Abstract
Objective. Total hip replacement (THA) surgery is a successful procedure, yet blacks in the United States undergo THA less often and reflect poorer outcomes than whites. The purpose of this study is to systematically review the literature on health-related quality of life after THA, comparing blacks and whites.
Methods. A librarian-assisted search was performed in Medline through PubMed, Embase, and Cochrane Library on February 27, 2017. Original cohort studies examining pain, function, and satisfaction in blacks and whites 1 year after elective THA were included. Using the Patient/Population–Intervention–Comparison/Comparator–Outcome (PICO) process format, our population of interest was US black adults, our intervention was elective THA, our comparator was white adults, and our outcomes of interest were pain, function, and satisfaction after elective THA. The protocol was registered under the PROSPERO international register, and the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines were followed.
Results. Of the articles, 4739 were screened by title, 180 by abstract, 25 by full text, and 4 remained for analysis. The studies represented 1588 THA patients, of whom 240 (15%) were black. All studies noted more pain and worse function for blacks; although differences were statistically significant, they were not clinically significant. One study sought and identified less satisfaction for blacks after THA, and 1 study showed worse fear and anxiety scores in blacks.
Conclusion. When measured, there are small differences in THA outcomes between blacks and whites, but most studies do not analyze/collect race. Future studies should address the effect of race and socioeconomic factors on healthcare disparities.
Total hip replacement (THA) surgery is one of the most successful procedures and has been referred to as the “operation of the century,” in light of its predictable results and limited morbidity1,2. Use of THA has continued to rise for the treatment of pain and to restore function caused by endstage arthritis3. Nonetheless, despite an equivalent prevalence of hip osteoarthritis (OA) in blacks and whites, lower use of THA for blacks in the United States has persisted over decades4,5.
The reason for the disparities in THA use is unclear. Physician bias in recommendations for surgery does not seem to play a major role, because black and white patients are equally likely to be referred for arthroplasty, and when differences in recommendations are noted, they appear to be determined by patient preference6,7. Expectations for arthroplasty outcomes are lower for blacks, and blacks prefer nontraditional interventions such as prayer or massage over arthroplasty, possibly contributing to the disparity between whites and blacks in THA use8,9. Complications, including risk of revision, are higher in blacks undergoing total knee arthroplasty (TKA)10, and while noninfectious- and infection–related 30-day complications are higher after TKA for blacks, no difference in complications was observed after THA11. However, blacks are more likely to receive arthroplasty in low-volume hospitals12,13 where the outcomes for common procedures such as arthroplasty are not as consistent as the outcomes in high-volume hospitals14,15. For blacks undergoing TKA, poorer health-related quality of life (HRQOL) outcomes have been reported, although literature on the difference in outcomes was sparse16,17. This suggests that poorer outcomes for blacks may in part mediate their decreased arthroplasty use. The purpose of our study was to systematically review the literature to compare pain, function, and satisfaction for US blacks versus whites undergoing THA.
MATERIALS AND METHODS
We performed a systematic literature review using guidelines contained in the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement to determine whether there are differences in pain, function, or satisfaction after THA between US blacks and whites. We registered the systematic review protocol under PROSPERO international register (www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD42017064320). As per institutional policy for research without human subjects or their unique information, this study did not require institutional review board approval and otherwise followed appropriate ethical standards. Using the Patient/Population–Intervention–Comparison/Comparator–Outcome (PICO) format, our Population of interest was US black adults, our Intervention was THA, our Comparator was white adults, and our Outcomes of interest were pain, function, and satisfaction after elective THA measured by validated instruments. We restricted our search to US cohorts given the differences in access to healthcare and differences in the involvement of race in different countries; we also restricted our search to studies published after 2000 to reflect recent protocols in surgery. We excluded studies that specifically selected patients with inflammatory arthritis, revisions, fractures, hemophilia, or studied a specific surgical technique or implant materials. We excluded expert-opinion review articles, systematic reviews, abstracts, duplicate articles, case series, and case reports. The minimal acceptable followup postoperative was 1 year, and our search was limited to humans and English-language literature. The included papers were assessed for quality and sources of bias using the GRADE system (Grading of Recommendations Assessment, Development, and Evaluation).
Search strategy
A librarian-assisted search was performed in Medline through PubMed, Embase, and Cochrane Library (including Cochrane Central Register of Controlled Trials, Health Technology Assessment Database, Cochrane Methodology Register and NHS Economic Evaluation Database) on February 27, 2017. We also reviewed journals not in these databases that focus on healthcare disparities by hand-searching the individual journals (grey literature). The PubMed search strategy is detailed in Table 1.
