The Patient - Patient-Centered Outcomes Research (2022) 15:537–550

https://doi.org/10.1007/s40271-022-00577-9

ORIGINAL RESEARCH ARTICLE

Factors That Patients Consider in Their Choice of Non‑Surgical

Management for Hip and Knee Osteoarthritis: Formative Qualitative
Research for a Discrete Choice Experiment
Bryanne L. Kennedy1 · Gillian R. Currie1,2,3,4 · Ania Kania‑Richmond1,5 · Carolyn A. Emery1,2,3,4,6,7 · Gail MacKean1 ·
Deborah A. Marshall1,3,4,7,8

Accepted: 17 February 2022 / Published online: 16 March 2022

© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022

Abstract
Background Approximately half of patients with hip and knee osteoarthritis have tried non-surgical management before
surgical consultation. Understanding the many factors affecting the uptake of recommended strategies is important to inform
future development of such management strategies.
Objectives The aim of this study was to explore and identify factors that patients with osteoarthritis consider when choosing
non-surgical management for hip and knee osteoarthritis, as formative research for a study of patient preferences for non-
surgical management programs for osteoarthritis.
Methods A qualitative research design was used. Participants were recruited using a combination of stratified and conveni-
ence sampling. Interviews were conducted, using a semi-structured interview guide, with English-speaking patients who
had self-reported hip and/or knee osteoarthritis and at least one joint that had not undergone replacement surgery. Data were
thematically analyzed.
Results Thirteen patients participated in these interviews. Sixteen factors that participants considered when choosing non-
surgical osteoarthritis management were identified. Eleven were extrinsic, relating to features of programs and services,
and are categorized as types of interventions, general program and service details, and program-specific details. Five were
intrinsic to the individual and influenced how decisions for osteoarthritis management were approached and the options
available to choose from. Three novel factors included participants’ desire for further management, their views about joint
replacement surgery, and whether they felt personal choice was available in osteoarthritis management strategies.
Conclusion Key factors were identified that patients considered when making decisions about non-surgical management
for their osteoarthritis that will be used to inform a discrete choice experiment (DCE) that aims to measure preferences for
these factors.

* Deborah A. Marshall
damarsha@ucalgary.ca
Extended author information available on the last page of the article

Vol.:(0123456789)
538 B. L. Kennedy et al.

with osteoarthritis who underwent total knee arthroplasty

Key Points for Decision Makers in Alberta. They found that 75.4% had used exercise,
44.3% had received physiotherapy, and 69.2% of those
Three novel factors that patients considered when mak- who were overweight or obese had tried weight loss [7].
ing decisions about their osteoarthritis management These results indicate that a substantial number of patients
included desire for further management, participants’ are referred for surgery without having tried recommended
views about joint replacement surgery (with those look- non-surgical management.
ing to delay surgery being open to trying anything), and Many factors affect whether patients use recommended
whether patients felt they had choice in management. strategies. This includes access to providers, provider
Knowledge and education emerged as a key thread in expertise, and whether options are publicly funded or
this research. Participants’ perceptions of whether they require out-of-pocket costs. For example, an Alberta study
had choice in management was linked to their knowledge by Miller et al. identified challenges accessing profession-
of management options. als with osteoarthritis expertise after diagnosis as a barrier
to the use of nonsurgical management strategies [9]. This
Additional research is needed to determine the relative gap in care results in difficulty accessing information on
importance of each identified factor in decision making appropriate management. The same study concluded that
for osteoarthritis management. patients require detailed knowledge and specific, action-
able information for self management of osteoarthritis
[9]. Most of the recommended treatment modalities for
non-surgical management of arthritis fall outside the pub-
licly funded health system in Alberta [9, 10]. This creates
system inequities based on affordability and may steer
1 Introduction patients from effective management options towards pub-
licly funded services, including surgery [10].
Osteoarthritis is the most common joint disease worldwide, Patients’ value various features of health interventions
affecting one in six adult Canadians [1]. It is characterized differently, including features of non-surgical management
by structural pathology and symptoms including pain, stiff- programs for osteoarthritis [11]. The experience of peo-
ness, and instability [1, 2]. Risk factors for osteoarthritis ple living with osteoarthritis needs to be drawn upon to
include previous joint injuries, a body mass index over 30, understand the factors that influence peoples’ choices and
age over 50 years, and being female [2]. Disability is most the barriers to trying recommended non-surgical manage-
commonly associated with the knee, and osteoarthritis is the ment strategies. To understand the state of the literature,
leading cause of hip and knee replacements in Canada [1, 2]. a review of the qualitative literature on patient or provider
Of knee replacements performed in Canada between 2017 views of non-surgical management of osteoarthritis was
and 2018, 98.8% were for osteoarthritis [3]. conducted [12–33]. Features of non-surgical management
As there is no cure for osteoarthritis, management aims presented in these studies included program-specific char-
to reduce pain and improve function and quality of life [4]. acteristics that would apply to all (such as the cost of the
Recommended management begins with first-line treatments program), and characteristics specific to an individual’s
including non-pharmacologic treatments such as physical situation (experiencing financial burden) [12–33]. Of 21
therapy, exercise, and weight management. Second-line identified studies, only two were conducted in Canada [27,
treatments include medications. Lastly, at the end stage of 28]. These articles focused on physical therapists’ perspec-
disease when all other options are exhausted, surgery is rec- tives on the management of knee osteoarthritis in Ontario,
ommended [5]. The Osteoarthritis Research Society Interna- Alberta, and British Columbia [27, 28]. This highlighted
tional (OARSI) guidelines for non-surgical management of a gap in understanding of osteoarthritis management in
osteoarthritis include exercise, self-management and educa- Alberta from the patient perspective. The objective of
tion, weight management, and corticosteroids [5, 6]. this study was to identify factors that patients consider
Patients with end-stage osteoarthritis, where articular in their choice of non-surgical osteoarthritis management
cartilage has broken down and bone on bone contact is for hip and/or knee osteoarthritis. This research provides
occurring, should be considered for referral to a surgeon an in-depth understanding of the patient perspective that
only if appropriate non-surgical management has been will inform the development of attributes and levels for a
unsuccessful after 6 months [5]. Despite guidelines, only subsequent discrete choice experiment (DCE) to measure
61% of knee osteoarthritis patients have tried non-surgical patient preferences for non-surgical management programs
management prior to being referred for a surgical consulta- for osteoarthritis.
tion [7, 8]. In a recent study, King et al. examined patients
Factors Affecting OA Patient Non-surgical Management: A Formative Qualitative Study 539

