 to 1.73) 0.31 < 0.001 1.10 (0.44 to 1.77) 0.34 0.001 2.16 (1.14 to 3.18) 0.52 < 0.001

Treatment: amitriptyline (vs. placebo) −0.22 (−0.71 to 0.26) 0.25 0.369 −0.20 (−0.75 to 0.34) 0.28 0.462 −0.88 (−1.71 to −0.06) 0.42 0.036

Baseline HADS-D score 0.62 (0.55 to 0.69) 0.04 < 0.001 0.72 (0.64 to 0.80) 0.04 < 0.001 0.66 (0.54 to 0.79) 0.06 < 0.001

IBS subtype (vs. IBS-M or IBS-U)

IBS-C −0.13 (−0.83 to 0.58) 0.36 0.724 −0.27 (−1.00 to 0.47) 0.38 0.476 −0.59 (−1.69 to 0.51) 0.56 0.293

IBS-D −0.40 (−0.93 to 0.13) 0.27 0.139 −0.33 (−0.94 to 0.28) 0.31 0.292 −0.50 (−1.38 to 0.37) 0.44 0.261

Recruitment hub (vs. Wessex)

West of England −0.01 (−0.54 to 0.52) 0.27 0.962 0.32 (−0.25 to 0.90) 0.29 0.274 0.03 (−0.85 to 0.92) 0.45 0.945

West Yorkshire 0.56 (−0.14 to 1.25) 0.36 0.119 0.28 (−0.48 to 1.05) 0.39 0.467 1.27 (0.23 to 2.30) 0.53 0.016

Health Technology Assessment 2024 Vol. 28 No. 66

Complete case

Intercept 1.17 (0.58 to 1.76) 0.30 < 0.001 1.05 (0.41 to 1.70) 0.33 0.001 2.04 (1.01 to 3.07) 0.52 < 0.001

Treatment: amitriptyline (vs. placebo) −0.27 (−0.75 to 0.21) 0.24 0.264 −0.37 (−0.89 to 0.15) 0.27 0.161 −0.85 (−1.63 to −0.07) 0.39 0.032

Baseline HADS-D score 0.62 (0.55 to 0.69) 0.04 < 0.001 0.71 (0.64 to 0.79) 0.04 < 0.001 0.68 (0.55 to 0.80) 0.06 < 0.001

IBS subtype (vs. IBS-M or IBS-U) 0.311 0.630 0.306

IBS-C −0.11 (−0.80 to 0.58) 0.35 0.750 −0.22 (−0.97 to 0.52) 0.38 0.558 −0.73 (−1.85 to 0.38) 0.57 0.195

IBS-D −0.40 (−0.93 to 0.12) 0.27 0.131 −0.27 (−0.84 to 0.30) 0.29 0.356 −0.54 (−1.39 to 0.32) 0.43 0.215

Recruitment hub (vs. Wessex) 0.493 0.386 0.368

West of England −0.01 (−0.53 to 0.52) 0.27 0.985 0.40 (−0.17 to 0.97) 0.29 0.170 0.10 (−0.77 to 0.96) 0.44 0.823

West Yorkshire 0.38 (−0.30 to 1.06) 0.34 0.275 0.24 (−0.52 to 0.99) 0.38 0.537 0.79 (−0.35 to 1.94) 0.58 0.175
143
 144

TABLE 55 Work and Social Adjustment Scale (3, 6 and 12 months): linear regression – primary (ITT, 12-month ITT) and sensitivity (complete case) analysis
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Appendix 1
3 months 6 months 12 months

Parameter estimates Parameter estimates Parameter estimates

(95% CI) SE p-value (95% CI) SE p-value (95% CI) SE p-value

Primary analysis

Intercept 2.46 (1.00 to 3.93) 0.75 0.001 2.17 (0.67 to 3.66) 0.76 0.005 2.33 (−0.09 to 4.75) 1.23 0.059

Treatment: amitriptyline (vs. placebo) −0.27 (−1.36 to 0.83) 0.56 0.633 −0.38 (−1.48 to 0.72) 0.56 0.499 −2.14 (−3.80 to −0.49) 0.84 0.011

Baseline WSAS score 0.46 (0.38 to 0.54) 0.04 < 0.001 0.49 (0.41 to 0.56) 0.04 < 0.001 0.45 (0.32 to 0.58) 0.06 < 0.001

IBS subtype (vs. IBS-M or IBS-U)

IBS-C −0.63 (−2.20 to 0.95) 0.80 0.435 −0.62 (−2.22 to 0.98) 0.81 0.435 0.38 (−2.00 to 2.75) 1.21 0.757

IBS-D 0.24 (−0.99 to 1.47) 0.63 0.698 0.22 (−0.98 to 1.43) 0.61 0.698 0.39 (−1.40 to 2.18) 0.91 0.669

Baseline HADS-D score 0.36 (0.20 to 0.53) 0.08 < 0.001 0.38 (0.22 to 0.55) 0.09 < 0.001 0.34 (0.07 to 0.60) 0.13 0.013

Recruitment hub (vs. Wessex)

West of England 0.95 (−0.23 to 2.12) 0.60 0.115 0.97 (−0.23 to 2.17) 0.61 0.114 2.11 (0.32 to 3.90) 0.91 0.021

West Yorkshire 0.17 (−1.39 to 1.72) 0.79 0.832 0.06 (−1.55 to 1.67) 0.82 0.943 2.79 (0.46 to 5.13) 1.19 0.019

Complete case

Intercept 1.73 (−0.01 to 3.46) 0.88 0.051 1.57 (−0.21 to 3.35) 0.90 0.083 1.78 (−0.51 to 4.07) 1.16 0.127

