Accepted Manuscript

A new clinical model for facilitating the development of pattern recognition skills in
clinical pain assessment

David M. Walton, James M. Elliott

PII: S2468-7812(18)30081-X
DOI: 10.1016/j.msksp.2018.03.006
Reference: MSKSP 175

To appear in: Musculoskeletal Science and Practice

Received Date: 28 August 2017

Revised Date: 10 March 2018
Accepted Date: 17 March 2018

Please cite this article as: Walton, D.M., Elliott, J.M., A new clinical model for facilitating the development
of pattern recognition skills in clinical pain assessment, Musculoskeletal Science and Practice (2018),
doi: 10.1016/j.msksp.2018.03.006.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

A new clinical model for facilitating the development of pattern recognition skills in clinical pain
assessment

David M. Walton PT PhD1, James M. Elliott PT PhD2

1: Faculty of Health Science, Western University Canada

PT
2. Faculty of Health Sciences, The University of Sydney, and the Kolling Institute, Royal North Shore
Hospital, NSW, Australia

RI
Corresponding Author:
David Walton

SC
Associate Professor, School of Physical Therapy
Rm. EC1443, Western University
1201 Western Rd.
London ON, Canada

U
N6G 0K8
Phone: +1-519-661-2111 x80145
AN
Email: dwalton5@uwo.ca

Keywords: clinical reasoning, radar plot, triangulation, pain

M

Disclosure: The model being proposed has been used to educate participants on for-profit continuing
D

professional development courses, but the authors hold no exclusive copyright over the use of a radar
plot or the concept of triangulation.
TE
C EP
AC
 ACCEPTED MANUSCRIPT

A new clinical model for facilitating the development of pattern recognition skills in clinical pain
assessment

Authors blinded

Corresponding Author: (blinded)

PT
The model being proposed has been used to educate participants on for-profit continuing professional
development courses, but the authors hold no exclusive copyright over the use of a radar plot or the

RI
concept of triangulation.

U SC
AN
M
D
TE
C EP
AC

1
 ACCEPTED MANUSCRIPT

Abstract

Common, enigmatic musculoskeletal conditions such as whiplash-associated disorder, myofascial pain

syndrome, low back pain, headache, fibromyalgia, osteoarthritis, and rotator cuff pathology, account for
significant social, economic, and personal burdens on a global scale. Despite their primacy (and shared

PT
sequelae) there remains a paucity of available and effective management options for patients with both
acute and chronic conditions. Establishing an accurate prognostic or diagnostic profile on a patient-by-
patient basis can challenge the insight of both novice and expert clinicians. Questions remain on how

RI
and when to choose the right tool(s), at the right time(s), for the right patient(s), for the right
problem(s).

SC
The aim of this paper is to introduce a new clinical reasoning framework that is simple in presentation
but allows interpretation of complex clinical patterns, and is adaptable across patient populations with

U
acute or chronic, traumatic or non-traumatic pain. The concepts of clinical phenotyping (e.g. identifying
AN
observable characteristics of an individual resulting from the interaction of his/her genotype and their
environment) and triangulation serve as the foundation for this framework. Based on our own clinical
and research programs, we present these concepts using two patient cases; a) whiplash-associated
M

disorder (WAD) following a motor vehicle collision and b) mechanical low back pain.
D
TE
C EP
AC

2
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC

3
 ACCEPTED MANUSCRIPT

Introduction

Personalized pain management is gaining momentum as a sound approach to clinical practice.1 This
movement is emerging from some recognizable shortcomings of rote application of evidence from
clinical trials to all patients in a ‘one size fits all’ style.2 Randomized clinical trials (RCTs) of rehabilitation

PT
interventions have habitually struggled to adequately mimic the bespoke approach to care delivered by
clinicians. Reasons for this are wide and varied, but the multifactorial and highly personal nature of the
pain experience contributes to the challenges of adequate design and interpretation, of traditional RCTs.

RI
Treating every patient only as supported by evidence drawn from comparisons of group means risks
under- or over-treatment of the individual person. Arguably, a more logical and achievable approach

SC
would be to 1) implement a clinically rigorous yet feasible and personalized multidimensional
assessment, 2) identify multisystem patterns in the patient profile that may be driving the pain

U
experience and 3) intervene in a targeted fashion based on the results of that assessment. At the heart
of such an approach are the pillars of evidence-based practice - sound empirical evidence, clinician
AN
experience, and patient values.3