2 Methods age, sex, gender, population centre size, education level,

time since diagnosis, joints affected by osteoarthritis, and
A qualitative research design was best suited to address the ethnicity. This information allowed the final interview
study objectives. Qualitative description seeks to provide sample to be characterized.
a rich account of an experience from the point of view of
participants [34], and was selected to meet the study goal 2.3 Semi‑structured Interviews
to stay close to the participant voice. Data were collected
using semi-structured interviews and demographic data were From the studies identified in a review of the qualitative
captured with an online survey. literature on patient or provider views of non-surgical
Ethics approval was obtained from The Conjoint Health management of osteoarthritis [12–33], a list of non-sur-
Research Ethics Board, at the University of Calgary gical management features was compiled and grouped
(REB20-0671) in January of 2021, and informed verbal according to commonalities (Table 2). Identified features
consent was obtained prior to data collection. pertained to program-specific characteristics that would
apply to all (program type, provider, mode of delivery, set-
2.1 Study Participants ting, supervision, duration of the program, length of each
program session, frequency of the program, and location
Participants were recruited through various online mecha- of the program appointment), and characteristics specific
nisms (Table 1) [35–38]. The inclusion criteria were Eng- to an individual’s situation (financial burden, difficulty
lish-speaking adults aged 18 years and older of both sexes obtaining an appointment, effectiveness, and endorsement
residing in Alberta that self-reported a professional diag- from the patient’s health care provider).
nosis of knee and/or hip osteoarthritis and had at least one The topic and study objective guided the development
joint that had not undergone replacement surgery at the time of a preliminary interview guide protocol, which was
of their participation in the study. Self-reported knee and/ based on the literature review (Table 2) and critical input
or hip osteoarthritis was assessed by asking if the partici- from patient research partners trained in the Patient and
pant had been diagnosed with osteoarthritis by a health pro- Community Engagement Research (PaCER) program at
fessional. These inclusion criteria aligned with the future the University of Calgary [39, 40]. The interview ques-
research context of informing policy in Alberta, and the aim tions aimed to explore the patient experience with non-sur-
to capture the preferences of patients who could benefit from gical osteoarthritis management, and factors that influence
non-surgical management strategies to reduce pain, improve participants’ non-surgical management choices. The litera-
function, and potentially delay or prevent a surgery related ture review and features in Table 2 informed the interview
to their osteoarthritis. Participants received a $10 gift card guide probes but the a priori features were not explicitly
to Starbucks™ or Tim Hortons™ as a token of appreciation shown to participants. The interview guide is available as
for their participation. Online Resource 1 (see electronic supplementary material
[ESM]). Semi-structured interviews were conducted and
2.2 Online Pre‑interview Survey recorded using the Zoom™ platform. Each interview took
approximately 1 h.
Before their interview, participants completed an online
demographic survey containing questions regarding their

Table 1  Recruitment Strategy Description

mechanisms
Twitter posts Tweets by the study team, McCaig Institute, Arthritis Research Canada,
and Alberta Strategy for Patient-Oriented Research Patient Engage-
ment Platform (AbSPORU PEP)
Facebook posts Posts by AbSPORU PEP
Posts by the Arthritis Society
Website postings University of Calgary’s ‘Participate in research’ website
AbSPORU PEP’s Albertans4Health Research website
The Arthritis Society website
The Arthritis Research Canada website
Newsletters The McCaig Institute’s weekly newsletter
540 B. L. Kennedy et al.

Table 2  A priori features of non-surgical management identified from the literature review
Topic Features of non-surgical management References)

Financial burden Cost incurred due to time off from work [12]
Cost of the program [12, 18, 19, 22, 24, 28–30]
Private coverage or no private coverage [12]
Program type Exercise [12–19, 24–33]
Weight management [13, 15, 17–20, 27, 29–31]
Education [13, 14, 16–20, 22, 24–30, 32]
Difficulty obtaining an appointment Long wait times to get an appointment [12, 16, 28]
Effectiveness Does the program improve the participant's condition [12, 14, 15, 18, 22, 26, 27, 29, 30, 32, 33]
Provider GP, physiotherapist, or dietician [20, 28, 31]
Endorsement from the patient's health Yes or no [17, 25, 29–31]
care provider
Mode of delivery Telephone, video conferencing, or in-person [18, 21, 23, 24, 29, 30]
Setting Individual or group [16, 25, 26, 29, 30]
Supervision Supervised by a health care provider or self-directed [22]
Duration of the program Number of weeks [16, 25, 26, 30, 31]
Length of each program session Number of hours [16]
Frequency of the program Number of times per week [16, 26, 27, 30, 31]
Location of the program appointment Proximity to the patient [12, 16, 18, 19, 21, 26, 28, 29]

2.4 Researcher Characteristics and Reflexivity after coding blocks of three to four interviews to ensure
agreement between coders was maintained. Saturation was
BK, a second-year master’s student, led the research and operationalized as inductive thematic saturation where
conducted the interviews. She has experience assisting with additional interviews yield little to no new information
pre-test interviews for a DCE and previously led and tran- as assessed by whether new codes emerged [43]. To cor-
scribed focus groups. She also has extensive experience roborate that data saturation had been reached, the method
working with older adults within the health system, and as a proposed by Guest et al. was employed [44]. Interviews
result developed communication skills with older adults. BK were conducted until the quotient of the saturation ratio was
had not met any of the participants before the interview. At ≤ 5%, and the sample included a mix of ages, sexes, and hip
the beginning of each interview session, she introduced her- and knee osteoarthritis [44].
self as a master’s student and outlined her role in the study. The manuscript and the results are reported per the
consolidated criteria for reporting qualitative research
2.5 Analysis Methods (COREQ), and the Hollin et al. guidelines for qualitative
research to inform stated preference studies [45, 46].
Descriptive statistics (counts and percentages) were used to
summarize the participant demographics. Interviews were
transcribed verbatim. Interview data were analyzed using 3 Results
NVivo 12™ [41]. The analysis followed the Braun and
Clarke method for thematic analysis [42]. Two researchers Thirteen participants with hip and/or knee osteoarthri-
(BK and GM) independently used a combination of deduc- tis were interviewed between January and March 2021.
tive and inductive coding strategies to independently develop Researcher consensus during the analytic process and
an analytic framework based on the interview guide, and application of the Guest et al. method confirmed data
then modified that framework as descriptive themes emerged saturation was reached at interview eleven (Table 3).
from the data. Peer debriefing between the coders took place However, given the relative novelty of the Guest et al.
after each had coded three interviews and facilitated discus- method, two additional interviews were analyzed to con-
sions and consensus regarding codes, descriptive themes, firm saturation. The sample was diverse, including three
and saturation. The analytic framework was refined, and men and ten women, whose ages ranged from 40 to 70
the researchers recoded the same three interviews and peer years (Table 4). Most participants were Caucasian (85%)
debriefing took place again. Agreement was reached and and had knee osteoarthritis (69%). Approximately half had
each researcher continued to code every interview, meeting received a diagnosis within the past 5 years (46%) and had
Factors Affecting OA Patient Non-surgical Management: A Formative Qualitative Study 541

Table 3  Saturation assessment Interview ­numbera 5 6 7 8 9 10 11 12 13

using the Guest et al. [46]
method New descriptive themes per interview 1 3 1 2 2 2 0 0 0
Number of base codes 79
New descriptive themes in run 4 4 3 4 4 2 0 0
Saturation ratio 5.06 5.06 3.80 5.06 5.06 2.53 0 0
a
The first four interviews encompass the base run

