Racial/ethnic disparities in outcomes among patients with hip fracture: A retrospective matched cohort study
Article History
-
Received: ,
Accepted: ,
Abstract
Abstract
This study investigated the impact of race/ethnicity on post-operative complications, revisions, costs, and time to admission among patients with hip fracture surgically managed with hip plating, nailing, hemiarthroplasty, total hip arthroplasty, and screw fixation. We hypothesize that patients from racial/ethnic minority backgrounds will have higher post-operative complications, revisions, and costs.
Patient data was collected from PearlDiver, a large commercial insurance database, from 2014 to 2017. Patients from racial/ethnic minority backgrounds were matched to White patients according to age and gender, as well as comorbid tobacco use, diabetes, and obesity. Patients were also matched by Charlson Comorbidity Index (CCI) and Elixhauser Comorbidity Index (ECI). Post-operative complications, revisions, and costs at 30, 90, and 365 days were evaluated among the racial/ethnic groups.
Data from 9,494 White (93.34 %), 500 Black (4.92 %), 94 Asian (0.92 %), and 82 Hispanic (0.81 %) patients with hip fracture who underwent hip surgery was analyzed. Patients from every minority group had statistically significant increased CCI and ECI. Unmatched Black patients experienced more complications. Unmatched Black patients also had higher costs at all time points (p < 0.0001). Unmatched Hispanic patients had higher costs at one year.
The results indicated that unmatched Black patients with hip fracture undergoing hip surgery had significantly increased rates of complications and total costs at every time point as compared to White patients. Matched Black patients had higher costs at 90 and 365 days. The results of the study partially support our hypothesis that minority patients would have greater complications and costs than White patients. A potential explanation for why these associations were not observed among Asian and Hispanic patients was because there were fewer patients from these racial/ethnic backgrounds in the database that met study criteria. Achieving health equity requires efforts to improve health disparities for all members of society despite racial/ethnic background.
Keywords
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.
1 Introduction
Over 300,000 hip fractures occur each year in the United States (US).1 By the year 2040, experts anticipate that the incidence of hip fracture will increase to over 500,000.2 Hip fractures among elderly patients typically occur from low-energy trauma, such as a fall from standing height, while those in younger patients are likely from high-energy trauma.3,4 Risk factors for hip fracture include previous fracture, advanced age, female sex, estrogen deficiency, decreased bone density, prior fall, assisted ambulation, and Parkinson's disease.4–7
Depending on the type of hip fracture, surgical treatment involves hip plating, nailing, hemiarthroplasty, total hip arthroplasty, and screw fixation.8 Although hip surgery facilitates healing and decreases complications in patients with hip fracture, racial/ethnic disparities persist in management and post-surgical outcomes.9 Among studies analyzing hip fracture outcomes, most include a majority population of White patients.9 Research indicates that African-American patients experience nearly four times the average wait times for imaging as compared to their White counterparts.10 African-American and Hispanic patients also have prolonged length of stay (LOS) and decreased follow-up than White patients.9
To the best of our knowledge, no studies have investigated the rate of post-operative complications and total costs of patients with hip fracture undergoing surgical intervention. The aim of this study is to analyze the rates of post-operative complications, revisions, total costs, and time to admission in patients with hip fracture treated with hip plating, nailing, hemiarthroplasty, total hip arthroplasty, and screw fixation from different racial/ethnic groups before and after matching at various time points. We predict that patients belonging to racial/ethnic minority groups will have increased rates of post-operative complications, revisions, total costs, and time to admission than White patients.
2 Materials and methods
2.1 Data source and study population
This matched retrospective cohort study utilized data from the Medicare Standard Analytic Files of the PearlDiver Patient Records Database (www.pearldiverinc.com, Colorado Springs, CO, USA) for all patients with hip fracture who underwent hip plating, nailing, hemiarthroplasty, total hip arthroplasty, and screw fixation from January 2014 to January 2017. Complications and revisions were identified from the International Statistical Classification of Diseases −9 (ICD-9) and −10 (ICD-10) codes. The Medicare Standard Analytic Files of the PearlDiver database contains inpatient and outpatient information on long-term health records, charges, and reimbursements for 51 million covered individuals. Given that the PearlDiver database utilizes de-identified patient in accordance with the Health Insurance Portability and Affordability Act (HIPAA), institutional review board (IRB) approval was waived.
Patients with hip fracture treated with hip plating, nailing, hemiarthroplasty, total hip arthroplasty, and screw fixation from 2014 to 2017 from the PearlDiver database with at least one-year of post-operative follow-up were included in this study. Patients without at least one year of post-operative follow-up were excluded from the study. Patients were categorized by race/ethnicity as White, Black, Asian, or Hispanic by self-categorization in the PearlDiver database.
2.1.1 Patient characteristics
Prior to matching, White, Black, Asian, and Hispanic patients were compared at baseline with factors such as the Charlson Comorbidity Index (CCI), an estimate of one-year mortality in hospitalized patients with co-morbid health conditions11 and the Elixhauser Comorbidity Index (ECI),12 a measure grouping patient comorbidities by ICD codes. Following this, the PearlDiver database generated three unique propensity-matched cohorts for White, Black, Asian, and Hispanic patients according to age and gender, as well as comorbid tobacco use, diabetes, and obesity. Patients were also matched according to the CCI and ECI. For the purposes of this study, obesity represents a body mass index (BMI) greater than 30 kg/m2.13
2.2 Post-operative complications
The surgical post-operative complications of interest for this study for each matched cohort were surgical site infection (SSI), deep venous thrombosis (DVT), wound disruption, hematoma, pulmonary embolism (PE), and blood transfusion. Medical post-operative complications for each matched cohort consisted of myocardial infarction (MI), anemia, arrhythmia, congestive heart failure (CHF), acute kidney injury (AKI), cardiac arrest, pneumonia, urinary tract infection (UTI), and readmission. Surgical and medical complications were evaluated at 30, 90, and 365 days.
The frequency of revisions was assessed for each matched cohort two years after surgery. Each complication of interest was selected prior to data collection and analysis. Odds ratios (OR) were utilized to measure the association of complications for each matched group. The OR was computed with the 95 % confidence interval (CI) and p-value for the post-operative complications. Data with less than 11 patients generated an output of “-1” from the PearlDiver system. Separate CCI Reports were run for patients with data less than 11 to obtain the exact values.
2.3 Total cost of care
Using the means and standard deviations (SD) of the total cost among the matched cohort, the difference of means, 95 % CI, and p-values for the total cost of care at 30, 90, and 365 days was calculated.
2.4 Statistical methods
Patient characteristics were reported according to the corresponding mean and SD for continuous variables and frequency (n) and proportion (%) for categorical variables. The two-independent sample t-test for unequal variances was utilized for continuous variables. The test for significance reflected if an association could be attributed to a specific cause. A statistically significant result had a p ≤ 0.05. Statistical analyses were performed using MedCalc's Odds Ratio Calculator and MedCalc's Comparison of Means Calculator.14,15
3 Results
Among patients with hip fracture underwent hip plating, nailing, hemiarthroplasty, total hip arthroplasty, and screw fixation from January 2014 to January 2017, 9,494 White patients, 500 Black patients, 94 Asian patients, and 82 Hispanic patients met inclusion criteria for this study. Black and Hispanic patients were respectively about seven and four times more likely to be younger than 65 years of age at the time of surgery respectively. All minority patients had increased CCI, ECI and higher rates of diabetes. Black patients had higher rates of obesity, while Asian patients had decreased rates of obesity. Black patients had more frequent tobacco use [Table 1].
