1 Introduction

Race, ethnicity, and social determinants of health (SDoH) have been strongly linked to health and patient outcomes and are important in achieving equitable access and care in patients with total hip (THA) and total knee arthroplasty (TKA). Hartnett et al. demonstrated that non-White patients and patients from areas of higher social deprivation were associated with significantly less THA utilization, underscoring these factors' influence on accessibility.¹ Similarly, a meta-analysis by Rudisill et al. showed that racial minority and socioeconomically deprived patients experience higher rates of 30-day complications and healthcare utilization when undergoing THA or TKA.² Such disparities emphasize the complex interplay of clinical, socioeconomic, and systemic factors that must be considered in contextualizing research surrounding hip and knee total joint arthroplasty (TJA).

Randomized control trials (RCTs) are considered the “gold standard” of medical research, with access to participation playing a key role in ensuring health equity. Results of an RCT with an inadequately represented cohort limit the confidence of researchers, clinicians, and policymakers to apply a study's findings to all patients.³ Various system-wide attempts have been made to improve such inequities in medical literature. The National Institutes of Health (NIH) Revitalization Act of 1993 mandated that all clinical trials receiving federal funding exhibit analysis of results according to demographics such as race, ethnicity, and gender.^4,5 In 2010, the Consolidated Standards of Reporting Trials (CONSORT) guidelines were introduced and recommended in medical literature to ensure the transparent reporting of clinical trials.⁴ The CONSORT guidelines strongly recommend reporting all pertinent demographic and clinical information, whether as a baseline table or in the initial description of a study population.^6,7 Despite these efforts, there is still underrepresentation in randomized control trials with minimal consideration of SDoH. In the past two decades, less than half of US clinical trials have reported data on race or ethnicity.⁸ Similarly, a study of 189 RCTs in high-impact sports medicine journals found that less than 10% reported on race, ethnicity, employment, insurance, and education.⁹ As the clinical utility of the published data depends on its generalizability across specific patient cohorts, greater emphasis must be placed on reporting demographics and SDoH in RCTs.¹⁰

Previous studies have indicated the low prevalence of reporting SDoH and demographic factors in orthopaedic RCTs. In 2014, Somerson et al. reported that of the 158 orthopaedic RCTs, only 20.3% accounted for race or ethnicity.³ A more recent study in arthroplasty revealed that only five of 72 RCTs included income, insurance, or education in their demographic tables.¹¹ Griffin et al. reported that RCTs published in the Journal of Arthroplasty between 2015 and 2019 provided little information regarding race and ethnicity and were reported at the highest frequency in 2019.⁶ However, these studies reviewed a limited cohort of journals, and no trend updates in arthroplasty RCTs have been reported since 2019. A broad-spectrum updated analysis of demographic and SDoH reporting in RCTs will accurately reflect the trends in transparent reporting and underline areas of improvement to ensure the publication and support of equitable gold-standard arthroplasty research. Therefore, this systematic review aims to investigate the prevalence of reporting and analysis of patient demographics and SDoH in hip and knee total joint arthroplasty-related RCTs published between 2019 and 2024.

2 Methods

2.1

2.1 Manuscript screening

The researchers searched PubMed to target all PubMed-indexed journals. Fig. 1 demonstrates the search term and study design used. Prospective total knee and/or hip arthroplasty were terms of interest. A language filter of English was applied. Articles included in this study were limited to publication dates from October 31st, 2019, to October 31st, 2024, to encompass randomized control trials published over the past five years. The search yielded 1612 publications. Duplicates were removed, resulting in 1597 articles. Screening of the abstract was conducted to determine if they met initial objective inclusion criteria, which consisted of English language RCTs pertaining only to THA or TKA. Articles not meeting these criteria were excluded and recorded with their corresponding criteria for exclusion. A total of 432 publications met the inclusion criteria.

Close

Fig. 1 Flowchart with inclusion criteria for systematic review from search term to full-text screening.

Export to PPT

These publications were further analyzed to confirm the availability of the full text and inclusion of Table 1 and/or an overview of patient demographics. Articles missing the full text or articles that were protocols of the study design without results were excluded. from further analysis. Questions or disagreements about including a study were discussed among the four reviewers. The final 391 manuscripts were included for data extraction. Our systematic review does not meet the inclusion criteria for the International Prospective Register of Systematic Reviews (PROSPERO) registration as they do not accept methodological reviews that only assess the quality of reporting. This systematic review was exempt from review by the Institutional Review Board.