Table 4  Participant Characteristic Categories N = 13

demographics
Age (years) 20–49 3 (23%)
50–64 3 (23%)
65+ 7 (54%)
Sex Male 3 (23%)
Female 10 (77%)
Gender Man 3 (23%)
Woman 10 (77%)
Osteoarthritic joint Knee 9 (69%)
Hip 2 (15.5%)
Both 2 (15.5%)
Time since diagnosis 0–5 years 6 (46%)
6–10 years 3 (23%)
20 years or more 4 (31%)
Education level High school diploma 1 (8%)
Diploma from trade school or college 4 (31%)
University degree 8 (61%)
Population centre of residence between 1000 and 29,999 2 (15.5%)
between 30,000 and 99,999 2 (15.5%)
100,000 and over 9 (69%)
Population group Caucasian 11 (85%)
Indigenous peoples 1 (7.5%)
Other: “white, small percentage Indigenous/Metis” 1 (7.5%)
(self-reported by one participant)

a university degree (45%). Most resided in a population 3.1 Extrinsic Factors

centre of at least 100,000 people (69%).
A total of 16 factors were identified that participants Eleven extrinsic factors emerged related to management
considered when making management decisions for their decisions for osteoarthritis. The factors are organized
osteoarthritis. Through the analytic process, these factors according to types of interventions, general program and
were further classified as extrinsic or intrinsic as a means service details, and program-specific details.
of grouping these diverse factors. Extrinsic factors were
characteristics of management activities that program 3.1.1 Types of Interventions
designers have control over. Intrinsic factors were quali-
ties that existed within the individual. Eleven factors were Medication, potential risks and side effects, and mobility
extrinsic and related to features of programs and services. aids were the most important aspects of an intervention
Five factors were intrinsic and influenced how participants that participants considered when choosing a non-surgi-
approached management decisions for osteoarthritis and cal management strategy. For example, some participants
the options available from which to choose. A description had positive perceptions of medication and were open to
of each of these 16 factors, accompanied by supportive using medication for their osteoarthritis if it was needed.
quotes, is included in Table 5. In contrast, most participants had negative perceptions of
542 B. L. Kennedy et al.

Table 5  Supportive quotes for identified factors

Descriptive theme Description Quotes

Extrinsic Factors
Types of interventions
Medication Positive and negative perceptions of medication “I don’t believe in a lot of drug intervention and
stuff like that. So I wanted to do something
that didn’t involve taking pain killers for the
rest of my life… so all of those seemed to be a
better way, for me anyways, to try managing.”
(Participant #1)
“I believe [medications are] there for a reason.
I’m the first one that says take the pain medica-
tion when you need it.” (Participant #12)
Potential risks and side effects associated Considering risks and side effects “I still have a mental thing because all of the
with medication drugs have side effects, whether you like it or
not, and it’s the side effects… you can have a
drug that helps with your arthritis, but makes
you even more tired.” (Participant #6)
“…are there potential side effects? and what are
those?” (Participant #5)
Mobility aids Perceptions of mobility aids, such as stigma “…there’s also been [pause] for me, a bit of
surrounding use embarrassment, because I am heavier, I’m
walking with a cane, and what do people think.
You know what kind of judgements are being
put on me, that kind of thing.” (Participant #9)
“[I have] a prescription for an offloading knee
brace, which I haven’t done anything with…
honestly, it’s for optics I think mostly. I just
think it looks a bit stupid you know you got this
big brace. I try so hard not to hobble and not to
let [the osteoarthritis] kind of affect my life I
guess in a way. I don’t know how to put it, but
just the idea of having this brace, was, I was
just kind of, I didn’t want to go there I guess.”
(Participant #2)
General program and service details
Cost Overall or out-of-pocket costs “With not being able to work and being on dis-
ability [cost] is definitely an issue. I went to
get the injection… and they didn’t tell me there
was gonna be a cost, and my insurance wouldn’t
cover any of it. So the doctor at that point
said… this injection is two hundred and some
dollars and… this other injection is seventy
five and we need two [vials], so I would save
you this much money and it’s about as good…
nobody told me up until I got there that cost
would have been an issue. So that comes into
some degree of consideration.” (Participant #9)
“I place a high value on my health so unless the
fees were astronomical, I would get it done.
Most of them were covered by my healthcare
anyway.” (Participant #3)
Factors Affecting OA Patient Non-surgical Management: A Formative Qualitative Study 543

Table 5  (continued)
Descriptive theme Description Quotes

Presence or absence of recommendations Recommendations of an option by peers who “The initial idea of injections came as one option
have experience with the option or from from my doctor saying well, we can do this,
trusted providers and he went over some of those pros and cons.
I knew my own brother had gone through it so
I asked him as well… so knowing that he had
gone through it and it worked, I said okay I
know this can work then.” (Participant #5)
“Whatever my doctor tells me to do I’ll do it. You
know if I have to go to physio I go to physio.”
(Participant #8)
Perceived effectiveness Effectiveness “I try and do all sorts of other things before I
resort to a medication… but if the medication
works… then I’m all about it. Like if I see the
results then I’m okay with taking it.” (Partici-
pant #4)
“I’m resistant to all medication, but when the
effectiveness is significant enough, I’m bal-
anced, and I’ll take it. I just haven’t found a
pain medication that I can take that worked well
enough to make it worth any problem at all.”
(Participant #7)
Program-specific details
Mode of delivery Whether the option is delivered in-person or “We could have the biggest and best equipment
remotely sitting in our house and I would not use it. I
need to go somewhere to do it. That’s the way I
am.” (Participant #5)
“As far as going to appointments, I’m okay with
that.” (Participant #4)
Location of appointments Proximity of in-person appointments “[The appointment] has to be close to me, espe-
cially in the winter.” (Participant #1)
“…[location] certainly does play a role, because
I do live out in ___ so we don’t have accessibil-
ity to as many services… I don’t have as much
access as somebody else in town.” (Participant
#12)
Type of session Whether sessions are group or individual-based “I always prefer individual sessions… for
personal training and working on my knee
I just like to have my own personal trainer.”
(Participant #13)
“I need to be there in-person… and I need the
other people around me who are doing the same
thing.” (Participant #1)
Level of supervision Whether it is supervised or self-directed “Even though the online videos are great, they’re
always saying don’t do this, watch this, be aware
of this, it’s just not the same as someone staring
at you and telling you that.” (Participant #2)
“I like to have somebody there saying you should
put your weight on your heel instead or what-
ever it is. I just like that personal direction.”
(Participant #13)
544 B. L. Kennedy et al.