Recent years have seen an increasing focus on identifying subgroups of patients with painful
M

musculoskeletal conditions such as neck and low back pain, intended to provide more guidance for
clinical decision making. Subgroups have been described to estimate risk of chronicity,4,5 response to
D

treatment,6,7 and specific pain mechanisms.8–10 In some cases this approach has shown promise; Hill and
TE

colleagues have provided preliminary evidence that prognosis-based subgrouping of patients with acute
low back pain may lead to improved outcomes and treatment efficiency.11 Few other pain conditions
enjoy the same evidentiary base,12,13 and most expert clinicians would agree that basing treatment
EP

decisions on a single tool or algorithm over-simplifies the complex patterns and interactions associated
with the personal experience of pain. Proponents of sub-classification have also yet to reconcile the
C

clinical reality that few patients fit neatly within distinct homogenous ‘boxes’.14–16
AC

Subgrouping can be a useful tool for novice clinicians, but clinical expertise appears to be associated
with a move beyond such algorithmic approaches. Benner has proposed theories on the development
of clinical experts as a process through which providers rely less on structured rules and procedures, and
more on past experiences, intuition, and heuristics.17 A recognized indicator of transition from novice to
expert clinician is the growing internal reference ‘archive’ for recognizing patterns of clinical
presentation and increased comfort with ambiguity, leading to treatment decisions that can be made
without a distinct patient classification.18 We are choosing ‘pattern recognition’ to describe this

4
 ACCEPTED MANUSCRIPT

competence, defining it as the perception and integration of information from multiple sources to arrive
at recognizable patterns. It would seem a reasonable pursuit to develop an academic model for
facilitating the development of pattern recognition skills in students and early-career clinicians to
accelerate their transition to expert-level practitioners. As the field of musculoskeletal pain continues to
grow in exciting scientific directions it also grows in complexity, rendering the utility of a

PT
multidimensional clinical reasoning framework even more valuable. This professional issue presents a
new framework that is simple in presentation but facilitates interpretation of complex clinical patterns,

RI
and appears to be adaptable across patient populations and professional disciplines. The outcomes of
such an approach should result in the exploration and development of more effective targeted

SC
interventions on a patient-by-patient basis while minimizing the likelihood that clinicians are paralyzed
by too much information.

U
The authors are leveraging their own experiences in clinical practice (combined >30 years), pre- and
post-professional teaching, mentorship, and basic and clinical research in the field of
AN
neuromusculoskeletal pain and trauma to propose this framework. Two illustrative case examples using
a patient with whiplash-associated disorder and another with mechanical low back pain are included as
M

appendices.

The Radar Plot and Triangulation

D

The radar plot and associated concept of triangulation are being presented as emerging concepts rather
TE

than empirically derived formulae. On the contrary, the value of characterizing pain in this way is to
endorse a move away from formulaic or algorithmic approaches that may carry unintended
consequences of reducing the clinical decision making skills of practitioners.19 A suggested radar plot
EP

displaying 7 domains as potential ‘pain drivers’ is shown in Figure 1. While not exhaustive, the seven
points represent different domains of a patient’s pain experience, offering potentially greater
C

granularity for clinical decision-making than do contemporary models of pain. 20,21 This framework is not
AC

meant to be diagnostic in nature, rather we present it as a complimentary tool to recent taxonomies of

chronic pain, such as that endorsed by the Analgesic, Anesthetic, and Addiction Clinical Trial Translations
Innovations Opportunities Network and the American Pain Society (ACTTION-APS)22,23 who have
described specific clinical pain diagnoses. Our new framework is presented as a tool to identify the
magnitude of the primary driver(s) of a pain experience without requiring a label on the condition. We
have found it applicable to a wide variety of both acute and chronic musculoskeletal conditions.

5
 ACCEPTED MANUSCRIPT

The seven points have been deliberately chosen as ones that should logically be amenable to different
interventions, even when a firm diagnosis cannot be reached, and have previously been associated with
the qualitative or quantitative experience of pain. They are:

• NOCICEPTIVE (PHYSIOLOGICAL) INPUT24, defined here as pain produced primarily through input

PT
from peripheral nociceptive afferents following transduction of noxious-level mechanical,
thermal or chemical stimuli that leads to an action potential volley from the periphery through
the central nervous system. In this case, the afferent volleys are initiated through depolarization

RI
of nociceptive end organs as a result of abnormal stress on (or injury to) peripheral tissue(s). In
other contexts these have been described as ‘mechanical’ or ‘inflammatory’ pain behaviors.