| White | Black | Comparison | CI | P Value | Asian | Comparison | CI | P Value | Hispanic | Comparison | CI | P Value | |
| Total n (%) | 9494 (93.34) | 500 (4.92) | 94 (0.92) | 82 (0.81) | |||||||||
| Under 65 n (%) | 661 (6.96) | 154 (32.8) | 6.6587 | 5.4000–8.2109 | < 0.0001 | 11 (11.70) | 1.771 | 0.940–3.338 | 0.0772 | 17 (22.1) | 3.495 | 2.037–5.996 | < 0.0001 |
| Male n (%) | 3153 (332) | 180 (36.0) | 1.131 | 0.938–1.365 | 0.1975 | 21 (22.34) | 0.579 | 0.355–0.942 | 0.0277 | 22 (26.8) | 0.737 | 0.452–1.204 | 0.2234 |
| CCI avg (SD) | 3.38 (3.39) | 4.21 (3.91) | 1.24556213 | 1.155–1.337 | < 0.0001 | 4.18 (3.32) | 1.236686391 | 1.033–1.440 | 0.0228 | 5.11 (4.1) | 1.51183432 | 1.293–1.730 | < 0.0001 |
| ECI avg (SD) | 8.03 (4.71) | 8.5 (4.93) | 1.058530511 | 1.006–1.111 | 0.0301 | 9.94 (5.01) | 1.225404732 | 1.106–1.345 | 0.0002 | 10.6 (5.24) | 1.320049813 | 1.545–3.595 | < 0.0001 |
| LOS avg (SD) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | |||||||
| Obesity n (%) | 3,225 (33.97) | 255 (51.0) | 2.023 | 1.689–2.423 | < 0.0001 | 116 (17.02) | 0.399 | 0.233–0.684 | 0.0008 | 29 (35.4) | 1.064 | 0.675–1.677 | 0.7903 |
| Diabetes n (%) | 5,518 (58.12) | 354 (70.8) | 1.747 | 1.435–2.128 | < 0.0001 | 73 (77.7) | 2.505 | 1.539–4.077 | 0.0002 | 68 (82.9) | 3.5 | 1.966–6.230 | < 0.0001 |
| Tobacco Use n (%) | 4,651 (48.99) | 270 (54.0) | 1.222 | 1.021–1.464 | 0.0292 | 25 (26.6) | 0.377 | 0.238–0.597 | < 0.0001 | 34 (41.5) | 0.738 | 0.475–1.147 | 0.1763 |
Regarding surgical complications in the unmatched cohort, 30 days following surgery, Black patients had more than double the odds of hematoma and transfusion, and Asian patients almost quadruple the odds of SSI. Ninety days after hip surgery, Black patients continued to have higher odds of hematoma and transfusion, with an added more than double increase in wound disruption. At 90 days, Asian patients had over twice the odds of SSI. At the one-year post-operative period, Black patients maintained significantly higher rates of wound disruption and transfusion, along with new onset increased odds of DVT. Furthermore, Black patients maintained a higher likelihood to be readmitted at all time points [Table 2].
| Complications at 30 Days | |||||||||||||
| White | Black | Odds Ratio | CI | P Value | Asian | Odds Ratio | CI | P Value | Hispanic | Odds Ratio | CI | P Value | |
| SSI n (%) | 196 (2.06) | 12 (2.4) | 1.167 | 0.647–2.104 | 0.609 | 7 (7.45) | 3.817 | 1.745–8.351 | 0.0008 | 2 (2.44) | 1.816 | 0.290–4.859 | 0.8126 |
| DVT n (%) | 37 (0.39) | 2 (0.4) | 0.971 | 0.247–4.271 | 0.9713 | 0 (0) | 1.334 | 0.081–21.892 | 0.8398 | 0 (0) | 1.529 | 0.093–25.102 | 0.7664 |
| Wound Disruption n (%) | 55 (0.58) | 6 (1.2) | 2.084 | 0.893–4.865 | 0.0894 | 1 (1.06) | 1.845 | 0.253–13.476 | 0.5459 | 0 (0) | 1.031 | 0.063–16.828 | 0.983 |
| Hematoma n (%) | 126 (1.33) | 14 (2.8) | 2.142 | 1.224–3.748 | 0.0076 | 1 (1.06) | 0.8 | 0.111–5.781 | 0.8245 | 1 (1.22) | 0.918 | 0.127–6.647 | 0.9324 |
| PE n (%) | 103 (1.08) | 6 (1.2) | 1.107 | 0.484–2.535 | 0.8092 | 0 (0) | 0.48 | 0.030–7.785 | 0.6057 | 0 (0) | 0.55 | 0.034–8.927 | 0.6741 |
| Transfusion n (%) | 194 (2.04) | 22 (4.4) | 2.206 | 1.406–3.462 | 0.0006 | 2 (2.13) | 1.042 | 0.255–4.261 | 0.9542 | 2 (2.44) | 1.199 | 0.293–4.910 | 0.8014 |
| Readmission n (%) | 856 (9.02) | 72 (14.4) | 1.698 | 1.309–2.200 | 0.0001 | 13 (13.83) | 1.62 | 0.898–2.921 | 0.1091 | 10 (12.20) | 1.402 | 0.721–2.726 | 0.3199 |
| Complications at 90 Days | |||||||||||||
| White | Black | Odds Ratio | CI | P Value | Asian | Odds Ratio | CI | P Value | Hispanic | Odds Ratio | CI | P Value | |
| SSI n (%) | 326 (3.43) | 19 (3.8) | 1.111 | 0.693–1.7870 | 0.6621 | 7 (7.45) | 2.263 | 1.039–4.926 | 0.0396 | 4 (4.88) | 1.442 | 0.525–3.963 | 0.4777 |
| DVT n (%) | 70 (0.74) | 4 (0.8) | 1.086 | 0.395–2.986 | 0.8734 | 0 (0) | 0.707 | 0.044–11.505 | 0.8077 | 0 (0) | 0.81 | 0.049–13.192 | 0.8824 |
| Wound Disruption n (%) | 87 (0.92) | 12 (2.4) | 2.659 | 1.444–4.895 | 0.0017 | 1 (1.06) | 1.163 | 0.160–8.436 | 0.8815 | 1 (1.22) | 1.335 | 0.184–9.701 | 0.7753 |
| Hematoma n (%) | 179 (1.89) | 16 (3.2) | 1.72 | 1.023–2.892 | 0.0407 | 1 (1.06) | 0.56 | 0.078–4.037 | 0.5647 | 1 (1.22) | 0.643 | 0.089–4.642 | 0.661 |
| PE n (%) | 142 (1.50) | 12 (2.4) | 1.613 | 0.889–2.928 | 0.116 | 0 (0) | 0.346 | 0.021–5.598 | 0.4548 | 0 (0) | 0.396 | 0.025–6.419 | 0.5147 |
| Transfusion n (%) | 290 (3.05) | 26 (5.2) | 1.741 | 1.153–2.628 | 0.0083 | 2 (2.13) | 0.69 | 0.169–2.814 | 0.6048 | 3 (3.66) | 1.25 | 0.378–3.84 | 0.7522 |
| Readmission n (%) | 1320 (13.90) | 103 (20.6) | 1.607 | 1.284–2.011 | < 0.0001 | 14 (14.89) | 1.084 | 0.612–1.918 | 0.7826 | 14 (17.07) | 1.275 | 0.715–2.273 | 0.4103 |
| Complications at One Year | |||||||||||||
| White | Black | Odds Ratio | CI | P Value | Asian | Odds Ratio | CI | P Value | Hispanic | Odds Ratio | CI | P Value | |
| SSI n (%) | 534 (5.62) | 28 (5.6) | 0.995 | 0.673–1.472 | 0.9814 | 9 (9.57) | 1.777 | 0.889–3.551 | 0.1039 | 5 (6.10) | 1.09 | 0.439–2.703 | 0.8533 |
| DVT n (%) | 196 (2.06) | 17 (3.4) | 1.67 | 1.009–2.764 | 0.0462 | 1 (1.06) | 0.51 | 0.071–3.678 | 0.5042 | 0 (0) | 0.287 | 0.018–4.641 | 0.3792 |
| Wound Disruption n (%) | 215 (2.26) | 23 (4.6) | 2.085 | 1.343–3.237 | 0.0011 | 1 (1.06) | 0.464 | 0.064–3.345 | 0.4462 | 3 (3.66) | 1.639 | 0.513–5.232 | 0.4042 |
| Hematoma n (%) | 233 (2.45) | 19 (3.8) | 1.57 | 0.975–5.529 | 0.0635 | 1 (1.06) | 0.427 | 0.059–3.079 | 0.3988 | 1 (1.22) | 0.491 | 0.068–3.541 | 0.48 |
| PE n (%) | 175 (1.84) | 14 (2.8) | 1.534 | 0.883–2.664 | 0.1287 | 0 (0) | 0.281 | 0.017–4.543 | 0.3713 | 0 (0) | 0.322 | 0.0200–5.210 | 0.4248 |
| Transfusion n (%) | 454 (4.78) | 44 (8.8) | 1.921 | 1.390–2.655 | 0.0001 | 6 (6.38) | 1.358 | 0.581–3.121 | 0.4716 | 5 (6.10) | 1.293 | 0.521–3.210 | 0.5797 |
| Readmission n (%) | 2758 (29.05) | 192 (38.4) | 1.523 | 1.265–1.833 | < 0.0001 | 26 (27.66) | 0.934 | 0.593–1.471 | 0.768 | 23 (28.05) | 0.952 | 0.587–1.545 | 0.842 |
Concerning medical complications in the unmatched cohort, 30 days after hip surgery, Black patients had significantly increased rates of AKI with decreased rates of arrhythmia. Asian patients had double the rate of pneumonia. Ninety days after hip surgery, Black patients maintained increased odds of AKI and decreased rates of arrhythmia, now with decreased rates of UTI as well. One year after surgery, Black patients retained higher odds of AKI, as well as lower odds of arrhythmia and UTI, with a new association of decreased rates of pneumonia. Asian patients experienced decreased odds of arrhythmia at one year [Table 3]. No significant difference was observed for revision rates two years after hip surgery in the unmatched cohort [Table 4].