Table 5  (continued)
Descriptive theme Description Quotes

Duration The length of the commitment to the option “I just feel I don’t wanna be, I guess, I’ll say it
bothered going every week or whatever it is to
[physio] for that hour or whatever.” (Participant
#2)
“I pay sometimes for physio advice if I feel like
I don’t know what else to try myself, but then
I want them to teach me to do it all at home by
myself. I don’t wanna go to physio all the time.”
(Participant #6)
Intrinsic Factors
Declining further management Further management is not needed or preferred “[the doctor] asked me have you ever thought
about getting a shot in your knee, a cortisone
shot… and I mean at this point I feel like I’m
still doing really well that I don’t need anything
like that.” (Participant #13)
“I know there’s like prescription medication that
you can take for [the osteoarthritis], but that’s
not, I, I just feel like I’m not there yet because
I’m still like able to do everything.” (Participant
#4)
Context of overall health Considering management decisions in the “Pools were going to be my main way of trying
context of their overall health and other health to work the hip and get some exercise… I had
conditions been doing a bit of [swimming]… the agora-
phobia had gotten to the point that I couldn’t
even go to the pool alone. So, my therapist and
I were working toward me being able to do that
and I now can do that, although it may be a bit
of a setback when [pools re-open].” (Participant
#9)
“I mean Tylenol is good enough… and if I want
to control my pain, I could take those pills,
[and] I could up my blood pressure medication.
I could do that… and I think it would work. I
just don’t want to right now.” (Participant #2)
Views about joint replacement surgery Some were waiting for surgery; some were try- “To me, surgery has to be the last of the last of
ing to prevent or delay it the last resort, where you literally can’t live
another minute without it… you’ve got to
exhaust your good periods and hit a time where
you don’t have any, any more good periods and
your pain is too high to live with.” (Participant
#6)
“my driving belief that I don’t want surgery made
it really easy to try [anything]… I’m an any-
thing but surgery person.” (Participant #7)
Level of understanding and knowledge Whether participants felt informed about the “I’ve tried all of the [options] that I’ve been told
disease and management about. There may be some I don’t know about,
that’s possible.” (Participant #3)
“I don’t really know much about actual osteoar-
thritis.” (Participant #13)
Choice in management Whether participants felt they have choice in “Do I have choice? Well for sure, it’s up to me
management right. I mean I can take the glucosamine and
turmeric, I can do my exercises, or not.” (Par-
ticipant #10)
“I think the choice is to cave in and give up like
some people would do… and [the osteoarthritis]
becomes worse and worse” (Participant #12)
Factors Affecting OA Patient Non-surgical Management: A Formative Qualitative Study 545

medication and actively avoided any option that included 3.1.3 Program‑Specific Details
medication. Reasons for avoiding medication included not
believing in drug intervention, wanting to stay more natu- Some of the factors that emerged were only discussed in the
ral, and a resistance to all medication in general. Partici- context of programs, including mode of delivery, location
pants also discussed considering and intentionally weigh- of appointments, type of session, level of supervision, and
ing the potential risks and side effects associated with a duration. Mode of delivery of programs appeared to play a
medication against potential benefits of medication-based role in participants’ decisions about osteoarthritis manage-
options. ment options. Some participants preferred in-person options
Another factor related to the type of intervention was over options that were delivered virtually. The ability to go
mobility aids, which included devices such as canes, braces, somewhere for in-person options had a motivational compo-
and walkers. Some participants discussed stigmas around nent that virtual options, which could be done from home,
using mobility aids, including their appearance, as well as lacked. All participants that identified the mode of deliv-
peers’ negative perceptions related to use of mobility aids. ery as a factor in their decision making preferred in-person
Others also discussed their own negative views of mobility programming options. For some, the location of in-person
aids, citing their bulk, inconvenience, and that they drew appointments was an important consideration, particularly
attention to impaired physical abilities. These perceptions/ those living in rural areas.
factors lead to a hesitancy or resistance towards using mobil- Session type, be it group or individual, also emerged as a
ity aids. relevant factor to participation. Some participants preferred
options in an individual setting, citing the one-on-one atten-
3.1.2 General Program and Service Details tion. Others preferred group sessions, citing the motivational
aspect of being surrounded by peers working on the same
Some of the factors that emerged were broad and applica- thing.
ble to both programs and services, including cost, absence Level of supervision was another factor. Some partici-
or presence of recommendations, and perceived effective- pants expressed that they always preferred to be supervised
ness. Participants conceptualized costs in different ways. For because of the direct guidance associated with supervision.
some, costs reflected the total cost of an option, which may In addition, some participants preferred options that had an
be paid for by different sources (e.g., some may be covered initial, supervised teaching session with the remainder being
by healthcare benefits). For others, costs reflected the out-of- self-led and unsupervised.
pocket cost incurred with an option. For some participants, A further consideration was the duration or the number of
cost had no bearing on management decisions, and this gave sessions associated with an option. For example, participants
them a greater number of options to choose from. However, expressed reservations towards physiotherapy, as most did
most participants expressed that cost influenced their osteo- not want an option that had a weekly commitment.
arthritis management decisions.
Recommendations from peers who had experience with 3.2 Intrinsic Factors
an option or from trusted providers emerged as influencing
management decisions. Options that came with a recom- Intrinsic factors included the desire for further management,
mendation from these groups were perceived more favour- the context of overall health, views about joint replacement
ably and increased the likelihood of trying an intervention surgery, level of understanding and knowledge, and choice
than those that were not recommended. Trusting the source in management. Perceived need for further management
was an important quality associated with change in opin- influenced participants’ management decisions. Most par-
ion or actions that resulted in trying a new approach or ticipants discussed that further management was not pre-
intervention. ferred because they did not believe they required it. This
Perceived effectiveness of an intervention was also part included participants whose current management strategies
of decision making for osteoarthritis management for some were working well. As a result, participants declined further
participants. Participants discussed effectiveness in terms management and did not consider any other options, outside
of their experience with how well an option had worked for of what they were currently doing. Declining further man-
them or their peers, their own research on the potential effec- agement was also related to the availability of information
tiveness of an option, and what a provider has shared about on the benefits of early management.
the potential effectiveness of an option. They also discussed Participants considered their overall health, including
effectiveness in relation to weighing potential risks and side their other health conditions, when making management
effects, with some saying that if it worked well enough it decisions. In some cases, osteoarthritis management options
may balance out negative side effects. had implications for patients’ other health conditions, and
they were forced to make trade-offs when deciding how to
546 B. L. Kennedy et al.

