Race and ethnic disparities arthroplasty trends and hotspots: Bibliometric analysis

Abstract

Introduction

Despite continual advancements in total joint arthroplasty and perioperative optimization, there remains national variability in outcomes. These outcome variabilities have been in part attributed to racial and ethnic disparities in healthcare quality and access to care. This study aims to identify arthroplasty racial and ethnic disparities research and to predict future hotspots.

Methods

Ethnic and racial disparities articles between 1992 and 2022 were queried from the Web of Science Core Collection of Clarivate Analytics. Bibliometric indicators in excel format were extracted and subsequently imported for further analysis. Bibliometrix and VOSviewer analyzed current and previous research.

Results

Database search yielded 234 total articles assessing racial and ethnic disparities between 1992 and 2022. Twenty-six countries published manuscripts with the United States producing the majority of publications. The Veterans Health Administration and University of Pittsburgh were the most relevant institutions. Ibrahim SA was the most relevant and influential author within this field. Visuals of thematic map and co-occurrences identified the basic, motor, and niche themes within the literature.

Conclusions

Racial and ethnic disparity within arthroplasty literature demonstrate growing traction with global contributions. United States authors and institutions are the largest contributors within this field. This bibliometric analysis identified previous, current, and future trends for prediction of future hotspots.

1. Introduction

Total joint arthroplasty (TJA) of the hip (THA) and knee (TKA) are successful treatment modalities to improve pain and function in end stage arthritis of the respective joints. Despite the continuous improvement efforts through advances in technology and perioperative optimization, among others, there remains variability in outcomes at the national level, in part attributed to racial and ethnic disparities in quality and access to care.⁹ An early report in 1985 by the U.S. Department of Health and Human Service Secretary's Task Force raised awareness to health care disparities between different racial and ethnic groups and implications on public health.⁷ In lieu of growing awareness, there remains inconsistent care and quality of care reported among minority groups.²⁶ Despite support at the federal, state, and local level over two decades, minimal progress in closing the disparity gap has been achieved, with evidence of worsening inequality in some health indicators.¹⁶ Within THA and TKA, there remains racial and ethnic disparities within different aspects of the procedures. A recent analysis of the American Joint Replacement Registry noted lower likelihood of utilization of advanced technology in THA and TKA recipients of minority racial and ethnic backgrounds.^18,27

Similarly, African American (AA) patients are reported to be hesitant regarding the procedure due to perceived suboptimal postoperative rehabilitation course, functional status, and pain 11. Moreover, there are consistent studies describing an over-representation of inadequate pain management among racial and ethnic minorities.⁶ Chen et al. noted that AA patients were less likely to receive opiates for chronic pain control.³ The same trend in disparities has been reported in postoperative outcomes, with higher 30 day complications and readmission rates, prolonged length of stay, and mortality among AA patients and increased complications and nonhome discharge for Hispanic patients.²¹

The aim of this bibliometric analysis is to highlight the trends in arthroplasty literature in research assessing racial and ethnic disparities over a 30-year period. The key countries, institutions, authors, journals, and most influential articles will be analyzed. Current themes and keywords will be assessed to understand the transition and evolution of the trending research. The authors seek to identify future topics of interest for further development and shed light on major themes that might help in overcoming these disparities.

2. Methods

2.1. Sources of data and search strategy

Web of Science (WOS) Core Collection of Clarivate Analytics is the most frequently used scientific information engine for bibliometric analysis and was utilized in this study. The source outputs detailed data for bibliometric analysis. Science Citation Index Expanded and Social Science Citation index were used for literature search with the key words "hip" OR "knee" (Topic) AND "arthroplasty" OR "replacement" (Topic) AND "race" OR "ethnic*" OR "racial" (Topic) AND 1992–2022 (Year Published) AND Article (Document Type). The query within the database used articles as the document type and index of SCI-EXPANDED and SSCI within the years 1992–2022 in efforts to minimize omissions. The search key terms were highlighted and selected on the basis of published literature identifying trends within racial and ethnic disparities.

2.2. Data extraction

Two authors performed the data collection and extraction. After the initial screen of the database query, information regarding articles and all pertinent data required for the following analysis were collected which included: title, publication year, journal published, authors, institutions, countries, abstracts, references, citations, and impact factors.

