Are There Racial and Ethnic Variations in Patient Attitudes Toward Hip and Knee Arthroplasty for Osteoarthritis? A Systematic Review

Abstract

Background

It is not clear why people who identify as Black or Hispanic are less likely to undergo discretionary musculoskeletal surgery such as arthroplasty for osteoarthritis of the hip or knee. Inequities and mistrust are important factors to consider. The role of socioeconomic factors and variation in values, attitudes, and beliefs regarding discretionary procedures are less well understood. A systematic review of the evidence regarding mindsets toward knee and hip arthroplasty among Black and Hispanic people could inform attempts to limit disparities in care.

Questions/purposes

In a systematic review of qualitative and quantitative evidence, we asked: (1) What factors are associated with racial and ethnic variations in attitudes toward discretionary hip and knee arthroplasty for osteoarthritis? (2) Do studies that investigate racial and ethnic variations in mindsets toward discretionary orthopaedic care control for potential confounding by socioeconomic factors?

Methods

A systematic search of PubMed, Cochrane, and Embase (last searched August 2023) for studies that addressed racial and ethnic variations in mindsets toward discretionary musculoskeletal care use was conducted. We excluded studies that were not published in English, lacked full-text availability, and those that documented patient approaches without comparing them to the willingness to undergo a discretionary procedure. Twenty-one studies were included—14 quantitative and seven qualitative—including 8472 patients. The Mixed Methods Appraisal Tool was used for quality assessment of included studies. The studies included demonstrated low risk of bias: five quantitative studies lacked detail regarding nonresponse bias and one qualitative study lacked details regarding the racial and ethnic composition of its cohort. To answer our first research question, we categorized themes associated with racial differences in mindsets toward discretionary care and recorded the presence of associations in quantitative studies. To answer our second question, we identified whether quantitative studies address potential confounding with socioeconomic factors. There were no randomized trials, so no meta-analysis was performed.

Results

In general, self-identified Black and Hispanic patients had a lower preference for hip and knee arthroplasty than self-identified White patients. Black patients were more likely to regard osteoarthritis as a natural and irremediable part of aging and prefer home remedies. Both Black and Hispanic patients valued support from religion and were relatively cost-conscious. Black and Hispanic patients had lower perception of benefit, were less familiar with the procedure, had higher levels of fear regarding surgery and recovery, and had more-limited trust in care. Generally, Black and Hispanic social networks tended to address these concerns, whereas White social networks were more likely to discuss the benefits of surgery. Thirteen of 14 quantitative studies considered and accounted for potential confounding socioeconomic variables in their analyses.

Conclusion

The observation that lower preference for discretionary arthroplasty among Black and Hispanic patients is independent from socioeconomic factors and is related to accommodation of aging, preference for agency (home remedies), greater consideration of costs, recovery concerns, and potential harms directs orthopaedic surgeons to find ways to balance equitable access to specialty care and discretionary surgery while avoiding undermining effective accommodation strategies. It is important not to assume that lower use of discretionary surgery represents poorer care or is a surrogate marker for discrimination.

Level of Evidence

Level III, therapeutic study.

Introduction

A systematic review of 82 studies found that patients who identify as Black race or Hispanic ethnicity had lower use of total joint arthroplasty, worse preoperative capability, and more major complications and readmissions than White patients did [2]. Disparities in healthcare in the United States are associated with income, insurance status, race and ethnicity, and place of residency [4, 12, 31]. These inequities can also vary according to clinician behavior, system factors, and personal factors such as trust in the healthcare system, geographic location, lack of transportation, and social support [8, 27]. For nondiscretionary care (care essential to maintaining health), these disparities can result in worse health [31].

For discretionary care, such as arthroplasty for osteoarthritis of the hip or knee, there is no “ideal” amount of care, and many patients choose to accommodate symptoms. The observed differences in arthroplasty by self-identified race and ethnicity might be due to variations in access. However, they might also be a consequence of variations in trust, beliefs, preferences, and expectations [2]. It is also important to consider economic disparities, given that socioeconomic disadvantage is not independent from other types of societal disadvantage, nor entities called “race” or “ethnicity.” Variations in care based on distrust, limited access, affordability, or limited awareness of the available health benefits of arthroplasty merit correction. Preferences for accommodating the natural aging process and concerns regarding recovery from surgery and potential adverse events are appropriate and potentially advantageous. The relative weight of societal disadvantages that should be eliminated compared with aspects of effective accommodation that might be worthy of emulation is an important topic of inquiry.