| Complications at 30 Days | |||||||||||||
| White | Black | Odds Ratio | CI | P Value | Asian | Odds Ratio | CI | P Value | Hispanic | Odds Ratio | CI | P Value | |
| MI n (%) | 263 (2.77) | 10 (2) | 0.716 | 0.379–1.356 | 0.3503 | 4 (4.26) | 1.56 | 0.569–4.279 | 0.3877 | 3 (3.66) | 1.333 | 0.418–4.249 | 0.6271 |
| Anemia n (%) | 326 (3.43) | 20 (4) | 1.453 | 0.913–2.310 | 0.1147 | 3 (3.19) | 0.927 | 0.292–2.944 | 0.8979 | 1 (1.22) | 0.347 | 0.048–2.503 | 0.2938 |
| Arrythmia n (%) | 1694 (17.84) | 66 (13.2) | 0.7 | 0.538–0.912 | 0.0082 | 12 (12.77) | 0.674 | 0.367–1.238 | 0.2032 | 12 (14.63) | 0.789 | 0.427–1.460 | 0.4506 |
| CHF n (%) | 618 (6.51) | 30 (6) | 0.917 | 0.628–1.338 | 0.6522 | 8 (8.51) | 1.336 | 0.645–2.770 | 0.4361 | 7 (8.54) | 1.341 | 0.615–2.921 | 0.4609 |
| AKI n (%) | 532 (5.60) | 46 (9.2) | 1.707 | 1.245–2.340 | 0.0009 | 4 (4.26) | 0.749 | 0.274–2.046 | 0.5726 | 4 (4.88) | 0.864 | 0.315–2.369 | 0.7762 |
| Cardiac Arrest n (%) | 15 (0.16) | 0 (0) | 0.611 | 0.037–10.226 | 0.7318 | 0 (0) | 3.236 | 0.192–54.478 | 0.415 | 0 (0) | 3.707 | 0.220–62.467 | 0.3633 |
| Pneumonia n (%) | 404 (4.23) | 14 (2.8) | 0.648 | 0.378–1.113 | 0.1159 | 8 (8.51) | 2.093 | 1.007–4.349 | 0.0477 | 5 (6.10) | 1.461 | 0.588–3.630 | 0.4142 |
| UTI n (%) | 1236 (13.02) | 37 (7.4) | 1.124 | 0.870–1.453 | 0.3723 | 13 (13.83) | 1.072 | 0.595–1.932 | 0.8162 | 7 (8.54) | 0.624 | 0.287–1.356 | 0.2335 |
| Complications at 90 Days | |||||||||||||
| White | Black | Odds Ratio | CI | P Value | Asian | Odds Ratio | CI | P Value | Hispanic | Odds Ratio | CI | P Value | |
| MI n (%) | 391 (4.12) | 12 (2.4) | 0.573 | 0.320–1.024 | 0.0602 | 4 (4.26) | 1.035 | 0.378–2.831 | 0.947 | 3 (3.66) | 0.884 | 0.278–2.813 | 0.8347 |
| Anemia n (%) | 425 (4.48) | 26 (5.2) | 1.171 | 0.779–1.758 | 0.448 | 3 (3.19) | 0.704 | 0.222–2.231 | 0.5503 | 2 (2.44) | 0.554 | 0.131–2.178 | 0.3812 |
| Arrythmia n (%) | 2036 (21.45) | 82 (16.4) | 0.719 | 0.564–0.915 | 0.0074 | 15 (15.96) | 0.696 | 0.400–1.211 | 0.1991 | 14 (17.07) | 0.754 | 0.423–1.343 | 0.3381 |
| CHF n (%) | 799 (8.42) | 44 (8.8) | 1.05 | 0.764–1.443 | 0.7632 | 6 (6.38) | 0.742 | 0.324–1.702 | 0.4811 | 8 (9.76) | 1.177 | 0.565–2.449 | 0.6639 |
| AKI n (%) | 648 (6.83) | 55 (11) | 1.687 | 1.261–2.258 | 0.004 | 4 (4.26) | 0.607 | 0.222–1.657 | 0.3297 | 5 (6.10) | 0.886 | 0.358–2.198 | 0.7947 |
| Cardiac Arrest n (%) | 20 (0.21) | 1 (0.2) | 0.949 | 0.127–7.088 | 0.9595 | 0 (0) | 2.445 | 0.147–40.730 | 0.5332 | 0 (0) | 2.801 | 0.168–46.703 | 0.4731 |
| Pneumonia n (%) | 574 (6.05) | 24 (4.8) | 0.784 | 0.516–1.191 | 0.2534 | 9 (9.57) | 1.645 | 0.824–3.288 | 0.1585 | 6 (7.32) | 1.227 | 0.532–2.829 | 0.6315 |
| UTI n (%) | 1824 (19.21) | 63 (12.6) | 0.606 | 0.463–0.793 | 0.0003 | 20 (21.28) | 1.137 | 0.692–1.867 | 0.6135 | 15 (18.29) | 0.941 | 0.537–1.652 | 0.8333 |
| Complications at One Year | |||||||||||||
| s | White | Black | Odds Ratio | CI | P Value | Asian | Odds Ratio | CI | P Value | Hispanic | Odds Ratio | CI | P Value |
| MI n (%) | 701 (7.39) | 35 (7) | 0.944 | 0.664–1.343 | 0.7489 | 5 (5.32) | 0.705 | 0.285–1.740 | 0.448 | 7 (8.54) | 1.171 | 0.538–2.550 | 0.6915 |
| Anemia n (%) | 617 (6.50) | 32 (6.4) | 0.984 | 0.681–1.420 | 0.9303 | 6 (6.38) | 0.981 | 0.427–2.252 | 0.9638 | 5 (6.10) | 0.934 | 0.377–2.317 | 0.8833 |
| Arrythmia n (%) | 2844 (29.96) | 127 (25.4) | 0.796 | 0.648–0.978 | 0.0301 | 19 (20.21) | 0.592 | 0.357–0.982 | 0.0422 | 21 (25.61) | 0.805 | 0.489–1.324 | 0.393 |
| CHF n (%) | 986 (10.39) | 55 (11) | 1.067 | 0.800–1.422 | 0.6611 | 9 (9.57) | 0.914 | 0.458–1.822 | 0.7976 | 8 (9.76) | 0.933 | 0.449–1.940 | 0.852 |
| AKI n (%) | 989 (10.42) | 78 (15.6) | 1.59 | 1.237–2.042 | 0.0003 | 7 (7.45) | 0.692 | 0.320–1.499 | 0.35 | 10 (12.20) | 1.194 | 0.614–2.322 | 0.6004 |
| Cardiac Arrest n (%) | 46 (0.48) | 2 (0.4) | 0.825 | 0.200–3.408 | 0.7902 | 0 (0) | 1.075 | 0.066–17.575 | 0.9595 | 0 (0) | 1.232 | 0.075–20.123 | 0.8839 |
| Pneumonia n (%) | 1503 (15.83) | 50 (10) | 0.591 | 0.439–0.795 | 0.0005 | 13 (13.83) | 0.853 | 0.474–1.537 | 0.597 | 10 (12.20) | 0.738 | 0.380–1.434 | 0.3706 |
| UTI n (%) | 2966 (31.24) | 108 (21.6) | 0.606 | 0.488–0.754 | < 0.0001 | 26 (27.66) | 1.136 | 0.740–1.744 | 0.5601 | 28 (34.15) | 1.141 | 0.722–1.805 | 0.5723 |
| White | Black | Odds Ratio | CI | P Value | Asian | Odds Ratio | CI | P Value | Hispanic | Odds Ratio | CI | P Value | |
| Revisions n (%) | 218 (2.30) | 17 (3.4) | 1.498 | 0.907–2.474 | 0.1148 | 3 (3.19) | 1.403 | 0.441–4.465 | 0.5667 | 0 (0) | 0.257 | 0.0160–4.162 | 0.3391 |
Black patients in the unmatched cohort had significantly greater costs at every time point, ranging from 20 to 68 %. Hispanic patients had 56 % increased costs at one year [Table 5]. There was no significant difference between the time from admission to surgery for any racial groups Table 6. Each patient was matched according to age, gender, CCI, and ECI, as well as comorbidities including obesity, diabetes, and tobacco usage. The method of matching is summarized by Figs. 1–3.
