Introduction

The National Institute of Arthritis and Musculoskeletal Diseases has designated racial disparities in orthopaedic surgery as a major priority for investigation and action [1]. Racial disparities are well established in total joint arthroplasty (TJA), particularly in the Black/African American community [2, 3]. Black patients have been shown to have lower procedure utilization rates, higher complication rates [4], lower functional outcomes, increased revision rates [5,6,7], and lower patient satisfaction [8]. These findings coupled with a rapidly growing demand in TJA [9, 10] beg the need for greater understanding of population-specific risk factors and trends in health outcomes to reduce racial disparities in TJA care.

Over the past decade, there have been several initiatives aimed at improving access to care and health outcomes among minority individuals. One notable example is the Affordable Care Act (ACA) of 2010 that expanded healthcare coverage to uninsured Americans in low- and middle-income households. The ACA also included provisions to improve the collection of race/ethnicity in healthcare databases and increased funding for various community-based initiatives to address racial/ethnic disparities. Emerging reports following the ACA have shown significant reductions in the number of uninsured minorities [11, 12]. For example, in the initial few months following the first ACA open enrollment period in 2014, the percentage of uninsured non-Hispanic Blacks declined from 25.5% in 2013 to 17.2% [12]. Along the same time, TJA practice has experienced tremendous advancements including less-invasive surgical approaches, improved perioperative recovery protocols, and adoption of bundled payment models that are intended to promote high-quality outcomes and lower costs. The impact of the aforementioned initiatives and advancements in TJA care on minority patients remain unclear. Furthermore, as value-based care takes hold, there is a growing emphasis to create risk stratification protocols that optimize surgical outcomes. Given the intricacies and complexities inherent to each racial group, population-specific risk stratification would allow more specific preoperative optimization to mitigate potential adverse events following surgery.

The objective of this study was to examine annual trends in procedure utilization, comorbidity profiles, hospital length of stay (LOS), 30-day outcomes, and risk factors for adverse events among Black patients undergoing primary total hip arthroplasty (THA) in a current time period reflecting greater access to care along with greater scrutiny on the value of care provided. The information obtained in this study will help us gauge how THA care in the Black community has evolved in light of the numerous recent legislative and surgical changes. In addition, population-specific risk stratification may allow more precise preoperative optimization in this racial group.

Methods

Institutional review board approval was not required. The American College of Surgeons National Surgery Quality Improvement Program (ACS-NSQIP) database was queried for all patients who self-identified as non-Hispanic Black/African American and underwent primary, elective THA between 2011 and 2017. The procedures were confirmed to be elective by cross-referencing associated ICD-9 and ICD-10 codes. The ACS-NSQIP is a national chart-based database with over 600 participating hospitals. It captures all the surgical cases for hospitals with lower volumes and uses a systemic sampling for those with larger volumes [13]. The database includes encounter data spanning hospital admission up to 30 days postoperatively.

Variables collected in the ACS-NSQIP include age, sex, body mass index (BMI), current smoking, chronic steroid use, American Society of Anesthesiologists (ASA) physical classification, functional status (independent vs. partially/completely dependent), living environment (admitted directly from home vs. facility), and select comorbidities: diabetes, hypertension, chronic obstructive pulmonary disease (COPD), heart failure, anemia, metastatic cancer, coagulopathy, dyspnea, ascites, and chronic kidney disease (CKD). Other variables include surgical diagnosis, operative time, and laterality (unilateral vs. bilateral).

The primary outcomes of the study were the annual trends in THA utilization, comorbidity profiles, LOS, and 30-day adverse events (AEs) during the study period. AEs included medical complications, surgical complications, readmissions, reoperations, and mortality. A secondary outcome of the study was identifying the risk factors associated with the development of AEs.

Values for continuous variables were presented as mean and standard deviation, and nominal variables as frequency and percentage. For temporal changes, the overall p values were determined from a univariate mixed effects logistic regression, which included an ordinal time variable for admission year. For simplicity, data was presented in time period cohorts; however, all analyses were performed across each individual year. For identification of risk factors for AEs, univariate analyses were first conducted to explore the association of each collected variable with development of any postoperative AE. For nominal variables, overall p values were determined using chi-squared tests. For continuous variables, overall p values were obtained from Welch’s t test. Multivariate logistic regression analyses were then utilized to yield odds ratios for developing any postoperative AE. Statistical significance was defined as p < 0.05. Data was analyzed using Stata 16.0 (State Statistical Software: Release 16. College Station, TX: StateCorp LLC).