SC
• PERIPHERAL NEUROPATHY25 defined here according to the definition from the International
Association for the Study of Pain as pain caused by a lesion or disease of the peripheral nervous

U
system (www.iasp-pain.org/Taxonomy#Neuropathicpain).
• CENTRAL NOCIPLASTIC CHANGE26 defined here as pain that can be traced to either a central
AN
facilitation of action potentials (amplification or disinhibition) from the periphery, or as ectopic
impulses generated within the central nervous system with no direct input from the periphery.
M

This is analogous to the IASP definition of central sensitization . The term nociplastic is endorsed
by Kosek and colleagues27 as an alternative to the more ambiguous ‘sensitization’, the
D

implication being that such mechanisms are potentially reversible (e.g. ‘plastic’).
• EMOTIONAL DYSREGULATION OR PATHOLOGY28 defined here as diagnosable psychopathology
TE

or affective dysregulation, conditions described in the Diagnostics and Statistical Manual – V.29
These could include depression, anxiety, or other mood or personality disorders. There is a long
EP

history of association between psychopathology and pain,30 and while causal mechanisms are
yet unclear it appears likely that pain magnifies negative mood while negative mood amplifies
pain.
C

• MALADAPTIVE COGNITIONS31 defined here as inaccurate or irrational beliefs, thoughts or

AC

behaviors about, or resulting from, the experience of pain. Similar to emotional distress, there
is a long history of association between exaggerated negative orientation towards pain (e.g.
fear, catastrophization, low self-efficacy) and pain,32,33 but the causal pathways have proven
elusive. The key difference between this and the prior category are that while maladaptive
cognitions may be a precursor of psychopathology, there are no defined diagnostic criteria for
them (i.e. there is no DSM-V entry for ‘pain catastrophizer’). The practical implication , and the

6
 ACCEPTED MANUSCRIPT

reason for their separation, is that, for example, a physical therapist may be well positioned to
address maladaptive beliefs or cognitions about pain,34 while addressing psychopathology
should be the domain of a mental health professional.
• SOCIOENVIRONMENTAL CONTEXT35 defined here as the very wide-ranging and amorphous
contextual factors that affect not only one’s experience of pain but also access to appropriate

PT
care, willingness to report, and the way in which pain is described. This could include relations
with important others, prevailing cultural beliefs or language about pain, socially-constructed

RI
gender roles, early life adversity, environmental demands, stressors, and many others.
• SENSORIMOTOR DYS-INTEGRATION36 defined here as discordance between the perceived self

SC
and the actual self. Alternatively described as a problem of interoception,37 the driver here is
one of a mismatch between two or more sensory inputs into the central nervous system, such
as optical input stating the head is in one position, and cervical proprioceptive input indicating it

U
is in a different position. While there is consistent evidence that such sensorimotor discordance
AN
exists with greater frequency in people with pain36, causal pathways are yet to be fully
elucidated.
M

These points are not likely to be exhaustive, however, we believe they satisfy the ACTTION-APS criteria
of being adequately exhaustive for clinical use, mutually exclusive, biologically plausible, reliable,
D

clinically useful and simple.23 Readers will note that this framework is meant to be applied after the
patient has passed screening for red flags or other systemic influences (comorbidities, medications) that
TE

may also contribute to their symptoms, and only after the patient has been deemed likely appropriate
for assessment and care by the health practitioner.
EP

Table 1 provides sample indicators of a patient’s status on each domain. However, these are meant to
be examples rather than endorse a clinical edict. Readers will also note that the radar plot tool is
C

intended to offer adequate direction for clinical decisions but not so much that it becomes burdensome
for the clinician, the patient, or other stakeholders. The status levels on each domain are limited to
AC

qualitative ranges of: very low, low, moderate, high, and very high. While tools exist to evaluate these
domains, our belief is that few are at the stage of development to allow endorsement to greater
precision than these 5 broad levels. The diagrammatic representation of relative patient location on
each of the 7 domains should make treatment decisions easier by quickly identifying the primary drivers
of the patient’s pain experience, even if two different clinicians may assign different absolute locations
on each.

7
 ACCEPTED MANUSCRIPT

A second concept is required here, that being triangulation, which is drawn largely from military or
geographic positional science. Figure 2 graphically depicts this concept using the contemporary
example of locating the global position of a mobile phone. While one source of information can provide
a very broad sense of position, two sources narrows possible position to the region of overlap, and three
sources all ‘pointing in the same direction’ leave only one possible position. This concept of

PT
triangulation can be applied to estimate the magnitude of contribution from each of the radar plot
domains, requiring use of at least 3 information sources before being confident in locating the patient

RI
on each. Especially when information gleaned from very different information sources all point in the
same general direction (e.g. patient self-report, imaging, and clinical tests) does confidence in a patient’s

SC
location increase. This analogy can be further understood through positive (+LR) or negative (-LR)
likelihood ratios and pre- and post-test probabilities. Box 1 demonstrates an example using 3 clinical
tests of low-to-moderate diagnostic validity (Sensitivities and Specificities ranging from 0.66 to 0.80).