Treatment: amitriptyline (vs. placebo) −1.04 (−2.30 to 0.23) 0.64 0.108 −1.29 (−2.61 to 0.02) 0.67 0.054 −1.70 (−3.24 to −0.15) 0.78 0.031

Baseline WSAS score 0.46 (0.37 to 0.55) 0.05 < 0.001 0.49 (0.40 to 0.58) 0.05 < 0.001 0.45 (0.34 to 0.57) 0.06 < 0.001

IBS subtype (vs. IBS-M or IBS-U) 0.594 0.993 0.910

IBS-C −0.32 (−2.13 to 1.49) 0.93 0.729 0.05 (−1.87 to 1.97) 0.98 0.957 −0.32 (−2.61 to 1.96) 1.16 0.782

IBS-D −0.10 (−1.46 to 1.27) 0.70 0.890 0.09 (−1.35 to 1.52) 0.73 0.905 0.19 (−1.49 to 1.87) 0.85 0.825

Baseline HADS-D score 0.44 (0.24 to 0.64) 0.10 < 0.001 0.42 (0.22 to 0.62) 0.10 < 0.001 0.37 (0.12 to 0.62) 0.13 0.004

Recruitment hub (vs. Wessex) 0.246 0.227 0.013

West of England 1.27 (−0.10 to 2.64) 0.70 0.069 1.26 (−0.18 to 2.70) 0.73 0.085 2.58 (0.86 to 4.29) 0.87 0.003

West Yorkshire 1.04 (−0.78 to 2.86) 0.93 0.264 0.65 (−1.28 to 2.57) 0.98 0.508 1.14 (−1.19 to 3.48) 1.18 0.336
DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

TABLE 56 Patient Health Questionnaire-12 (6 months): linear regression – primary (ITT) and sensitivity (complete case)
analysis

6 months

Parameter estimates (95% CI) SE p-value

Primary analysis

Intercept 2.47 (1.68 to 3.25) 0.40 < 0.001

Treatment: amitriptyline (vs. placebo) −0.04 (−0.58 to 0.49) 0.27 0.877

Baseline PHQ-12 score 0.50 (0.41 to 0.59) 0.04 < 0.001

IBS subtype (vs. IBS-M or IBS-U)

IBS-C −0.26 (−1.01 to 0.48) 0.38 0.488

IBS-D −0.54 (−1.11 to 0.02) 0.29 0.061

Baseline HADS-D score 0.16 (0.08 to 0.25) 0.04 < 0.001

Recruitment hub (vs. Wessex)

West of England −0.13 (−0.69 to 0.43) 0.29 0.639

West Yorkshire −0.31 (−1.08 to 0.46) 0.39 0.427

Sex: males vs. females −0.11 (−0.68 to 0.46) 0.29 0.704

Complete case

Intercept 2.46 (1.67 to 3.25) 0.40 < 0.001

Treatment: amitriptyline (vs. placebo) −0.22 (−0.73 to 0.29) 0.26 0.400

Baseline PHQ-12 score 0.51 (0.42 to 0.60) 0.04 < 0.001

IBS subtype (vs. IBS-M or IBS-U) 0.180

IBS-C −0.26 (−0.99 to 0.47) 0.37 0.489

IBS-D −0.54 (−1.11 to 0.03) 0.29 0.064

Baseline HADS-D score 0.17 (0.08 to 0.25) 0.04 0.000

Recruitment hub (vs. Wessex) 0.557

West of England −0.20 (−0.76 to 0.36) 0.29 0.483

West Yorkshire −0.39 (−1.13 to 0.35) 0.38 0.306

Sex: males vs. females −0.15 (−0.72 to 0.42) 0.29 0.601

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 145
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Appendix 1

Secondary end point: weekly relief

Number of participants responding and reporting adequate relief at each week

1 132 (72.5%) 50 (27.5%) 50 (26.2%) 141 (73.8%)

2 130 (65.0%) 70 (35.0%) 58 (29.1%) 141 (70.9%)
3 96 (49.5%) 98 (50.5%) 59 (30.6%) 134 (69.4%)
4 124 (62.0%) 76 (38.0%) 69 (35.2%) 127 (64.8%)
5 111 (55.2%) 90 (44.8%) 78 (40.2%) 116 (59.8%)
6 85 (45.9%) 100 (54.1%) 79 (39.3%) 122 (60.7%)
7 104 (52.0%) 96 (48.0%) 76 (40.6%) 111 (59.4%)
8 91 (47.2%) 102 (52.8%) 63 (34.1%) 122 (65.9%)
9 87 (47.8%) 95 (52.2%) 78 (41.3%) 111 (58.7%)
10 98 (50.5%) 96 (49.5%) 76 (40.2%) 113 (59.8%)
In the past 7 days have you
11 87 (46.8%) 99 (53.2%) 68 (38.6%) 108 (61.4%)
had adequate relief of your
12 89 (46.8%) 101 (53.2%) 77 (40.3%) 114 (59.7%) IBS symptoms?
Week