Table 1 Summaries of demographics and social determinant variable inclusion (n = 391).

	ReportedN, (%)	References
Race	7 (1.8)	Haider et al., 202412
		Haffar et al., 202213
		Mishra et al., 202114
		Visperas et al., 202115
		Buttacavoli et al., 202016
		Tang et al., 202217
		Pizzi et al., 202018
Ethnicity	2 (0.5)	Mishra et al., 202114
		Buttacavoli et al., 202016
Insurance	1 (0.3)	Lu et al., 202123
Income	1 (0.3)	Zhang et al., 202022
Education	6 (1.5)	Visperas et al., 202115
		Nguyen et al., 202219
		Power et al., 202220
		Chen et al., 202021
		20 Zhang et al., 202022
		Lu et al., 202123
Employment	2 (0.5)	Visperas et al., 202115
		Lu et al., 202123
Any excluding gender	12	Haider et al., 202412
		Haffar et al., 202213
		Mishra et al., 202114
		Visperas et al., 202115
		Buttacavoli et al., 202016
		Tang et al., 202217
		Pizzi et al., 202018
		Nguyen et al., 202219
		Power et al., 202220
		Chen et al., 202021
		20 Zhang et al., 202022
		Lu et al., 202123

3 Results

A total of 382 trials (97.7%) reported gender in their study. In contrast, 12 (3.1%) considered race, ethnicity, insurance, income, education, and employment in their research (Table 1). Seven (1.8%) papers reported race ^12–18. However, none were discussed in the discussion portion of the manuscript. Categories of race reported included White/Caucasian and Black/African American, Asian, American Indian/Alaskan Native, and Native Hawaiian/other Pacific Islanders. Four (14%) papers included other race categories (Table 2). Two studies reported Hispanic or non-Hispanic ethnicity.^14,16 The second most reported social determinant variable was education (n = 6; 1.5%) ^15,19–23. All manuscripts reporting social determinant variables included them in the demographics table.

No significant differences were found when comparing the articles that reported a social determinant variable based on journal type, publication year, and type of surgery (Table 3). No more than four (0–5.5%) of randomized control trials reported social determinant variables in a year-long period. Three of the studies were published in the Journal of Arthroplasty (Impact Factor (IF) = 3.4), and three studies were published in the following non-arthroplasty orthopaedic journals: Orthopaedic Nursing (IF = 0.7), Orthopedics (IF = 1.1), and Journal of Orthopaedic Surgery and Research (IF = 2.8). No differences were found in randomized control trials focusing on THA, TKA, or both THA and TKA.

Among the trials reporting social determinant variables, the most common interventions in question were related to postoperative rehabilitation management (n = 5), analgesia/anesthesia protocol (n = 3), preoperative information/education, and postoperative intervention (n = 2 each; Fig. 2. Commonly measured outcomes were pain and analgesia consumption (n = 6; 34%), functional outcome (n = 4; 22%), and other (n = 4; 22%) (Fig. 3). Outcomes categorized as ‘other’ included patient recall, sleep patterns, kinesophobia, and postoperative deep vein thrombosis (DVT)/pulmonary embolism (PE).

Close

Fig. 2 Exposures/interventions of manuscripts that report social determinant variables.

Export to PPT

Close

Fig. 3 Outcomes measured of manuscripts that report social determinants.

Export to PPT

Of the total RCTs reviewed, 25 (6.4%) were multi-center RCTs. Guidelines for quality reporting in RCTs, the majority of which were Consolidated Standards of Reporting Trials (CONSORT) guidelines, were mentioned in 196 (50.1%) of the manuscripts. Of the 12 RCTs reporting race, ethnicity, or SDoH variables, two (16.7%) were multi-center studies (P = 0.175). Seven (58.3%) of 12 RCTs explicitly mentioned the quality reporting guidelines in their methodology (P = 0.771). No statistical difference in study quality was demonstrated between the RCTs that reported SDoH and those that did not (Table 4).