U
For ease, the example displays findings of all negative or all positive results, but it should be noted that
AN
negative results in one domain (e.g. central nociplastic) may be considered positive in another (e.g.
nociceptive).
M

Through use of the multi-domain radar plot, complex data sources built on triangulated findings can be
visualized. This approach encourages appropriate implementation and interpretation of sound
D

measurement tools and tests, and a subtle but important paradigmatic shift in treatment planning
where sound assessment of modifiable domains (rather than categorical labels) becomes the priority
TE

from which treatment strategies can naturally flow. As demonstrated in the sample cases below (Box 2),
the patient’s subjective history should also be considered a source of information for triangulation.
EP

While anecdotal, our experience suggests the radar plot and concept of triangulation appear to resonate
with students and novice or mid-career clinicians across professional disciplines. It appears to function
C

adequately well as a teaching tool, but experienced readers will recognize that, like any such approach,
it is arguably too reductionistic. For example, it assumes clear distinctions between domains of the pain
AC

experience that likely overlap (e.g. central nociplastic and sensorimotor dysintegration). Further,
consistent with the current state of research in the field, the separate domains ignore interactions
between, for example, nociceptive input and sensorimotor dysintegration in the presence of
maladaptive cognitions of, say, middle-aged East-Asian females. As the research progresses, so too can
this teaching tool evolve.

8
 ACCEPTED MANUSCRIPT

We hope that the radar plot will facilitate communication across professional disciplines, between
patients and providers, and with third-party payors. Three tools per domain, not all of which need to be
exhaustive measures (e.g. a well-formed single direct question could serve as a useful discriminatory
tool for one or more domain(s)) appear to be reasonable educational targets for trainees and educators,
offering a scaffold for education in pain rehabilitation disciplines.

PT
The intention of publishing this tool is to permit further development by educators, researchers, and
clinicians in the spirit of facilitating professional development towards optimizing patient outcomes.

RI
Whether use of this tool expedites transition towards expert-level clinical reasoning or improves clinical
behaviors/outcomes is a reasonable direction for future study. We hope others will find value in this

SC
line of reasoning by substituting the domains in our example with those that are relevant for fields other
than pain, and that it helps novice clinicians across any number of disciplines to make sense of a highly

U
complex and often confusing field of research. AN
M
D
TE
C EP
AC

9
 ACCEPTED MANUSCRIPT

1. Woolf CJ, Physicians AC of, Society AP. Pain: moving from symptom control toward mechanism-
specific pharmacologic management. Ann Intern Med. 2004;140(6):441-451.

2. Rothwell PM. External validity of randomised controlled trials: “To whom do the results of this
trial apply?” Lancet. 2005;365(9453):82-93. doi:10.1016/S0140-6736(04)17670-8.

PT
3. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS. Evidence based medicine: what
it is and what it isn’t. BMJ. 1996;312(7023):71-72.

RI
4. Ritchie C, Hendrikz J, Kenardy J, Sterling M. Derivation of a clinical prediction rule to identify both
chronic moderate/severe disability and full recovery following whiplash injury. Pain.

SC
2013;154(10):2198-2206. doi:10.1016/j.pain.2013.07.001; 10.1016/j.pain.2013.07.001.

5. Hill JC, Dunn KM, Lewis M, et al. A primary care back pain screening tool: identifying patient

U
subgroups for initial treatment. Arthritis Rheum. 2008;59(5):632-641. doi:10.1002/art.23563.
AN
6. Fritz JM, Brennan GP. Preliminary examination of a proposed treatment-based classification
system for patients receiving physical therapy interventions for neck pain. Phys Ther.
2007;87(5):513-524. doi:10.2522/ptj.20060192.
M

7. O’sullivan P. Diagnosis and classification of chronic low back pain disorders: Maladaptive
D

movement and motor control impairments as underlying mechanism. Man Ther. 2005;10(4):242-
255. doi:S1356-689X(05)00110-4 [pii]; 10.1016/j.math.2005.07.001 [doi].
TE

8. Freynhagen R, Baron R, Gockel U, Tolle TR. painDETECT: a new screening questionnaire to

identify neuropathic components in patients with back pain. Curr Med Res Opin.
EP

2006;22(10):1911-1920. doi:10.1185/030079906X132488.

9. Arendt-Nielsen L, Yarnitsky D. Experimental and clinical applications of quantitative sensory

C

testing applied to skin, muscles and viscera. J Pain. 2009;10(6):556-572.

AC

doi:10.1016/j.jpain.2009.02.002.

10. Nordin M, Carragee EJ, Hogg-Johnson S, et al. Assessment of neck pain and its associated
disorders: results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and Its
Associated Disorders. Spine (Phila Pa 1976). 2008;33(4 Suppl):S101-22.
doi:10.1097/BRS.0b013e3181644ae8.

11. Hill JC, Whitehurst DG, Lewis M, et al. Comparison of stratified primary care management for low

10
 ACCEPTED MANUSCRIPT

back pain with current best practice (STarT Back): a randomised controlled trial. Lancet.
2011;378(9802):1560-1571. doi:10.1016/S0140-6736(11)60937-9.

12. Michaleff ZA, Maher CG, Lin CW, et al. Comprehensive physiotherapy exercise programme or
advice for chronic whiplash (PROMISE): a pragmatic randomised controlled trial. Lancet.

