A patient-centered perspective on surgery avoidance for hip or knee arthritis: Lessons for the future


  • Peri J. Ballantyne,

    Corresponding author
    1. University of Toronto, Toronto, Ontario, Canada
    • Faculty of Pharmacy, University of Toronto, 19 Russell Street, Toronto, Ontario, Canada M5S 2S2
    Search for more papers by this author
  • Monique A. M. Gignac,

    1. Arthritis Community Research and Evaluation Unit, University Health Network, Toronto, Ontario, Canada
    2. University of Toronto, Toronto, Ontario, Canada
    Search for more papers by this author
  • Gillian A. Hawker

    1. Women's College Hospital at the University of Toronto, Toronto, Ontario, Canada
    2. Arthritis Community Research and Evaluation Unit, Toronto, Ontario, Canada
    3. University Health Network Research Institute, Toronto, Ontario, Canada
    4. Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada
    Search for more papers by this author



Research indicates that there is a discrepancy between need and patient preference for total joint arthroplasty (TJA), an efficacious and cost-effective treatment for severe hip or knee arthritis. To understand this discrepancy, we conducted qualitative research to assess the illness perceptions and preferred accommodations and coping strategies of patients with advanced osteoarthritis who had expressed a preference to avoid TJA.


In-depth interviews were conducted with a community sample of 29 men and women who were medically assessed as appropriate candidates for TJA but who had expressed a preference to avoid surgery. Inductive content analysis of text data was used to examine how patients' illness perceptions and preferred coping strategies related to surgery preference.


Participants frequently rejected the medicalization of arthritis, normalizing the experience of functional decline and defining it as age normative. Participants drew on a broad set of previous experiences with informal and formal care to make decisions about how to manage their condition. Previous negative encounters in medical and surgical care, including those from a distant past or those experienced vicariously, combined with the perception (reinforced by physicians and others) that doing nothing was a viable option deterred arthritis-related help seeking in the health care system.


Individuals with arthritis may benefit from additional counseling regarding effective medical and surgical treatments. Physicians may better meet patient needs by gauging patient preferences for a combination of self-management strategies and medical interventions.


Arthritis of the hip and knee is a major cause of long-term disability (1, 2). When medical therapy fails to control pain or physical functioning reaches an unacceptable level, total joint arthroplasty (TJA) of the hip or knee is indicated (3, 4). Although TJA results in substantial reduction in pain and disability (3–7), studies indicate that candidates for surgery can be reluctant to consider TJA. For example, in a US study, many African Americans with advanced knee osteoarthritis (OA) reported being unwilling to have surgery, reflecting a belief that it would not significantly improve their current health (8). In another Canadian study, among potential candidates for TJA, only one-third were willing to consider TJA as a treatment option (9, 10). Key factors found to be associated with willingness to consider TJA include expectations of pain relief and improved walking, restoration of quality of life (11, 12), perceptions of arthritis severity and candidacy for TJA, perceived acceptability of TJA risks, opinions of others (9, 13), and fear of surgery (8, 14). Willingness for surgery has been found to vary by age and sex. For example, among TJA candidates ages 80 years and older, women were less likely to agree to TJA than men (15). Some investigators have interpreted sex differences as reflecting the priority women give to their care-taking roles and their concerns that surgery would interfere with these roles, as well as women's poorer expectations of surgery outcomes and greater fear of postsurgical pain (16).

These findings support decision-making and behavior change theories that emphasize the importance of expectations of treatment outcomes, external resources, and perceptions of others in health decisions (17–22). However, there is a need for research investigating the interrelationships among potential explanatory variables, such as outcome expectations, previous experience, and the role of others, among older adults. For example, in one study, investigators examined the adaptations and coping strategies that potential candidates for TJA use in lieu of surgery to try to reduce pain and disability (14). The authors reported that among African Americans in the US, unwillingness to undergo TJA was related to a preference for natural remedies and reliance on religious belief for strength and guidance (14). To further an understanding of the treatment decisions surrounding TJA, we conducted a qualitative study of the illness perceptions and preferred coping strategies among individuals unwilling to consider TJA, despite potential eligibility. The findings have the potential to help clinicians address the concerns of patients unwilling to undergo surgery by illuminating their preferences and alternative coping strategies in greater detail.