13 86 (46.7%) 98 (53.3%) 71 (40.1%) 106 (59.9%) Yes

14 84 (45.9%) 99 (54.1%) 72 (38.7%) 114 (61.3%) No
15 78 (43.3%) 102 (56.7%) 70 (39.1%) 109 (60.9%)
16 85 (47.2%) 95 (52.8%) 67 (39.4%) 103 (60.6%)
17 80 (44.2%) 101 (55.8%) 76 (44.4%) 95 (55.6%)
18 84 (47.5%) 93 (52.5%) 81 (46.8%) 92 (53.2%)
19 84 (49.1%) 87 (50.9%) 76 (45.5%) 91 (54.5%)
20 88 (50.9%) 85 (49.1%) 71 (42.0%) 98 (58.0%)
21 90 (49.7%) 91 (50.3%) 78 (46.2%) 91 (53.8%)
22 81 (46.8%) 92 (53.2%) 75 (47.2%) 84 (52.8%)
23 85 (50.9%) 82 (49.1%) 70 (43.5%) 91 (56.5%)
24 77 (46.1%) 90 (53.9%) 73 (47.1%) 82 (52.9%)
25 88 (50.9%) 85 (49.1%) 57 (37.0%) 97 (63.0%)

240 220 200 180 160 140 120 100 80 60 40 20 0 20 40 60 80 100 120 140 160 180 200 220 240

Amitriptyline arm (N = 232) Number of participants Placebo arm (N = 231)

FIGURE 23 Number and proportion of participants reporting adequate relief each week.

146

NIHR Journals Library www.journalslibrary.nihr.ac.uk

 TABLE 57 Number and proportion of participants reporting adequate relief each week and model estimates
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
Observed N (%) satisfactory relief Model estimatesa

Amitriptyline (n = 232) Placebo (n = 231) Proportion satisfactory relief (95% CI)b

% of % of Odds ratio (95%

Week N % of completed % of randomised Missing N completed randomised Missing Amitriptyline Placebo CI) p-valuec

1 50 27.5 21.6 50 50 26.2 21.6 40 0.26 (0.20 to 0.35) 0.25 (0.19 to 0.33) 1.06 (0.67 to 1.68) 0.805

2 70 35.0 30.2 32 58 29.1 25.1 32 0.33 (0.26 to 0.41) 0.28 (0.21 to 0.36) 1.30 (0.84 to 1.99) 0.238

3 98 50.5 42.2 38 59 30.6 25.5 38 0.49 (0.40 to 0.57) 0.30 (0.23 to 0.38) 2.20 (1.45 to 3.34) < 0.001

4 76 38.0 32.8 32 69 35.2 29.9 35 0.38 (0.30 to 0.47) 0.35 (0.27 to 0.43) 1.17 (0.78 to 1.75) 0.448

5 90 44.8 38.8 31 78 40.2 33.8 37 0.44 (0.36 to 0.53) 0.39 (0.31 to 0.47) 1.26 (0.85 to 1.87) 0.255

6 100 54.1 43.1 47 79 39.3 34.2 30 0.54 (0.46 to 0.62) 0.39 (0.31 to 0.47) 1.87 (1.26 to 2.78) 0.002

7 96 48.0 41.4 32 76 40.6 32.9 44 0.48 (0.39 to 0.56) 0.40 (0.32 to 0.48) 1.37 (0.92 to 2.04) 0.116

8 102 52.8 44.0 39 63 34.1 27.3 46 0.51 (0.43 to 0.60) 0.34 (0.26 to 0.42) 2.09 (1.41 to 3.12) < 0.001

9 95 52.2 40.9 50 78 41.3 33.8 42 0.50 (0.41 to 0.59) 0.39 (0.31 to 0.48) 1.56 (1.05 to 2.32) 0.029

10 96 49.5 41.4 38 76 40.2 32.9 42 0.49 (0.40 to 0.57) 0.39 (0.31 to 0.47) 1.46 (0.99 to 2.16) 0.054

Health Technology Assessment 2024 Vol. 28 No. 66

11 99 53.2 42.7 46 68 38.6 29.4 55 0.52 (0.43 to 0.61) 0.37 (0.29 to 0.46) 1.84 (1.24 to 2.73) 0.003

12 101 53.2 43.5 42 77 40.3 33.3 40 0.53 (0.44 to 0.61) 0.39 (0.31 to 0.48) 1.77 (1.19 to 2.63) 0.005

13 98 53.3 42.2 48 71 40.1 30.7 54 0.53 (0.45 to 0.62) 0.39 (0.31 to 0.48) 1.76 (1.18 to 2.61) 0.005

14 99 54.1 42.7 49 72 38.7 31.2 45 0.53 (0.45 to 0.61) 0.36 (0.28 to 0.44) 2.03 (1.36 to 3.04) < 0.001

15 102 56.7 44.0 52 70 39.1 30.3 52 0.54 (0.46 to 0.63) 0.37 (0.29 to 0.45) 2.05 (1.37 to 3.06) < 0.001

16 95 52.8 40.9 52 67 39.4 29.0 61 0.51 (0.42 to 0.60) 0.38 (0.30 to 0.46) 1.71 (1.14 to 2.58) 0.010

17 101 55.8 43.5 51 76 44.4 32.9 60 0.54 (0.45 to 0.63) 0.45 (0.36 to 0.54) 1.45 (0.97 to 2.17) 0.068

18 93 52.5 40.1 55 81 46.8 35.1 58 0.52 (0.44 to 0.61) 0.44 (0.35 to 0.52) 1.43 (0.96 to 2.13) 0.080

19 87 50.9 37.5 61 76 45.5 32.9 64 0.50 (0.41 to 0.59) 0.41 (0.33 to 0.50) 1.41 (0.94 to 2.12) 0.100

20 85 49.1 36.6 59 71 42.0 30.7 62 0.47 (0.39 to 0.56) 0.40 (0.32 to 0.49) 1.33 (0.89 to 2.01) 0.167

continued
147
 148
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Appendix 1
TABLE 57 INumber and proportion of participants reporting adequate relief each week and model estimates (continued)