PT
2014;384(9938):133-141. doi:10.1016/S0140-6736(14)60457-8; 10.1016/S0140-6736(14)60457-
8.

RI
13. Jull G, Kenardy J, Hendrikz J, Cohen M, Sterling M. Management of acute whiplash: a randomized
controlled trial of multidisciplinary stratified treatments. Pain. 2013;154(9):1798-1806.

SC
doi:10.1016/j.pain.2013.05.041; 10.1016/j.pain.2013.05.041.

14. Kamper SJ, Maher CG, Hancock MJ, Koes BW, Croft PR, Hay E. Treatment-based subgroups of low
back pain: a guide to appraisal of research studies and a summary of current evidence. Best Pract

U
Res Clin Rheumatol. 2010;24(2):181-191. doi:10.1016/j.berh.2009.11.003.
AN
15. Sun X, Briel M, Walter SD, Guyatt GH. Is a subgroup effect believable? Updating criteria to
evaluate the credibility of subgroup analyses. BMJ. 2010;340:c117.
M

http://www.ncbi.nlm.nih.gov/pubmed/20354011. Accessed August 22, 2017.

16. Hancock M, Herbert RD, Maher CG. A guide to interpretation of studies investigating subgroups
D

of responders to physical therapy interventions. Phys Ther. 2009;89(7):698-704.

TE

doi:10.2522/ptj.20080351.

17. Benner P. From Novice to Expert, Excellence and Power in Clinical Nursing Practice. 1st ed. Menlo
EP

Park, CA: Addison-Wesley Publishing Company; 1984. doi:10.1002/nur.4770080119.

18. Jones MA. Clinical reasoning in manual therapy. Phys Ther. 1992;72(12):875-884.
C

http://www.ncbi.nlm.nih.gov/pubmed/1454863. Accessed April 22, 2017.

AC

19. Cabitza F, Rasoini R, Gensini GF. Unintended Consequences of Machine Learning in Medicine.
JAMA. 2017;318(6):517-518. doi:10.1001/jama.2017.7797.

20. Melzack R. From the gate to the neuromatrix. Pain. 1999;Suppl 6:S121-6.

21. Gifford L. Pain, the tissues and the nervous system: a conceptual model. Physiotherapy.
1998;84(1):27-36.

22. Kent ML, Tighe PJ, Belfer I, et al. The ACTTION-APS-AAPM Pain Taxonomy (AAAPT)

11
 ACCEPTED MANUSCRIPT

Multidimensional Approach to Classifying Acute Pain Conditions. Pain Med. 2017;18(5):947-958.

doi:10.1093/pm/pnx019.

23. Fillingim RB, Bruehl S, Dworkin RH, et al. The ACTTION-American Pain Society Pain Taxonomy
(AAPT): An Evidence-Based and Multidimensional Approach to Classifying Chronic Pain

PT
Conditions. J Pain. 2014;15(3):241-249. doi:10.1016/j.jpain.2014.01.004.

24. Perl ER. Cutaneous polymodal receptors: characteristics and plasticity. Prog Brain Res.

RI
1996;113:21-37. http://www.ncbi.nlm.nih.gov/pubmed/9009726. Accessed June 9, 2017.

25. Hsieh JC, Belfrage M, Stone-Elander S, Hansson P, Ingvar M. Central representation of chronic

SC
ongoing neuropathic pain studied by positron emission tomography. Pain. 1995;63(2):225-236.
doi:030439599500048W [pii].

U
26. Flor H, Braun C, Elbert T, Birbaumer N. Extensive reorganization of primary somatosensory cortex
in chronic back pain patients. Neurosci Lett. 1997;224(1):5-8. doi:S0304394097134413 [pii].
AN
27. Kosek E, Cohen M, Baron R, et al. Do we need a third mechanistic descriptor for chronic pain
states? doi:10.1097/j.pain.0000000000000507.
M

28. Blozik E, Laptinskaya D, Herrmann-Lingen C, et al. Depression and anxiety as major determinants
D

of neck pain: a cross-sectional study in general practice. BMC Musculoskelet Disord. 2009;10:13.
doi:10.1186/1471-2474-10-13.
TE

29. American Psychiatric Association., American Psychiatric Association. DSM-5 Task Force.
Diagnostic and Statistical Manual of Mental Disorders : DSM-5. American Psychiatric Association;
EP

2013.

30. Dimitriadis Z, Kapreli E, Strimpakos N, Oldham J. Do psychological states associate with pain and
C

disability in chronic neck pain patients? J Back Musculoskelet Rehabil. 2015;28(4):797-802.

AC

doi:10.3233/BMR-150587.