We therefore systematically reviewed studies that assessed the attitudes of patients who identify as Black race or Hispanic ethnicity toward musculoskeletal specialty care. We asked: (1) What factors are associated with racial and ethnic variations in attitudes toward discretionary hip and knee arthroplasty for osteoarthritis? (2) Do studies that investigate racial and ethnic variations in mindsets toward discretionary orthopaedic care control for potential confounding by socioeconomic factors?

Materials and Methods

This mixed-methods, systematic review was performed following the Preferred Reporting Items for Systematic Review and Meta-Analyses guidelines. We searched PROSPERO to identify any recent or unpublished systematic reviews on this topic. The study protocol was registered with PROSPERO (CRD42023413761).

Eligibility Criteria

Peer-reviewed studies of variations in mindsets—defined as attitudes and beliefs—regarding discretionary orthopaedic surgery by self-identified race and ethnicity were included. We excluded studies not published in English and those lacking full-text availability. Qualitative and quantitative studies were analyzed separately.

Information Sources and Search Strategy

We searched three databases (PubMed, Embase, and Cochrane) from July 2023 to August 2023 using a comprehensive list of search terms (Supplemental Digital Content 1; http://links.lww.com/CORR/B280).

Selection Process

The selection process was facilitated using the Rayyan internet-based application designed for systematic reviews (Qatar Computing Research Institute). Two independent authors (YA and JH) reviewed articles from the search for inclusion. The search also included identifying additional studies from the reference and related articles list of each study. The senior author (DR) reviewed any discrepancies or disputes, although no issues arose that necessitated arbitration.

Data Collection Process

For the quantitative studies, measures of attitudes and behavior of specific racial or ethnic groups were identified and grouped by theme: willingness to consider, expectations of arthroplasty, fear of arthroplasty, and surgeon offer of knee arthroplasty.

For the qualitative studies, the two reviewers noted areas of variation in attitudes toward treatment of osteoarthritis of the knee and hip by race or ethnicity. Similar themes were consolidated. The analysis of categories underwent further discussion and refinement in collaboration with the senior author (DR). Descriptive quotes representing specific themes were collected where available (Supplemental Digital Content 2; http://links.lww.com/CORR/B280).

Data Items: Primary and Secondary Study Outcomes

Our primary outcome was differences in perspectives, opinions, sentiments, and attitudes regarding treatment of osteoarthritis of the hip and knee and discretionary arthroplasty by self-identified race and ethnicity. Our secondary outcome was whether quantitative studies accounted for confounding socioeconomic factors.

Risk of Bias and Study Quality

Two researchers (YA and JH) independently assessed the quality of the studies using the validated Mixed Methods Appraisal Tool (MMAT). This tool assesses the methodologic quality of various study designs, including quantitative, qualitative, and mixed-methods studies. The MMAT uses a three-tier response system: “Yes” indicates that the criterion is met, “no” indicates that it is not met, and “can't tell” implies insufficient information in the paper for judgment [16, 26]. Five quantitative studies lacked detail regarding nonresponse bias and one qualitative study lacked details regarding the racial and ethnic composition of its cohort (Supplemental Digital Content 3; http://links.lww.com/CORR/B280). Otherwise, the risk of bias was low.

Study Selection

The search yielded 10,551 articles, which included 10,544 articles from the database searches and seven additional articles after review of references and related articles. After removing 4038 duplicates, we used the Rayyan system to filter articles for relevance. After applying the filtering criteria, 6026 articles were excluded because they did not include a racial or ethnic descriptor (such as White, Black, or Hispanic) in combination with a disease of aging (such as osteoarthritis) or a discretionary orthopaedic procedure. This left 487 articles that were screened independently by two researchers (YA and JH) based on title and abstract. Upon examination of the abstracts, 411 articles were excluded because they did not address mindsets regarding discretionary surgery. After reviewing the full text of 76 articles, we excluded quantitative studies that cataloged the approaches used by Black and Hispanic patients to accommodate osteoarthritis but did not investigate a potential preference for these strategies over arthroplasty. A combined 55 qualitative and quantitative studies that cataloged accommodative approaches and did not address mindsets related to discretionary care were excluded. Twenty-one articles were analyzed (Fig. 1).