Results

A total of 11,574 Black patients were included in the analysis accounting for 7.4% of all THAs performed during the study period compared with 74.5% in non-Hispanic Whites. Between 2011 and 2017, there was a 109% increase in the number of THAs performed among Black patients. There were also reductions in the rates of tobacco smoking (1.4%, p = 0.024), anemia (6.6%, p < 0.0001), dyspnea (1.0%, p < 0.0001), and osteonecrosis (3.32%, p < 0.0001). Table 1 summarizes the trends in demographics, comorbidity, and perioperative characteristics for the study cohort.

Table 1 Trends in baseline and perioperative characteristics of the study cohort
Full size table

Postoperatively, the frequency of inpatient stays > 2 days decreased from 68.4 to 36.8% (p < 0.0001). There were 647 patients (5.6%) who developed 30-day AEs. Except for a reduction in medical complications (largely myocardial infarction, p = 0.014), there were no differences in rates of readmissions, reoperations, surgical complications (including DVT/PE), or mortality over the study period. Table 2 summarizes the trends in 30-day postoperative outcomes.

Table 2 Trends in length of stay and 30-day outcomes of the study cohort
Full size table

After controlling for all baseline and perioperative differences, independent risk factors for 30-day AEs were BMI > 35 (OR 1.28; 95% CI, 1.07–1.54; p = 0.007), ASA classification > 2 (OR 1.65; 95% CI 1.31–1.93; p < 0.0001), dependent functional status (OR 1.65; 95% CI, 1.12–2.43; p = 0.011), bleeding disorders (OR 1.62; 95% CI 1.05–2.50; p = 0.031), CKD (OR 2.10; 95% CI, 1.59–2.76; p < 0.0001), and osteonecrosis (OR 1.52; 95% CI, 1.19–1.95; p = 0.001). Figure 1 summarizes the results of the multivariate logistic regression analyses.

Fig. 1
figure 1

Multivariate logistic regression analysis for development of a 30-day postoperative event. A multivariate logistic regression was utilized to yield odds ratios for development of a postsurgical event, defined as any combination of a medical complication, surgical complication, mortality, readmission, or reoperation. Patient factors previously demonstrating (p < 0.05) or approaching significance (p < 0.10) between the patient groups (Table 1) were controlled for in the regression analysis. Continuous variables (age, BMI, ASA score) were converted to nominal variables according to Youden’s index so as to maximize the discriminative capacity of the cutoff value. COPD, chronic obstructive pulmonary disease; BMI, body mass index; ASA, American Society of Anesthesiologists physical classification system

Full size image

Discussion

To our knowledge, this is the first study to examine the time trends in health profiles and outcomes pertaining to primary THA among Black patients. Specifically, the study explored the annual trends in procedure utilization, comorbidity profiles, LOS, and 30-day outcomes between 2011 and 2017. Over the study period, there was a twofold increase in the number of Black patients undergoing THA although remained disproportionally low compared with their White counterparts. This was accompanied by reductions in the rates of anemia, dyspnea, tobacco smoking, and osteonecrosis. Despite such favorable trends in comorbidity profiles, the overall rates of 30-day postoperative adverse events (AEs) remained unchanged. Only the rate of medical complications showed improvement, which was primarily driven decreased incidence of myocardial infarction. The percentage of Black patients requiring prolonged hospitalization decreased by nearly one-half. Independent risk factors for AEs were BMI > 35, ASA classification > 2, dependent functional status, diabetes, bleeding disorders, CKD, and osteonecrosis as the THA indication.

The observed trend toward shorter LOS from 2011 to 2017 mirrors trends in the general population as fast-track and outpatient THAs have become more prevalent. Prior studies on short stay TJA have demonstrated significant cost savings without compromising safety. However, patients with certain major medical comorbidities may not be appropriate for these accelerated protocols. The historically higher comorbidity burden in Black patients may pose a barrier to the adoption of short stay TJA in this racial group. Moreover, we are unable to comment on discharge disposition due to limitations in our data, and it is possible that some of this trend could be explained by earlier discharge to post-acute facilities. Nevertheless, the observed trends for improved comorbidity profiles and lack of increased postoperative AEs associated with the simultaneous decline in LOS over the study period are encouraging.