31. George SZ, Hirsh AT. Psychologic influence on experimental pain sensitivity and clinical pain
intensity for patients with shoulder pain. J Pain. 2009;10(3):293-299.
doi:10.1016/j.jpain.2008.09.004; 10.1016/j.jpain.2008.09.004.

32. Sullivan MJ, Stanish W, Sullivan ME, Tripp D. Differential predictors of pain and disability in
patients with whiplash injuries. Pain ResManag. 2002;7(2):68-74.

12
 ACCEPTED MANUSCRIPT

33. Crombez G, Vlaeyen JW, Heuts PH, Lysens R. Pain-related fear is more disabling than pain itself:
evidence on the role of pain-related fear in chronic back pain disability. Pain. 1999;80(1-2):329-
339.

34. Bennell KL, Ahamed Y, Jull G, et al. Physical Therapist-Delivered Pain Coping Skills Training and

PT
Exercise for Knee Osteoarthritis: Randomized Controlled Trial. Arthritis Care Res (Hoboken).
2016;68(5):590-602. doi:10.1002/acr.22744.

RI
35. Raichle KA, Romano JM, Jensen MP. Partner responses to patient pain and well behaviors and
their relationship to patient pain behavior, functioning, and depression. Pain. 2011;152(1):82-88.

SC
doi:10.1016/j.pain.2010.09.015.

36. de Vries J, Ischebeck BK, Voogt LP, et al. Joint position sense error in people with neck pain: A
systematic review. Man Ther. 2015;20(6):736-744. doi:10.1016/j.math.2015.04.015.

37.
U
Di Lernia D, Serino S, Riva G. Pain in the body. Altered interoception in chronic pain conditions: A
AN
systematic review. Neurosci Biobehav Rev. 2016;71:328-341.
doi:10.1016/j.neubiorev.2016.09.015.
M
D
TE
C EP
AC

13
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE

Figure 1: Example of a radar plot with 7 distinct (but potentially overlapping) domains of the pain
EP

experience that can be useful for clinical evaluation and treatment decisions for people in pain.
C
AC

14
 ACCEPTED MANUSCRIPT

PT
RI
Figure 2: A graphical depiction of the concept of triangulation. Starting from left to right, if the distance

SC
from a center (e.g. a cell phone tower) is known to be 2km, the object (e.g. a phone) can be anywhere
within that 2 km radius. When a second tower, 3 km away is also able to ‘see’ the device, there are two

U
possible points that those two radii overlap at which the device can be located. If a third tower, in this
example a less powerful one, can see the device 500m away, there is only one possible location that all 3
AN
radii overlap where the device could be located.
M
D
TE
C EP
AC

15
 ACCEPTED MANUSCRIPT

Box 1: Sample Triangulation using Likelihood Ratios

Assumptions for this example:
Pre-test probability that condition exists (odds)
50% (1:2)
Discriminative Validity Likelihood Ratios Post-test Post-test

PT
probability probability
All tests positive All tests negative
Clinical test 1 Sn 0.80, Sp 0.80 +LR 4.00, -LR 0.25 67% (2:1) 11% (1:8)

RI
Clinical test 2 Sn 0.75, Sp 0.75 +LR 3.00, -LR 0.33 86% (6:1) 4% (1:25)
Clinical test 3 Sn 0.66, Sp 0.67 +LR 2.00, -LR 0.51 92% (12:1) 2% (1:50)

SC
With 3 tests, each of modest to low discriminative validity, but each positive, likelihood that a condition
exists (or that a particular mechanism on the radar plot is important) goes from 50% to 92%, suggesting
that domain should be in the high to very high range of the plot. If all 3 tests are negative, the likelihood

U
a domain is a strong driver goes from 50% to 2%, moving that domain to low or very low. Both
calculations are conducted accepting some likely inflation due to ignoring the prior odds fallacy.
AN
M
D
TE
C EP
AC

16
 ACCEPTED MANUSCRIPT

Appendices: Sample triangulation approach*

Case 1: Jan
Jan is a 39 year-old Caucasian female with a history of persistent whiplash-associated disorder arising as
a result of a motor vehicle crash 11 months prior. She was the belted driver of a 4-door sedan that was
stopped when it was impacted from behind on the right side by a sport utility vehicle (SUV) traveling an
estimated 30 km/h (~ 19 mph). Her headrest was well adjusted. She denies loss of consciousness.

PT
Imaging in the Emergency Department has ruled out significant pathology. Neck pain, stiffness and
headaches began the following day and worsened within 48 hours motivating her to see her family
physician. She received a diagnosis of whiplash associated disorder grade II. She has received physical

RI
therapy (modalities and unsupervised exercise), massage, NSAIDs (acetaminophen) and wage indemnity
benefits since that time. She has yet to return to her pre-collision job of floor supervisor for an auto
parts plant, complaining of ongoing neck pain and headaches as well as sensitivity to light and difficulty

SC
concentrating that worsens after a couple hours of work. She is married with one teenaged son and
prior to her injury she contributed equally to the family’s finances.