Fig. 1.

This PRISMA flowchart shows the systematic, three-database search and inclusion strategy.

Study Characteristics

The included articles comprised seven qualitative studies and 14 quantitative studies with a total of 8472 patients. Eight quantitative studies (two addressing people with hip and knee symptoms [1, 19] and six evaluating knee symptoms alone [5, 6, 11, 23, 28, 30]) measured willingness to consider total joint arthroplasty and factors associated with willingness. Of these, one conjoint analysis (a market research method that has been used in clinical research to better understand patient preferences in different aspects of healthcare services [6]) and one willingness-to-pay analysis (a market research method that assesses the maximum amount of money individuals are willing to spend on a particular product or service [5]) were included. Five quantitative studies investigated expectations of arthroplasty [3, 6, 13, 17, 18], one studied the fear of arthroplasty [24], and one evaluated surgeon recommendation of arthroplasty [15]—all in the context of knee osteoarthritis (Table 1).

Table 1.

Quantitative studies addressing the preferences of people with hip and knee osteoarthritis

Study	Measures	Joint	Study design	Participants	White	Black	Hispanic	Mean age in years	Women
Ibrahim et al. [19]	Willingness to consider	Hip and knee	Cross-sectional	596	334	262		65	0%
Allen et al. [1]	Willingness to consider	Hip and knee	Cross-sectional	1522	1049	473		68	67%
Figaro et al. [11]	Willingness to consider	Knee	Cross-sectional	94		94		71	89%
Kwoh et al. [23]	Willingness to consider	Knee	Cross-sectional	799	514	285		63	64%
Suarez-Almazor et al. [28]	Willingness to consider	Knee	Cross-sectional	198	66	66	66	64	63%
Vina et al. [30]	Willingness to consider	Knee	Cross-sectional	799	514	285		63	64%
Byrne et al. [6]	Willingness to consider, expectations of arthroplasty	Knee	Cross-sectional	391	130	131	130	55	58%
Byrne et al. [5]	Willingness to consider	Knee	Cross-sectional	193	64	65	64	48	53%
Ang et al. [3]	Expectations of arthroplasty	Hip and knee	Cross-sectional	691	428	263		64	41%
Groeneveld et al. [13]	Expectations of arthroplasty	Hip and knee	Cross-sectional	909	459	450		60	0%
Ibrahim et al. [17]	Expectations of arthroplasty	Hip and knee	Cross-sectional	593	332	261		66	0%
Ibrahim et al. [18]	Expectations of arthroplasty	Hip and knee	Cross-sectional	596	334	262		66	0%
Lavernia et al. [24]	Fear of arthroplasty	Hip and knee	Cross-sectional	128	109	19		65	66%
Hausmann et al. [15]	Surgeon recommendation of arthroplasty	Hip and knee	Cross-sectional	457	337	120		NA	5%

NA = not available.

In the qualitative studies, eight themes were found, including social networks [7, 10, 21, 22, 25], spirituality [10, 14, 20, 21, 25], fear [10, 21, 25], candidacy for surgery [10, 25], accommodation of arthritis [10, 21], self-efficacy [22], finances [7, 22], and trust [7, 10, 22] (Table 2).

Table 2.

Qualitative study characteristics and identified themes among people with hip and knee osteoarthritis

Study	Joint	Design	Coding method	Themes identified	Subjects	Mean age in years	Women	White	Black	Hispanic	Race or ethnicity not specified
Chang et al. [7]	Knee	Focus groups	Thematic content analysis	Social networks, finances, trust	37	60	68%	20	17
Katz et al. [21]	Knee	Focus groups	Grounded theory	Spirituality, fear, accommodation, social networks	39	60	90%	14		25
Kroll et al. [22]	Knee	Focus groups	Grounded theory	Social networks, finances, trust, self-efficacy	37	64	62%
Figaro et al. [10]	Knee	Structured field interviews	Grounded theory	Spirituality, fear, accommodation, social networks, trust, candidacy for surgery	94	71	84%		94
Parks et al. [25]	Hip and knee	Interviews	Grounded theory	Spirituality, fear, social networks, candidacy for surgery	36	68	81%		28	6	2
Ibrahim et al. [20]	Hip and knee	Focus groups	Thematic content analysis	Spirituality	75	62	72%		75
Harvey and Silverman [14]	Hip and knee	Interviews	Thematic content analysis	Spirituality	88	76	47%	41	47