2.3. Bibliometric analysis

Bibliometric indicators in excel format were extracted from the WOS database. The data was subsequently imported for bibliometric analysis. Apparent data deficiencies were crosschecked within the WOS database. Country categorization was obtained from information regarding various regions of interest. The visualization of data with knowledge maps of the bibliometric indicators specific to scientific production, co-authorship, co-citation, topic trend, thematic map, dual-map overlay, and thematic evolution was performed using VOSviewer (1.6.19.0) (Leiden University, Leiden, the Netherlands) and Bibliometrix (University of Naples Federico, Naples, Italy) software.

3. Results

3.1. Publication data

After review of the initial search output, the WOS database search yielded 234 articles assessing racial and ethnic disparities in arthroplasty between years 1992–2022. The number of references identified totaled 4979 with an average 27.78 citations per document. There is an increasing trend of annual growth within the literature pertaining to racial and ethnic disparities with a growth rate of 13.7 % per year. Global annual publication is demonstrated in Fig. 1. The publication output was largest in the years 2017 and 2021 with an absolute publication of 11 manuscripts, followed by 2013 and 2018 with 5 manuscripts.

Fig. 1.

Global annual scientific production with the greatest increase in 2017 and 2021.

3.2. Countries

Table 1 demonstrates worldwide research productivity. The included articles originated from a total of 26 different countries, with the US being the largest contributor with 76.82 % of the global output. The top 10 countries in productivity were 5 from Asia/Oceania, 3 From Europe, and 2 from North America, and ranked, after the US, as follows: China at 5.3 %, UK at 2.35 %, Canada at 1.97 %, and South Korea at 1.85 % of total publications. The US had the greatest total citations with 5740 references and ranked first in annual productivity as highlighted in Fig. 2.

Table 1.

Country Productivity in worldwide research in Racial and Ethnic Disparity arthroplasty.

Region	Freq
USA	623
CHINA	43
UK	19
CANADA	16
SOUTH KOREA	15
INDIA	13
SINGAPORE	13
FRANCE	9
JAPAN	9
SWEDEN	7
AUSTRALIA	6
NEW ZEALAND	6
MALAYSIA	5
SWITZERLAND	4
EGYPT	3
SOUTH AFRICA	3
BRAZIL	2
DENMARK	2
GERMANY	2
INDONESIA	2
NIGERIA	2
SAUDI ARABIA	2
SPAIN	2
IRAN	1
PHILIPPINES	1
SUDAN	1

Fig. 2.

Country trends with a) production over time and b) total citations.

3.3. Institutions/authors

A total of 369 institutions published at least an article regarding racial and ethnic disparities in TJA. The affiliation metric is a measure of all the contributing authors institutional affiliations. Fig. 3a demonstrates the top 5 institutions in productivity, which are all from the US with the Veterans Health Administration (VHA)/US Department of Veteran's Affairs (VA) contributing the most with 987 author affiliations, followed by 380 from University of Pittsburgh, University of Pennsylvania/Pennsylvania Medicine with 444 authors. The greatest annual production for the top institutions were University of Pennsylvania in 2017 with 7 articles each followed by VHA/VA in 2017 and 2013 with 6 and 5 articles respectively, and University of Pittsburg with 5 articles in 2005.

Fig. 3A).

Total affiliated author production within racial and ethnic disparities over time. 3b) Most relevant author contributions and 3c) total author citations. 3d) Lotka's law illustrating the inverse relationship between increasing articles written and decreasing authors achieving total number of articles. 3e) Three-field plots highlighting the authoring paper (AU) citation of the reference paper (CR) utilizing key words (DE).

A total of 1006 authors contributed to research into racial and ethnic disparities in THA and TKA. All top 10 authors were based in the US as demonstrated in Fig. 3b with the top three authors being Ibrahim SA from Weill Cornell University with 27 contributions, Kwoh CK from University of Arizona-Tucson with 17 contributions, and Parks ML from Hospital of Special Surgery at 8 contributions. The most cited authors are illustrated in Fig. 3c with Ibrahim SA as the most with 292 citations, followed by Kwoh CK at 233 citations, and Burant CJ and Siminoff LA at 124 and 117 references, respectively. The authors with the highest impact as assessed by h-index were Ibrahim SA at an index of 20, Kwoh CK at 16, Burant CJ at 7, and Ang DC, Parks ML, and Siminoff LA with an index of 6 for each.

