Abstract
Background:
Vulnerable populations, including patients from a lower socioeconomic status, are at an increased risk for infection, revision surgery, mortality, and complications after total joint arthroplasty (TJA). An effective metric to quantify and compare these populations has not yet been established in the literature. The Area Deprivation Index (ADI) provides a composite area-based indicator of socioeconomic disadvantage consisting of 17 U.S. Census indicators, based on education, employment, housing quality, and poverty. We assessed patient risk factor profiles and performed multivariable regressions of total complications at 30 days, 90 days, and 1 year.
Methods:
A prospectively collected database of 3,024 patients who underwent primary elective total knee arthroplasty or total hip arthroplasty performed by 3 fellowship-trained orthopaedic surgeons from January 1, 2015, through December 31, 2021, at a tertiary health-care center was analyzed. Patients were divided into quintiles (ADI ≤20 [n = 555], ADI 21 to 40 [n = 1,001], ADI 41 to 60 [n = 694], ADI 61 to 80 [n = 396], and ADI 81 to 100 [n = 378]) and into groups based on the national median ADI, ≤47 (n = 1,896) and >47 (n = 1,128).
Results:
Higher quintiles had significantly more females (p = 0.002) and higher incidences of diabetes (p < 0.001), congestive heart failure (p < 0.001), chronic obstructive pulmonary disease (p < 0.001), hypertension (p < 0.001), substance abuse (p < 0.001), and tobacco use (p < 0.001). When accounting for several confounding variables, all ADI quintiles were not associated with increased total complications at 30 days, but age (p = 0.023), female sex (p = 0.019), congestive heart failure (p = 0.032), chronic obstructive pulmonary disease (p = 0.001), hypertension (p = 0.003), and chronic kidney disease (p = 0.010) were associated. At 90 days, ADI > 47 (p = 0.040), female sex (p = 0.035), and congestive heart failure (p = 0.001) were associated with increased total complications.
Conclusions:
Balancing intrinsic factors, such as patient demographic characteristics, and extrinsic factors, such as social determinants of health, may minimize postoperative complications following TJA. The ADI is one tool that can account for several extrinsic factors, and can thus serve as a starting point to improving patient education and management in the setting of TJA.
Level of Evidence:
Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.
Social determinants of health, defined as the conditions in which people are born, grow, live, work, and age, have been associated with disparities in outcomes following total joint arthroplasty (TJA)1-3. Vulnerable populations, including patients from racial minorities and from a lower socioeconomic status, are at an increased risk for infection, revision surgery, mortality, and complications following TJA4-10. Decreased access to health care, implicit racial bias, an interplay of socioeconomic factors such as neighborhood wealth and education, and patient perceptions have been offered as possible but not definitive explanations for the perpetuation of these disparities2,11-16.
One tool that has enabled quantification of socioeconomic status into a usable metric is the Area Deprivation Index (ADI). The ADI provides a composite area-based indicator of socioeconomic disadvantage consisting of 17 U.S. Census indicators, based on education, employment, housing quality, and poverty17. The national ADI provides normalized percentile scores ranging from 1 to 100, with higher scores suggesting increased social disadvantage. Several studies have examined the impact of ADI thresholds that may portend worse outcomes following TJA and have found mixed results overall. Grits et al.18 and Khlopas et al.19 both found that an ADI > 60 was associated with increased odds of nonhome discharge and prolonged length of stay. Mehta et al. showed that ADI > 75 was associated with discharge to an institution rather than home for postoperative care and rehabilitation after total hip arthroplasty (THA)20. One recent analysis found that the ADI did not predict 90-day postoperative emergency department visits after total knee arthroplasty (TKA)21.
Given the lack of consensus with regard to the impact of the ADI on outcomes following TJA, we aimed to analyze patients who underwent TJA at a single institution to examine the potential differences on the basis of ADI represented as quintiles of ADI ≤ 20, ADI 21 to 40, ADI 41 to 60, ADI 61 to 80, and ADI 81 to 100, as well as ADI above and below the national median: ADI ≤ 47 and ADI > 47. We assessed patient risk factor profiles and performed multivariable regressions of total complications at 30 days, 90 days, and 1 year. We hypothesized that higher ADI quintiles would have disparate patient demographic characteristics as well as higher rates of total complications at all time points.
Materials and Methods
Patient Selection
This study involved a prospectively collected database of patients who underwent primary elective TKA or THA performed by 3 fellowship-trained orthopaedic surgeons from January 1, 2015, through December 31, 2021, at a tertiary health-care center. A total of 3,024 patients had complete data, including the ADI and demographic variables, and were included. Another 200 patients were excluded because of missing demographic variables and 50 were excluded because the ADI was not available because of changing living situations. Patients were divided into quintiles of ADI ≤ 20 (n = 555), ADI 21 to 40 (n = 1,001), ADI 41 to 60 (n = 694), ADI 61 to 80 (n = 396), and ADI 81 to 100 (n = 378). This distribution is consistent with a mean of 50 in the United States, which has also been shown to be consistent across geographic areas18. Institutional review board approval with exempt status was given because of the retrospective nature of the study.
ADI
The ADI is based on a measurement created by the U.S. Health Resources & Services Administration, which has been refined and validated down to the Census Block Group neighborhood level21,22. The ADI takes into account theoretical domains of education, income and/or employment, housing, and household characteristics. The variables carrying the most weight include the percentage of the population below 150% of the poverty level (0.1037), median family income (−0.0977), percentage of families below the poverty level (0.0977), percentage of the population ≥25 years of age with no high school education (−0.0970), income disparity (0.0936), and percentage of the population without a telephone (0.0877) (Table I). The weights are based on the methodologies of 2 prior studies used to quantify the ADI23,24. These domains are then ranked from lowest (0) to highest (100), with higher scores suggestive of more disadvantaged groups in a region of interest at the state or national level.