Results

Racial and Ethnic Variations in Attitudes About Arthroplasty

Across the studies reviewed, factors associated with a preference to accommodate osteoarthritis included regard for osteoarthritis as a natural and irremediable part of aging among Black patients (two qualitative studies [10, 21]), self-assessment of candidacy for surgery based on comorbidities (two qualitative studies [10, 25]), religious faith as support for self-management and acceptance among both Hispanic (one qualitative study [21]) and Black patients (four qualitative studies [10, 14, 20, 25]), and a preference for home remedies among Black patients (one qualitative study [10]). One qualitative study found that when discussing the causes of their illness, Black patients focused on internal causes (such as aging), whereas Hispanic and White patients were more prone to discuss external causes (such as attributing their osteoarthritis to past accidents or injuries) [22].

The conjoint analysis (a market research method that assesses the maximum amount of money individuals are willing to spend on a particular product/service) found that Black patients were less likely than Hispanic and non-Hispanic White patients to choose TKA [6]. In a study that found Black patients were offered TKA less frequently than White patients regardless of osteoarthritis severity, the only factor associated with surgeon recommendations was patient preference [15].

In the willingness-to-pay analysis, Black and Hispanic patients were willing to pay less for arthroplasty, and Black patients were willing to pay the least [5]. One qualitative study found that Black patients were more likely than Hispanic and White patients to mention financial concerns when discussing arthroplasty [22], but financial security irrespective of race and ethnicity was not considered. A second qualitative study found that White women, Black men, and Black women were more likely to have financial concerns than White men [7], and again, financial security in general was not considered.

Quantitative associations with lower willingness to consider arthroplasty included lower perception of benefit in four studies [11, 19, 23, 28] and more limited familiarity with the procedure and trust in care in one study [28]. Five quantitative studies found that patients who are Black were less likely than White patients to perceive beneficial outcomes of arthroplasty [3, 6, 13, 17, 18]. Three qualitative studies found that trust in the medical system or in physicians were concerns of Black patients, Hispanic patients, and White women [7, 10, 22].

One study found that Black patients had higher levels of fear of surgery perioperatively than White patients [24]. Three qualitative studies found that fears about anesthesia (intubation, “going to sleep,” and death), fear of pain, fear of medical fitness, and fear of unfavorable surgical outcomes were more common among Black and Hispanic patients [10, 21, 25]. In the quantitative studies, Black patients expected more problems such as longer hospital courses, difficulty walking, and greater pain [3, 6, 13, 17, 18].

Three qualitative studies found that Black and Hispanic patients tended to rely on social networks to consider timeframes for comfort and capability during recovery, whereas social networks among White patients were more likely to discuss the benefits of surgery [7, 10, 25]. Three studies also highlighted the influence of social networks on fear and anxiety regarding the procedure and expectations for recovery in Black and Hispanic patients [10, 21, 25]. Additionally, one study found that social networks positively influenced Black and White participants’ trust in physicians who were recommended by their social circles, but not for Hispanic participants [22].

Confounding Variables to Race and Ethnicity in the Evidence Base

Thirteen of 14 quantitative studies considered and accounted for potential confounding socioeconomic variables in their analyses (Supplemental Digital Content 4; http://links.lww.com/CORR/B280). The sole quantitative study—associated with the theme of fear of arthroplasty—that did not explicitly mention controlling for confounding factors [24] yielded results consistent with the findings of the qualitative studies, suggesting reliable findings.