Observed N (%) satisfactory relief Model estimatesa

Amitriptyline (n = 232) Placebo (n = 231) Proportion satisfactory relief (95% CI)b

% of % of Odds ratio (95%

Week N % of completed % of randomised Missing N completed randomised Missing Amitriptyline Placebo CI) p-valuec

21 91 50.3 39.2 51 78 46.2 33.8 62 0.49 (0.40 to 0.57) 0.42 (0.34 to 0.51) 1.30 (0.87 to 1.94) 0.194

22 92 53.2 39.7 59 75 47.2 32.5 72 0.53 (0.44 to 0.61) 0.44 (0.36 to 0.53) 1.42 (0.95 to 2.13) 0.084

23 82 49.1 35.3 65 70 43.5 30.3 70 0.50 (0.42 to 0.59) 0.40 (0.31 to 0.48) 1.56 (1.03 to 2.36) 0.035

24 90 53.9 38.8 65 73 47.1 31.6 76 0.51 (0.42 to 0.60) 0.41 (0.33 to 0.50) 1.50 (1.00 to 2.26) 0.052

25 85 49.1 36.6 59 57 37.0 24.7 77 0.48 (0.39 to 0.56) 0.32 (0.24 to 0.41) 1.92 (1.25 to 2.95) 0.003

Overall 1.56 (1.20 to 2.03) < 0.001

a Covariates: West Yorkshire vs. Wessex OR 1.07 (0.74, 1.53), p = 0.726; West of England vs. Wessex OR 0.85 (0.63, 1.15), p = 0.295; HADS-D score OR 0.96 (0.93, 1.00), p = 0.075;
IBS-C vs. IBS-M or IBS-U OR 1.04 (0.70, 1.55), p = 0.844; IBS-D vs. IBS-M or IBS-U OR 1.14 (0.86, 1.52), p = 0.358.
b ‘Typical’ participant, Wessex, IBS-M, HADS-D score = 4.
c Rows are emphasised in bold where good evidence (p < 0.05) of effect was observed, and in italics where weak evidence (p < 0.1) of effect was observed.
 Secondary end point: tolerability
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
TABLE 58 Participant-reported tolerability on the ASEC: number of participants reporting symptoms (mild, moderate or severe) at 3, 6 and 12 months for participants on trial medication

Month 3 Month 6 Month 12

Amitriptyline Placebo Total (n = 385) Amitriptyline Placebo (n = 152) Total (n = 318) Amitriptyline Placebo Total
(n = 193) (%) (n = 192) (%) (%) (n = 166) (%) (%) (%) (n = 61) (%) (n = 56) (%) (n = 117) (%)

1. Dry mouth 122 (63.2) 87 (45.3) 209 (54.3) 90 (54.2 56 (36.8) 146 (45.9) 33 (54.1) 24 (42.9) 57 (48.7)

Linked to trial 84 (43.5) 45 (23.4) 129 (33.5) 57 (34.3) 26 (17.1) 83 (26.1) 22 (36.1) 9 (16.1) 31 (26.5)
medication

2. Drowsiness 128 (66.3) 67 (34.9) 195 (50.6) 88 (53.0) 52 (34.2) 140 (44.0) 28 (45.9) 25 (44.6) 53 (45.3)

Linked to trial 95 (49.2) 27 (14.1) 122 (31.7) 56 (33.7) 13 (8.6) 69 (21.7) 16 (26.2) 8 (14.3) 24 (20.5)
medication

3. Insomnia (diffi- 78 (40.4) 108 (56.3) 186 (48.3) 77 (46.4) 96 (63.2) 173 (54.4) 25 (41.0) 35 (62.5) 60 (51.3)
culty sleeping%)

Linked to trial 10 (5.2) 13 (6.8) 23 (6.0) 4 (2.4) 5 (3.3) 9 (2.8) 5 (8.2) 2 (3.6) 7 (6.0)
medication

4. Blurred vision 29 (15.0) 24 (12.5) 53 (13.8) 28 (16.9) 14 (9.2) 42 (13.2) 9 (14.8) 10 (17.9) 19 (16.2)

Health Technology Assessment 2024 Vol. 28 No. 66

Linked to trial 5 (2.6) 2 (1.0) 7 (1.8) 4 (2.4) 1 (0.7) 5 (1.6) 4 (6.6) 0 (0.0) 4 (3.4)
medication

5. Headache 74 (38.3) 85 (44.3) 159 (41.3) 78 (47.0) 80 (52.6) 158 (49.7) 21 (34.4) 23 (41.1) 44 (37.6)

Linked to trial 14 (7.3) 14 (7.3) 28 (7.3) 7 (4.2) 5 (3.3) 12 (3.8) 4 (6.6) 0 (0.0) 4 (3.4)
medication

6. Constipation 110 (57.0) 89 (46.4) 199 (51.7) 93 (56.0) 78 (51.3) 171 (53.8) 30 (49.2) 31 (55.4) 61 (52.1)

Linked to trial 36 (18.7) 21 (10.9) 57 (14.8) 22 (13.3) 13 (8.6) 35 (11.0) 8 (13.1) 3 (5.4) 11 (9.4)
medication

7. Diarrhoea 117 (60.6) 126 (65.6) 243 (63.1) 98 (59.0) 103 67.8) 201 (63.2) 37 (60.7) 40 (71.4) 77 (65.8)

Linked to trial 21 (10.9) 16 (8.3) 37 (9.6) 15 (9.0) 7 (4.6) 22 (6.9) 5 (8.2) 2 (3.6) 7 (6.0)
medication