Participants were drawn from an ongoing multiphase community cohort study of older adults with OA (9, 10). From 372 individuals meeting the criteria for arthroplasty but identified as definitely or probably unwilling to consider TJA, we selected a convenience sample of 30 individuals stratified by urban/rural region (an audio recording malfunction prevented the inclusion of 1 interview in the analysis). The smaller convenience sample allowed researchers to generate detailed information on the perceptions and behaviors of older adults unwilling to consider TJA and to lay the groundwork for the refinement of models of decision making and selection of variables for future quantitative research (23, 24). As shown in Table 1, the qualitative sample was similar to the larger subcohort from which it was drawn. Some exceptions were that the qualitative study participants were less likely to report annual incomes <$20,000, had somewhat lower arthritis disability scores, and had a greater number of comorbidities than members of the larger unwilling subcohort.

Table 1. Sample characteristics of qualitative study participants and a cohort of participants not considering total joint arthroplasty*
 Qualitative study participants (n = 29)Cohort of participants not considering surgery (n = 372)
  • *

    Values are the percentage unless otherwise indicated. OA = osteoarthritis; RA = rheumatoid arthritis; WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index.

  • The WOMAC scale (39) scores for pain, stiffness, and difficulties with function were transformed into a summary score that ranged from 0 (no pain or disability) to 100 (the most severe pain and disability).

Age, mean (range) years67 (59–86)69 (55–100)
Female sex6973.5
Education level  
 Primary or less3833
 Some/completed secondary4550
 Some/completed postsecondary1717
Rural region5551
Knee arthritis62.037.0
Hip arthritis1711.0
Both knee and hip arthritis2152.0
RA and OA1711
Mean Arthritis Severity Score (WOMAC)46.4551.62
Comorbidities, no.  

Potential participants were contacted by telephone and asked to participate in an interview focused on assessing the illness perceptions and preferred accommodations and coping strategies of patients with advanced arthritis who had indicated an unwillingness to consider TJA. Unwillingness for surgery was an inclusion criterion. (Un)willingness had been established in the cohort population up to 18 months prior to the initiation of the qualitative study and, on invitation to participate in the qualitative study, was reconfirmed by individuals' agreement with the statement, “You have been identified as someone who has significant hip and/or knee problems, who understands the risks and benefits of joint replacement surgery, but who is not willing to consider joint replacement surgery.” The study received ethics approval from the University of Toronto. Written and verbal consent to participate was obtained from all participants.

The interview guide was developed by the first author (PJB), a health sociologist with expertise in qualitative methods, in collaboration with the cohort study team (headed by GAH). Interviews were audio recorded and transcribed verbatim. Interviews lasted from 1 to 2 hours and were conducted by the first author and a second trained, experienced interviewer. Interview questions served as a guide aimed at encouraging participants to discuss issues relevant to the impact and management of their arthritis. Topics included perceptions of general health, health treatment histories, daily activities, support resources, role demands, and current and expected future adaptations and accommodations to arthritis (Table 2). The interview guide ensured that consistent topics were covered across interviews. In keeping with qualitative interviewing methods, participants' own priorities and concerns determined the depth, emphasis, and order of topics in each interview (25).

Table 2. Semistructured interview guide
1. Current status: age, medical status (other health problems and current treatments), marital status, living arrangements, employment status/main occupation, level of disability or impairment.
2. Background: arthritis history: since when? Courses of action or treatment and effectiveness? History of other health conditions and treatments.
3. Personal response to arthritis: What does/did arthritis mean to you/to your life? Self-treatment (i.e., diet, exercise, health care products); perceived effects (positive and negative, unknown); source of information about treatments.
4. Life changes in response to arthritis: When? What effect? (Such as moving, employment changes or accommodations, purchasing services, and affordability of these.)
5. Informal network: What are daily routines? Describe negotiation of major daily roles in household; any conflict or tension in carrying these out? (Telephone contact with others?)
 Probe: Who is in network; ages of network members (spouse, family, friends); what kind of support provided; outcomes of support provided; reciprocity in exchange, and what does it involve? Issues of control, negotiation, conflict in support network? Does or has conflict emerged because of arthritis disability and how is/was this managed?
6. Formal resources available and used (institutional help and assistance); previous experiences with arthritis experts and outcomes (negative and positive); who has been identified as potential carers and have these sources been pursued? Is treatment available when needed (waiting period)? Have you used alternative treatments?
7. Work history, and any effect of arthritis on employment. Effect on income security?
8. What support would be helpful to you at this time?
9. What accommodations are you willing to make around this problem, if things get worse? Have any plans been made? (i.e., plan on joint replacement surgery?)