Discussion

There are disparities in the use of discretionary orthopaedic surgical care such as hip and knee arthroplasty for osteoarthritis by race and ethnicity [9, 29, 32]. What is unclear is the degree to which these differences arise from factors worthy of correction such as distrust, limited access, affordability, or limited awareness of the available health benefits of arthroplasty. We should also be curious about the degree to which the observed differences in use of discretionary tests and treatments arise from factors that should be encouraged and emulated, such as acceptance and accommodation of aging, agency (home remedies), and community and spiritual support. In the current systematic review of quantitative and qualitative studies of variation in attitudes toward discretionary knee and hip arthroplasty by race and ethnicity, we found that both greater adaptiveness and greater wariness accounted for lower preferences for discretionary arthroplasty among Black and Hispanic patients, and these differences were not accounted for by socioeconomic variations, although general financial security and other factors were not well addressed in the qualitative studies. The observation that disparities can signal effective health strategies such as accommodation of aging, preference for agency (home remedies), and greater consideration of costs, recovery concerns, and potential harms reminds us not to assume that lower use of discretionary surgery represents poorer care or that it is a surrogate marker for discrimination.

Limitations

First, although we used comprehensive search strategies and adhered closely to the PRISMA guidelines, it is possible that some relevant studies were not identified. This should have little or no impact on the findings because the themes were fairly consistent across studies. Second, the quantitative observations in the qualitative studies are somewhat atypical for that type of study and may need corroboration with quantitative studies. A relative prevalence of themes identified in qualitative research suggests hypotheses that can be tested in quantitative research. The relative concordance between the qualitative and quantitative studies in this systematic review suggests that such studies would likely corroborate the findings to date. Third, the long period during which these studies were conducted may not accurately capture the current perspectives of specific racial and ethnic groups. Many of the identified factors may represent relatively static cultural factors, and we think that the studies mostly identified factors that will be consistent for at least a generation. Fourth, the qualitative studies did not always identify details about concepts such as “fear,” and there is room for interpretation about what people were thinking based on the quotes that were provided. It seems sufficient for this study that differences by race were observed, and these can be further explored in new studies. Fifth, attitudes toward aging and discretionary treatment of age-related pathophysiology are complex, as are concepts of race and ethnicity and confounding factors such as socioeconomic status, gender, or other factors. Although most of the quantitative studies attempted to account for socioeconomic factors, one consideration for future studies is when self-identified race and ethnicity are the interesting factors to study, and we might be better off measuring thoughts and behaviors that are adaptive to age-related pathophysiology independent of social constructions related to race and ethnicity.

Discussion of Key Findings

The findings from both quantitative and qualitative studies that preferences and tactics for acceptance and accommodation of aging, more cautious regard for and fear of discretionary surgery, and differences in trust may account for variations in discretionary arthroplasty by self-identified race and ethnicity, controlling for socioeconomic factors, reminds us that variations can indicate helpful adaptations as well as unjustified variations in access. The degree to which the observed variations in use of, and attitudes toward, discretionary arthroplasty reflect positive attributes worth reinforcing or negative attributes that arise because of racism is unresolved. The themes that arose in the studies we reviewed seemed to reflect wisdom, critical thinking, and caution regarding surgeries with notable potential for harm and disappointment as well as resiliency and accommodation of aging. Variations in use by race and ethnicity may be due to remediable levels of mistrust and unawareness of treatment options. However, variations also seem to reflect accurate and valid judgments of the difficulties of recovery and the imperfections and potential harms of a major surgery. Because surgical treatment of osteoarthritis is discretionary, the identified variations in attitudes toward surgery may or may not be a problem to be solved. Surgeons may have a bias that greater use of discretionary surgery is desirable. If so, we might consider reorienting those sentiments given their potential to undermine effective health strategies such as accommodation of sensations associated with musculoskeletal senescence (programmed biological deterioration with age).

Conclusion

The observation that lower preference for discretionary arthroplasty among Black and Hispanic patients is independent from socioeconomic factors and is related to accommodation of aging, preference for agency, and greater wariness of costs, recovery concerns, and potential harms suggests that disparities in discretionary surgery can arise from positive and negative factors. Because of limited patient interest in discretionary surgery, we should not assume there are disparities in the use of arthroplasty based on socially constructed racial and ethnic categories. On the contrary, health strategies may be more effective to the degree they are informed by cultural variation in effective accommodative strategies. Additional research is merited to ensure that variations in care by social strata represent choices consistent with what matters most to patients (their values) rather than to disadvantages they may face. In daily practice, orthopaedic surgeons can find ways to balance equitable access to specialty care and discretionary surgery while avoiding undermining effective accommodation strategies.