4 Discussion

Our study found that less than two percent of all randomized control trials related to total hip and knee arthroplasty between 2019 and 2024 reported race and/or ethnicity. Insurance, income, education, and/or employment were minimally reported. No significant differences in reporting patterns were found across journal types or year of publication. These findings point out the paucity of reporting on social determinants of health in RCTs being conducted in patients who undergo hip and knee total joint arthroplasty. The impact of SDoH on orthopaedic surgery outcomes has been widely investigated over recent years. Race, ethnicity, neighborhood deprivation, insurance, income, and education play a role in patient satisfaction, complications, and healthcare utilization after TJA ^1,2,24–27. In one study, patients with Medicaid were found to undergo TJA at lower-volume hospitals and had higher rates of DVT, PE, and periprosthetic joint infection.²⁵ An RCT by Butler et al. determined that ethnicity, education, poverty, and income were more influential of dissatisfaction and hip score after THA than implant factors.²⁸ As race, ethnicity, and SdoH strongly impact TJA outcomes, their inclusion in RCTs is of high importance.

Despite their influence, this study reports a low volume and quality reporting of such variables. While 12 studies (3.1%) were found to report either racial, ethnic, or SDoH demographics, none discussed the relevance of these variables in the manuscript. The minimal reporting concurs with previous literature across orthopaedic subspecialties.^3,9,11,29,30 A systematic review of TJA RCTs published in four journals between 2017 and 2019 by Donnelly et al. found that only 4.2% and 1.4% of studies included race and ethnicity, respectively, in their demographic table. Only six (1.5%) studies included income, insurance, and/or education in their tables.¹¹ Comparing the results of our study to that by Griffin et al., who reviewed RCTs published in the Journal of Arthroplasty, our study reported race (6 vs. 1.8%) and ethnicity less frequently (4 vs. 0.5%). Griffin et al. did not investigate the prevalence of reporting socioeconomic factors.⁶ Two key differences between our study and the aforementioned studies should be highlighted: the study period and the selection of journals reviewed. In contrast to Donnelly et al., who reported findings based on RCTs between 2017 and 2019, and Griffin et al.‘s four-year analysis between 2015 and 2019, our study investigated the recent trends between 2019 and 2024. In addition, we expanded our search to include journals outside of the four reviewed by these studies to provide a more comprehensive analysis. Despite expanding our search and investigating more recently published literature between 2019 and 2024, our study reports a lower prevalence of race, ethnicity, and SDoH variables in RCTs compared to previous systematic reviews that investigate time periods up to 2019. This could be attributed to the greater scope of our study, which includes all journals publishing on RCTs related to TJA. This may suggest that non-major and/or non-orthopaedic journals fail to emphasize the importance of reporting these variables. Nevertheless, this study supports evidence from previous observations to confirm that reporting these essential variables remains critically insufficient.

Of the RCTs that reported race, most failed to report specified races other than White/Caucasian and Black/African American. Similarly, a systematic review by Paul et al. recorded four of 35 orthopaedic RCTs to have included at least five specified races in their report.²⁹ This contrasts the guidelines by the NIH and the Food and Drug Administration, which recommends RCT reporting on racial and ethnic minority prevalence with a minimum of five specified races to improve the quality and standardization of reporting.^5,31 Patients of different races and ethnicities may have differences in genomic type and cultural and community norms that influence the efficacies, utilization rates, and risks of interventions. There is an abundance of arthroplasty literature that has demonstrated associations between race and rates of postoperative complications and healthcare utilization ^32–37. As such, the clear and specific categories of race are needed in identifying such differences. It is important to note, however, that race alone does not explain these associations. A study by Hadad et al., found that confounding factors such as poverty, education, and access to care plays a greater part in driving these disparities.³⁸ Given the strong influence of social determinants on arthroplasty access, utilization, and outcomes, addressing the shortfall in reporting not just race and ethnicity but also SDoH in these RCTs is imperative.