manage their osteoarthritis. For example, one participant 24, 28–30], mode of delivery [18, 21, 23, 24, 29, 30], loca-
stopped taking a medication that was helping their osteoar- tion of in-person appointments [12, 16, 18, 19, 21, 26, 28,
thritis because it increased their blood pressure. Considering 29], type of session [16, 25, 26, 29, 30], duration [16, 25, 26,
the context of their overall health encompassed trying to 30, 31], perceived effectiveness [12, 14, 15, 18, 22, 26, 27,
manage the osteoarthritis while also living with other health 29, 30, 32, 33], and the level of supervision [22]. In addition,
concerns, and how osteoarthritis management decisions our findings are consistent with the broader literature on
were not made in isolation, they were made by considering mobility aids for older adults that recognizes that attitudes
the implications on those other health concerns. and beliefs strongly affect the decision to use a mobility aid
Participants in this study all had at least one osteoarthritic [47]. The effect of endorsement from one’s health care pro-
joint that had not undergone joint replacement surgery, and vider on decision making has also been widely described in
their views about joint replacement surgery influenced their the literature [17, 25, 29–31]. Selten et al. conducted inter-
management decisions. Some participants had previously views to understand reasons for osteoarthritis management
undergone surgical consultations, some were waiting for sur- choices and discussed that family and friends having positive
gery, and some had not yet considered surgery. Participants experiences with a management option was a reason patients
who were actively avoiding or delaying surgery expressed cited for trying new options [29]. Lastly, decision making
a preference to exhaust all non-surgical options before con- specific to osteoarthritis is made within the broader con-
sidering surgery. text of one’s overall health and other specific comorbidities
Whether participants felt informed about the disease and or health issues. Selten et al. also identified that personal
its management, or their level of understanding and knowl- circumstances, such as comorbidities, influenced patients’
edge, was another factor impacting decision making. Most management choices [29].
participants discussed how they were unsure if they were Whether an option included medication, and the potential
aware of all the available options to manage their osteoar- side effects and risks associated with medication, influenced
thritis or associated symptoms. Some also felt they lacked decision making about osteoarthritis management. A DCE
knowledge about the disease itself. Many factors influenced estimating the relative influence of medication-related fac-
their level of understanding and knowledge, including access tors on the choices of patients with osteoarthritis found that
to healthcare professionals with expertise in osteoarthritis side effects, out-of-pocket costs, and mode of action were
and general availability of information. Interestingly, the driving management decisions for osteoarthritis [11]. This
degree to which understanding and knowledge impacted was also consistent with a qualitative study investigating
participants’ choices in managing osteoarthritis appeared patient and physician perceptions of osteoarthritis care by
to be variable: some felt that they were knowledgeable and Miller et al. that reported participants preferred options
that they had choice while others felt they did not and con- without the side effects associated with medications [48].
versely, some felt they lacked knowledge but that they had Level of understanding and knowledge was an intrinsic
choice while others felt they did not. Choice was more than factor that influenced management decisions. Miller et al.
just having management options to choose from and encom- also reported participants feeling like they had insufficient
passed adherence or non-adherence to recommendations and education on osteoarthritis and treatment options [48]. The
mindsets about living with the disease. patient–patient research study referenced earlier [9], which
aimed to understand what quality care means to Albertans
with osteoarthritis and what is most helpful for managing
4 Discussion their osteoarthritis, identified the right knowledge as one
of three components of a “quality system that supports
This study identified several factors that patients consider osteoarthritis patients”. Preference for specific information
in their choice of non-surgical management for hip and about osteoarthritis and how to manage it to “help us choose
knee osteoarthritis. Decision making regarding manage- self-management strategies appropriately” emerged as an
ment options for a chronic condition is a complex process. important component [9]. Understanding the knowledge that
These findings indicate that there are several aspects of one’s patients need to make informed decisions to manage their
environment (extrinsic factors) as well as internal qualities osteoarthritis could improve person-centred osteoarthritis
(intrinsic factors) that inform an individual’s perspective care. This requires co-creating with patients a patient-cen-
and, ultimately, how they proceed to manage their osteoar- tred knowledge component for any osteoarthritis manage-
thritis and associated symptoms. Each factor presented has ment strategy.
some degree of relevance to the decision-making process, The desire for further management was an emergent novel
which is determined at the individual level. factor that was found to influence osteoarthritis management
Several of the identified factors were consistent with the decision making, with most declining further management.
previously identified literature (Table 2): cost [12, 18, 19, 22, Novel factors were those not previously identified in the
Factors Affecting OA Patient Non-surgical Management: A Formative Qualitative Study 547

literature. Most participants believed that they were not at a the study population [11]. Policy-relevant attributes reflect
point where more management was needed and as a result, essential characteristics of the interventions that the patients
they did not consider other options at that time. This factor is will evaluate in the DCE. A literature review on the policy
not about the characteristics of an option and is amenable to topic, as was conducted in the present study, can lead to
change, for instance if symptoms get worse. It also presents the identification of a list of conceptual attributes. The tar-
an opportunity for education about the potential benefits of get population must then consider the conceptual attributes
non-surgical management at any point during the osteoar- important for their inclusion in the DCE, which requires
thritis journey. This factor is related to the right manage- a rigorous qualitative study within the local context, as in
ment at the right time based on the participant’s situation the present study. Qualitative studies can pick up context-
and could have implications for providing person-centred specific attributes that may be absent in the literature. Their
care for osteoarthritis. Understanding why a patient is refus- inclusion consequently reduces the chance of omitting rel-
ing an option could help clinicians address concerns and evant attributes. The extrinsic factors that emerged through
provide information so that patients have the tools to ensure analysis are all reasonable candidates as attributes in a DCE
appropriate osteoarthritis care over time and with varying because they are policy relevant, important to the patients,
symptoms. and reflect characteristics of osteoarthritis management
Participants’ views regarding joint replacement were also strategies.
a novel factor that was found to influence management deci- The next steps in this research are to refine this list of fac-
sion making about osteoarthritis, with those who wanted to tors in the development of the specific attributes and levels
avoid or delay surgery wishing to exhaust all non-surgical for a subsequent DCE. Both the factors identified in this
options. Studies have described patients avoiding or delay- manuscript and the a priori features identified from the lit-
ing surgery, however, the implications of this on manage- erature (Table 2) will be used in the development process
ment decision making have not been explored [49–52]. which will follow best practices for attribute and level selec-
Understanding the effect that participants’ views about joint tion from the checklist by Bridges et al. [53]. First attributes
replacement surgery have on osteoarthritis decision making will be selected based on the following criteria presented
could provide further insight into what drives decision mak- by Bridges et al.: “relevance to the research question, rel-
ing and identify gaps in knowledge or gaps in care. evance to the decision context, and whether attributes are
Whether participants felt like they had choice in manag- related to one another”, and then levels will be selected
ing their osteoarthritis emerged as a novel factor. To make [53]. The factors presented above should be investigated in
management decisions for osteoarthritis, those options must the DCE to understand the preference weights. This would
be known to the patient. The options that patients can choose provide an understanding of the most important factors for
from were in the context of the information available to them osteoarthritis decision making, and how patients make trade-
and, as such, were linked to their level of understanding and offs. Intrinsic factors such as the influence that participants’
knowledge. Understanding the patient’s level of knowledge views about joint replacement surgery have on osteoarthritis
and understanding, and whether they feel like they have management decisions, desire for further management, and
choices, could help program designers better address barri- whether participants felt like they had choice in management
ers to non-surgical management. are novel factors that require further exploration in subse-
quent research. Targeted questions about these novel factors
4.1 Informing a DCE within the DCE survey could provide additional insight for
decision makers into the relationships with the other factors
The following factors relate to individual beliefs and are identified, and their role in patient decision making.
internal considerations and not choices that a participant
would make. For example, a participant may or may not 4.2 Strengths and Limitations
be considering surgery, which influences how they proceed
with management decision making for osteoarthritis. These This study’s qualitative design was well aligned with our
factors are, however, still important for the subsequent DCE objectives. Semi-structured interviews allowed for in-depth
because they could help understand responses to the DCE’s understanding of the experience of patients living with
tasks. Collecting information about these factors, by asking and managing their hip and knee osteoarthritis in Alberta.
targeted questions, would provide the research team with a Another strength was the focus on patients with lived experi-
better understanding of the motivations behind the choices ence as experts in osteoarthritis, including engaging patient
that participants make and where the participant is in their partners in the design and refinement of the interview guide.
management journey. Conducting interviews via video conferencing likely reduced
Attributes for a DCE on non-surgical osteoarthritis man- barriers to participation such as travel and mobility. Further,
agement options should be policy relevant and important to the ability to have visual contact on screen likely enhanced
548 B. L. Kennedy et al.