8. Increased 54 (28.0) 44 (22.9) 98 (25.5) 45 (27.1) 34 (22.4) 79 (24.8) 21 (34.4) 15 (26.8) 36 (30.8)
appetite

continued
149
 150
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Appendix 1
TABLE 58 Participant-reported tolerability on the ASEC: number of participants reporting symptoms (mild, moderate, or severe) at 3, 6, and 12 months for participants on trial
medication (continued)

Month 3 Month 6 Month 12

Amitriptyline Placebo Total (n = 385) Amitriptyline Placebo (n = 152) Total (n = 318) Amitriptyline Placebo Total
(n = 193) (%) (n = 192) (%) (%) (n = 166) (%) (%) (%) (n = 61) (%) (n = 56) (%) (n = 117) (%)

Linked to trial 20 (10.4) 17 (8.9) 37 (9.6) 16 (9.6) 4 (2.6) 20 (6.3) 6 (9.8) 1 (1.8) 7 (6.0)
medication

9. Decreased 34 (17.6) 28 (14.6) 62 (16.1) 17 (10.2) 22 (14.5) 39 (12.3) 5 (8.2) 6 (10.7) 11 (9.4)
appetite

Linked to trial 10 (5.2) 5 (2.6) 15 (3.9) 3 (1.8) 2 (1.3) 5 (1.6) 1 (1.6) 1 (1.8) 2 (1.7)
medication

10. Nausea or 35 (18.1) 26 (13.5) 61 (15.8) 26 (15.7) 26 (17.1) 52 (16.4) 9 (14.8) 6 (10.7) 15 (12.8)
vomiting

Linked to trial 11 (5.7) 7 (3.6) 18 (4.7) 3 (1.8) 3 (2.0) 6 (1.9) 1 (1.6) 1 (1.8) 2 (1.7)
medication

11. Problems with 31 (16.1) 23 (12.0 54 (14.0) 36 (21.7) 20 (13.2) 56 (17.6) 11 (18.0) 10 (17.9) 21 (17.9)
urination

Linked to trial 9 (4.7) 3 (1.6) 12 (3.1) 11 (6.6) 3 (2.0) 14 (4.4) 5 (8.2) 1 (1.8) 6 (5.1)
medication

12. Problems with 29 (15.0) 23 (12.0) 52 (13.5) 24 (14.5) 16 (10.5) 40 (12.6) 9 (14.8) 8 (14.3) 17 (14.5)
sexual function

Linked to trial 6 (3.1) 3 (1.6) 9 (2.3) 4 (2.4) 0 (0.0) 4 (1.3) 3 (4.9) 0 (0.0) 3 (2.6)
medication

13. Palpitations 56 (29.0) 37 (19.3) 93 (24.2) 41 (24.7) 38 (25.0) 79 (24.8) 14 (23.0) 18 (32.1) 32 (27.4)

Linked to trial 18 (9.3) 4 (2.1) 22 (5.7) 6 (3.6) 1 (0.7) 7 (2.2) 3 (4.9) 2 (3.6) 5 (4.3)
medication

14. Feeling light- 73 (37.8) 63 (32.8) 136 (35.3) 69 (41.6) 54 (35.5) 123 (38.7) 22 (36.1) 22 (39.3) 44 (37.6)
headed on standing

Linked to trial 19 (9.8) 7 (3.6) 26 (6.8) 13 (7.8) 6 (3.9) 19 (6.0) 2 (3.3) 2 (3.6) 4 (3.4)
medication

15. Feeling like the 29 (15.0) 24 (12.5) 53 (13.8) 20 (12.0) 19 (12.5) 39 (12.3) 6 (9.8) 6 (10.7) 12 (10.3)
room is spinning
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
TABLE 58 Participant-reported tolerability on the ASEC: number of participants reporting symptoms (mild, moderate, or severe) at 3, 6, and 12 months for participants on trial
medication (continued)

Month 3 Month 6 Month 12

Amitriptyline Placebo Total (n = 385) Amitriptyline Placebo (n = 152) Total (n = 318) Amitriptyline Placebo Total
(n = 193) (%) (n = 192) (%) (%) (n = 166) (%) (%) (%) (n = 61) (%) (n = 56) (%) (n = 117) (%)

Linked to trial 8 (4.1) 3 (1.6) 11 (2.9) 5 (3.0) 1 (0.7) 6 (1.9) 1 (1.6) 0 (0.0) 1 (0.9)
medication

16. Sweating 71 (36.8) 60 (31.3) 131 (34.0) 54 (32.5) 49 (32.2) 103 (32.4) 19 (31.1) 23 (41.1) 42 (35.9)

Linked to trial 20 (10.4) 8 (4.2) 28 (7.3) 9 (5.4) 6 (3.9) 15 (4.7) 5 (8.2) 2 (3.6) 7 (6.0)
medication

17. Increased body 56 (29.0) 48 (25.0) 104 (27.0) 35 (21.1) 36 (23.7) 71 (22.3) 13 (21.3) 10 (17.9) 23 (19.7)
temperature

Linked to trial 14 (7.3) 15 (7.8) 29 (7.5) 7 (4.2) 3 (2.0) 10 (3.1) 2 (3.3) 2 (3.6) 4 (3.4)
medication

18. Tremor 17 (8.8) 13 (6.8) 30 (7.8) 13 (7.8) 11 (7.2) 24 (7.5) 8 (13.1) 2 (3.6) 10 (8.5)

Linked to trial 1 (0.5) 2 (1.0) 3 (0.8) 4 (2.4) 2 (1.3) 6 (1.9) 2 (3.3) 0 (0.0) 2 (1.7)
medication