TABLE I.
| Category | Concept | U.S. Census Bureau ACS 5-Year Variable Group | 2000 Singh Coefficient |
|---|---|---|---|
| Poverty | Median family income | B19113 | −0.0977 |
| Income disparity | B19001 | 0.0936 | |
| Percentage of families below the poverty level | B17010 | 0.0977 | |
| Percentage of the population <150% of the poverty level | C17002 | 0.1037 | |
| Percentage of single-parent households with dependents <18 years of age | B09002 | 0.0719 | |
| Percentage of households without a motor vehicle | B25044 | 0.0694 | |
| Percentage of households without a telephone | B25043 | 0.0877 | |
| Percentage of occupied housing units without complete plumbing | B25016 | 0.0510 | |
| Housing | Percentage of owner-occupied housing units | B25003 | −0.0615 |
| Percentage of households with >1 person per room | B25014 | 0.0556 | |
| Median monthly mortgage | B25088 | −0.0770 | |
| Median gross rent | B25064 | −0.0781 | |
| Median home value | B25077 | −0.0688 | |
| Employment | Percentage of employed persons ≥16 years of age in white-collar jobs | C24010 | −0.0874 |
| Percentage of the civilian labor force (≥16 years of age) unemployed | B23025 | 0.0806 | |
| Education | Percentage of the population ≥25 years of age who did not graduate from high school | B15003 | −0.0970 |
| Percentage of the population ≥25 years of age with at least a high school education | B15003 | 0.0849 |
ACS = American Community Survey.
Outcomes
The primary outcomes of the present study were 30-day, 90-day, and 1-year total complications, which comprised emergency department visits, readmissions, aseptic loosening, dislocations, deep venous thromboses, pulmonary emboli, manipulations under anesthesia, periprosthetic joint infections, periprosthetic fractures, and surgical site infections.
Patient Demographic Characteristics
Demographic variables included age, body mass index (BMI), sex, race, alcohol abuse, tobacco use, substance abuse, chronic obstructive pulmonary disease, chronic kidney disease, American Society of Anesthesiologists (ASA) class, hypertension, and congestive heart failure.
Statistical Analysis
Continuous variables were compared using Student t tests. Categorical variables were compared using Pearson chi-square tests. Significance was set at p < 0.05. Data analyses were performed using R software, version 4.1.1 (R Foundation for Statistical Computing).
Results
Patient Demographic Characteristics by ADI Quintile
BMI (p = 0.060), ASA class (p = 0.800), and chronic kidney disease (p = 0.070) were similar among the ADI quintiles. Higher quintiles had more females (p = 0.002) and higher incidences of diabetes (p < 0.001), congestive heart failure (p < 0.001), chronic obstructive pulmonary disease (p < 0.001), hypertension (p < 0.001), substance abuse (p < 0.001), and tobacco use (p < 0.001) (Table II).
TABLE II.
Baseline Characteristics by ADI Quintile
| Variable | ADI | ||||
|---|---|---|---|---|---|
| ≤20 (N = 555) | 21 to 40 (N = 1,001) | 41 to 60 (N = 694) | 61 to 80 (N = 396) | 81 to 100 (N = 378) | |
| Age* (yr) | 64.2 ± 11.1 | 63.3 ± 11.6 | 62.1 ± 11.0 | 63.5 ± 11.0 | 61.7 ± 11.0 |
| BMI group† | |||||
| <20 kg/m2 | 8 (1.4%) | 12 (1.2%) | 9 (1.3%) | 2 (0.5%) | 7 (1.9%) |
| 20 to <30 kg/m2 | 176 (31.7%) | 319 (31.9%) | 189 (27.2%) | 103 (26.0%) | 106 (28.0%) |
| 30 to <40 kg/m2 | 244 (44.0%) | 447 (44.7%) | 310 (44.7%) | 193 (48.7%) | 183 (48.4%) |
| ≥40 kg/m2 | 48 (8.6%) | 95 (9.5%) | 105 (15.1%) | 52 (13.1%) | 48 (12.7%) |
| Unknown | 79 (14.2%) | 128 (12.8%) | 81 (11.7%) | 46 (11.6%) | 34 (9.0%) |
| Sex† | |||||
| Female | 305 (55.0%) | 600 (59.9%) | 438 (63.1%) | 270 (68.2%) | 253 (66.9%) |
| Male | 250 (45.0%) | 401 (40.1%) | 256 (36.9%) | 126 (31.8%) | 125 (33.1%) |
| Race† | |||||
| American Indian or Alaska Native | 3 (0.5%) | 3 (0.3%) | 1 (0.1%) | 1 (0.3%) | 0 (0.0%) |
| Asian | 10 (1.8%) | 12 (1.2%) | 3 (0.4%) | 0 (0.0%) | 0 (0.0%) |
| Black or African American | 100 (18.0%) | 403 (40.3%) | 490 (70.6%) | 305 (77.0%) | 348 (92.1%) |
| White | 420 (75.7%) | 552 (55.1%) | 183 (26.4%) | 79 (19.9%) | 21 (5.6%) |
| Native Hawaiian, other Pacific Islander | 1 (0.2%) | 2 (0.2%) | 0 (0.0%) | 1 (0.3%) | 0 (0.0%) |
| Declined to answer | 6 (1.1%) | 10 (1.0%) | 5 (0.7%) | 2 (0.5%) | 1 (0.3%) |
| Multiracial | 15 (2.7%) | 19 (1.9%) | 11 (1.6%) | 8 (2.0%) | 8 (2.1%) |
| Unknown | 0 (0.0%) | 0 (0.0%) | 1 (0.1%) | 0 (0.0%) | 0 (0.0%) |
| Alcohol abuse† | 240 (43.2%) | 398 (39.8%) | 253 (36.5%) | 137 (34.6%) | 150 (39.7%) |
| Tobacco use† | 137 (24.7%) | 304 (30.4%) | 222 (32.0%) | 161 (40.7%) | 186 (49.2%) |
| Substance abuse† | 33 (5.9%) | 83 (8.3%) | 84 (12.1%) | 57 (14.4%) | 84 (22.2%) |
| Chronic obstructive pulmonary disease† | 18 (3.2%) | 31 (3.1%) | 39 (5.6%) | 32 (8.1%) | 33 (8.7%) |
| Congestive heart failure† | 19 (3.4%) | 59 (5.9%) | 39 (5.6%) | 40 (10.1%) | 31 (8.2%) |
| Hypertension† | 258 (46.5%) | 540 (53.9%) | 418 (60.2%) | 252 (63.6%) | 260 (68.8%) |
| Chronic kidney disease† | 27 (4.9%) | 60 (6.0%) | 53 (7.6%) | 21 (5.3%) | 14 (3.7%) |
| Diabetes† | 68 (12.3%) | 144 (14.4%) | 132 (19.0%) | 84 (21.2%) | 103 (27.2%) |
| ASA class† | |||||
| 1 | 11 (2.0%) | 13 (1.3%) | 8 (1.2%) | 3 (0.8%) | 3 (0.8%) |
| 2 | 279 (50.3%) | 477 (47.7%) | 327 (47.1%) | 195 (49.2%) | 198 (52.4%) |
| 3 | 255 (45.9%) | 492 (49.2%) | 350 (50.4%) | 189 (47.7%) | 174 (46.0%) |
| 4 | 7 (1.3%) | 13 (1.3%) | 6 (0.9%) | 6 (1.5%) | 3 (0.8%) |
| 2E | 2 (0.4%) | 3 (0.3%) | 0 (0.0%) | 2 (0.5%) | 0 (0.0%) |
| 3E | 1 (0.2%) | 2 (0.2%) | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) |
| 4E | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) |
| Unknown | 0 (0.0%) | 1 (0.1%) | 3 (0.4%) | 1 (0.3%) | 0 (0.0%) |
The values are given as the mean and the standard deviation.