3.4. Journals

In total, 78 journals published manuscripts related to racial and ethnic disparities in arthroplasty between years 1995–2022. Fig. 4a demonstrates the journal with the greatest production and highest citation of articles which was the Journal of Arthroplasty with 39 articles. This was followed by Clinical Orthopaedics and Related Research with 18 articles, and Arthritis Care & Research and Journal of Racial and Ethnic Health Disparities both at 11 articles. The greatest annual production from Journal of Arthroplasty was in 2021with 9 published articles. When co-citation analysis was assessed with a minimum of 25 citations, it resulted in a total of 45 journals. Among those journals, the greatest citations were for the Journal of Arthroplasty with 784 references, Journal of Bone and Joint Surgery at 601 and Clinical Orthopaedics and Related Research at 560 citations. The most impactful journals by h-index in Fig. 4b were Clinical Orthopaedics and Related Research at an index of 15, Journal of Arthoplasty at 13, Arthritis & Rheumatism and Arthritis Care & Research at an index of 9, and Journal of Bone and Joint Surgery at an h-index of 8.

Fig. 4A).

Source journal publication output throughout time period and b) journal impact by h-index.

3.5. Most influential manuscripts

Fig. 5 illustrates the top 10 most cited articles at the global stage with the most influential article from 2003. The largest burst reference was for Skinner et al. at 411, with Singh et al. and Ibrahim et al. following at 206 and 161, respectively.

Fig. 5.

Most globally cited articles with author, year published, and published journal.

3.6. Keywords

The keywords co-occurrence visualization is illustrated in Fig. 6 with a tree-map for trending keywords within racial and ethnic disparities in arthroplasty research. Among the 412 keywords reported, 55 keywords occurred at least 8 times and had sufficient link strength for determination of occurrences clusters. Fig. 6a demonstrates the occurrence visualization with size correlating to occurrence frequency. Fig. 6b illustrates the trend in the evolution of key words from 1995 to 2022. Between years 1995 and 2010 the main keywords were arthroplasty and osteoarthritis with changes in 2011–2015 with alignment, quality of life, and African-Americans becoming the main occurring keywords. This trend changed again between 2016 and 2020 with evolution of themes to outcomes, design, racial-differences, morphology, and racial/ethnic disparities. As trends evolve, more recent literature is focused on outcomes, length of stay, population health, and intraoperative risk factors. Fig. 6d represent the density clusters of the different topics. Fig. 7 is an illustration of the thematic map as organized by density and centrality. In that figure, density represents the cluster development and the strength of the topic, while centrality is strength of the association whereby keywords link clusters together. In that representation, Motor themes (high centrality and density) are mainstream themes highly linked by keywords, Basic themes (high centrality, low density) are themes central to racial/ethnic disparities without full development, but have significant cluster linking keywords such as anatomy and femoral component rotation in Fig. 7 which is a developing cluster for patient specific implant design. Niche themes (low centrality, high density) are clusters that are focused and developed, but have yet to significantly impact racial/ethnic disparities research.

Fig. 6A).

Network co-occurrence overlay for specific cluster development. B) Thematic evolution over 4 time periods and the shift of keywords. C) Co-occurrence overlay with time correlation to theme development. D) Co-occurrence overlay with density clusters.

Fig. 7.

Thematic map by density and centrality with keyword linkage and cluster determination.

4. Discussion

In this manuscript, the literature is analyzed to highlight the trending focus points of racial and ethnic disparities research output within arthroplasty. The most pertinent countries, author, institutions, collaboration, journal impact, and keywords were reported to identify current and future hotspots within the literature.

4.1. Publication production

The global annual production demonstrated a positive trend with a growth rate from 1995 to 2022 of 13.7 %. The greatest increase in production was from 2021 to 2022 with 32 published articles. This trend highlights a growing interest within arthroplasty literature aiming to better evaluate and understand racial and ethnic disparities.

4.2. Countries, institutions, and authors

The most productive country was the United States with leading metrics at the level of the most published articles, annual productivity, and total global citations. The United States had 76.82 % of all total global articles and 5240 total citations. Global collaboration mainly occurred between the United States (60 %) and other countries including Canada, Japan, and United Kingdom.