Health Technology Assessment 2024 Vol. 28 No. 66

19. Disorientation 24 (12.4) 8 (4.2) 32 (8.3) 13 (7.8) 10 (6.6) 23 (7.2) 2 (3.3) 4 (7.1) 6 (5.1)

Linked to trial 10 (5.2) 1 (0.5) 11 (2.9) 4 (2.4) 1 (0.7) 5 (1.6) 2 (3.3) 0 (0.0) 2 (1.7)
medication

20. Yawning 67 (34.7) 68 (35.4) 135 (35.1) 63 (38.0) 50 (32.9) 113 (35.5) 18 (29.5) 25 (44.6) 43 (36.8)

Linked to trial 19 (9.8) 10 (5.2) 29 (7.5) 13 (7.8) 3 (2.0) 16 (5.0) 3 (4.9) 1 (1.8) 4 (3.4)
medication

21. Weight gain 72 (37.3) 59 (30.7) 131 (34.0) 73 (44.0) 49 (32.2) 122 (38.4) 28 (45.9) 25 (44.6) 53 (45.3)

Linked to trial 22 (11.4) 14 (7.3) 36 (9.4) 20 (12.0) 10 (6.6) 30 (9.4) 10 (16.4) 3 (5.4) 13 (11.1)
medication
151
 152

Secondary end points: 12-month irritable bowel syndrome with constipation and subjective global assessment
NIHR Journals Library www.journalslibrary.nihr.ac.uk

Appendix 1
TABLE 59 Twelve-month total IBS-SSS score: linear regression – primary (12-month ITT) and sensitivity (complete case) analysis

Primary analysis (n = 291) Complete case (n = 225)

Parameter estimates (95% CI) Std error p-value Parameter estimates (95% CI) Std error p-value

Intercept 15.37 (−34.48 to 65.22) 25.41 0.545 11.24 (−42.15 to 64.63) 27.09 0.679

Treatment: amitriptyline (vs. placebo) −22.59 (−49.35 to 4.16) 13.60 0.098 −24.34 (−50.49 to 1.81) 13.27 0.068

Baseline IBS-SSS score 0.49 (0.35 to 0.63) 0.07 < 0.001 0.50 (0.34 to 0.65) 0.08 < 0.001

IBS subtype (vs. IBS-M or IBS-U) 0.893

IBS-C 8.27 (−26.68 to 43.22) 17.81 0.643 4.08 (−33.06 to 41.22) 18.84 0.829

IBS-D −5.46 (−32.94 to 22.01) 14.00 0.696 −4.63 (−33.45 to 24.18) 14.62 0.752

Baseline HADS-D score 7.20 (3.31 to 11.09) 1.98 < 0.001 7.13 (3.01 to 11.24) 2.09 0.001

Recruitment hub (vs. Wessex) 0.077

West of England 22.84 (−6.15 to 51.83) 14.77 0.122 26.30 (−2.81 to 55.40) 14.77 0.076

West Yorkshire −15.79 (−56.90 to 25.32) 20.78 0.449 −9.95 (−48.24 to 28.35) 19.43 0.609
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
TABLE 60 Twelve-month SGA of relief of IBS symptoms: logistic regression – primary and sensitivity analysis

Primary analysis (n = 291) Complete case (n = 225)

P. est. SE p-value Odds ratio (95% CI) P. est. SE p-value Odds ratio (95% CI)

Intercept −0.01 0.35 0.984 0.20 0.36 0.577

Treatment: amitriptyline (vs. placebo) 0.46 0.26 0.083 1.58 (0.94 to 2.64) 0.55 0.27 0.046 1.73 (1.01 to 2.95)

IBS subtype (vs. IBS-M or IBS-U) 0.870

IBS-C −0.14 0.39 0.720 0.87 (0.40 to 1.88) −0.13 0.39 0.732 0.88 (0.41 to 1.88)

Health Technology Assessment 2024 Vol. 28 No. 66

IBS-D −0.06 0.29 0.839 0.94 (0.53 to 1.67) −0.15 0.30 0.619 0.86 (0.48 to 1.55)

Baseline HADS-D score −0.05 0.04 0.216 0.95 (0.88 to 1.03) −0.06 0.04 0.164 0.94 (0.87 to 1.02)

Recruitment hub (vs. Wessex) 0.231

West of England −0.14 0.29 0.635 0.87 (0.49 to 1.54) −0.21 0.30 0.479 0.81 (0.45 to 1.46)

West Yorkshire 0.31 0.39 0.433 1.36 (0.63 to 2.92) 0.46 0.41 0.262 1.58 (0.71 to 3.52)
153
 154

Further exploratory analysis

NIHR Journals Library www.journalslibrary.nihr.ac.uk

Appendix 1
Reduction in total irritable bowel syndrome with constipation score and item scores at 3 and 6 months

TABLE 61 ≥ 50-point reduction in total IBS-SSS score at 3 and 6 months: logistic regression (n = 463, multiple imputation)a

3 months 6 months

Parameter estimate Std error p-value Odds ratio (95% CI) Parameter estimate Std error p-value Odds ratio (95% CI)

Intercept −1.39 0.40 0.001 −1.33 0.41 0.001

Treatment: amitriptyline (vs. placebo) 0.40 0.22 0.068 1.49 (0.97 to 2.28) 0.39 0.22 0.068 1.48 (0.97 to 2.27)

Baseline total IBS-SSS score 0.01 0.00 < 0.001 1.01 (1.01 to 1.01) 0.01 0.00 < 0.001 1.01 (1.00 to 1.01)

IBS subtype (vs. IBS-M or IBS-U)

IBS-C −0.67 0.31 0.028 0.51 (0.28 to 0.93) −0.56 0.30 0.059 0.57 (0.32 to 1.02)