Electronic search strategy.
A sensitive search strategy was applied to identify all relevant data related to race. The included studies used instruments validated for use after THA, for example, The Western Ontario and McMaster Universities Arthritis Index (WOMAC), Hip Injury and Osteoarthritis Outcome Score, Oxford Hip Score, and Harris Hip Score (HHS). The screening was performed by 2 authors (SG, BM) who independently screened 4739 articles by titles and abstracts, and identified articles for further review. If there was a conflict, the 2 reviewers discussed the paper and reached an agreement. We confirmed with the same author of 3 of the studies that they were performed in separate cohorts of patients.
RESULTS
We identified 6991 articles (3372 Medline through PubMed + 554 Cochrane + 3065 Embase). Duplicates were removed, leaving 4739 articles screened by title. One hundred eighty papers were selected for abstract review. We excluded 63 abstracts that reported on non-US cohorts, 45 that failed to analyze/report race, 21 that were on a wrong topic/procedure, 17 that reported outcomes not of interest, and 9 that were review articles. The remaining 25 studies underwent full-text review. Twenty-one studies were eliminated because of inadequate analysis of the outcomes of interest by race or wrong outcome (Figure 1). Four studies were included in the final analysis after consensus of the 2 reviewing authors. Two studies were prospective in design and 2 were retrospective analysis of prospectively acquired data. The studies included 1588 THA patients, including 240 (15%) blacks (Table 2)18,19,20,21.
Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2009 flow diagram.
Included studies.
Lavernia, et al18 performed a prospective study in which 331 patients with endstage OA underwent primary or revision hip or knee arthroplasty between October 2000 to March 2002. All had patient data collected at baseline and were followed for a minimum of 3 years (average 5 yrs; range 3–8 yrs). All surgeries were performed by 3 surgeons and used a cemented technique. Regardless of time, blacks scored worse than whites for all measures except for the Medical Outcomes Study Short Form-36 physical function and general health scores. Blacks had worse WOMAC scores (scale 1–100, higher worse) for physical function, pain, and total scores. Total scores of blacks were higher (44.31 ± 2.06) compared to whites (34.51 ± 0.87). Using the Quality of Well-being index as well as the Pain Anxiety Symptom Scale (PASS) in the study, blacks presented at the time of surgery with a higher fear score (p = 0.002), cognitive subscale, and total PASS score (p = 0.04) compared with whites.
Allen Butler, et al19 published a prospective randomized study that enrolled 102 patients who were randomized to receive 1 of 2 different implants. Detailed patient data were collected preoperatively and included race, diagnosis, age, sex, insurance status, medical comorbidities, tobacco and alcohol use, household income, educational level, and history of treatment for lumbar spine pathology. Detailed baseline and followup data at (minimally) 2 years included HHS, Medical Outcomes Study Short Form-12, WOMAC, pain drawing, and UCLA activity rating and satisfaction questionnaire. They found no difference in outcome between the 2 implant types, but found that blacks, those with less education, those with income under the poverty level, and those with poorer baseline hip scores were at higher risk for a poor HHS (< 90), greater thigh pain on a pain drawing, and a visual analog scale (VAS) ≥ 3 at 2 years.
Lavernia, et al20 performed a retrospective review of prospectively acquired data of 1749 patients receiving total joint arthroplasty (739 hips and 1010 knees) between 1992 and 2007. The data were analyzed to determine the influence of race and ethnicity on well-being, pain, and function 2–16 years after total joint arthroplasty was performed. Blacks consisted of 32 (10.8%) of the non-Hispanic men and 54 (12.2%) of the non-Hispanic women undergoing THA, and among the black Hispanics undergoing THA, 15 (5.1%) were men and 17 (3.8%) were women. Blacks presented with worse pain and poorer function across multiple measures, and the disparity in outcomes persisted for WOMAC pain and stiffness after surgery. Blacks had worse WOMAC pain scores (mean range 2.55–3.79) compared to whites (mean range 0.92–1.16) on a 4-point scale. This study collected extensive quality-of-life data, but did not include education or income in the analysis.