communication between the interviewer and interviewee. Availability of data and material The ethics approval for this study
Using video conferencing can also limit participation as it precludes the sharing of the raw data.
requires a web-enabled device and a good-quality internet Code availability Not applicable.
connection and as such participants were offered the option
of using the telephone, though none opted to use this method Ethics approval Ethics approval was obtained from The Conjoint
[54]. The study also had limitations. The inclusion crite- Health Research Ethics Board, UCalgary (REB20-0671) in January
of 2021.
ria required participants to be English-speaking, and this
excluded Albertans who spoke other languages. Also, all Consent Informed verbal consent was obtained prior to conducting
the participants identified as male and female sexes, and interviews.
men and women genders, respectively, and as such no other
Author contributions GRC and DAM conceived and designed the
perspectives were captured. It is common that participants study with input from BLK and feedback from AKR and CAE. BLK
in health research are often female, highly educated, and conducted and transcribed all the interviews. BLK worked with GM to
come from higher socioeconomic backgrounds and our study analyze and interpret the data with input from AKR, and feedback from
shares this limitation [55]. GRC, CAE, and DAM. BLK drafted the manuscript and all authors
provided critical feedback and approved the submitted version.

5 Conclusions
References
This research identified 16 factors that patients consider
when making management decisions for their osteoarthri- 1. Arden N, Nevitt MC. Osteoarthritis: epidemiology. Best Pract Res
tis. Three of these were novel factors that require further Clin Rheumatol. 2006;20(1):3–25. https://​doi.​org/​10.​1016/j.​berh.​
research to understand the implications for osteoarthritis 2005.​09.​007.
2. Zhang Y, Jordan JM. Epidemiology of osteoarthritis. Clin Geriatr
care. The study contributes to a better understanding of
Med. 2010;26(3):355–69. https://​doi.​org/​10.​1016/j.​cger.​2010.​03.​
management decision making for patients with hip and knee 001.
osteoarthritis and provides a foundation for future research 3. Hip and Knee Replacements in Canada, 2017–2018: Cana-
to measure patients’ preferences. It also provides a list of dian Joint Replacement Registry Annual Report. Ottawa, ON:
CIHI2019.
extrinsic factors that can be used in a DCE, with the goal
4. Hunter DJ, Lo GH. The management of osteoarthritis: an over-
of further understanding patient preferences for the non- view and call to appropriate conservative treatment. Med Clin
surgical management of osteoarthritis. North Am. 2009;93(1):127–43. https://​doi.​org/​10.​1016/j.​mcna.​
2008.​07.​009.
5. Hunter DJ, Bierma-Zeinstra S. Osteoarthritis. Lancet (London,
England). 2019;393(10182):1745–59. https://​doi.​org/​10.​1016/​
s0140-​6736(19)​30417-9.
Supplementary Information The online version contains supplemen- 6. Bannuru RR, Osani MC, Vaysbrot EE, Arden NK, Bennell K,
tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 40271-0​ 22-0​ 0577-9. Bierma-Zeinstra SMA, et al. OARSI guidelines for the non-sur-
gical management of knee, hip, and polyarticular osteoarthritis.
Acknowledgements Thank you to Jean Miller and Sylvia Teare, trained Osteoarthr Cartil. 2019;27(11):1578–89. https://d​ oi.o​ rg/1​ 0.1​ 016/j.​
patient researchers who received training in the Patient and Community joca.​2019.​06.​011.
Engagement Research (PaCER) program at the University of Calgary, 7. King LK, Marshall DA, Faris P, Woodhouse L, Jones CA, Nose-
for bringing a patient perspective to the design of this project. Thank worthy T, et al. Use of recommended non-surgical knee osteoar-
you to the Alberta Strategy for Patient-Oriented Research Support Unit thritis management in patients prior to totalv knee arthroplasty: a
Patient Engagement Platform (AbSPORU PEP), the McCaig Institute cross-sectional study. J Rheumatol. 2019. https://d​ oi.o​ rg/1​ 0.3​ 899/​
for Bone and Joint Health, Arthritis Research Canada (ARC), and the jrheum.​190467.
Arthritis Society for their support with recruitment of participants. 8. Marshall DA, Faris P, Jones A, Noseworthy T, Dick D, Squire
Howden J, et al. Relationship between appropriateness and
arthroplasty recommendation. Int J Technol Assess Health Care.
Declarations 2019;34(1):58–9.
9. Miller JL, Teare SR, Marlett N, Shklarov S, Marshall DA. Support
Funding Canadian Institutes of Health Research grant number 407498 for living a meaningful life with osteoarthritis: a patient-to-patient
“Towards a patient-centred system: Integrating preferences of patients research study. Patient. 2016;9(5):457–64. https://d​ oi.o​ rg/1​ 0.1​ 007/​
with osteoarthritis into evaluation of health services interventions to s40271-​016-​0169-9.
improve patient outcomes and health system efficiency”. DAM is sup- 10. Alberta Bone & Joint Institute. The Osteoarthritis Crisis in
ported by the Arthur J.E. Child Chair in Rheumatology Research and Alberta: Access, Quality, and Long-Term Planning. Alberta, Can-
Canada Research Chair in Health Systems and Services Research. ada: AHS Bone and Joint Health Strategic Clinical Network2019.
11. Laba TL, Brien JA, Fransen M, Jan S. Patient preferences for
Conflict of interest The authors have no conflicts of interest relevant to adherence to treatment for osteoarthritis: the MEdication Deci-
the content of this article. sions in Osteoarthritis Study (MEDOS). BMC Musculoskelet
Disord. 2013;14:160. https://​doi.​org/​10.​1186/​1471-​2474-​14-​160.
Factors Affecting OA Patient Non-surgical Management: A Formative Qualitative Study 549