The values are given as the number of patients, with the percentage in parentheses.
Multivariable Regression for Total Complications by ADI Quintile
When accounting for several confounding variables, all ADI quintiles were not associated with increased total complications at 30 days, but age (p = 0.023), female sex (p = 0.014), congestive heart failure (p = 0.032), chronic obstructive pulmonary disease (p = 0.001), hypertension (p = 0.003) and chronic kidney disease (p = 0.010) were associated (Table III).
TABLE III.
Multivariable Logistic Regression for 30-Day Outcomes Based on ADI Quintiles*
| Predictor | OR† | P Value‡ |
|---|---|---|
| ADI quintile§ | ||
| 5 vs. 1 | 1.42 (0.80 to 2.52) | 0.150 |
| 4 vs. 1 | 1.31 (0.75 to 2.29) | 0.357 |
| 3 vs. 1 | 1.16 (0.69 to 1.94) | 0.874 |
| 2 vs. 1 | 0.87 (0.53 to 1.42) | 0.055 |
| Race | ||
| American Indian or Alaska Native vs. White | NA | 0.977 |
| Asian vs. White | 0.91 (0.12 to 7.13) | 0.934 |
| Black or African American vs. White | 1.38 (0.95 to 2.00) | 0.927 |
| Declined to answer vs. White | NA | 0.959 |
| Multiracial vs. White | 0.26 (0.03 to 1.90) | 0.954 |
| Native Hawaiian, other Pacific Islander vs. White | NA | 0.981 |
| Age, per year | 0.98 (0.97 to 1.00) | 0.023 |
| Male sex vs. female sex | 0.69 (0.50 to 0.94) | 0.019 |
| ASA class# | ||
| 4 vs. 1 | 2.51 (0.25 to 24.69) | 0.713 |
| 3 vs. 1 | 1.55 (0.20 to 11.89) | 0.393 |
| 2E vs. 1 | 4.97 (0.25 to 98.50) | 0.343 |
| 2 vs. 1 | 1.93 (0.25 to 14.59) | 0.840 |
| BMI group** | ||
| 3 vs. 0 | NA | 0.897 |
| 2 vs. 0 | NA | 0.910 |
| 1 vs. 0 | NA | 0.907 |
| Presence vs. absence of health issue | ||
| Alcohol use | 1.28 (0.95 to 1.74) | 0.106 |
| Tobacco use | 0.80 (0.58 to 1.11) | 0.187 |
| Substance abuse | 1.33 (0.85 to 2.06) | 0.210 |
| Chronic obstructive pulmonary disease | 2.37 (1.44 to 3.88) | 0.001 |
| Congestive heart failure | 1.68 (1.05 to 2.71) | 0.032 |
| Hypertension | 1.70 (1.20 to 2.40) | 0.003 |
| Chronic kidney disease | 1.98 (1.18 to 3.31) | 0.010 |
| Diabetes | 1.05 (0.73 to 1.49) | 0.801 |
NA = not applicable due to limitations in the database.
The values are given as the odds ratio (OR), with the 95% confidence interval in parentheses.
Significant values are shown in bold.
In this category, 5 = ADI 81 to 100, 4 = ADI 61 to 80, 3 = ADI 41 to 60, 2 = ADI 21 to 40, and 1 = ADI 0 to 20.
In this category, ASA 4 = a patient with severe systemic disease that is a constant threat to life, ASA 3 = a patient with severe systemic disease, ASA 2E = a patient with mild systemic disease who needs emergency surgery, ASA 2 = a patient with mild systemic disease, and ASA 1 = a normal healthy patient.
In this category, 3 = BMI ≥ 50 kg/m2, 2 = BMI 40 to <50 kg/m2, 1 = BMI 30 to <40 kg/m2, and 0 = BMI 20 to <30 kg/m2.
At 90 days, female sex (p = 0.033) and congestive heart failure (p = 0.008) were associated with increased total complications (Table IV).
TABLE IV.