| White | Black | Comparison | CI | P Value | Asian | Comparison | CI | P Value | Hispanic | Comparison | CI | P Value | |
| 30 days avg (SD) | 63,815.54 (±51,593.67) | 76,726.99 (±60,765.85) | 1.202 | 1.13–1.28 | < 0.0001 | 63,793.53 (±62,426.37) | 1.000 | 0.84–1.16 | 0.9967 | 61,310.87 (±54,409.08) | 0.961 | 0.78–1.14 | 0.6618 |
| 90 days avg (SD) | 75, 758.78 (±67,282.80) | 101,173.81 (±99,433.73) | 1.335 | 1.25–1.42 | < 0.0001 | 74,413.55 (±70,121.03) | 0.982 | 0.80–1.16 | 0.8471 | 78,904.98 (±69,830.27) | 1.042 | 0.77 to 1.16 | 0.7084 |
| 1 year avg (SD) | 133,754.41 (±181,640.20) | 224,673.71 (±378,393.70) | 1.680 | 1.548–1.812 | < 0.0001 | 147,404.19 (±238,893.80) | 1.102 | 0.825–1.379 | 0.47 | 208,788.40 (±355,520.00) | 1.561 | 1.26–1.86 | 0.0002 |
| White | Black | Comparison | CI | P Value | Asian | Comparison | CI | P Value | Hispanic | Comparison | CI | P Value | |
| avg (SD) | 261.96 (679.14) | 246.02 (677.97) | 0.939 | 0.43–1.44 | 0.7997 | 220.32 (684.69) | 0.841 | −0.03–1.65 | 0.6578 | 164.25 (498.07) | 0.627 | −0.72–1.53 | 0.3015 |



The matching protocol produced 265 White and 265 Black patients, 61 White and 61 Asian patients, and 39 White and 39 Hispanic patients [Table 7]. Regarding post-operative surgical complications at 90 days, matched Black patients had nearly four times the odds of requiring a blood transfusion [Table 8]. Concerning post-operative medical complications at 30 days, matched Black and Hispanic patients had decreased rates of UTI. At the 90-day post-operative period, matched Black patients had almost four times the likelihood of developing anemia. Matched Black patients had decreased odds of MI and UTI. At the one-year post-operative mark, matched Black patients developed protection against pneumonia and maintained decreased odds of UTI [Table 9]. After matching there was no significant differences in the rate of revision surgery when comparing White patients and minority patients [Table 10]. At 90 and 365 days respectively, matched Black patients had a significant 19 % and 46 % increase in total costs compared to White patients [Table 11]. Following the matching protocol, no statistically significant differences were noted for time to admission [Table 12].
| White | Black | Comparison | CI | PValue | White | Asian | Comparison | CI | PValue | White | Hispanic | Comparison | CI | P Value | |
| Total n (%) | 265 (36.3) | 265 (36.3) | 61 (8.4) | 61 (8.4) | 39 (5.3) | 39 (5.3) | |||||||||
| Under 65 n (%) | 57 (21.5) | 57 (21.5) | 1.000 | 0.661–1.514 | 1.0000 | N/A | N/A | N/A | N/A | ||||||
| Male n (%) | 73 (27.5) | 73 (27.5) | 1.000 | 0.683–1.464 | 1.0000 | 10 (16.4) | 10 (16.4) | 1.000 | 0.383–2.608 | 1.0000 | 4 (10.3) | 4 (10.3) | 1.000 | 0.232–4.319 | 1.0000 |
| CCI avg (SD) | 2.57 (2.35) | 2.57 (2.35) | 1.000 | 0.844–1.156 | 1.0000 | 3.25 (2.72) | 3.25 (2.72) | 1.000 | 0.700–1.300 | 1.0000 | 4.26 (3.36) | 4.26 (3.36) | 1.0000 | 0.644–1.356 | 1.000 |
| ECI avg (SD) | 6.8 (4.15) | 6.8 (4.15) | 1.000 | 0.896–1.104 | 1.0000 | 8.69 (4.45) | 8.69 (4.45) | 1.000 | 0.816–1.184 | 1.0000 | 9.82 (4.35) | 9.82 (4.35) | 1.0000 | 0.800–1.998 | 1.000 |
| LOS avg (SD) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | |||||||||
| Obesity n (%) | 123 (46.4) | 123 (46.4) | 1.000 | 1.162–2.443 | 1.0000 | N/A | N/A | 12 (30.8) | 12 (30.8) | 1.000 | 0.382–2.616 | 1.0000 | |||
| Diabetes n (%) | 177 (66.8) | 177 (66.8) | 1.000 | 0.697–1.436 | 1.0000 | 45 (73.8) | 45 (73.8) | 1.000 | 0.446–2.241 | 1.0000 | 32 (82.1) | 32 (82.1) | 1.000 | 0.315–3.179 | 1.0000 |
| Tobacco Use n (%) | 134 (50.6) | 134 (50.6) | 1.000 | 0.711–1.406 | 1.0000 | 12 (19.7) | 12 (19.7) | 1.000 | 0.410–2.442 | 1.0000 | 15 (38.5) | 15 (38.5) | 1.000 | 0.402–2.490 | 1.0000 |
| Complications at 30 Days | |||||||||||||||
| White | Black | Odds Ratio | CI | P Value | White | Asian | Odds Ratio | CI | P Value | White | Hispanic | Odds Ratio | CI | P Value | |
| SSI n (%) | 7 (2.64) | 5 (1.89) | 0.7088 | 0.2221 to 2.2622 | 0.561 | 0 (0) | 5 (8.20) | 11.9735 | 0.6474 to 221.4612 | 0.0953 | 1 (2.56) | 1 (2.56) | 1 | 0.0603 to 16.5783 | 1 |
| DVT n (%) | 2 (0.75) | 1 (0.38) | 0.4981 | 0.0449 to 5.5269 | 0.5703 | 2 (3.28) | 0 (0) | 0.1935 | 0.0091 to 4.1155 | 0.2923 | 1 (2.56) | 0 (0) | 0.3249 | 0.0128 to 8.2239 | 0.4953 |
| Wound Disruption n (%) | 4 (1.51) | 2 (0.75) | 0.4962 | 0.0901 to 2.7326 | 0.4208 | 0 (0) | 0 (0) | 1 | 0.0195 to 51.2084 | 1 | 1 (2.56) | 0 (0) | 0.3249 | 0.0128 to 8.2239 | 0.4953 |
| Hematoma n (%) | 1 (0.38) | 6 (2.26) | 6.1158 | 0.7312 to 51.1545 | 0.0947 | 0 (0) | 1 (1.64) | 3.0496 | 0.1218 to 76.3471 | 0.4974 | 0 (0) | 1 (2.56) | 3.0779 | 0.1216 to 77.9101 | 0.4953 |
| PE n (%) | 1 (0.38) | 3 (1.13) | 3.0229 | 0.3124 to 29.2499 | 0.3394 | 2 (3.28) | 0 (0) | 0.1935 | 0.0091 to 4.1155 | 0.2923 | 0 (0) | 0 (0) | 1 | 0.0194 to 51.6625 | 1 |
| Transfusion n (%) | 2 (0.75) | 7 (2.64) | 3.5678 | 0.7343 to 17.3364 | 0.1148 | 0 (0) | 1 (1.64) | 3.0496 | 0.1218 to 76.3471 | 0.4974 | 0 (0) | 1 (2.56) | 3.0779 | 0.1216 to 77.9101 | 0.4953 |
| Complications at 90 Days | |||||||||||||||
| White | Black | Odds Ratio | CI | P Value | White | Asian | Odds Ratio | CI | P Value | White | Hispanic | Odds Ratio | CI | P Value | |
| SSI n (%) | 11 (4.15) | 7 (2.64) | 0.6265 | 0.2391 to 1.6417 | 0.3414 | 1 (1.64) | 6 (9.84) | 6.5455 | 0.7636 to 56.1035 | 0.0865 | 1 (2.56) | 1 (2.56) | 1 | 0.0603 to 16.5783 | 1 |
| DVT n (%) | 2 (0.75) | 3 (1.13) | 1.5057 | 0.2495 to 9.0853 | 0.6554 | 2 (3.28) | 0 (0) | 0.1935 | 0.0091 to 4.1155 | 0.2923 | 1 (2.56) | 0 (0) | 0.3249 | 0.0128 to 8.2239 | 0.4953 |
| Wound Disruption n (%) | 4 (1.51) | 5 (1.89) | 1.2548 | 0.3332 to 4.7254 | 0.7372 | 0 (0) | 0 (0) | 1 | 0.0195 to 51.2084 | 1 | 1 (2.56) | 0 (0) | 0.3249 | 0.0128 to 8.2239 | 0.4953 |
| Hematoma n (%) | 4 (1.51) | 7 (2.64) | 1.7703 | 0.5121 to 6.1206 | 0.3668 | 1 (1.64) | 1 (1.64) | 1 | 0.1937 to 5.1614 | 1 | 0 (0) | 1 (2.56) | 3.0779 | 0.1216 to 77.9101 | 0.4953 |
| PE n (%) | 4 (1.51) | 6 (2.26) | 1.5116 | 0.4216 to 5.4192 | 0.5259 | 2 (3.28) | 0 (0) | 0.1935 | 0.0091 to 4.1155 | 0.2923 | 0 (0) | 0 (0) | 1 | 0.0194 to 51.6625 | 1 |
| Transfusion n (%) | 3 (1.13) | 11 (4.15) | 3.7822 | 1.0430 to 13.7156 | 0.043 | 1 (1.64) | 3 (4.92) | 3.1034 | 0.3137 to 30.7019 | 0.3328 | 1 (2.56) | 1 (2.56) | 1 | 0.0603 to 16.5783 | 1 |