IBS-D 0.03 0.23 0.912 1.03 (0.65 to 1.63) −0.06 0.23 0.805 0.94 (0.60 to 1.49)

Baseline HADS-D score −0.07 0.03 0.019 0.93 (0.87 to 0.99) 0.02 0.03 0.640 1.02 (0.95 to 1.08)

Recruitment hub (vs. Wessex)

West of England −0.09 0.23 0.684 0.91 (0.58 to 1.43) −0.28 0.23 0.215 0.75 (0.48 to 1.18)

West Yorkshire 0.31 0.32 0.325 1.37 (0.73 to 2.54) −0.45 0.32 0.162 0.64 (0.34 to 1.20)

a Complete case analysis of ≥ 50-point reduction in 6-month total IBS-SSS score gives treatment effect: OR 1.58 (95% CI 1.04 to 2.41), p = 0.034 at 3 months and OR 1.62 (95% CI
1.06 to 2.47), p = 0.025 at 6 months.
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use,
Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health

DOI: 10.3310/BFCR7986
TABLE 62 ≥ 30% reduction in IBS-SSS abdominal pain at 3 and 6 months: logistic regression (n = 463, multiple imputation)a

3 months 6 months

Parameter estimate Std error Odds ratio (95% CI) p-value Parameter estimate Std error Odds ratio (95% CI) p-value

Intercept −0.54 0.37 0.138 −0.63 0.37 0.089

Treatment: amitriptyline (vs. placebo) 0.17 0.20 1.19 (0.81 to 1.75) 0.377 0.51 0.20 1.66 (1.12 to 2.46) 0.012

Baseline IBS-SSS score 0.00 0.00 1.00 (1.00 to 1.01) 0.006 0.00 0.00 1.00 (1.00 to 1.00) 0.101

IBS subtype (vs. IBS-M or IBS-U)

IBS-C 0.00 0.28 1.00 (0.58 to 1.74) 0.992 −0.38 0.29 0.68 (0.39 to 1.20) 0.188

IBS-D 0.22 0.22 1.25 (0.82 to 1.90) 0.305 −0.07 0.23 0.93 (0.59 to 1.46) 0.751

Health Technology Assessment 2024 Vol. 28 No. 66

Baseline HADS-D score −0.08 0.03 0.93 (0.87 to 0.98) 0.009 −0.02 0.03 0.98 (0.93 to 1.04) 0.559

Recruitment hub (vs. Wessex)

West of England −0.36 0.22 0.70 (0.46 to 1.06) 0.094 −0.15 0.22 0.86 (0.56 to 1.31) 0.482

West Yorkshire −0.03 0.29 0.97 (0.55 to 1.70) 0.914 0.21 0.29 1.23 (0.70 to 2.18) 0.472

a Complete case analysis of ≥ 30% reduction in IBS-SSS abdominal pain gives treatment effect: OR 1.28 (95% CI 0.870 to 1.883), p = 0.2106 at 3 months and 1.739 (95% CI 1.165 to
2.597), p = 0.0068 at 6 months.
155
 Appendix 1

TABLE 63 ≥ 30% reduction in IBS-SSS abdominal distention: logistic regression (n = 463, multiple imputation)a

3 months 6 months

Parameter Std Odds ratio Parameter Std Odds ratio

estimate error (95% CI) p-value estimate error (95% CI) p-value

Intercept −0.83 0.37 0.024 −0.93 0.39 0.017

Treatment: amitripty- 0.21 0.20 1.23 (0.83 to 0.303 0.28 0.21 1.33 (0.89 to 0.171
line (vs. placebo) 1.82) 1.99)

Baseline IBS-SSS 0.00 0.00 1.00 (1.00 to 0.124 0.00 0.00 1.00 (1.00 to 0.115
score 1.00) 1.00)

IBS subtype (vs. IBS-M or IBS-U)

IBS-C 0.23 0.28 1.26 (0.73 to 0.404 0.01 0.29 1.01 (0.57 to 0.966
2.17) 1.78)

IBS-D −0.09 0.22 0.91 (0.60 to 0.677 0.10 0.22 1.11 (0.71 to 0.650
1.40) 1.71)

Baseline HADS-D −0.02 0.03 0.98 (0.93 to 0.588 −0.01 0.03 0.99 (0.94 to 0.819
score 1.04) 1.05)

Recruitment hub (vs. Wessex)

West of England −0.04 0.22 0.96 (0.63 to 0.863 −0.25 0.22 0.78 (0.51 to 0.269
1.47) 1.21)

West Yorkshire 0.01 0.28 1.01 (0.59 to 0.958 −0.07 0.29 0.93 (0.52 to 0.805
1.75) 1.65)

a Complete case analysis of ≥ 30% reduction in IBS-SSS abdominal distention gives treatment effect of: OR 1.24 (95% CI
0.844 to 1.821), p = 0.2725 at 3 months and 1.468 (95% CI 0.982 to 2.195), p = 0.061 at 6 months.