Multivariable Logistic Regression for 90-Day Outcomes Based on ADI Quintiles*
| Predictor | OR† | P Value‡ |
|---|---|---|
| ADI quintile§ | ||
| 5 vs. 1 | 1.48 (0.82 to 2.66) | 0.627 |
| 4 vs. 1 | 1.63 (0.93 to 2.86) | 0.235 |
| 3 vs. 1 | 1.64 (0.98 to 2.75) | 0.149 |
| 2 vs. 1 | 1.21 (0.74 to 1.98) | 0.332 |
| Race | ||
| American Indian or Alaska Native vs. White | NA | 0.969 |
| Asian vs. White | 0.86 (0.11 to 6.68) | 0.954 |
| Black or African American vs. White | 1.36 (0.96 to 1.91) | 0.946 |
| Declined to answer vs. White | 0.64 (0.09 to 4.90) | 0.959 |
| Multiracial vs. White | 1.05 (0.36 to 3.02) | 0.950 |
| Native Hawaiian, other Pacific Islander vs. White | NA | 0.975 |
| Age, per year | 0.99 (0.98 to 1.01) | 0.417 |
| Male sex vs. female sex | 0.72 (0.53 to 0.97) | 0.033 |
| ASA class# | ||
| 4 vs. 1 | 3.39 (0.34 to 34.07) | 0.952 |
| 3 vs. 1 | 2.44 (0.32 to 18.50) | 0.959 |
| 2E vs. 1 | NA | 0.962 |
| 2 vs. 1 | 2.34 (0.31 to 17.52) | 0.960 |
| BMI group** | ||
| 3 vs. 0 | 3.34 (0.44 to 25.20) | 0.303 |
| 2 vs. 0 | 3.66 (0.49 to 27.43) | 0.166 |
| 1 vs. 0 | 3.23 (0.43 to 24.35) | 0.359 |
| Presence vs. absence of health issue | ||
| Alcohol use | 0.83 (0.61 to 1.11) | 0.212 |
| Tobacco use | 1.32 (0.97 to 1.79) | 0.082 |
| Substance abuse | 1.19 (0.78 to 1.82) | 0.428 |
| Chronic obstructive pulmonary disease | 1.06 (0.60 to 1.86) | 0.846 |
| Congestive heart failure | 1.92 (1.19 to 3.09) | 0.008 |
| Hypertension | 1.05 (0.77 to 1.44) | 0.748 |
| Chronic kidney disease | 0.82 (0.44 to 1.52) | 0.522 |
| Diabetes | 0.79 (0.54 to 1.15) | 0.211 |
NA = not applicable due to limitations in the database.
The values are given as the odds ratio (OR), with the 95% confidence interval in parentheses.
Significant values are shown in bold.
In this category, 5 = ADI 81 to 100, 4 = ADI 61 to 80, 3 = ADI 41 to 60, 2 = ADI 21 to 40, and 1 = ADI 0 to 20.
In this category, ASA 4 = a patient with severe systemic disease that is a constant threat to life, ASA 3 = a patient with severe systemic disease, ASA 2E = a patient with mild systemic disease who needs emergency surgery, ASA 2 = a patient with mild systemic disease, and ASA 1 = a normal healthy patient.
In this category, 3 = BMI ≥ 50 kg/m2, 2 = BMI 40 to <50 kg/m2, 1 = BMI 30 to <40 kg/m2, and 0 = BMI 20 to <30 kg/m2.
At 1 year, only female sex (p = 0.001) and Black or African American race compared with White race (p < 0.001) were associated with increased total complications (Table V).
TABLE V.
Multivariable Logistic Regression for 1-Year Outcomes Based on ADI Quintiles*
| Predictor | OR† | P Value‡ |
|---|---|---|
| ADI quintile§ | ||
| 5 vs. 1 | 1.24 (0.79 to 1.94) | 0.289 |
| 4 vs. 1 | 1.11 (0.72 to 1.73) | 0.829 |
| 3 vs. 1 | 1.18 (0.79 to 1.75) | 0.428 |
| 2 vs. 1 | 0.92 (0.63 to 1.33) | 0.115 |
| Race | ||
| American Indian or Alaska Native vs. White | NA | 0.960 |
| Asian vs. White | 0.49 (0.06 to 3.72) | 0.956 |
| Black or African American vs. White | 1.52 (1.15 to 2.02) | <0.001 |
| Declined to answer vs. White | 1.42 (0.41 to 4.95) | 0.932 |
| Multiracial vs. White | 1.17 (0.51 to 2.67) | 0.937 |
| Native Hawaiian, other Pacific Islander vs. White | NA | 0.968 |
| Age, per year | 0.99 (0.98 to 1.00) | 0.082 |
| Male sex vs. female sex | 0.67 (0.53 to 0.86) | 0.001 |
| ASA class# | ||
| 4 vs. 1 | 4.83 (0.51 to 45.91) | 0.941 |
| 3 vs. 1 | 3.98 (0.53 to 29.86) | 0.946 |
| 2E vs. 1 | NA | 0.953 |
| 2 vs. 1 | 4.31 (0.58 to 32.10) | 0.944 |
| BMI group** | ||
| 3 vs. 0 | 1.63 (0.48 to 5.54) | 0.463 |
| 2 vs. 0 | 1.69 (0.50 to 5.73) | 0.325 |
| 1 vs. 0 | 1.52 (0.45 to 5.14) | 0.742 |
| Presence vs. absence of health issue | ||
| Alcohol use | 1.10 (0.87 to 1.39) | 0.437 |
| Tobacco use | 1.20 (0.94 to 1.55) | 0.151 |
| Substance abuse | 1.06 (0.74 to 1.50) | 0.767 |
| Chronic obstructive pulmonary disease | 1.24 (0.78 to 1.96) | 0.369 |
| Congestive heart failure | 1.44 (0.94 to 2.20) | 0.090 |
| Hypertension | 1.18 (0.91 to 1.52) | 0.212 |
| Chronic kidney disease | 0.82 (0.49 to 1.38) | 0.461 |
| Diabetes | 1.11 (0.83 to 1.49) | 0.469 |
NA = not applicable due to limitations in the database.
The values are given as the odds ratio (OR), with the 95% confidence interval in parentheses.
Significant values are shown in bold.
In this category, 5 = ADI 81 to 100, 4 = ADI 61 to 80, 3 = ADI 41 to 60, 2 = ADI 21 to 40, and 1 = ADI 0 to 20.
In this category, ASA 4 = a patient with severe systemic disease that is a constant threat to life, ASA 3 = a patient with severe systemic disease, ASA 2E = a patient with mild systemic disease who needs emergency surgery, ASA 2 = a patient with mild systemic disease, and ASA 1 = a normal healthy patient.
In this category, 3 = BMI ≥ 50 kg/m2, 2 = BMI 40 to <50 kg/m2, 1 = BMI 30 to <40 kg/m2, and 0 = BMI 20 to <30 kg/m2.