Interview transcripts were analyzed using inductive content analysis (26, 27). This entailed repeated close readings of the transcripts to 1) identify the major issues that emerged and establish a coding framework, 2) generate a complete list of themes and subthemes, and 3) review and synthesize the themes to illuminate patterns and relationships among them. Text data were coded by PJB with face and content validity, and reliability of codes was further assessed by the second interviewer and an additional reviewer (GAH). Ethnograph software was used for data management (28).


Four overarching themes emerged from the interviews: the assessment of arthritis in the context of comorbid conditions, personal meaning and identity changes related to arthritis, the impact of others and socialization on illness, and experiences with the health care system as influencing treatment decision making.

Assessment of arthritis severity in the context of comorbid conditions

In all cases, participants' descriptions of their functional limitations and pain were consistent with medical assessments of advanced disease. However, for most participants, treatment of arthritis was not considered in isolation, but was spontaneously discussed in relation to treatment for other conditions such as diabetes, heart disease, and chronic back pain. In general, although arthritis was perceived as debilitating, it was not the primary health concern in participants' lives, with decisions for treatment being weighed against the overall perceived benefit of the outcome. This was illustrated by one woman, age 62 years, who commented: “After an accident in 1975, I had 3 operations on my left ankle… . So, I have very, very limited movement in my left ankle. I have degeneration in the spine … I had my back x-rayed a while ago and the doctor says that the discs are very bad… . It's hitting the sciatic nerve, so my leg from here to below my knee is numb … they're talking about maybe getting my knee replaced but I still wouldn't be able to walk, so why go through the pain of having the knee done and still not be able to walk?” Discussions among this respondent and others pointed to a hierarchy of health impairments and concerns. Treatment of arthritis was not the highest priority, and was sometimes seen as contraindicated when other conditions were taken into account. Additional examples of the comparison of arthritis treatment with that of other conditions are outlined in Table 3.

Table 3. Hierarchy of health impairments and concerns
Respondent: “I was taking some, ahm, high blood pressure pills that my body slowly developed a problem with … ah, the … result was the pills built up in me and I didn't realize it, and it was leaving me totally fatigued. I was just so tired all the time, and then I'd get severe periods of fatigue. And 2 days after I stopped taking these blasted pills, which I never wanted to take, I felt better. I'm on a different kind now, but they seem to be all right.”
Interviewer: “Has your doctor suggested you have a hip replacement?”
Respondent: “He's a new doctor and I haven't really talked–he's aware of it and, ah, I haven't gotten into the arthritis with him very much because I was dealing with this other problem.”
Interviewer: “I'm asking you all these questions because I'm trying to find out what is the underlying feeling [about total joint arthroplasty]. Do you feel your health is … as good as or worse or … ?”
Respondent: “I always think it's going to get better. Isn't that dumb? But, ahm, I feel better than I did last year. But that was mainly because of those pills… .” (woman, age 68 years)
Respondent: “I have all degenerated disks all the way down my back and ah, arthritis has set in my lower back, my hip, both knees and my left foot.”
Interviewer: “Now what courses of action or treatment have you attempted … ?”
Respondent: “Well, my MD sent me to an orthopedic specialist in X Hospital. He put me through all kinds of tests, and ahm, did the whole business with the idea that perhaps some sort of surgery would ease the pain on my lower back and therefore ease the pain in my legs. But the report he sent to my doctor, ah, indicated that there were too many spots that were … he'd have to operate on too many spots to be logical. So they quoted it as being inoperable. And that's when my doctor, put me on these pain killers. So now I just–I have to go along with it, that's all I can do. I expect, I really truly expect to some day be in a wheelchair unless maybe I get knees or something but we discussed it, ahm, like I say, I'm a pretty heavy fellow, and unless I lose a lot of weight, I don't think a knee would be practical for me. It's too much weight on it.” (man, age 71 years)
Respondent: “I had a heart attack about a year ago. And then 3 months ago I had some more pains, I was in for another 3, 4 days. Since I came home from the hospital … I started walking every day, and losing weight … what happened was … my heart doctor was away and this new guy come in and … and he went like this (grabs stomach) and he said, ‘What’s this? What's this?' And he embarrassed me. So when I come home I thought well I got to get rid of some of that. So I cut back. I like to eat and that. I tried to cut back and tried to do some exercise to get rid of it, and I did, I got rid of maybe 15 pounds … and I'd like to get another 15 …”
Interviewer: “And in terms of arthritis, what's happening? You've had significant knee and both hips and both knees and some in your hands and wrists …”
Respondent: “I feel good in that too. I used to just have somebody like I had a knife in my hips. But now, I mean I feel it, it aches in there a little bit, but not like before. Right now I envision myself as lucky that I'm as in as good health as I am. That I got over everything, I got over my heart attack and I don't feel any different now than I did before I had a heart attack … as far as walking I just can't believe how … how this walking–I mean, if walking does it; before I just felt static, I couldn't walk, but not now … I mean everything's loosening up. I used to go round the block, and I'd stop 3 or 4 times. But … like I left this morning and I went a kilometer.” (man, age 65 years).