| Complications at One Year | |||||||||||||||
| White | Black | Odds Ratio | CI | P Value | White | Asian | Odds Ratio | CI | P Value | White | Hispanic | Odds Ratio | CI | P Value | |
| SSI n (%) | 17 (6.42) | 9 (3.40) | 0.5129 | 0.2244 to 1.1722 | 0.1134 | 2 (3.28) | 6 (9.84) | 3.2182 | 0.6230 to 16.6228 | 0.163 | 1 (2.56) | 2 (5.13) | 2.0541 | 0.1785 to 23.6326 | 0.5636 |
| DVT n (%) | 4 (1.51) | 8 (3.02) | 2.0311 | 0.6041 to 6.8287 | 0.252 | 3 (4.92) | 0 (0) | 0.1359 | 0.0069 to 2.6882 | 0.19 | 1 (2.56) | 0 (0) | 0.3249 | 0.0128 to 8.2239 | 0.4953 |
| Wound Disruption n (%) | 4 (1.51) | 6 (2.26) | 1.5116 | 0.4216 to 5.4192 | 0.5259 | 0 (0) | 0 (0) | 1 | 0.0195 to 51.2084 | 1 | 1 (2.56) | 2 (5.13) | 2.0541 | 0.1785 to 23.6326 | 0.5636 |
| Hematoma n (%) | 4 (1.51) | 7 (2.64) | 1.7703 | 0.5121 to 6.1206 | 0.3668 | 1 (1.64) | 1 (1.64) | 1 | 0.1937 to 5.1614 | 1 | 1 (2.56) | 2 (5.13) | 3.0779 | 0.1216 to 77.9101 | 0.4953 |
| PE n (%) | 8 (3.02) | 7 (2.64) | 0.8716 | 0.3115 to 2.4390 | 0.7935 | 2 (3.28) | 0 (0) | 0.1935 | 0.0091 to 4.1155 | 0.2923 | 0 (0) | 0 (0) | 1 | 0.0194 to 51.6625 | 1 |
| Transfusion n (%) | 8 (3.02) | 17 (6.42) | 2.2021 | 0.9335 to 5.1949 | 0.0714 | 3 (4.92) | 4 (6.56) | 1.3567 | 0.2906 to 6.3347 | 0.698 | 1 (2.56) | 2 (5.13) | 2.0541 | 0.1785 to 23.6326 | 0.5636 |
| Complications at 30 Days | |||||||||||||||
| White | Black | Odds Ratio | CI | P Value | White | Asian | Odds Ratio | CI | P Value | White | Hispanic | Odds Ratio | CI | P Value | |
| MI n (%) | 6 (2.26) | 4 (1.51) | 0.6616 | 0.1845 to 2.3718 | 0.5259 | 3 (4.92) | 4 (6.56) | 1.3567 | 0.2906 to 6.3347 | 0.698 | 2 (5.13) | 0 (0) | 0.1899 | 0.0088 to 4.0867 | 0.2887 |
| Anemia n (%) | 2 (0.75) | 9 (3.40) | 0.5259 | 0.9893 to 21.6042 | 0.0516 | 4 (6.56) | 3 (4.92) | 0.7371 | 0.1579 to 3.4414 | 0.698 | 1 (2.56) | 0 (0) | 0.3249 | 0.0128 to 8.2239 | 0.4953 |
| Arrhythmia n (%) | 36 (13.58) | 25 (9.43) | 0.6626 | 0.3856 to 1.1387 | 0.1363 | 11 (18.03) | 6 (9.84) | 0.4959 | 0.1708 to 1.4398 | 0.1971 | 10 (25.64) | 6 (15.38) | 0.5273 | 0.1706 to 1.6297 | 0.2663 |
| CHF n (%) | 13 (4.91) | 9 (3.40) | 0.6815 | 0.2862 to 1.6226 | 0.3863 | 4 (6.56) | 4 (6.56) | 1 | 0.2384 to 4.1943 | 1 | 4 (10.26) | 4 (10.26) | 1 | 0.2315 to 4.3188 | 1 |
| AKI n (%) | 23 (8.67) | 17 (6.42) | 0.7212 | 0.3760 to 1.3836 | 0.3255 | 2 (3.28) | 3 (4.92) | 1.5259 | 0.2459 to 9.4699 | 0.6501 | 3 (7.69) | 0 (0) | 0.132 | 0.0066 to 2.6441 | 0.1854 |
| Cardiac Arrest n (%) | 1 (0.38) | 0 (0) | 0.3321 | 0.0135 to 8.1891 | 0.5002 | 0 (0) | 0 (0) | 1 | 0.0195 to 51.2084 | 1 | 0 (0) | 0 (0) | 1 | 0.0194 to 51.6625 | 1 |
| Pneumonia n (%) | 12 (4.53) | 4 (1.51) | 0.3231 | 0.1028 to 1.0151 | 0.0531 | 0 (1.16) | 6 (9.84) | 14.4054 | 0.7932 to 261.6208 | 0.0713 | 1 (2.56) | 1 (2.56) | 1 | 0.0603 to 16.5783 | 1 |
| UTI n (%) | 42 (15.85) | 15 (5.66) | 0.3186 | 0.1720 to 0.5902 | 0.0003 | 9 (14.75) | 7 (11.48) | 0.749 | 0.2598 to 2.1588 | 0.5925 | 8 (20.51) | 1 (2.56) | 0.102 | 0.0121 to 0.8601 | 0.0359 |
| Readmission n (%) | 21 (7.92) | 35 (13.21) | 1.7681 | 0.9997 to 3.1272 | 0.0501 | 4 (6.56) | 9 (14.75) | 2.4663 | 0.7163 to 8.4915 | 0.1524 | 2 (5.13) | 1 (2.56) | 0.4868 | 0.0423 to 5.6013 | 0.5636 |
| Complications at 90 Days | |||||||||||||||
| White | Black | Odds Ratio | CI | P Value | White | Asian | Odds Ratio | CI | P Value | White | Hispanic | Odds Ratio | CI | P Value | |
| MI n (%) | 14 (5.28) | 4 (1.51) | 0.2748 | 0.0892 to 0.8460 | 0.0244 | 4 (6.56) | 4 (6.56) | 1 | 0.2384 to 4.1943 | 1 | 2 (5.13) | 0 (0) | 0.1899 | 0.0088 to 4.0867 | 0.2887 |
| Anemia n (%) | 3 (1.13) | 11 (4.15) | 3.7822 | 1.0430 to 13.7156 | 0.043 | 4 (6.56) | 3 (4.92) | 0.7371 | 0.1579 to 3.4414 | 0.698 | 1 (2.56) | 0 (0) | 0.3249 | 0.0128 to 8.2239 | 0.4953 |
| Arrhythmia n (%) | 36 (13.58) | 31 (11.70) | 0.8427 | 0.5042 to 1.4085 | 0.5137 | 12 (19.67) | 8 (13.11) | 0.6164 | 0.2324 to 1.6345 | 0.3308 | 1 (2.56) | 0 (0) | 0.6569 | 0.2312 to 1.8665 | 0.4303 |
| CHF n (%) | 14 (5.28) | 14 (5.28) | 1 | 0.4671 to 2.1408 | 1 | 5 (8.20) | 6 (9.84) | 1.2218 | 0.3522 to 4.2382 | 0.7522 | 11 (28.21) | 8 (20.51) | 1.2868 | 0.3183 to 5.2022 | 0.7235 |
| AKI n (%) | 23 (8.68) | 21 (7.92) | 0.9056 | 0.4883 to 1.6795 | 0.7529 | 3 (4.92) | 3 (4.92) | 1 | 0.1937 to 5.1614 | 1 | 4 (10.26) | 5 (12.82) | 0.0999 | 0.0052 to 1.9208 | 0.1267 |
| Cardiac Arrest n (%) | 1 (0.38) | 0 (0) | 0.3321 | 0.0135 to 8.1891 | 0.5002 | 0 (0) | 0 (0) | 1 | 0.0195 to 51.2084 | 1 | 0 (0) | 0 (0) | 1 | 0.0194 to 51.6625 | 1 |
| Pneumonia n (%) | 16 (6.04) | 7 (2.64) | 0.4222 | 0.1708 to 1.0439 | 0.0619 | 5 (8.20) | 6 (9.84) | 1.2218 | 0.3522 to 4.2382 | 0.7522 | 3 (7.69) | 1 (2.56) | 0.3158 | 0.0314 to 3.1772 | 0.3278 |
| UTI n (%) | 42 (15.85) | 26 (9.81) | 0.5776 | 0.3427 to 0.9735 | 0.0393 | 14 (22.95) | 13 (21.31) | 0.9092 | 0.3865 to 2.1387 | 0.8274 | 14 (35.90) | 7 (17.95) | 0.3906 | 0.1371 to 1.1133 | 0.0786 |
| Readmission n (%) | 43 (16.23) | 50 (18.88) | 1.2006 | 0.7666 to 1.8806 | 0.4244 | 7 (11.48) | 10 (16.39) | 1.5126 | 0.5352 to 4.2750 | 0.435 | 4 (10.26) | 1 (2.56) | 0.2303 | 0.0245 to 2.1606 | 0.1986 |
| Complications at One Year | |||||||||||||||
| White | Black | Odds Ratio | CI | P Value | White | Asian | Odds Ratio | CI | P Value | White | Hispanic | Odds Ratio | CI | P Value | |
| MI n (%) | 18 (6.79) | 11 (4.15) | 0.5943 | 0.2751 to 1.2838 | 0.1854 | 4 (6.56) | 4 (6.56) | 1 | 0.2384 to 4.1943 | 1 | 3 (7.69) | 3 (7.69) | 1 | 0.1891 to 5.2893 | 1 |
| Anemia n (%) | 11 (4.15) | 13 (4.91) | 1.1912 | 0.5238 to 2.7091 | 0.6764 | 5 (8.20 | 4 (6.56) | 0.786 | 0.2006 to 3.0793 | 0.7296 | 2 (5.13) | 1 (2.56) | 0.4868 | 0.0423 to 5.6013 | 0.5636 |
| Arrhythmia n (%) | 64 (24.15) | 50 (18.87) | 0.9302 | 0.6048 to 1.4307 | 0.7419 | 17 (27.87) | 11 (18.03) | 0.5694 | 0.2410 to 1.3454 | 0.1993 | 13 (33.33) | 12 (30.77) | 0.8889 | 0.3432 to 2.3023 | 0.8083 |
| CHF n (%) | 18 (6.79) | 20 (7.55) | 1.1202 | 0.5785 to 2.1692 | 0.7364 | 7 (11.48) | 6 (9.84) | 0.8416 | 0.2656 to 2.6665 | 0.7694 | 4 (10.26) | 5 (12.82) | 1.2868 | 0.3183 to 5.2022 | 0.7235 |
| AKI n (%) | 32 (12.08) | 27 (10.19) | 0.826 | 0.4799 to 1.4219 | 0.4903 | 3 (4.92) | 3 (4.92) | 1 | 0.1937 to 5.1614 | 1 | 7 (17.95) | 3 (7.69) | 0.381 | 0.0908 to 1.5981 | 0.1871 |