The unchanged reporting patterns over the five years of publication were striking. Our findings are consistent with those of Donnelly et al., who also noted no difference in SDoH variable reporting in TJA RCTs over three years between 2017 and 2019.¹¹ Without reporting these factors, there is a major limitation in generalizability of the study results and implement interventions for all patients. The lack of generalizability of these RCTs also reflects the disparities observed in various studies and the unchanging inequity trends in orthopaedic surgery over time ^34,39–42. Studies focused on neighborhood deprivation demonstrated its association with prolonged length of stay, discharge disposition, and readmission after TJA.^41,42 While RCTs are the gold standard of research, they can only be interpreted in the context of the patient population studied. Optimizing protocols and interventions based on RCTs that did not provide transparency on how a patient's race, ethnicity, income, insurance, and education impacted their research are likely to introduce bias, thereby marginalizing underrepresented patient populations and hindering equitable care. Study sponsors and journal guidelines should require the inclusion of demographic differences to ensure a more diverse population being studied. Factors related to race, ethnicity, and SDoH should be included in future RCTs to improve the quality of arthroplasty research and orthopaedic care to patients of all backgrounds.

There are several potential theories for the lack of reporting on SDoH variables in RCTs. The National Institute of Health encourages reporting racial and ethnic identities but acknowledges that trials may not have sufficient statistical power for subgroup comparisons. This could also be the case for other SDoH variables. Participants may not volunteer SDoH information. As we included all RCTs regardless of the country where the study was conducted, the perceived importance of SDoH variables may vary. Discussions surrounding SDoH are often linked to health equity in the United States. Researchers in other countries may sense income and insurance status to have a less direct impact on health outcomes than in the United States, where healthcare access is more stratified. Some countries may have relatively homogeneous populations in the context of socioeconomic or cultural norms, which may also diminish the impact of SDoH on outcomes. There may be a paucity of studies linking some of the outcomes in RCTs with SDoH variables, which may prompt investigators to defer reporting these factors. In light of the various hypotheses, we encourage authors of RCTs to be transparent in reporting how homogenous or heterogenous the patient cohorts are regarding SDoH factors to allow readers the opportunity to interpret results appropriately and determine whether they are applicable. Investigators should survey publications for relevant health disparities and SDoH based on specific study questions and base the decision to include or exclude SDoH variables on relevance, necessary statistical power, and the demographic context in which trials are conducted. Goals for representation should be defined at the beginning of the study period. If goals are met and there is sufficient power, we recommend a sub-cohort analysis of the results in the context of different SDoH variables to assess for any differences across cohorts. In cases where the goals fail to be met, this should be discussed as a limitation point of the study.

This study has several potential limitations. Firstly, this study was limited to RCTs published within a five-year timeframe. However, we believe that this is enough time period to represent the most current trends in SDoH reporting and provides a longer timeframe than previously published studies on this topic. Secondly, the study does not account for the various structural systems across countries in accessing care for total hip and knee arthroplasty that may influence SDoH, such as universal healthcare. The SDoH factors analyzed in this study were not exhaustive; however, we believe we have included the most prevalent ones based on prior systematic reviews. Our study included all orthopaedic and non-orthopaedic PubMed-indexed journals that fit our inclusion criteria. Our research findings highlight no difference in the reporting of SDoH variables between orthopaedic and non-orthopaedic journals, indicating that these deficiencies are not limited to major orthopaedic journals and call for further investigation across different medical journals. Finally, there is a lack of standardized categorization of social determinant variables across the studies that reported them, making it difficult to make conclusions about each variable (Appendix B).

5 Conclusion

In conclusion, our study highlights a critical gap in reporting race, ethnicity, and social determinants of health in randomized controlled trials related to total hip and knee arthroplasty over the past five years. The unchanged minimal reporting of these variables limits the generalizability of RCT results and may perpetuate health inequities by failing to account for the influences of social determinants on arthroplasty outcomes. Given their impact on developing interventions and health policy, we underscore the importance of including and reporting these variables in RCTs.

Informed consent

N/A.

Authors contribution

All authors contributed to the study design. Michelle Shimizu: Conceptualization, Methodology, Investigation, Writing – Original Draft, Writing – Review & Editing; Muhammad Hamza Ilyas: Conceptualization, Investigation, Formal Analysis; Isaiah Freeman: Investigation, Data Curation, Writing – Review & Editing; Hahn Kang: Data Curation, Formal Analysis; William Sampson: Data Curation, Formal Analysis; Carlo Mannina: Investigation, Data Curation; Young-Min Kwon: Resources, Supervision, Project Administration. All authors have reviewed and approved the final article.

Institutional ethical committee approval

This study was exempt from Institutional Review Board approval.

Informed consent (patient/guardian)

Mandatory only for case reports/clinical images and not applicable to this research.

Funding/sponsorship

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.