Personal meaning and identity changes related to arthritis

Similar to other research, arthritis was described as affecting not only physical functioning, but also the patient's sense of self (29), particularly as a result of changes in participation in valued roles and relationships. Three different perspectives on self-concept emerged: arthritis as nonproblematic to identity or as age normative, arthritis as compromising identity, and arthritis as an invasive feature of a “spoiled” identity (30). Each perspective influenced the ways in which treatment was perceived. The different perspectives are illustrated with quotes in Table 4.

Table 4. Differences in personal meaning and orientation to arthritis and arthritis disability
1. Arthritis is nonproblematic to identity; arthritis is age-normative:
 “As I say, I've never really pushed the fact that I've got the arthritis because it was just basic well you've got it so you live with it. You know, it's the way we've thought about it, that you've got it, there's not a lot can be done about it, so you basically live with it.” (woman, age 65 years)
Respondent: “You know, I'm going to be 80 in August so I can be thankful I had this much mileage.”
Interviewer: “Do you have any other illnesses or health problems besides the arthritis?”
Respondent: (Chuckles) “At this age you have all kinds of problems… .” (woman, age 79 years)
2. Arthritis compromises identity:
“What seems to be the problem is if you're feeling well, you think–you forget you've got this, you know. It's like you go out and you overdo and then you come home and, it's very painful.” (woman, age 65 years)
3. Arthritis and a “spoiled” identity:
“He (husband) has to cut my food; he has to walk me to the washroom, back out here again; it's so horrible–especially someone who was always used to doing for myself. You know, I've been working since I was 12; and to have to depend on somebody so totally, it's really horrible. I'd like to be able to get up and kick my heels up and you know, just do all the things other people do.” (woman, age 71 years)

In several cases, arthritis was viewed as an inconvenience but did not challenge participants' identities. As the first quote in Table 4 illustrates, “you live with it.” Throughout her interview, this 65-year-old woman reported carrying out important roles, despite limitations and pain. Her treatment of arthritis included the occasional use of acetaminophen and alternative medications, which she considered sufficient given that arthritis was just part of life. She was aware of others who had experienced varying degrees of success with TJA and was cautious about the procedure: “I think it works all right, and eventually I may have to have it done. I don't want to undergo surgery until I have to. With any kind of surgery you're running a risk, and I just don't particularly like surgery. I don't like being put to sleep.”

Other participants viewed arthritis as a normal and expected aspect of aging. They modified social roles and activities so that they were “age appropriate.” One consequence of this perspective was that these participants preferred to leave well enough alone and manage without TJA: “I've had pretty good use of my limbs, and I'm almost 80, I'd rather leave that for younger people that have years ahead of them” (woman, age 79 years).