| Cardiac Arrest n (%) | 2 (0.75) | 0 (0) | 0.1985 | 0.0095 to 4.1545 | 0.2974 | 0 (0) | 0 (0) | 1 | 0.0195 to 51.2084 | 1 | 0 (0) | 0 (0) | 1 | 0.0194 to 51.6625 | 1 |
| Pneumonia n (%) | 32 (12.08) | 15 (5.66) | 0.4369 | 0.2306 to 0.8275 | 0.0111 | 9 (14.75) | 8 (13.11) | 0.8721 | 0.3125 to 2.4340 | 0.7939 | 5 (12.82) | 3 (7.69) | 0.5667 | 0.1257 to 2.5553 | 0.4598 |
| UTI n (%) | 69 (26.03) | 45 (16.98) | 0.581 | 0.3810 to 0.8861 | 0.0117 | 23 (37.70) | 23 (37.70) | 1 | 0.4808 to 2.0798 | 1 | 20 (51.28) | 13 (33.33) | 0.475 | 0.1902 to 1.1862 | 0.1109 |
| Readmission n (%) | 91 (34.34) | 90 (33.96) | 0.9834 | 0.6867 to 1.4082 | 0.927 | 11 (18.03) | 18 (29.51) | 1.9027 | 0.8102 to 4.4683 | 0.1397 | 10 (18.01) | 4 (10.26) | 0.3314 | 0.0940 to 1.1680 | 0.0857 |
| White | Black | Odds Ratio | CI | P Value | White | Asian | Odds Ratio | CI | P Value | White | Hispanic | Odds Ratio | CI | P Value | |
| Revisions n (%) | 6 (2.26) | 8 (3.02) | 1.349 | 0.4615 to 3.9427 | 0.5844 | 1 (1.64) | 3 (4.92) | 3.1034 | 0.3137 to 30.7019 | 0.3328 | 0 (0) | 0 (0) |
| White | Black | Comparison | CI | P Value | White | Asian | Comparison | CI | P Value | White | Hispanic | Comparison | CI | P Value | |
| 30 days avg (SD) | 64,868.99 (±48,716.73) | 72,124.03 (±44,027.79) | 1.112 | 0.77–1.01 | 0.0727 | 65,678.61 (±49,608.78) | 60,877.87 (±61,367.30) | 0.927 | 0.77–1.38 | 0.6355 | 64,926.33 (±74,129.00) | 53,402.03 (±46,617.25) | 0.823 | 0.75–1.61 | 0.4137 |
| 90 days avg (SD) | 74,269.69 (±56,568.46) | 88,578.33 (±71,291.44) | 1.193 | 0.66–0.96 | 0.0108 | 80,768.92 (±74,025.59) | 68,361.67 (±65,523.34) | 0.846 | 0.84–1.46 | 0.3289 | 74,051.31 (±75,268.29) | 65,511.26 (±58,366.31) | 0.885 | 0.71–1.53 | 0.5772 |
| 1 year avg (SD) | 113,200.90 (±84,538.04) | 165,522.07 (±293,407.70) | 1.462 | 0.212–0.863 | 0.0055 | 115,354.20 (±101,714.00) | 98,480.41 (±89,523.17) | 0.854 | 0.849–1.444 | 0.3327 | 121,174.46 (±113,122.20) | 159,560.72 (±244,532.10) | 1.317 | −0.03–1.39 | 0.3764 |
| White | Black | Comparison | CI | P Value | White | Asian | Comparison | CI | P Value | White | Hispanic | Comparison | CI | P Value | |
| avg (SD) | 283.116 (±790.848) | 220.628 (±648.063) | 0.779 | 0.72–1.84 | 0.3203 | 484.096 (±819.656) | 217.838 (±720.756) | 0.450 | 0.95–3.49 | 0.0591 | 384.767 (±616.741) | 170.638 (±559.629) | 0.443 | 0.70–3.81 | 0.1125 |
4 Discussion
We analyzed data on 10,170 patients who underwent hip fracture repair in the PearlDiver database and investigated any potential links between their self-reported race and particular complications, revisions, and total costs. Our findings revealed that there were fewer Hispanic patients who underwent hip fracture repair in the database when compared to Asian patients. This observation was intriguing since Hispanic individuals account for approximately 19 % of the US population, while Asians comprise only 6 %.16 Upon analyzing comorbidities across different racial groups, we observed that Black, Asian, and Hispanic patients all had higher CCI and ECI. This suggests that underrepresented minorities who fracture their hip may have higher rates of comorbid conditions compared to White patients. Additionally, Black patients demonstrated higher rates of obesity, diabetes, and tobacco use, whereas Asian patients had higher diabetes rates, but lower obesity and tobacco use rates. Hispanic patients only had increased diabetes rates. Contrary to our results Elsharydah et al.17 reported that Black patients had a lower CCI score than White patients, and Nayar et al.18 found Black and Asian patients had lower CCI values than White patients. These findings both contrast with the results of previous studies that reported higher comorbidities among racial minority patients.19–22
Previous research has highlighted the necessity of modifying comorbidity indices to better reflect the health status of Asian patients.23 This stems from the possibility of overlooking higher body fat percentages in Asians due to the BMI cutoff for obesity at 30kg/m2.24 It may also be due to excluding non-English speaking Asians can result in an underestimation of smoking prevalence.25 Additionally, our findings showed that Black and Hispanic patients had a 6.7 and 3.5 times higher likelihood of being under 65 years of age, respectively. This observation is significant because it was seen in other studies and suggests there may be factors that decrease overall bone health and advocate for advanced screening protocols in these patient populations.9,26,27
At every time point before matching, Black patients had a higher incidence of transfusion, readmission, and AKI, but a lower incidence of arrhythmia. Within 30 days and 90 days after hip fracture repair, Asian patients had a higher incidence of SSI. At 90 days and one year of surgical repair, Black patients had a higher incidence of wound complications and lower incidence of UTI. The goal of patient matching is to minimize differences in the cohort so that outcome of hip fracture repair is not influenced by baseline variables. As a result, differences in age, gender, ECI, CCI, as well as obesity, diabetes, and tobacco use between minority patient groups and White patients at baseline were matched. Following this process, Black patients had significantly lower rates of UTI at every examined time point. Black patients had higher rates of transfusion and anemia, but decreased rates of MI at 90 days. At the one-year post-operative mark, Black patients also had decreased rates of pneumonia. Previous studies have also reported higher rates of AKI, transfusion, and DVT in Black patients, which is consistent with our findings.28–31 We observed a significantly lower rate of UTI in Black patients, while the opposite was seen in a study by Jarman et al.31 Furthermore, a study by Azin et al.32 found no significant difference in wound complications between White and Black patients following surgery. However, we found an increase in wound complications for Black patients. Finally, we saw that Black patients had a higher rate of hematoma and Asian patients had a higher rate of SSI. Both observations were also found in a previous study done by Coffey et al.33