The VHA/VA were the most relevant institutions in productivity for racial and ethnic disparities within arthroplasty likely that the VA system is a large and integrated system with a population that is “equitable” to minimize confounders during data query.¹⁹ Ibrahim SA was the most relevant and influential author with a total of 27 articles published and an h-index of 20. Fig. 3d represent all authors productivity through the time period and demonstrating Lotka's Law with decreasing total author numbers with increasing published articles. Fig. 3e is three-fields plot highlighting the most relevant authors and their associated commonly cited articles in addition to corresponding keywords for those studies.

4.3. Journals and relevant articles

The Journal of Arthroplasty published the most articles and was the most cited, followed by Clinical Orthopaedic and Related Research with 18 publications, and Arthritis Care & Research and Journal of Racial and Ethnic Health Disparities with each publishing 11 articles. The most influential journal was Clinical Orthopaedics and Related Research.

The most relevant and influential articles with the greatest citation burst contribute substantially to the analysis of current trends and the identification of future hotspots.²⁸ The most influential articles in this analysis were “Racial, Ethnic, and Geographical Disparities in Rates of Knee Arthroplasty among Medicare Patients” by Skinner et al.,²⁴ “Racial disparities in knee and hip total joint arthroplasty: an 18-year analysis of nation Medicare data” by Singh et al.,²³ and “Differences in expectations of outcome mediate African American/white patient differences in ‘willingness’ to consider joint replacement” by Ibrahim et al..¹¹ In a query of Medicare claims data for TKA, Skinner et al. observed a disparity amongst AA males in every hospital region, and in some regions less than a third as compared to white males.²⁴ This disparity as described by the authors cannot be explained by socioeconomic status or geography areas as the discrepancy persisted while controlling for geographical variables. This article raises the question of “which rate is right?” that is a call to action towards not only eliminating disparities, but to provide an adequate choice for these patients. Singh et al. analyzed Medicare provider data over 18 years to identify trends in utilization rates for total joint arthroplasty for potential healthcare disparities among AA patients.²³ In addition to lower utilization rates for AA, that patient population also had longer length of stay, were less likely to be discharged home, and minimal reductions in racial disparities. Singh and colleagues address the need for locally based interventions as an avenue towards addressing community barriers. Ibrahim et al. surveyed elderly male patients in 2001 at the VA and identified hesitation amongst AA patients towards considering joint replacement due in part to post-operative expectations of pain, hospital course, and functional status.¹¹ The authors then suggested that more research that further work will be required to identify if there is a trend towards decreased utilization for AA and for further efforts towards addressing expectations and utilization of utilization of joint arthroplasty.

4.4. Keywords

Fig. 6a is the analysis of keywords co-occurrence overlay and Fig. 6d is the density map highlighting individual clusters based on associated keywords. The thematic evolution representation in Fig. 6b illustrates three transitions from 1995 to 2010 to 20111-2015 with emphasis on quality of life and AA, while the transition from 2011 to 2015 to 2016–2020 stratified upon outcomes with racial differences and racial/ethnic disparities. Another transition from 2016 to 2020 to 2021–2022 with evolution towards outcomes, risk factors, and population-based research. The morphology of thematic evolution corresponds to an increasing emphasis on perioperative optimization with attention to racial and ethnic factors as potential contributors to postoperative outcomes. The co-occurrence overlay in Fig. 6c illustrates the time period with its relevant cluster. As an example, disparity was a cluster of focus in 2016 with nodal connections to “quality-of-life” in 2010, “ethnic differences” in 2010, and “access” in 2014. Another cluster of interest was “outcomes” in 2018 with node connections to keywords of “ethnic differences” in 2010, “African-American” in 2014, and “obesity” in 2016.

The relevant keyword search with analysis of cluster theme identified 55 keywords that constituted 5 clusters falling into the following categories: patient specific implant design, comparative analysis, care delivery, pain management, and outcomes assessment.