Not all participants had integrated arthritis into their self-concept. Some respondents discussed arthritis as challenging their sense of self. For them, the disease was separate and foreign from their identities (see Table 4). Many participants attempted to carry out normal activities and ignored the disease as much as possible until it intruded into their lives. On these occasions, participants responded with treatment that was focused on short-term pain management and curbing current activities. Long-term responses to arthritis were avoided. As one respondent noted, despite being able to do only approximately one-third of what she wanted, she wasn't “bad enough” to inquire about surgery:

Interviewer: “Would you initiate a conversation about this (TJA)?”

Respondent: “Oh sure. Yeah, sure, I would if I felt it was that, you know, if it was that bad. I wouldn't feel that at this point… .”

Interviewer: “How bad does it have to be in terms of your own experience?”

Respondent: “Well, I guess if you're immobilized …” (woman, age 65 years).

The comments of most participants fell into 1 of the 2 groups described above. However, one 71-year-old participant described a total loss of her identity and self as independent and autonomous. Her arthritis and its resulting incapacity had become a pervasive feature of her identity (see Table 4). She described multiple comorbidities and an inability to cope with her physical limitations. Throughout her interview, she made frequent reference to her past self-identity and roles, and lamented the losses she had incurred. She believed that her overall health was too poor for TJA. Her spouse (and caregiver, who attended her interview) stated: “Her hip certainly isn't, ah, bad enough to have hip surgery, that's for sure … and she doesn't want it because it could kill her, because she does have other problems.” Instead of surgery, the couple focused on her daily caregiving needs and growing physical and psychological dependence. Their interview reflected a collective learned helplessness, the burden of caregiving, and the belief that nothing could be done.

In summary, advanced OA sometimes challenged or threatened the self-identities of older adults. Despite these challenges, none of the participants sought TJA as a means to regain a threatened identity. Instead, participants minimized or normalized their situation by discussing modifications to activities in terms of what they believed were age-normative expectations, they ignored their disease and tried to carry on as normal despite experiencing exacerbated symptoms, or they became overwhelmed by health and activity changes and felt powerless to change their situation.

Impact of others and socialization on illness

Informal social networks made up of family, friends, and neighbors were critical to participants' management and coping. Marital relationships were particularly important and decisions were often made on the basis of the couple's capacity to cope rather than the individual's capacity, reinforcing the idea found in other research that health professionals may need to consider the couple as the patient when considering disease management options (31–33). The excerpt below reflects this as a 68-year-old man with arthritis whose wife also had arthritis discusses household activities:

Interviewer: “How do you manage the household?”

Respondent: “We do it between the two of us. If we don't do it today it gets done tomorrow or some other day. Don't worry too much about it.”

Interviewer: “Who does the groceries, or how do you do it?”

Respondent: “Both of us. We go: we go with the Wheel-Trans [i.e., publicly subsidized transportation service for persons with disability], that way she doesn't have to do much walking and I have this wheelchair anyway.”

In general, family, friends, and close others contributed to the decision not to have TJA in 4 ways: they provided instrumental support (i.e., help with tasks) that minimized the effect of the patients' physical limitations, reinforcing patients' beliefs that they did not need surgery; they provided information about possible strategies of care such as home support services or alternative health care; they provided emotional support and helped keep the participants socially involved and connected to others despite physical limitations, reinforcing the idea that surgery is avoidable; and they provided standards and norms that participants drew upon to justify their decision not to have surgery. The first 3 types of support have been discussed at length in the literature and have been labeled instrumental, informational, and emotional/affective support (34, 35).

The fourth type of support reflects findings in other OA research (9, 10, 13); however, the specific ways that significant persons in a social network provide norms for responding to illness are illustrated in our qualitative study. These norms appear to influence individuals' coping with arthritis and their decisions to forgo surgery. Both current and historically distant relationships influenced respondents' expectations of the appropriate illness-related decisions and behaviors, as illustrated in Table 5. In cognitive psychology, the concept of availability bias (36), referring to the influence of a selective reference group of others, is useful for describing this phenomenon. The excerpts in Table 5 suggest that individuals with arthritis learn about appropriate responses to their illness vicariously, or through the experience of significant others, through socialization and social learning (37, 38) that occurs over their lifetime.