Patient Demographic Characteristics by ADI > 47 Versus ≤ 47
Compared with the ADI ≤ 47 cohort, the ADI > 47 cohort had higher incidences of BMI 30 to <40 kg/m2 (52.15% compared with 44.99%; p = 0.001), BMI ≥ 40 kg/m2 (15.38% compared with 10.23%; p < 0.001), female sex (66.13% compared with 59.07%; p = 0.001), Black or African American race (81.03% compared with 38.63%; p < 0.001), tobacco use (41.49% compared with 28.59%; p < 0.001), substance abuse (15.96% compared with 8.49%; p < 0.001), chronic obstructive pulmonary disease (8.60% compared with 4.01%; p < 0.001), and congestive heart failure (6.83% compared with 4.80%; p = 0.020) (Table VI).
TABLE VI.
Baseline Characteristic by High Versus Low ADI
| Variable | ADI > 47 (N = 1,128) | ADI ≤ 47 (N = 1,896) | P Value* |
|---|---|---|---|
| Age† (yr) | 62.1 ± 10.88 | 64.2 ± 11.24 | 0.760 |
| BMI groupठ| |||
| <20 kg/m2 | 13 (1.30%) | 25 (1.32%) | 0.960 |
| 20 to <30 kg/m2 | 312 (31.17%) | 581 (30.64%) | 0.760 |
| 30 to <40 kg/m2 | 522 (52.15%) | 855 (45.09%) | 0.001 |
| ≥40 kg/m2 | 154 (15.38%) | 194 (10.23%) | <0.001 |
| Sex§ | |||
| Female | 746 (66.13%) | 1,120 (59.07%) | 0.001 |
| Male | 382 (33.87%) | 776 (40.93%) | 0.001 |
| Race§ | |||
| American Indian or Alaska Native | 2 (0.18%) | 6 (0.32%) | 0.470 |
| Asian | 2 (0.18%) | 23 (1.21%) | 0.003 |
| Black or African American | 914 (81.03%) | 733 (38.66%) | <0.001 |
| White | 182 (16.13%) | 1,073 (56.62%) | <0.001 |
| Native Hawaiian, other Pacific Islander | 1 (0.09%) | 3 (0.16%) | 0.610 |
| Multiracial | 19 (1.68%) | 42 (2.22%) | 0.310 |
| Declined to answer | 8 (0.71%) | 16 (0.84%) | 0.700 |
| Health issues with available data§ | |||
| Alcohol abuse | 415 (36.79%) | 763 (40.24%) | 0.060 |
| Tobacco use | 468 (41.49%) | 542 (28.59%) | <0.001 |
| Substance abuse | 180 (15.96%) | 161 (8.49%) | <0.001 |
| Chronic obstructive pulmonary disease | 97 (8.60%) | 76 (4.01%) | <0.001 |
| Congestive heart failure | 77 (6.83%) | 91 (4.80%) | 0.020 |
| Diabetes | 266 (23.58%) | 265 (13.98%) | 0.900 |
Significant values are shown in bold.
The values are given as the mean and the standard deviation.
In this category, there were 1,001 patients with data in the ADI > 47 group and 1,655 patients with data in the ADI ≤ 47 group.
The values are given as the number of patients, with the percentage in parentheses.
Multivariable Regression for Total Complications by ADI > 47 Versus ADI ≤ 47
At 30 days, female sex (p = 0.019), congestive heart failure (p = 0.037), chronic obstructive pulmonary disease (p < 0.001), hypertension (p = 0.002), and chronic kidney disease (p = 0.010) were associated with increased total complications (Table VII). At 90 days, ADI > 47 (p = 0.047), female sex (p = 0.035), and congestive heart failure (p = 0.010) were associated with increased total complications (Table VIII). At 1 year, male sex (p = 0.002) was associated with increased total complications (Table IX).
TABLE VII.
Multivariable Logistic Regression for 30-Day Outcomes Based on High Versus Low ADI*
| Predictor | OR† | P Value‡ |
|---|---|---|
| High vs. low ADI§ | 1.32 (0.97 to 1.80) | 0.079 |
| Race | ||
| American Indian or Alaska Native vs. White | NA | 0.977 |
| Asian vs. White | 0.91 (0.12 to 7.10) | 0.934 |
| Black or African American vs. White | 1.44 (1.01 to 2.05) | 0.926 |
| Declined to answer vs. White | NA | 0.959 |
| Multiracial vs. White | 0.26 (0.04 to 1.96) | 0.953 |
| Native Hawaiian, other Pacific Islander vs. White | NA | 0.981 |
| Age, per year | 0.98 (0.97 to 1.00) | 0.025 |
| Male sex vs. female sex | 0.69 (0.50 to 0.94) | 0.019 |
| ASA class# | ||
| 4 vs. 1 | 2.51 (0.26 to 24.63) | 0.704 |
| 3 vs. 1 | 1.56 (0.20 to 11.95) | 0.411 |
| 2E vs. 1 | 4.75 (0.24 to 94.69) | 0.367 |
| 2 vs. 1 | 1.96 (0.26 to 14.81) | 0.886 |
| BMI group** | ||
| 3 vs. 0 | NA | 0.897 |
| 2 vs. 0 | NA | 0.911 |
| 1 vs. 0 | NA | 0.908 |
| Presence vs. absence of health issue | ||
| Alcohol use | 1.28 (0.94 to 1.72) | 0.115 |
| Tobacco use | 0.80 (0.58 to 1.11) | 0.187 |
| Substance abuse | 1.35 (0.87 to 2.09) | 0.178 |
| Chronic obstructive pulmonary disease | 2.42 (1.47 to 3.97) | 0.001 |
| Congestive heart failure | 1.65 (1.03 to 2.66) | 0.037 |
| Hypertension | 1.71 (1.21 to 2.42) | 0.002 |
| Chronic kidney disease | 1.96 (1.18 to 3.28) | 0.010 |
| Diabetes | 1.06 (0.74 to 1.51) | 0.767 |
NA = not applicable due to limitations in the database.
The values are given as the odds ratio (OR), with the 95% confidence interval in parentheses.
Significant values are shown in bold.
In this category, high = ADI > 47 and low = ADI ≤ 47.
In this category, ASA 4 = a patient with severe systemic disease that is a constant threat to life, ASA 3 = a patient with severe systemic disease, ASA 2E = a patient with mild systemic disease who needs emergency surgery, ASA 2 = a patient with mild systemic disease, and ASA 1 = a normal healthy patient.