4.5. Patient specific implant design

The cluster of patient specific implant design included key words of alignment, anthropometric measure, component, design, distal femur, and morphology. Osteoarthritis and gait variations have been noted to be associated and vary at the racial and patient specific level. Previous studies noted white female patients when compared to AA female patients had better kinematics with faster speed, longer length of stride, and greater duration within double support phase of walking.²⁵ The disparity in osteoarthritis symptomology and decreased gait variations may be attributed to anthropometric differences between Caucasians and African Americans with regards to size and shape in the natural anatomy of the knee.¹⁷ Total joint arthroplasty systems have predetermined sizes, however minimizing discrepancies between component fit and patient anatomy could theoretically minimize postoperative complications while optimizing outcomes and improving quality of life.⁸ The component design cluster has trending key words of “prosthesis” and “design” from 2016 to “morphology” in 2020 and continues to be a developing cluster that hasn't completely matured (Fig. 6c). While correlation of patient specific instrumentation, was it at the racial or gender scale, with postoperative outcomes and functionality continues to be a topic for debate, it appears to be more relevant in the TKA, and less so THA, literature. These arguments could stem from the lower satisfaction rates with TKA compared to THA, and constitute a path in the continuous search towards improving outcomes with the procedure. Future trends in this cluster could substantially change with the introduction of robotics and the higher level of collected data in TKA, which is expected to improve the understanding of potential contributors of satisfaction, and might boom or bust this cluster accordingly.

4.6. Comparative analysis

Comparative analysis cluster included keywords such as decision making, cost effectiveness, discharge destination, ethnic differences, quality, and socioeconomic status. This cluster of literature output contributed critical evidence highlighting existing disparities at the racial and ethnic level. The evolution of themes towards length of stay and outcome comparisons is witnessed in recent literature in 2021–2022 (Fig. 6b). The access to arthroplasty among minorities exhibit a level of discrepancy when compared to White race, with reports of lower rates of the procedure among AA and Hispanic patients.⁵ Notably, black women are noted to more likely delay arthroplasty and present with significantly decreased function at time of surgery when compared to white women.² With continued efforts and a trend towards home disposition following TKA and THA, black women demonstrate a higher likelihood towards disposition to inpatient rehabilitation and nursing facilities, and a concomitant higher complications and readmissions rate.^2,15 Other studies have highlighted that older black patients were less likely to be discharged home, due to a variety of suboptimal deterministic variables such as need of more assistance, socioeconomic status, pre-operative functional status, and presence of family support.^2,22 Comparative analysis is critical to identify the presence of disparities and highlight at which level and variables these disparities exist. As we continue to understand these disparities and design processes to overcome them, there will be a need for continued future research that falls within the comparative analysis cluster to assess progress made at the level of overcoming disparities. While comparative analysis has been highlighted as a separate cluster within the bibliometric analysis of the literature due to the use of specific keywords, the visual representation highlights a center-role of this cluster that plays a substantial role in, and links, the clusters of outcomes assessment, care delivery, and pain management.

4.7. Care delivery

Care delivery cluster included key words of access, disability, disparities, and preferences. Racial and ethnic disparities as a theme have been evolving since 2011 from the key words “African-American” to “racial and ethnic disparities” in 2016–2020 (Fig. 6b). Disparities in ethnic and racial care has been attributed to the deleterious effects of inherent social determinants of health.²⁰ Such determinants are situations and conditions in which individuals progress through life. Access to high quality primary care and referral network may present difficulty due in part to the differences in the insurance and healthcare system leading to more late-stage osteoarthritis.¹⁰ Interestingly, while the Veterans Health Administration is an integrated healthcare system with equitable access, there is still reported incidences of racial disparities with arthroplasty access and complications within that system.²⁰ Despite equitable healthcare system, discrepancy still exist and is likely multifactorial. While there are patient specific factors, implicit bias has been identified among physicians as a source for discrepancy. Oliver et al. performed a study among primary care providers and demonstrated that there is a strong clinician implicit bias for greater preference for white patients with associated medical cooperativeness, and likely could influence decisions.¹⁴ In addition to such potential bias, added factors that could fall under the umbrella of social determinants of health, such as environmental, financial, social, and cultural aspects could contribute to these disparities. The future of the care delivery cluster is trending towards a general focus on integrating these social determinants of health, with aims to better understand those variables, and designing care delivery protocols to overcome such variables, beyond the simple walls of the healthcare system.