Lavernia and Villa21 performed another retrospective review of prospectively collected data of 2142 primary THA and TKA (1665 patients) performed by the senior author (CJL) in a single institution from May 2003 to November 2012. The cases included 105 blacks, 46 of whom underwent THA. Only 39/105 blacks (37%) had 1-year followup. Patients undergoing TKA and THA were analyzed together. More blacks than whites underwent surgery for avascular necrosis (15% vs 3%) and inflammatory arthritis (7% vs 3%). Blacks were younger than whites at the time of surgery (mean age 63 ± 13.2 vs 70 ± 9.9 yrs), but there was no difference in the Charlson Comorbidity Index or the American Society of Anesthesiologists score between groups. Small differences were noted in preoperative VAS pain (8 ± 1.8 vs 8 ± 2.0; mean difference 0.76, SD 0.34–1.1; p < 0.001) and preoperative WOMAC function (42 ± 13 vs 38 ± 12; mean difference 4.9, SD 2.2–7.5; p < 0.001) between black and white patients, but these differences, while statistically significant, are not clinically significant. Postoperative differences were noted in pain, WOMAC, and overall well-being and the differences were again statistically but not clinically significant. The higher proportion of blacks diagnosed with inflammatory arthritis and avascular necrosis may have also skewed the results, and no information regarding education or income was included in the analysis.
Synthesis of results
All 4 studies included in this systematic literature review noted more pain and poorer scores for function for blacks after THA, and although differences were statistically significant, they were not clinically significant. One study sought and identified less satisfaction for blacks after THA, and 1 study showed worse fear and anxiety scores in blacks. The data could not be metaanalyzed because of the heterogeneity in outcome measures. There were no randomized controlled trials, and all observational studies were considered low to moderate quality.
DISCUSSION
Our systematic review aimed to compare quality-of-life outcomes of pain, function, and satisfaction after THA between blacks and whites in the United States after 2000. However, of 4739 publications identified, only 4 contained adequate information to evaluate racial disparities in THA outcomes in the United States. These 4 studies reported on a total of 1588 THA, and included data on 240 blacks (15%), describing greater pain and poorer function and less satisfaction after THA for US blacks. The largest differences in quality-of-life outcomes were in the study by Allen Butler, et al19 that included a high proportion of patients living in poverty, and identified race, income below poverty, and less education as significant risk factors for poor outcomes. It is striking that despite the documented persistence of racial disparities in the use of arthroplasty4, and the emphasis by the US Surgeon General and the Institute of Medicine on the importance of eliminating racial disparities in health, the study of racial disparities in THA outcomes has been quite limited22,23. Similar to our previous systematic literature review addressing differences in TKA outcomes between whites and blacks17, our current study highlights the omission of race and socioeconomic status in most studies of THA outcomes.
Overall, patients undergoing THA are more satisfied and show greater improvement than similar patients undergoing TKA24,25, yet there is no difference in the disparity in use of THA compared to TKA, indicating that disparities in arthroplasty use are multifactorial. Minority patients who prefer “culturally alike” physicians may have difficulty finding them because black and Hispanic physicians account for < 10% of all physicians in the United States26. Prior studies have identified poorer baseline pain and function as a risk factor for poor outcomes after THA; however, race and socioeconomic status have not been analyzed27,28. Previous studies analyzing race have revealed that black patients undergoing TKA have worse pain and function both at baseline and at followup17,29,30,31,32,33. However, the racial disparity in outcomes after TKA is seen only in neighborhoods with high poverty levels, demonstrating an interaction of race and poverty that contributes to healthcare outcomes16. The risk of TKA revision has also been linked to race. A recent metaanalysis revealed that race was a significant risk factor for TKA revision, although in studies using the Medicare (health insurance for the elderly) 5% sample, insurance status (Medicaid eligibility; health insurance for low-income people, a surrogate for socioeconomic status) was a confounder10. Because more blacks live in poverty than whites34, and poverty is associated with poorer health outcomes35, attribution of mediators of risk in healthcare outcomes can be difficult. While poorer outcomes after THA may contribute to disparities in THA use, the differences are small, and multiple other factors are likely to contribute to THA use.
Our conclusions are limited by the paucity of studies available for inclusion. An additional weakness to our analysis is that all 4 studies were performed in southern cities; 3 of the studies reflect a single surgeon’s experience in Miami, Florida, and the fourth study describes a small cohort collected in New Orleans, Louisiana, limiting the generalizability of this systematic review. In addition, 1 study combined THA and TKA in the final analysis.
Race and socioeconomic status are infrequently included in studies of THA outcomes. Understanding and addressing the reasons for persistent racial disparities in the use and outcomes of successful procedures such as THA will require inclusion of both race and other significant socioeconomic variables, such as education and poverty in studies of THA. Future studies should address these deficiencies so that the multidimensional factors leading to healthcare disparities can be addressed.
- Accepted for publication November 3, 2017.