In this category, 3 = BMI ≥ 50 kg/m2, 2 = BMI 40 to <50 kg/m2, 1 = BMI 30 to <40 kg/m2, and 0 = BMI 20 to <30 kg/m2.
TABLE VIII.
Multivariable Logistic Regression for 90-Day Outcomes Based on High Versus Low ADI*
| Predictor | OR† | P Value‡ |
|---|---|---|
| High vs. low ADI§ | 1.36 (1.00 to 1.83) | 0.047 |
| Race | ||
| American Indian or Alaska Native vs. White | NA | 0.969 |
| Asian vs. White | 0.84 (0.11 to 6.46) | 0.954 |
| Black or African American vs. White | 1.40 (1.01 to 1.96) | 0.946 |
| Declined to answer vs. White | 0.64 (0.08 to 4.86) | 0.959 |
| Multiracial vs. White | 1.06 (0.37 to 3.03) | 0.950 |
| Native Hawaiian, other Pacific Islander vs. White | NA | 0.975 |
| Age, per year | 1.00 (0.98 to 1.01) | 0.444 |
| Male sex vs. female sex | 0.72 (0.54 to 0.98) | 0.035 |
| ASA class# | ||
| 4 vs. 1 | 3.51 (0.35 to 35.16) | 0.952 |
| 3 vs. 1 | 2.51 (0.33 to 19.01) | 0.959 |
| 2E vs. 1 | NA | 0.962 |
| 2 vs. 1 | 2.39 (0.32 to 17.94) | 0.960 |
| BMI group** | ||
| 3 vs. 0 | 3.38 (0.45 to 25.48) | 0.292 |
| 2 vs. 0 | 3.68 (0.49 to 27.64) | 0.163 |
| 1 vs. 0 | 3.23 (0.43 to 24.34) | 0.369 |
| Presence vs. absence of health issue | ||
| Alcohol use | 0.83 (0.61 to 1.11) | 0.209 |
| Tobacco use | 1.30 (0.95 to 1.77) | 0.098 |
| Substance abuse | 1.19 (0.78 to 1.82) | 0.426 |
| Chronic obstructive pulmonary disease | 1.06 (0.60 to 1.87) | 0.836 |
| Congestive heart failure | 1.88 (1.17 to 3.03) | 0.010 |
| Hypertension | 1.06 (0.77 to 1.44) | 0.739 |
| Chronic kidney disease | 0.84 (0.45 to 1.56) | 0.573 |
| Diabetes | 0.79 (0.54 to 1.15) | 0.211 |
NA = not applicable due to limitations in the database.
The values are given as the odds ratio (OR), with the 95% confidence interval in parentheses.
Significant values are shown in bold.
In this category, high = ADI > 47 and low = ADI ≤ 47.
In this category, ASA 4 = a patient with severe systemic disease that is a constant threat to life, ASA 3 = a patient with severe systemic disease, ASA 2E = a patient with mild systemic disease who needs emergency surgery, ASA 2 = a patient with mild systemic disease, and ASA 1 = a normal healthy patient.
In this category, 3 = BMI ≥ 50 kg/m2, 2 = BMI 40 to <50 kg/m2, 1 = BMI 30 to <40 kg/m2, and 0 = BMI 20 to <30 kg/m2.
TABLE IX.
Multivariable Logistic Regression for 1-Year Outcomes Based on High Versus Low ADI*
| Predictor | OR† | P Value‡ |
|---|---|---|
| High vs. low ADI§ | 1.20 (0.94 to 1.53) | 0.148 |
| Race | ||
| American Indian or Alaska Native vs. White | NA | 0.960 |
| Asian vs. White | 0.48 (0.06 to 3.71) | 0.956 |
| Black or African American vs. White | 1.56 (1.19 to 2.05) | 0.930 |
| Declined to answer vs. White | 1.42 (0.41 to 4.96) | 0.932 |
| Multiracial vs. White | 1.18 (0.51 to 2.69) | 0.937 |
| Native Hawaiian, other Pacific Islander vs. White | NA | 0.968 |
| Age, per year | 0.99 (0.98 to 1.00) | 0.089 |
| Male sex vs. female sex | 0.67 (0.53 to 0.86) | 0.002 |
| ASA class# | ||
| 4 vs. 1 | 4.81 (0.51 to 45.65) | 0.941 |
| 3 vs. 1 | 3.98 (0.53 to 29.84) | 0.945 |
| 2E vs. 1 | NA | 0.952 |
| 2 vs. 1 | 4.33 (0.58 to 32.26) | 0.943 |
| BMI group** | ||
| 3 vs. 0 | 1.62 (0.48 to 5.48) | 0.459 |
| 2 vs. 0 | 1.67 (0.49 to 5.64) | 0.337 |
| 1 vs. 0 | 1.49 (0.44 to 5.05) | 0.772 |
| Presence vs. absence of health issues | ||
| Alcohol use | 1.10 (0.87 to 1.40) | 0.426 |
| Tobacco use | 1.20 (0.93 to 1.54) | 0.162 |
| Substance abuse | 1.07 (0.75 to 1.52) | 0.708 |
| Chronic obstructive pulmonary disease | 1.26 (0.79 to 1.99) | 0.330 |
| Congestive heart failure | 1.41 (0.93 to 2.16) | 0.108 |
| Hypertension | 1.18 (0.92 to 1.53) | 0.200 |
| Chronic kidney disease | 0.83 (0.50 to 1.39) | 0.481 |
| Diabetes | 1.12 (0.84 to 1.50) | 0.446 |
NA = not applicable due to limitations in the database.
The values are given as the odds ratio (OR), with the 95% confidence interval in parentheses.
Significant values are shown in bold.
In this category, high = ADI > 47 and low = ADI ≤ 47.
In this category, ASA 4 = a patient with severe systemic disease that is a constant threat to life, ASA 3 = a patient with severe systemic disease, ASA 2E = a patient with mild systemic disease who needs emergency surgery, ASA 2 = a patient with mild systemic disease, and ASA 1 = a normal healthy patient.
In this category, 3 = BMI ≥ 50 kg/m2, 2 = BMI 40 to <50 kg/m2, 1 = BMI 30 to <40 kg/m2, and 0 = BMI 20 to <30 kg/m2.