U
On clinical examination cervical mobility is limited in all planes with no other obvious mechanical
pattern. She appears exquisitely tender to palpate anywhere in the neck or shoulder girdle region, even
AN
flinching at times in response to light touch. Pressure pain detection threshold testing reveals
widespread sensory hypersensitivity (local to the neck and over tibialis anterior) and she is
hypersensitive to cold stimuli. Her pain thresholds decrease (more sensitive) following 3 minutes of
M

moderately vigorous stationary cycling. Joint position sense error (nominating the centre of a target
after returning from cervical rotation with the eyes closed) and two-point discrimination are both
impaired compared to population norms but still within the high ends of normal. Smooth pursuit neck
D

torsion reveals no signs of saccadic eye movements.

TE

Her self-report measures indicate poorly-localized widespread pain on a body diagram, severe disability
according the Neck Disability Index score of 35/50, a Pain Catastrophizing Scale score of 32/52 (high), a
score on the self-report version of the Leeds Assessment of Neuropathic Signs and Symptoms of 11/24
EP

(one point under the cut score of 12/24), and scores on the Patient Health Questionnarie-9 are over-
threshold for a potential depressive disorder.
C

She reports a generally good relationship with her insurer and family doctor, and her husband and son
are supportive. However, she also describes feeling pressured and scrutinized by her employer and
AC

coworkers during an earlier failed attempt to return to work. She is also experiencing increasing financial
hardship due to medical expenses and lost wages.

17
 ACCEPTED MANUSCRIPT

Triangulating:
DOMAIN LEVEL
Nociceptive Low
Peripheral Low-Moderate
Neuropathic

PT
Central Nociplastic High
Cognitive High
Emotional Moderate-High

RI
Socioenvironmental Mod
Sensorimotor Low-Moderate

SC
Create pain profile:

U
AN
M
D
TE
C EP
AC

Case 2: Alex
Alex is a 31 year-old African-American male with a history of intermittent low back pain that he
attributes to his job standing 7.5 hours per day in a retail electronics store. The pain has been present
for the 3 or 4 months with no definable etiology. He is otherwise healthy and enjoys playing tennis twice

18
 ACCEPTED MANUSCRIPT

weekly. He has remained at work but reports that he requires non-steroidal anti-inflammatories to help
manage the low back pain on average 3 out of 5 shifts per week. He describes pain and stiffness that is
worse in the morning, improves through the mid-part of the day but worsens again in the last few hours
of most shifts. This is his first time seeking formal rehabilitation care on the recommendation of a
coworker. He is also the primary breadwinner for his family that includes a wife and one young son.

PT
On clinical examination lumbar mobility is nearly full in all planes though he describes a ‘pinching’ type
pain during extension and combined multiplanar movements of extension/side-bend especially to the
right side. He describes local tenderness to palpation over the right lower lumbar / lumbosacral region.

RI
Pressure pain threshold testing reveals mechanical hypersensitivity in that same right lumbar area but
normal sensitivity elsewhere. Neurological testing appears normal with no obvious signs of motor
weakness or fatigue, though straight leg raising is somewhat limited on the right side. Two-point

SC
discrimination is within normal limits and he is easily able to complete a line drawing of his back that is
proportionate to objective reality.

U
His self-report measures indicate well-defined localized pain over the right lower lumbar/lumbosacral
region on a body diagram with no indication of numbness or paraesthesia. Scores on the Roland Morris
AN
Disability Questionnaire (RMDQ) are 6/24, noting issues with standing for long period, occasionally
bending over, rising in the morning, and sometimes moving more slowly. His Pain Catastrophizing Scale
score is 4/52 or low, and he shows no clinical indications of an emotional pathology so formal screening
M

is not conducted.

He reports that he generally enjoys his job and has a good group of coworkers who tend to have fun at
D

work, though his manager is not always supportive of his need to sit down at times. His family doctor is
the one that suggested NSAIDs but otherwise Alex feels as though she sort of ‘waived-off’ his questions
TE

about his low back. He is not one to let on that he is in pain when around home as he feels compelled to
be the ‘man of the house’ and a ‘good father’ when not at work.
EP

Triangulating:
DOMAIN LEVEL
Nociceptive High
C

Peripheral Very Low

Neuropathic
AC

Central Nociplastic Low

Cognitive Low
Emotional Low
Socioenvironmental Low-Moderate
Sensorimotor Very Low

Create pain profile:

19
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE

Resultant radar plots from example data obtained from clinical examinations of two cases detailed in
Box 2.
EP

*This is meant to be an illustrative exercise and the tests and their interpretation described are based on the
authors own experience and expertise in the field. They are not meant to be an endorsement of those specific
C

tests or interpretations. Such questions are better left for formal systematic reviews.
AC