Table 5. Use of social norms to justify decisions regarding total joint arthroplasty
“My mother had arthritis, too. But she wouldn't let it get her down either. My mother was the most marvelous knitter and crocheter. And … she made the most beautiful pastry, but her hands were twisted with arthritis … and ah, she still continued until the last 2 years of her life.” (woman, age 62 years)
“I lost my first son to leukemia 35 years ago last March. He'd be 42 now. He was quite a man. He was 7 years old when he died, but my god, that child went through more agony in one year than most human beings go through in their lifetime. That's why I don't give in. A lot of people wouldn't be able to, but I could.” (woman, age 62 years)
Interviewer: “Have you thought of what kind of, what services you may need in this area and enquired into what is available?”
Respondent: “You know, I come from the old school. When you're this old, you grew up with nothing and you learn to do with nothing, and you learn to look after yourself, you know, as long as you could anyway, so I guess that's what we do. We just … I don't know … I'm only going from one daughter-in-law, the slightest thing … she goes and has it fixed, eh. And I'm not saying it's not right, but she's sure using the medical business and she's good and healthy and all that now, maybe that's good.” (woman, age 76 years)

Experiences with the health care system as influencing arthritis treatment decision making

In addition to their informal support system, participants also spontaneously made reference to their past experiences with hospitals and formal health care. For some, previous nonarthritis-related surgical experiences, either their own or others' experiences, created a fear and mistrust of surgery that contributed to the avoidance of TJA, as illustrated in the excerpt below:

Respondent: “I was told in 1971 that… . [Dr. X] told me I'd be in a wheelchair in 10 years, guaranteed. I fooled him.”

Interviewer: “Was he recommending something you didn't like?”

Respondent: “Well, he was recommending that operation on my spine. Well, my sister in Barrie, she had an operation on her back … oh, 15 years ago maybe. And it never worked out. And XX had an operation on his back that never worked out either. And he had a second operation. It—he had a steel plate in the bottom of his spine 2 and a half inches long and he's still sore (unclear). I guess … you tolerate it but it never worked out to a point where he said he was satisfied with it. This is one of the reasons I'm reluctant about having any surgery … like the 3 operations the other ones have had just never worked, so I don't want to take a chance and ending up worse” (man, age 64 years).

Others noted that previous experiences with physicians, particularly around prescribing medications, had undermined their trust in their physicians and often left them believing that their interests came second. This, too, contributed to decisions concerning TJA. One 64-year-old man commented: “I don't like going to a doctor for anything. I don't like … I don't like to take a pill if I don't have to, I don't even take an aspirin… . I've had a few, ah, you know, instances with doctors that didn't work out and to avoid something if you can is the way I look at it; because of the … downfalls I've had in the past.” Another 68-year-old man stated: “Well, I haven't seen the doctor or anyone for years about my arthritis because I've learned to—he is aware that I have it, but like, I'm not one for taking a lot of pills or anything like this … and I just learned they can't do anything for me.”

A second theme that emerged in recounting experiences with health professionals was related to the timing of TJA discussions. Several participants noted that their family physician had never discussed surgery with them. Because physicians were recognized as the experts in treatment, participants assumed not only that surgery was not possible, but also that doing nothing was a viable option. A 71-year-old woman said: “… I don't think she [the doctor] thinks there's much we can do about it now. I think she thinks I'm 71, maybe she thinks I'm too old and maybe it's dangerous to have an operation.” Another woman, age 61 years, stated that her physician had not suggested surgery: “… I follow along pretty well what they've [the doctors] suggested. And as I say, I function so. … I figure they know what they're doing.”

In instances where health professionals had mentioned surgery, it was often described as a last resort. This left many participants wanting to try all other alternatives before TJA. One 64-year-old man reported, “if I get to the point where I can't get around no more, if I can't go for groceries, I can't … you know, walk in the store. If I got to that point I probably would have to. But right now I wouldn't consider it.” A 68-year-old man recounted an incident that took place more than 15 years earlier: “One doctor told me, ‘You’ll be on prednisone the rest of your life.' He gave me a prescription for that, he says, ‘I don’t want to see you now until you need joints replaced.'” One effect of health professionals' portrayal of surgery as a final resort was that it gave license to participants to infer a meaning that may or may not have been intended by health professionals, specifically, that surgery was something to be avoided.