12. Ackerman IN, Livingston JA, Osborne RH. Personal perspectives for knee osteoarthritis-a qualitative study. Arthritis Care Res.
on enablers and barriers to accessing care for hip and knee osteo- 2019;71(4):545–57. https://​doi.​org/​10.​1002/​acr.​23618.
arthritis. Phys Ther. 2016;96(1):26–36. https://​doi.​org/​10.​2522/​ 26. Lee F-KI, Lee T-FD, So WK-W. Effects of a tailor-made exercise
ptj.​20140​357. program on exercise adherence and health outcomes in patients
13. Allison K, Setchell J, Egerton T, Delany C, Bennell KL. In theory, with knee osteoarthritis: a mixed-methods pilot study. Clin Inter-
yes; in practice, uncertain: a qualitative study exploring physical ven Aging. 2016;11:1391–402.
therapists’ attitudes toward their roles in weight management for 27. MacKay C, Hawker GA, Jaglal SB. Qualitative study exploring
people with knee osteoarthritis. Phys Ther. 2019;99(5):601–11. the factors influencing physical therapy management of early knee
https://​doi.​org/​10.​1093/​ptj/​pzz011. osteoarthritis in Canada. BMJ Open. 2018;8(11): e023457. https://​
14. Brembo EA, Kapstad H, Eide T, Mansson L, Van Dulmen S, Eide doi.​org/​10.​1136/​bmjop​en-​2018-​023457.
H. Patient information and emotional needs across the hip osteo- 28. MacKay C, Hawker GA, Jaglal SB. How do physical therapists
arthritis continuum: a qualitative study. BMC Health Serv Res. approach management of people with early knee osteoarthritis? A
2016;16:88. https://​doi.​org/​10.​1186/​s12913-​016-​1342-5. Qualitative Study. Phys Ther. 2020;100(2):295–306. https://​doi.​
15. Bunzli S, O’Brien P, Ayton D, Dowsey M, Gunn J, Choong P, org/​10.​1093/​ptj/​pzz164.
et al. Misconceptions and the acceptance of evidence-based non- 29. Selten EM, Vriezekolk JE, Geenen R, van der Laan WH, van
surgical interventions for knee osteoarthritis a qualitative study. der Meulen-Dilling RG, Nijhof MW, et al. Reasons for treatment
Clin Orthop Relat Res. 2019;477(9):1975–83. https://​doi.​org/​10.​ choices in knee and hip osteoarthritis: a qualitative study. Arthritis
1097/​CORR.​00000​00000​000784. Care Res. 2016;68(9):1260–7. https://​doi.​org/​10.​1002/​acr.​22841.
16. Carmona-Teres V, Moix-Queralto J, Pujol-Ribera E, Lumillo- 30. Selten EMH, Geenen R, van der Laan WH, van der Meulen-
Gutierrez I, Mas X, Batlle-Gualda E, et al. Understanding knee Dilling RG, Schers HJ, Nijhof MW, et al. Hierarchical structure
osteoarthritis from the patients’ perspective: a qualitative study. and importance of patients’ reasons for treatment choices in knee
BMC Musculoskelet Disord. 2017;18(1):225. https://​doi.​org/​10.​ and hip osteoarthritis: a concept mapping study. Rheumatology
1186/​s12891-​017-​1584-3. (Oxford). 2017;56(2):271–8. https://​doi.​org/​10.​1093/​rheum​atolo​
17. Cuperus N, Smink AJ, Bierma-Zeinstra SMA, Dekker J, Schers gy/​kew409.
HJ, de Boer F, et al. Patient reported barriers and facilitators to 31. Selten EMH, Vriezekolk JE, Nijhof MW, Schers HJ, van der
using a self-management booklet for hip and knee osteoarthritis Meulen-Dilling RG, van der Laan WH, et al. Barriers imped-
in primary care: results of a qualitative interview study. BMC Fam ing the use of non-pharmacological, non-surgical care in hip and
Pract. 2013;14:181. https://​doi.​org/​10.​1186/​1471-​2296-​14-​181. knee osteoarthritis: the views of general practitioners, physical
18. Egerton T, Nelligan R, Setchell J, Atkins L, Bennell KL. General therapists, and medical specialists. J Clin Rheumatol Pract Rep
practitioners’ perspectives on a proposed new model of service Rheum Musculoskeletal Dis. 2017;23(8):405–10. https://​doi.​org/​
delivery for primary care management of knee osteoarthritis: a 10.​1097/​RHU.​00000​00000​000562.
qualitative study. BMC Fam Pract. 2017;18(1):85. https://​doi.​org/​ 32. Thorstensson CA, Roos EM, Petersson IF, Arvidsson B. How do
10.​1186/​s12875-​017-​0656-7. middle-aged patients conceive exercise as a form of treatment for
19. Gay C, Eschalier B, Levyckyj C, Bonnin A, Coudeyre E. Motiva- knee osteoarthritis? Disabil Rehabil. 2006;28(1):51–9.
tors for and barriers to physical activity in people with knee osteo- 33. Wallis JA, Webster KE, Levinger P, Singh PJ, Fong C, Taylor NF.
arthritis: a qualitative study. Jt Bone Spine. 2018;85(4):481–6. Perceptions about participation in a 12-week walking program
https://​doi.​org/​10.​1016/j.​jbspin.​2017.​07.​007. for people with severe knee osteoarthritis: a qualitative analysis.
20. Holden MA, Waterfield J, Whittle R, Bennell K, Quicke JG, Disabil Rehabil. 2019;41(7):779–85. https://​doi.​org/​10.​1080/​
Chesterton L, et al. How do UK physiotherapists address weight 09638​288.​2017.​14087​10.
loss among individuals with hip osteoarthritis? A mixed-methods 34. Bradshaw C, Atkinson S, Doody O. Employing a Qualitative
study. Musculoskeletal Care. 2019;17(1):133–44. https://​doi.​org/​ Description Approach in Health Care Research. Glob Qual Nurs
10.​1002/​msc.​1383. Res. 2017. https://​doi.​org/​10.​1177/​23333​93617​742282.
21. Hurley M, Dickson K, Hallett R, Grant R, Hauari H, Walsh N, 35. University of Calgary. Participate in Research at UCalgary. 2018.
et al. Exercise interventions and patient beliefs for people with https://​www.​ucalg​ary.​ca/​resea​rch/​parti​cipate/. Accessed 19 Jun
hip, knee or hip and knee osteoarthritis: a mixed methods review. 2021.
Cochrane Database Syst Rev. 2018;4: CD010842. https://​doi.​org/​ 36. AbSPORU. Albertans4HealthResearch.ca. 2021. https://a​ pp.b​ ette​
10.​1002/​14651​858.​CD010​842.​pub2. rimpa​ct.​com/​Publi​cOrga​nizat​ion/​25809​ea0-​7311-​40db-​bfac-​
22. Kanavaki AM, Rushton A, Efstathiou N, Alrushud A, Klocke R, aaa0e​28ba5​18/1. Accessed 19 Jun 2021.
Abhishek A, et al. Barriers and facilitators of physical activity 37. Arthritis Society. Study Recruitment. 2021. https://​arthr ​itis.​ca/​
in knee and hip osteoarthritis: a systematic review of qualitative resea​rchers/​patie​nt-​engag​ement/​study-​recru​itment. Accessed 19
evidence. BMJ Open. 2017;7(12): e017042. https://​doi.​org/​10.​ Jun 2021.
1136/​bmjop​en-​2017-​017042. 38. Arthritis Research Canada. Participate in Research. 2021. https://​
23. Lawford BJ, Delany C, Bennell KL, Bills C, Gale J, Hinman RS. www.​arthr​itisr​esear​ch.​ca/​parti​cipate-​in-​resea​rch/. Accessed 19
Training physical therapists in person-centered practice for people Jun 2021.
with osteoarthritis: a qualitative case study. Arthritis Care Res. 39. Marlett N, Shklarov S, Marshall D, Santana MJ, Wasylak T. Build-
2018;70(4):558–70. https://​doi.​org/​10.​1002/​acr.​23314. ing new roles and relationships in research: a model of patient
24. Lawford BJ, Delany C, Bennell KL, Hinman RS. “I was really engagement research. Quality Life Res. 2015;24(5):1057–67.
skeptical…But it worked really well”: a qualitative study of https://​doi.​org/​10.​1007/​s11136-​014-​0845-y.
patient perceptions of telephone-delivered exercise therapy by 40. Shklarov S, Marshall DA, Wasylak T, Marlett NJ. “Part of the
physiotherapists for people with knee osteoarthritis. Osteoarthr Team”: mapping the outcomes of training patients for new roles in
Cartil. 2018;26(6):741–50. https://​doi.​org/​10.​1016/j.​joca.​2018.​ health research and planning. Health Expect. 2017;20(6):1428–36.
02.​909. https://​doi.​org/​10.​1111/​hex.​12591.
25. Lawford BJ, Delany C, Bennell KL, Hinman RS. “I was really 41. QSR-International. NVivo 12 Qualitative Data Analysis Soft-
pleasantly surprised”: firsthand experience and shifts in physi- ware. 1999. https://​www.​qsrin​terna​tional.​com/​nvivo-​quali​tative-​
cal therapist perceptions of telephone-delivered exercise therapy data-​analy​sis-​softw​are/​home.
550 B. L. Kennedy et al.