Discussion
The interplay of socioeconomic factors may contribute to disparate outcomes following TJA for patients from low socioeconomic status groups15,16. The ADI offers a reproducible way of capturing several components of socioeconomic disadvantage, with the potential to tailor management to patients who need additional support. A consensus on a threshold for the ADI following TJA has not yet been determined in the literature. Our major findings were that higher ADI quintiles were associated with worse behavioral risk factor profiles than lower ADI quantiles were and that higher ADI quintiles were not associated with increased risk of total complications at any of the time points, although ADI > 47 was associated with increased total complications at 90 days only.
We acknowledge the limitations that were present in the study. The ADI may have missed some components of an all-encompassing socioeconomic status metric, including health literacy, immigration status, racial segregation, area crime rates, green space, and transportation19,22,23. Because we studied only an American population, our results may not be generalizable globally. Other metrics such as the Social Vulnerability Index may be more useful for larger geographic areas but have an inability to target smaller areas at the level of neighborhoods, which is an advantage of the ADI24. Because the ADI is a composite of 17 elements of deprivation, we were limited in our ability to analyze the effects of individual factors, and, furthermore, such an analysis might not have accounted for overlapping determinants, such as insurance and race. Nevertheless, a breakdown of the individual components of the ADI has been presented in the interests of transparency. The small sample size may have missed potential associations. There were additional confounding variables that may have influenced postoperative outcomes but were not taken into account in our multivariable regression, such as health-care insurance, income, housing insecurity, transportation, education, and nutrition25. A lack of generalizability could have resulted from conducting the study in a single regional health-care system. Although discrepancies in TJA outcomes may exist among the surgeons, strict preoperative, perioperative, and postoperative protocols in our hospital, as well as similar training, mitigate the concern regarding differences in management affecting postoperative outcomes. Another area of focus is the role of TJA access and TJA utilization as important factors in mediating disparities in outcomes following TJA, but that was not a goal of this study. The strength of this study is a single institution’s novel inclusion of 2 different ADI categories (quintiles and the national mean cutoff) that, to our knowledge, have not been analyzed with regard to TJA.
Risk factors for patients from deprived neighborhoods have been well-described in the literature. In 1 analysis of 27,121 patients undergoing THA, the authors found that higher ADI was associated with increased risk of multimorbidity (≥2 chronic conditions) and that ADI may be a relevant proxy for socioeconomic status when an individual’s socioeconomic status is not available26. Kamath et al. found that female sex, non-White race, education of high school or less, current smoking, BMI of >30 kg/m2, more limitations in instrumental activities of daily living, and an ASA class of >2 were characteristics of patients from the most deprived neighborhoods, as represented by the highest quintile of the ADI (80 to 100)27. In our study, we found female sex and several risk factors associated with more deprived neighborhoods (including diabetes, congestive heart failure, chronic obstructive pulmonary disease, hypertension, substance abuse, and tobacco use) to be independent risk factors for total complications at 30 days, 90 days, and 1 year. We suggest that individual-level disadvantage and neighborhood-level disadvantage, as measured by the ADI, have a dynamic relationship in influencing perioperative health status and, ultimately, postoperative outcomes.
A similar, dynamic relationship can be found between intrinsic factors, such as female sex, BMI, and patient comorbidities, and extrinsic factors, such as socioeconomic status (including marital status), access to tobacco stores, and living in “food deserts.” Both intrinsic and extrinsic factors have led to longer length of stay and worse outcomes following TJA1,4,28-31. This is consistent with our findings that patient demographic characteristics, namely female sex and congestive heart failure, and the ADI, a proxy for neighborhood socioeconomic status disadvantage, are both associated with increased total complications following TJA. Appropriate and effective intervention may better serve patients by addressing both intrinsic and extrinsic factors, such as by nutritional support programs that could minimize the effect of BMI and living in a food desert in the same effort32. Additionally, we recommend clinical intervention in the form of preoperative risk factor management programs. A recent article reviewed the advantages and disadvantages of 10 different risk stratification tools to predict readmission and discharge status following TJA33. The authors concluded that individual metrics, such as race, insurance status, income, social support, housing status, and access to care, should be uniformly assessed in the preoperative setting. Our recommendation is the inclusion of multidimensional variables, such as the ADI, in risk assessment programs because of their unique ability to capture comprehensive measures of socioeconomic factors as well as more granular social determinants of health. Once high-risk patients are identified, clinicians can work toward better managing nonmodifiable and modifiable risk factors prior to TJA with the intent of minimizing postoperative complications. For instance, a reversal in differences between Black and White patients in the odds of readmission from 2009 to 2016 that resulted in a lower rate of readmission in 2015 occurred secondary to hospital-based quality improvement initiatives, including publicly available quality measures and efforts by surgeons and hospitals to prepare for value-based contracts and enhance physician cultural competency34.
In several studies, authors have reported thresholds for the ADI that negatively influence outcomes following TJA, but the heterogeneity in study designs limits their generalizability and utility. Our analysis found ADI > 47 to be an independent risk factor for increased total complications at 90 days. Shaw et al. found that the ADI did not influence 90-day postoperative emergency department visit after TKA in a study with similar cohort sizes (3,024 in our cohort compared with 2,655 in their cohort). However, our patient population (in Baltimore, Maryland) was inherently different from that in the study by Shaw et al. (in Detroit, Michigan) and represented more disadvantaged neighborhoods (as indicated by the differences in male versus female sex, insurance type, ASA class, and congestive heart failure), which may have led to the dissimilar outcomes21. Similar to our study, Khlopas et al. also studied a more disadvantaged population, as shown by the number of female patients, patients who smoked, African American patients, and younger patients. They found that an ADI of 61 to 70 compared with an ADI of 31 to 40 was associated with higher odds of a ≥3-day length of stay and of nonhome discharge, but not with 90-day emergency department visits or reoperations19. Grits et al. showed that individuals in the ADI 61 to 80 quintile had higher odds of nonhome discharge compared with the ADI 21 to 40 group18. These findings are consistent with our finding that an ADI threshold of >47 was associated with a higher total rate of postoperative complications, albeit at different time intervals. In order to maximize patient care within this group with higher-than-average ADI, improvements can be made in better identifying vulnerable patient populations, prioritizing patient education, providing home health, and setting appropriate patient expectations35-37.