We used qualitative methods to explore the ways in which unwilling candidates for TJA perceived and managed arthritis. Previous studies have found factors associated with willingness for TJA to include future expectations of benefit, fears about surgery, and support from others. The present study adds to this research by highlighting that, among individuals unwilling to have TJA, past negative experiences with surgery, unwanted treatment (especially medications), and health professionals' advice or apathy toward surgery were discussed to a greater extent than future outcome expectations or worries. This finding suggests that there may be qualitative differences in the importance of factors related to decisions among those willing and unwilling to have surgery. Specifically, individuals unwilling to have surgery may focus more on their past to reinforce decisions than on expectations of benefits in the future. To more comprehensively assess factors related to decision making, future studies should assess in detail previous experiences, as well as future expectations.

A second finding of interest was related to participants' current self-perceptions and how these were used to support decisions regarding TJA. Previous studies have found that older adults often normalize OA, seeing it as a part of aging. The present study also found support for that perception and illustrated how, by normalizing their disease, participants minimized its role in their lives and justified avoidance of surgery. Other ways in which participants minimized the importance of OA and its treatment included comparing OA with other conditions assessed as more serious, distancing one's self from the disease, and promising one's self to consider surgery when health problems became “really bad.” This latter state was often so extreme it was unlikely to ever occur. The results underscore the benefits of continued education of patients regarding the impact and consequences of OA and potential efficacy of early treatment.

Other individuals reinforced participants' decisions to avoid TJA. In many cases, other individuals influenced participants' decisions by providing instrumental help with activities that were difficult to perform, thereby negating the apparent need for surgery, and by actively supporting participants' decisions to avoid surgery. More insidious and difficult for clinicians to deal with are participants' interpretations of discussions with health professionals. When physicians omitted TJA as a course of treatment or described it as a last resort, it was often interpreted as expert opinion of an absence of need for surgery. Participants also drew on past experiences of others to support decisions. These experiences were offered as norms or standards for care despite their potential divergence from current knowledge and practices. The scenario that emerges from these efforts is one where participants actively construct a current reality out of disparate sources and amounts of information. These findings, pointing to the complexity of information that patients draw upon to make decisions, require further investigation.

There are several limitations to this study. Our sample was small and nonrandom; the experiences of this group may not reflect those of all persons unwilling to consider TJA. For example, the fact that the qualitative study participants had somewhat better (lower) arthritis disability scores and a higher proportion had multiple comorbidities compared with members of the larger unwilling subcohort may indicate that for this particular group, surgery is deferred because priority is given to other conditions. Therefore, although the qualitative sample was diverse and generally reflected the characteristics of participants from the larger community cohort, replications of this study would ascertain the stability of our findings, and perhaps additional qualitative aspects of unwillingness for TJA would emerge. Longitudinal research would also allow greater insight into individuals' decision making and how it relates to disease, personal, and environmental factors. This study included only persons unwilling to undergo TJA; the perceptions, experiences, and coping strategies of this subgroup may differ substantially from those who were willing to undergo surgery. Future research exploring potential differences between the 2 groups may point to important implications for clinical treatment of patients with OA. Despite its limitations, this study illuminates the perceptions and behaviors of older adults unwilling to consider TJA and points to directions for future research to refine models of decision making and select variables for future research.


Dr. Ballantyne had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Drs. Ballantyne and Hawker.

Acquisition of data. Drs. Ballantyne and Hawker.

Analysis and interpretation of data. Drs. Ballantyne, Gignac, and Hawker.

Manuscript preparation. Drs. Ballantyne, Gignac, and Hawker.

Statistical analysis. Drs. Ballantyne and Hawker.


We thank the 2 anonymous reviewers for their insightful comments and suggestions for the article. Detelina Radoeva assisted with interviews and with preliminary coding of the text data. Angela Wall provided detailed descriptors of the subcohort data.