Moderator analysis of 6-month total irritable bowel syndrome with constipation

score

TABLE 64 Mean (SD) total IBS-SSS score at baseline and 6 months by IBS subtype and recruitment hub

Baseline Month 6

Amitriptyline, Placebo, mean Amitriptyline, Placebo, mean Total, mean

mean (SD) (SD) Total, mean (SD) mean (SD) (SD) (SD)

IBS subtype

IBS-C 253.5 (98.36) 298.1 (102.92) 274.9 (102.40) 165.3 (102.52) 239.0 (119.84) 199.4
(116.08)

IBS-D 269.9 (86.94) 263.4 (88.65) 266.7 (87.60) 163.8 (103.69) 199.0 (110.84) 180.9
(108.33)

IBS-M or 284.6 (89.80) 270.3 (86.13) 277.3 (88.01) 178.6 (113.99) 187.5 (113.92) 183.1
IBS-U (113.72)

Recruitment hub

West 288.4 (88.29) 290.4 (104.31) 289.4 (96.17) 138.5 (101.39) 196.1 (127.84) 166.9
Yorkshire (117.91)

Wessex 260.6 (95.54) 271.2 (94.51) 265.8 (94.93) 168.2 (104.93) 201.9 (112.10) 185.1
(109.58)

West of 279.9 (85.16) 264.3 (77.66) 272.1 (81.65) 185.3 (111.16) 199.9 (112.58) 192.3
England (111.73)

156

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Fit for total score

With 95% confidence limits

200
Linear predictor

alloc
Amitriptyline
Placebo
150

100

West Yorkshire West of England Southampton

Recruitment hub

Fit computed at total score base = 280 HADSRandF07P1 = 4 StoolRand = mixed and unclassified stool pattern

FIGURE 24 Moderating effect of recruitment hub on the total IBS-SSS score treatment effect at 6 months.

IBS-SSS score with unadjusted 95% CI – by recruitment hub

Baseline Month 3 Month 6 Month 12

300

250
Total IBS-SSS score

Recruitment hub
West Yorkshire
200
Southampton
West of England

150

100

Amitriptyline Placebo Amitriptyline Placebo Amitriptyline Placebo Amitriptyline Placebo

FIGURE 25 Total IBS-SSS score by treatment arm and recruitment hub.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 157
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Appendix 1

Fit for total score

With 95% confidence limits

250
Linear predictor

alloc
Amitriptyline
Placebo
200

150

Constipation predominant Diarrhoea predominant Mixed and unclassified stool pattern

Stool pattern

Fit computed at total score base = 280 HADSRandF07P1 = 4 RecHub F07P1 = Southampton

FIGURE 26 Moderating effect of IBS subtype on the total IBS-SSS score treatment effect at 6 months.

IBS-SSS score with unadjusted 95% CI - by stool type

Baseline Month 3 Month 6 Month 12

300

250
Total IBS-SSS score

Stool pattern
Constipation predominant
Diarrhoea predominant
Mixed and unclassified
200 stool pattern

150

100
Amitriptyline Placebo Amitriptyline Placebo Amitriptyline Placebo Amitriptyline Placebo

FIGURE 27 Total IBS-SSS score by treatment arm and IBS subtype.

158

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Fit for total score

With 95% confidence limits

300
Linear predictor

200 alloc
Amitriptyline
Placebo

100

0
0 100 200 300 400 500
Total IBS-SSS score (baseline)

Fit computed at HADSRandF07P1 = 4 RecHubF07P1 = Southampton StoolRand = mixed and unclassified stool pattern

FIGURE 28 Moderating effect of baseline IBS-SSS score on the total IBS-SSS score treatment effect at 6 months.

Fit for total score

With 95% confidence limits
300

250
Linear predictor

alloc
200 Amitriptyline
Placebo

150

100
0 5 10 15 20
HADS-A total scores

Fit computed at total score base = 280 HADSRandF07P1 = 4 RecHubF07P1 = Southampton StoolRand = mixed and
unclassified stool pattern

FIGURE 29 Moderating effect of baseline HADS-A score on total IBS-SSS score at 6 months.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 159
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
 Appendix 1

Fit for total score

With 95% confidence limits
300

250
Linear predictor

alloc
Amitriptyline
Placebo
200

150

0 5 10 15
HADS-D score
Fit computed at total score base = 280 RecHubF07P1 = Southampton StoolRand = mixed and unclassified stool pattern

FIGURE 30 Moderating effect of baseline HADS-D score on total IBS-SSS score at 6 months.

Moderator analysis of 6-month subjective global assessment of relief of irritable bowel

syndrome symptoms

Predicted probabilities for Pricat = responder (score 1–3)

With 95% confidence limits
1.0

0.8

0.6
Probability

alloc
Amitriptyline
Placebo
0.4

0.2

0.0
Constipation predominant Diarrhoea predominant Mixed and unclassified stool pattern
Stool pattern

Fit computed at HADSRandF07P1 = 4 RecHubF07P1 = Southampton

FIGURE 31 Moderating effect of IBS subtype on SGA of relief of IBS symptoms treatment effect at 6 months.

160

NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/BFCR7986 Health Technology Assessment 2024 Vol. 28 No. 66

Subjective global assessment of relief of IBS symptoms by randomised treatment arm

Constipation predominant Diarrhoea predominant Mixed and unclassified stool pattern
110

100

90

80
SGA of relief of IBS symptoms
70
1 – Completely relieved
Frequency

60 2 – Considerably relieved
3 – Somewhat relieved
50 4 – Unchanged
5 – Worse
40
Missing
30

20

10

0
Amitriptyline Placebo Amitriptyline Placebo Amitriptyline Placebo

FIGURE 32 Subjective global assessment of relief of IBS symptoms at 6 months by treatment arm and stool type.

Copyright © 2024 Wright-Hughes et al. This work was produced by Wright-Hughes et al. under the terms of a commissioning contract issued by the Secretary of State for Health
and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, 161
distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For
attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.
EME
HSDR
HTA
PGfAR
PHR
Part of the NIHR Journals Library
www.journalslibrary.nihr.ac.uk

This report presents independent research funded by the National Institute for Health and Care Research (NIHR).
The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the
Department of Health and Social Care

Published by the NIHR Journals Library