Black patients are often found to have higher baseline creatinine levels regardless of their kidney function,34 which could lead to over-diagnosis of AKI.35 Some physicians use a “race correction” to adjust creatinine levels, however this can further complicate the measurement of kidney health.36 More research is needed to understand the relationship between creatinine levels and kidney health. Moreover, racial minority patients have higher risk factors for arrhythmia but lower rates of arrhythmia,37–39 therefore further research is needed to better understand this phenomenon.
Black patients had no significant difference in revision rates before or after matching. A study by Klemt et al. observed the opposite trend with an underutilization of revision total joint arthroplasty by Black patients.40 Prior to being matched, Black patients had notably higher costs at all-time intervals, while Hispanic patients had higher costs at the one-year mark. After matching, Black patients had considerably higher total care costs within 90 and 365 days following hip fracture repair. Neither Asian nor Hispanic patients showed significant differences in total costs after hip fracture repair. Other studies have also documented increased hospitalization, emergency room visits, and end-of-life costs for Black patients.41–43
In our study, we observed no significant difference in time to admission among the racial groups. Our results contrast with other studies, which found that Black patients often experience longer times to surgery as compared to White patients.10,18,44 As the incidence of hip fractures continues to increase and the geriatric population continues to increase in both number and proportion, it is paramount to reduce any delay from admission to surgery.45 We observed that unmatched Black patients had a higher likelihood of being readmitted than White patients at every time point measured. This correlates with previous studies that also found Black patients had higher rates of readmission.44,46
We acknowledge that this study has its limitations, as it is a retrospective analysis that involves data collected from multiple institutions. As a result, there may have been discrepancies in diagnostic procedures, treatment approaches, and physician expertise. Furthermore, despite controlling for variables such as age, gender, CCI, ECI, diabetes, obesity, and tobacco use, there are still potential confounding factors that could have influenced our findings, such as insurance coverage or the distance from healthcare facilities.47,48 Since retrospective studies rely on the accuracy of the recorded data, any errors in data entry, especially those based on non-standard or subjective criteria, may have had an impact on the credibility of our results. Moreover, a significant number of patients who underwent hip fracture repair were labeled as “unknown” or “other”, which limited the size of our study. Furthermore, it should be noted that the data gathered from the PearlDiver Database was limited to CPT and ICD-9 codes, which may have resulted in incomplete data retrieval. Additionally, socioeconomic status (SES) could be a significant factor in the racial differences observed, as the two are frequently interconnected.49 Regrettably, the PearlDiver Database did not provide a method for controlling for SES, and we acknowledge its potential impact on our results.
Nevertheless, the PearlDiver Database has a large number of patient records, which increases the study's predictive power. Additionally, the list of complications and total costs investigated in the study were predetermined before data collection began. Finally, the study authors did not evaluate or create the matched patient groups, which helps to minimize potential selection bias.
5 Conclusion
The results of this study indicate that there are notable differences in the outcomes of hip fracture repair, particularly among Black patients. Prior to matching, Black patients had significantly higher rates of AKI, transfusion, and readmission with lower rates of arrhythmia at all time points compared to White patients. Unmatched Asian patients had significantly increased rates of SSI at 30 and 90 days. Both before and after matching, there were no significant differences in the rate of revision surgery at any time period. Unmatched Black patients had significantly higher total costs of hip fracture treatment at all time points, and matched Black patients had significantly increased costs at 90 and 365 days compared to White patients. Unmatched Hispanic patients had higher costs at one year. No differences in cost were observed when comparing matched cohorts of White vs. Asian patients. The purpose of this study is to assist surgeons in identifying patients who are at a higher risk for specific complications and to promote the implementation of preventative measures to minimize healthcare disparities among ethnic minority groups. By taking such measures, we can work towards ensuring equal healthcare opportunities for all patients, regardless of their racial background.
Funding
No funding was received for this work.
Intellectual property
We confirm that we have given due consideration to the protection of intellectual property associated with this work and that there are no impediments to publication, including the timing of publication, with respect to intellectual property. In so doing we confirm that we have followed the regulations of our institutions concerning intellectual property.
Ethics
This project was conducted in accordance with the Journal of Orthopaedics's ethical policies.
Patient Consent
This study used de-identified data, so no personal identifiers were present. As a result, formal consent was not obtained, in accordance with ethical guidelines for non-identifiable information.
Authorship
All listed authors meet the ICMJE criteria. We attest that all authors contributed significantly to the creation of this manuscript, each having fulfilled criteria as established by the ICMJE.
We confirm that the manuscript has been read and approved by all named authors.
We confirm that the order of authors listed in the manuscript has been approved by all named authors.
Contact with the editorial office
We understand that this Corresponding Author is the sole contact for the Editorial process (including EVISE and direct communications with the office). He/she is responsible for communicating with the other authors about progress, submissions of revisions and final approval of proofs.