4.8. Pain management

Pain management is noted to be a weaker cluster. The US Institute of Medicine released a consensus study report in 2003 “Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care”. This report was a culmination of over a hundred studies that highlighted a persistent discrepancy between the care received by racial/ethnic minorities compared to nonminority after adjusting for care delivery factors. Beyond the level of care, there also exist a discrepancy in pain management with some reports detailing African American and Hispanic patients are at risk for inadequate pain control.¹³ Moreover, Hispanic patients were more likely to have shorter visits, less likely to receive opiates, and more likely to receive nonopioids for pain management.¹² Ongoing discussion regarding the keyword “pain” and the associated cluster has been of recent trend since 2014 with association to “management” in the co-occurrence overlay (Fig. 6c). This cluster seems to be weakening in recent literature and is expected to be assimilated in a general trend of outcomes assessment and improvement, as pain control constitute a major variable in determining patient satisfaction and quality of delivered care.

4.9. Outcome assessment

Outcomes assessment cluster included keywords such as complications, epidemiology, ethnic disparities, hospital volume, length of stay, and risk factors. Outcomes and risk mitigation have been a trending them in the arthroplasty literature in general, and within racial and ethnic disparities since 2016 according to the thematic evolution visual in Fig. 6b. The move towards a value-based care system contributed to the development of perioperative optimization and outpatient protocols. These efforts subsequently allowed for identification of higher complications and worst post-operative functional outcomes among minority AA and Hispanic patients.^1,29 Additionally, non-white patients are noted to undergo arthroplasty procedure at low volume centers with higher complication and mortality rates.²¹ Moreover, non-white minority patients, excluding Asian and Pacific Islander, were more likely to require a longer post operative length of stay and underwent a higher rate of reoperations.⁴ Such findings have been partially explained by patient with low socioeconomic status lacking insurance or the insight to obtain care at a specialty center and may refuse to travel to “unfamiliar” high volume hospitals due to various restraints.²⁹ These differences are a representation of a multifactorial etiology that spans access to care, socioeconomic and support systems, geographic and environmental differences, among others.²¹ As we understand these multifactorial conditions of racial and ethnic disparity, such inherent patient characteristic can be further assessed as non-medical variables. This is an opportunity to push beyond and designing and optimizing these racial and ethnic variables. Future research in outcome analysis should transition from the current state of understanding the status-quo to designing approaches to optimize and overcome these barriers, and hence transform the “outcomes assessment” cluster to an “outcomes modulation and improvement” cluster.

5. Conclusion

Racial and ethnic disparity within arthroplasty literature has been increasing at a growing rate since 1995 with continual contribution from various countries, institutions, and collaborations. United States is the largest contributor to the global scientific body regarding disparity research. The VHA/VA were the most relevant institution. Ibrahim SA, Kwoh CK, and Parks ML were the most relevant authors within racial and ethnic disparity arthroplasty research. Journal of Arthroplasty was the most relevant and influential journal within disparity research. As progress is made with understanding the multifactorial aspect of racial and ethnic disparities, there are opportunities to push beyond the assessment of disparities. There will be demand t towards overcoming such disparities and for further analysis towards progression of this goal. This bibliometric analysis identified previous, current, and future trends within arthroplasty racial and ethnic disparities with the goal of prediction of future hotspots.

Guardian/patient consent

Consent is not applicable for this study, review of literature.

Ethical statement

Not applicable, no patient data was involved in this study.

Funding statement

The authors have no financial disclosures or conflicts of interests.

Funding is not applicable for this study.

CRediT authorship contribution statement

Fong H. Nham: Conceptualization, Methodology, Investigation, Resources, Writing – original draft, Writing – review & editing. Eliana Kassis: Conceptualization, Methodology, Software, Validation, Formal analysis, Resources, Data curation, Visualization. Winnie Xu: Conceptualization, Methodology, Software, Resources, Data curation, Visualization. Mouhanad M. El-Othmani: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – review & editing, Visualization, Supervision, Project administration. Nana O. Sarpong: Conceptualization, Methodology, Validation, Investigation, Resources, Visualization, Supervision, Project administration.

Declaration of competing interest

Each author certifies that there are no funding or commercial associations that might pose a conflict of interest in connection with the submitted article related to the author or any immediate family members.