Our study suggests flattening in the rates of worsening arthroplasty outcomes among Black patients in recent years compared with older reports. In an 18-year analysis of Medicare beneficiaries undergoing THA, Singh et al. [4] showed an increased rate of hospital readmission in Black patients between 1991 and 2008 compared with Whites. However, consistent with our findings, a more recent retrospective analysis of a state arthroplasty registry by Aseltine et al. [14] showed narrowing disparities in 30-day readmission rates in Black patients from 2005 to 2015. It is likely that the receding tide in postsurgical adverse events is influenced by improvements in comorbidity profiles and/or optimization prior to surgery given our observation of the decreasing rates of tobacco smoking, anemia, dyspnea, and osteonecrosis at the time of surgery. Those factors have been shown to impact THA outcomes. For example, Kapadia et al. [15] demonstrated in a retrospective matched cohort study that tobacco smoking was associated with higher complication and revision rates following THA. The disconnect between improving comorbidity profiles and frank improvements in readmissions, reoperations, and mortality is somewhat surprising. Unfortunately, root cause analysis was not possible due to limitations of the ACS-NSQIP database.

While the number of THAs performed in Black patients has doubled over the recent years, Blacks made up only 7.4% of all THAs performed captured in the NSQIP database during the study period with the vast majority of procedures performed in White patients. This utilization rate appears to be disproportionate to 2018 United States Census showing that 12.7% of the population identified as non-Hispanic Black [16] and that the prevalence of both radiographic and symptomatic osteoarthritis among Black individuals were similar to Whites [17]. Clearly, despite encouraging trends in procedure utilization in the Black community, efforts are still needed to improve access to THA.

While numerous studies have demonstrated racial disparities in TJA, few have proposed interventions to bridge this gap. This likely reflects the complex and multifactorial nature of racial disparities. Based on our analysis of Black patients who experienced postoperative adverse events, one potential target for intervention is the early identification and mitigation of high-risk patients through preoperative counseling, medical optimization, referral to specialized surgery centers, and/or close postoperative follow-up. This is especially important in patients with osteonecrosis, ASA classification > 2, chronic kidney disease, and bleeding disorders, and those who are functionally debilitated at baseline as identified in the present study. The latter patient factor deserves a closer look as it is likely related to the tendency among minority patients to seek treatment at more advanced stages of arthritis [6]. Within the Black community, patients are more likely to be underinsured, have insurance plans that require referral from their primary care physicians, or be part of underdeveloped referral networks [18, 19]. This means more time is spent in the referral process, further delaying presentation to specialized arthritis care while pain and disability worsen [20]. To add to this challenge, caring for minority patients has traditionally been complicated by physician distrust, higher perception of increased postoperative pain [21, 22], and racial differences in physician pain management practices [23] leading to fewer patients seeking surgery. Our data lends support to promote efforts toward expedient reconstructive treatment when nonoperative management has failed and before functional capacity has severely deteriorated.

There are multiple limitations to this study. First, the ACS-NSQIP database collects from a disproportionately high number of academic hospitals. Academic centers are more likely to take on complex and high-risk cases compared with community hospitals, and hence, this could have potentially skewed the rates of adverse events. Second, the ACS-NSQIP database does not contain a comprehensive list of patient characteristics and is subject to abstraction errors. For example, socioeconomic factors known to affect health outcomes were not available. Third, we lack pain and functional outcome scores that reflect on the quality of care perceived by patients. Fourth, the surgical cases captured by the ACS-NSQIP database represent only a small sample of all THAs performed in the USA. While the database is often used to measure and improve the quality of surgical care, it is often criticized as lacking data from non-academic hospitals and surgical centers where the larger percentage of THAs are performed. However, the knowledge derived from this study should still be applicable regardless of the hospital type and is consistent with previous reports.

In conclusion, we observed favorable trends in procedure utilization, comorbidity profiles, and LOS among Black patients undergoing primary THA between 2011 and 2017. However, the rates of 30-day adverse health outcomes remained steady during the same period, suggesting flattening in the previously reported worsening outcomes. Still, more work remains to be done as the overall rates of procedure utilization, LOS, and outcomes remain lower compared with Whites. Patients with BMI > 35, ASA classification > 2, dependent functional status, diabetes, bleeding disorders, CKD, and preoperative diagnosis of osteonecrosis appear to be higher risk for developing AEs. Future research should look at the multi-faceted aspects including socioeconomic determinants that impact Black patients undergoing THA with a focus on providing the necessary resources to improve procedure utilization and outcomes.