We confirm that the email address shown below is accessible by the Corresponding Author, is the address to which Corresponding Author's EVISE account is linked, and has been configured to accept email from the editorial office of American Journal of Ophthalmology Case Reports: [Insert email address you wish to use for communication with the journal here].
We understand that this author is the sole contact for the Editorial process (including EVISE and direct communications with the office). He/she is responsible for communicating with the other authors, including the Corresponding Author, about progress, submissions of revisions and final approval of proofs.
CRediT authorship contribution statement
Emmanuella Oduguwa: Methodology, Software, Formal analysis, Investigation, Data curation, Writing – original draft. Vikram A. Aggarwal: Methodology, Software, Formal analysis, Investigation, Data curation, Writing – original draft. Senthil Sambandam: Conceptualization, Supervision, Reviewing and Editing, Project administration. Dane K. Wukich: Conceptualization, Supervision, Reviewing and Editing, Project administration.
References
- Trends in hip fracture mortality in Wisconsin and the United States, 1999-2017. Wis Med J. Mar 2020;119(1):48-51.
- [Google Scholar]
- Hip fracture care and national systems: the United States and Canada. OTA Int. Mar 2020;3(1)
- [Google Scholar]
- Impact of body mass index on adjacent segment disease after lumbar fusion for degenerative spine disease. Neurosurgery. Apr 2015;76(4):396-401.
- [Google Scholar]
- Obesity may be associated with adjacent-segment degeneration after single-level transforaminal lumbar interbody fusion in spinopelvic-mismatched patients with a minimum 2-year follow-up. J Neurosurg Spine Oct 9 2020:1-6.
- [Google Scholar]
- Hip fracture care in Parkinson disease: a retrospective analysis of 1,239 patients. Geriatr Orthop Surg Rehabil. 2022;13
- [Google Scholar]
- Comparison of differences in complications and revision after conversion to total hip arthroplasty from plating vs. Nailing vs. Hemiarthroplasty. Arch Bone Jt Surg. Sep 2022;10(9):785-790.
- [Google Scholar]
- Gender and race/ethnicity differences in hip fracture incidence, morbidity, mortality, and function. Clin Orthop Relat Res. Jul 2011;469(7):1913-1918.
- [Google Scholar]
- Racial disparities are present in the timing of radiographic assessment and surgical treatment of hip fractures. Clin Orthop Relat Res. Mar 2020;478(3):455-461.
- [Google Scholar]
- Charlson comorbidity index and a composite of poor outcomes in COVID-19 patients: a systematic review and meta-analysis. Diabetes Metabol Syndr. Nov-Dec 2020;14(6):2103-2109.
- [Google Scholar]
- Identifying increased risk of readmission and in-hospital mortality using hospital administrative data: the AHRQ elixhauser comorbidity index. Med Care. Jul 2017;55(7):698-705.
- [Google Scholar]
- Obesity: Epidemiology, pathophysiology, and therapeutics. Front. Endocrinol.. Sep 2021;12(1)
- [Google Scholar]
- Racial disparity in the perioperative care for patients undergoing total knee and hip arthroplasty: a retrospective propensity-matched cohort study. J Racial Ethn Health Disparities. Jun 2018;5(3):632-637.
- [Google Scholar]
- Racial disparity in time to surgery and complications for hip fracture patients. Clin Orthop Surg. Dec 2020;12(4):430-434.
- [Google Scholar]
- Racial disparity in surgical complications in New York State. Ann Surg. Aug 2005;242(2):151-155.
- [Google Scholar]
- Does race affect outcomes in total joint arthroplasty? Clin Orthop Relat Res. Nov 2015;473(11):3535-3541.
- [Google Scholar]
- Where do we stand today on racial and ethnic health disparities? An analysis of primary total hip arthroplasty from a 2011-2017 national database. Arthroplast Today. Dec 2020;6(4):872-876.
- [Google Scholar]
- Variations in postoperative complications according to race, ethnicity, and sex in older adults. J Am Geriatr Soc. Sep 2013;61(9):1499-1507.
- [Google Scholar]
- Recalibration and validation of the Charlson comorbidity index in an asian population: the national health insurance service-national sample cohort study. Sci Rep. Aug 13 2020;10(1)
- [Google Scholar]
- Weighing in on the hidden Asian American obesity epidemic. Prev Med. Apr 2015;73:6-9.
- [Google Scholar]
- Smoking prevalence among asian americans: findings from the national latino and asian American study (NLAAS) Publ Health Rep. Nov-Dec 2006;121(6):755-763.
- [Google Scholar]
- The impact of race/ethnicity on preoperative time to hip stabilization procedure after hip fracture. South Med J. May 2010;103(5):414-418.
- [Google Scholar]
- Socioeconomic Status and Dissatisfaction with Health Care Among Chronically Ill African Americans. 2003:742-748.
- [Google Scholar]
- Explaining the racial difference in AKI incidence. J Am Soc Nephrol. Aug 2014;25(8):1834-1841.
- [Google Scholar]
- Racial differences in acute kidney injury of hospitalized adults with diabetes. J Diabet Complicat. Aug 2016;30(6):1129-1136.
- [Google Scholar]
- Racial disparities in the use of blood transfusion in major surgery. BMC Health Serv Res. Mar 11 2014;14:121.
- [Google Scholar]
- The impact of delayed management of fall-related hip fracture management on health outcomes for African American older adults. J Trauma Acute Care Surg. Jun 1 2021;90(6):942-950.
- [Google Scholar]
- Racial disparities in surgery: a cross-specialty matched Comparison between black and white patients. Can J Surg. 2020;64
- [Google Scholar]
- Racial, ethnic, and socioeconomic disparities in estimates of AHRQ patient safety indicators. Med Care. Mar 2005;43(3 Suppl):I48-I57.
- [Google Scholar]
- Higher serum creatinine concentrations in black patients with chronic kidney disease: beyond nutritional status and body composition. Clin J Am Soc Nephrol. Jul 2008;3(4):992-997.
- [Google Scholar]
- Acute kidney injury 2016: diagnosis and diagnostic workup. Crit Care. Sep 27 2016;20(1):299.
- [Google Scholar]
- A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med. Mar 16 1999;130(6):461-470.
- [Google Scholar]
- Incident atrial fibrillation among Asians, Hispanics, blacks, and whites. Circulation. Dec 3 2013;128(23):2470-2477.
- [Google Scholar]
- Differences by race/ethnicity in the prevalence of clinically detected and monitor-detected atrial fibrillation: mesa. Circ Arrhythm Electrophysiol. Jan 2020;13(1)
- [Google Scholar]
- Racial differences in the incidence of and risk factors for atrial fibrillation in older adults: the cardiovascular health study. J Am Geriatr Soc. Feb 2013;61(2):276-280.
- [Google Scholar]
- Minority race and ethnicity is associated with higher complication rates after revision surgery for failed total hip and knee joint arthroplasty. J Arthroplasty. Apr 2021;36(4):1393-1400.
- [Google Scholar]
- Medicare expenditures by race/ethnicity after hospitalization for heart failure with preserved ejection fraction. JACC Heart Fail. May 2018;6(5):388-397.
- [Google Scholar]
- Is 30-day posthospitalization mortality lower among racial/ethnic minorities?: a reexamination. Med Care. Aug 2018;56(8):665-672.
- [Google Scholar]
- US health care spending by race and ethnicity, 2002-2016. JAMA. Aug 17 2021;326(7):649-659.
- [Google Scholar]
- Do disparities in wait times to operative fixation for pathologic fractures of the long bones and 30-day complications exist between black and white patients? A study using the nsqip database. Clin Orthop Relat Res. Jan 1 2022;480(1):57-63.
- [Google Scholar]
- Timing of hip-fracture surgery in elderly patients: literature review and recommendations. Hip Pelvis. Mar 2020;32(1):11-16.
- [Google Scholar]
- Racial and ethnic differences in 30-day hospital readmissions among US adults with diabetes. JAMA Netw Open. Oct 2 2019;2(10)
- [Google Scholar]
- Socioeconomic factors affecting outcomes in total knee and hip arthroplasty: a systematic review on healthcare disparities. Arthroplasty. Oct 3 2022;4(1):36.
- [Google Scholar]
- The role of socioeconomic status on hip fracture. Osteoporos Int. Oct 2006;17(10):1562-1568.
- [Google Scholar]
- Understanding associations among race, socioeconomic status, and health: patterns and prospects. Health Psychol. Apr 2016;35(4):407-411.
- [Google Scholar]