42. Braun V, Clarke V. Using thematic analysis in psychology. Qual the future. Arthritis Rheum. 2007;57(1):27–34. https://d​ oi.o​ rg/1​ 0.​
Res Psychol. 2006;3(2):77–101. https://​doi.​org/​10.​1191/​14780​ 1002/​art.​22472.
88706​qp063​oa. 50. Hawker GA, Wright JG, Badley EM, Coyte PC. Perceptions of,
43. Saunders B, Sim J, Kingstone T, Baker S, Waterfield J, Bartlam B, and willingness to consider, total joint arthroplasty in a popu-
et al. Saturation in qualitative research: exploring its conceptual- lation-based cohort of individuals with disabling hip and knee
ization and operationalization. Qual Quant. 2018;52(4):1893–907. arthritis. Arthritis Rheum. 2004;51(4):635–41. https://​doi.​org/​10.​
https://​doi.​org/​10.​1007/​s11135-​017-​0574-8. 1002/​art.​20524.
44. Guest G, Namey E, Chen M. A simple method to assess and 51. Hawker GA, Wright JG, Coyte PC, Williams JI, Harvey B, Glazier
report thematic saturation in qualitative research. PLoS ONE. R, et al. Differences between men and women in the rate of use of
2020;15(5):e0232076-e. https://​doi.​org/​10.​1371/​journ​al.​pone.​ hip and knee arthroplasty. N Engl J Med. 2000;342(14):1016–22.
02320​76. https://​doi.​org/​10.​1056/​nejm2​00004​06342​1405.
45. Hollin IL, Craig BM, Coast J, Beusterien K, Vass C, DiSantoste- 52. Hawker GA, Wright JG, Coyte PC, Williams JI, Harvey B, Gla-
fano R, et al. Reporting formative qualitative research to support zier R, et al. Determining the need for hip and knee arthroplasty:
the development of quantitative preference study protocols and the role of clinical severity and patients’ preferences. Med Care.
corresponding survey instruments: guidelines for authors and 2001;39(3):206–16. https://​doi.​org/​10.​1097/​00005​650-​20010​
reviewers. Patient. 2020;13(1):121–36. https://​doi.​org/​10.​1007/​ 3000-​00002.
s40271-​019-​00401-x. 53. Bridges JF, Hauber AB, Marshall D, Lloyd A, Prosser LA, Regier
46. Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting DA, et al. Conjoint analysis applications in health–a checklist: a
qualitative research (COREQ): a 32-item checklist for interviews report of the ISPOR Good Research Practices for Conjoint Analy-
and focus groups. Int J Qual Health Care. 2007;19(6):349–57. sis Task Force. Value Health. 2011;14(4):403–13. https://​doi.​org/​
https://​doi.​org/​10.​1093/​intqhc/​mzm042. 10.​1016/j.​jval.​2010.​11.​013.
47. Resnik L, Allen S, Isenstadt D, Wasserman M, Iezzoni L. Perspec- 54. Carter SM, Shih P, Williams J, Degeling C, Mooney-Somers
tives on use of mobility aids in a diverse population of seniors: J. Conducting qualitative research online: challenges and solu-
implications for intervention. Disabil Health J. 2009;2(2):77–85. tions. Patient. 2021;14(6):711–8. https:// ​ d oi. ​ o rg/ ​ 1 0. ​ 1 007/​
https://​doi.​org/​10.​1016/j.​dhjo.​2008.​12.​002. s40271-​021-​00528-w.
48. Miller KA, Osman F, Baier ML. Patient and physician perceptions 55. Cheung KL, ten Klooster PM, Smit C, de Vries H, Pieterse
of knee and hip osteoarthritis care: a qualitative study. Int J Clin ME. The impact of non-response bias due to sampling in pub-
Pract (Esher). 2020;74(12): e13627. https://​doi.​org/​10.​1111/​ijcp.​ lic health studies: a comparison of voluntary versus mandatory
13627. recruitment in a Dutch national survey on adolescent health.
49. Ballantyne PJ, Gignac MA, Hawker GA. A patient-centered per- BMC Public Health. 2017;17(1):276. https://​doi.​org/​10.​1186/​
spective on surgery avoidance for hip or knee arthritis: lessons for s12889-​017-​4189-8.

Authors and Affiliations

Bryanne L. Kennedy1 · Gillian R. Currie1,2,3,4 · Ania Kania‑Richmond1,5 · Carolyn A. Emery1,2,3,4,6,7 · Gail MacKean1 ·
Deborah A. Marshall1,3,4,7,8

1 6
Department of Community Health Sciences, Cumming Sport Injury Prevention Research Center, Faculty
School of Medicine, University of Calgary, Alberta, Canada of Kinesiology, University of Calgary, Alberta, Canada
2 7
Department of Pediatrics, Cumming School of Medicine, McCaig Institute for Bone and Joint Health, University
University of Calgary, Alberta, Canada of Calgary, Alberta, Canada
3 8
Alberta Children’s Hospital Research Institute, University Health Research Innovation Centre, University of Calgary,
of Calgary, Alberta, Canada Room 3C56, 3280 Hospital Drive NW, Calgary,
4 AB T2N 4Z6, Canada
O’Brien Institute of Public Health, University of Calgary,
Alberta, Canada
5
Bone and Joint Health Strategic Clinical Network, Alberta
Health Services, Alberta, Canada