20
 ACCEPTED MANUSCRIPT

Assessment Nociceptive Peripheral Central Nociplastic Emotional Maladaptive Socioenvironmental Sensorimotor

Domain (Physiological) Neuropathy Mechanisms Dysregulation Cognitions or Context Dys-integration
input Beliefs
History of the • Complaints are • Mechanism of • More difficult to draw • History of • No defined • May be more likely • More likely to
complaint proportionate to onset consistent connection between psychopathology pattern, can be when pathogenesis manifest in
the mechanism with trauma of a mechanism of onset especially if acute or chronic, has occurred in a chronic problems
peripheral nerve and current complaints temporally related traumatic or non- compensable
to other symptom traumatic environment or

PT
onset linked to other
stressors

RI
Patient narrative • Well-localized pain • Spontaneous or • Resting pain (local or • Symptoms • Examples: Belief • Feels under constant • Describes the
complaints ‘ectopic’ pain, widespread), may be consistent with that hurt = harm, scrutiny or injured body region
allodynia and local related to mood or psychopathology or that 100% relief surveillance (e.g. as though it is

SC
hyperalgesia emotional status (e.g. DSM-V is required before medicolegal detached from self
criteria) resuming activity involvement)

U
Standardized self- • Responses do not • Self-report • Self-report diagnostic • Self-report • Self-report • Self-report evaluative • Few available, but
4 12 13
report evaluations support other diagnostic tools tools (e.g. CSI ) diagnostic tools evaluative tools tools (e.g. SRI , IEQ ) may struggle to

AN
1 6 7 9 10
drivers in the (e.g. SLANSS ) (e.g. PHQ-9 , PCL ) (e.g. PCS , TSK , identify painful
11
framework FABQ ) areas on a body
diagram

M
Standardized clinical • Consistent and • Clinical signs of • Non-mechanical and • Pain not • Exaggerated or • Signs suggestive of • Signs of
evaluations and predictable pain or impaired non-predictable consistent with inconsistent pain intentional somatosensory

D
signs movement-related neural patterns of pain predictable behaviours out of exaggeration may reorganization
14
pain behaviour transmission along reproduction, mechanical proportion to provide a clue, but (e.g. 2PD ,
15

TE
the course of a with/without patterns magnitude of careful interpretation JPSE )
known sensory dysfunctional testing is encouraged
nerve descending pain
5
modulation
EP
Other observations • Responsive to • Not responsive to • Not responsive to • Small to no effect • Preference for • Counseled to avoid • May require
routine front-line NSAIDs, may be routine front-line on pain from avoidant or passive activity or ‘straining’ exploration and
pharmacotherapy responsive to therapies, nay be front-line coping methods, until after case is exclusion of a CNS
2 3
C

TCAs , SNRIs , responsive to opioids, pharmacotherapy, ‘all or none’-type settled disorder

pregabalin or TCAs and/or SSRIs may see effect thinking
AC

2
gabapentin from TCAs or
8 3
SSRIs /SNRIs

Table 1: Examples of tools or clinical signs currently available for estimating magnitude of dysfunction / impact in each of the 7 domains described by the sample
radar plot. NSAIDs = Non-Steroidal Anti-Inflammatories, TCA = Tricyclic Antidepressants, SSRI = Selective Serotonin Reuptake Inhibitors, SNRI = Serotonin &
Norepinephrine Reuptake Inhibitors.
1: SLANSS = Self-report version of the Leeds Assessment of Neuropathic Signs and Symptoms
 ACCEPTED MANUSCRIPT

2: TCAs = Tricyclic Antidepressants

3: SNRIs = Serotonin and Norepinephrine Reuptake Inhibitors. For a review of pharmacotherapy in neuropathic pain see Finnerup and colleagues 20151
4: CSI = Central Sensitivity Index
5: Dysfunctional pain modulation may manifest as exercise-induced hyperalgesia or abnormal conditioned pain modulation.
6: PHQ-9 = Patient Health Questionnaire 9-item version
7: PCL = Post-traumatic distress scale checklist
8: SSRIs = Selective Serotonin Reuptake Inhibitors
9: PCS = Pain Catastrophizing Scale

PT
10: TSK = Tampa Scale for Kinesiophobia
11: FABQ = Fear Avoidance Beliefs Questionnaire
12: SRI = Spousal Response Inventory

RI
13: IEQ = Injustice Experience Questionnaire
14: 2PD = two-point discrimination
15: JPSE = Joint Position Sense Error

U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Highlights

• A new radar plot-based clinical reasoning framework is presented

• The framework uses clinical phenotyping and triangulation for musculoskeletal pain
• The framework appears to resonate with clinicians
• The new framework requires formal empirical study to determine its overall usefulness

PT
RI
U SC
AN
M
D
TE
C EP
AC