Improving outcomes following TJA may be approached by better balancing intrinsic factors, such as patient demographic characteristics, and extrinsic factors, such as social determinants of health. The ADI is 1 tool that can account for several of the extrinsic factors and can be applied to different institutions. It can serve as a starting point to improving patient education and management in the setting of TJA. In a clinical setting, physicians can correlate patients’ 9-digit ZIP code with their ADI in order to obtain the percentile into which these patients fall. Then patients can be risk-stratified on the basis of their ADI score, and physicians can discuss their individual level of risk based on previous data. If a vulnerable patient is identified, interventions can be identified to mitigate the risk associated with undergoing TJA. For this to occur, the 17 indicators within the ADI need to be assessed for each patient and discussed in order to provide the patient with the best possible care. The ease of use, reproducibility, and multidimensional nature of the ADI enable its effective implementation as a marker in patient optimization. The consistent use of a single tool to measure neighborhood disadvantage works toward achieving these goals.
Footnotes
Investigation performed at the Rubin Institute for Advanced Orthopedics, Sinai Hospital of Baltimore, Lifebridge Health, Baltimore, Maryland
Disclosure: No external funding was received for the work. The Disclosure of Potential Conflicts of Interest forms are provided with the online version of the article (http://links.lww.com/JBJSOA/A616).
Contributor Information
Jeremy A. Dubin, Email: dubinjeremy@gmail.com.
Sandeep S. Bains, Email: sbains@lifebridgehealth.org.
Daniel Hameed, Email: dhameed@lifebridgehealth.org.
Rubén Monárrez, Email: monarrezrg@gmail.com.
Ruby Gilmor, Email: rgilmor@lifebridgehealth.org.
Zhongming Chen, Email: zchen2@lifebridgehealth.org.
James Nace, Email: nace9184@yahoo.com.
References
- 1.Delanois RE, Tarazi JM, Wilkie WA, Remily E, Salem HS, Mohamed NS, Pollack AN, Mont MA. Social determinants of health in total knee arthroplasty: are social factors associated with increased 30-day post-discharge cost of care and length of stay? Bone Joint J. 2021. Jun;103-B(6)(Supple A):113-8. [DOI] [PubMed] [Google Scholar]
- 2.Rubenstein WJ, Harris AHS, Hwang KM, Giori NJ, Kuo AC. Social determinants of health and patient-reported outcomes following total hip and knee arthroplasty in veterans. J Arthroplasty. 2020. Sep;35(9):2357-62. [DOI] [PubMed] [Google Scholar]
- 3.Delanois RE, Sax OC, Wilkie WA, Douglas SJ, Mohamed NS, Mont MA. Social determinants of health in total hip arthroplasty: are they associated with costs, lengths of stay, and patient reported outcomes? J Arthroplasty. 2022. Jul;37(7S):S422-7. [DOI] [PubMed] [Google Scholar]
- 4.Arroyo NS, White RS, Gaber-Baylis LK, La M, Fisher AD, Samaru M. Racial/ethnic and socioeconomic disparities in total knee arthroplasty 30- and 90-day readmissions: a multi-payer and multistate analysis, 2007-2014. Popul Health Manag. 2019. Apr;22(2):175-85. [DOI] [PubMed] [Google Scholar]
- 5.Ingleshwar A, Barbo A, Landon GC, Siff SJ, De Achaval S, Suarez-Almazor ME. Ethnic variations at time of surgery and during follow-up in patients undergoing total knee arthroplasty [Abstract]. In: 2013 ARC/ARHP Annual Meeting. 2012. Oct 25-30. San Diego, CA. Abstract no. 830. [Google Scholar]
- 6.Lavernia CJ, Villa JM. Does race affect outcomes in total joint arthroplasty? Clin Orthop Relat Res. 2015. Nov;473(11):3535-41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ong KL, Kurtz SM, Lau E, Bozic KJ, Berry DJ, Parvizi J. Prosthetic joint infection risk after total hip arthroplasty in the Medicare population. J Arthroplasty. 2009. Sep;24(6)(Suppl):105-9. [DOI] [PubMed] [Google Scholar]
- 8.Dy CJ, Marx RG, Bozic KJ, Pan TJ, Padgett DE, Lyman S. Risk factors for revision within 10 years of total knee arthroplasty. Clin Orthop Relat Res. 2014. Apr;472(4):1198-207. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Adelani MA, Archer KR, Song Y, Holt GE. Immediate complications following hip and knee arthroplasty: does race matter? J Arthroplasty. 2013. May;28(5):732-5. [DOI] [PubMed] [Google Scholar]
- 10.Ibrahim SA, Stone RA, Han X, Cohen P, Fine MJ, Henderson WG, Khuri SF, Kwoh CK. Racial/ethnic differences in surgical outcomes in veterans following knee or hip arthroplasty. Arthritis Rheum. 2005;52(10):3143-51. [DOI] [PubMed] [Google Scholar]
- 11.Dehon E, Weiss N, Jones J, Faulconer W, Hinton E, Sterling S. A systematic review of the impact of physician implicit racial bias on clinical decision making. Acad Emerg Med. 2017. Aug;24(8):895-904. [DOI] [PubMed] [Google Scholar]
- 12.Hall WJ, Chapman MV, Lee KM, Merino YM, Thomas TW, Payne BK, Eng E, Day SH, Coyne-Beasley T. Implicit racial/ethnic bias among health care professionals and its influence on health care outcomes: a systematic review. Am J Public Health. 2015. Dec;105(12):e60-76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hinman A, Bozic KJ. Impact of payer type on resource utilization, outcomes and access to care in total hip arthroplasty. J Arthroplasty. 2008. Sep;23(6)(Suppl 1):9-14. [DOI] [PubMed] [Google Scholar]
- 14.Lavernia CJ, Alcerro JC, Rossi MD. Fear in arthroplasty surgery: the role of race. Clin Orthop Relat Res. 2010. Feb;468(2):547-54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Goodman SM, Mehta B, Zhang M, Szymonifka J, Nguyen JT, Lee L, Figgie MP, Parks ML, Dey SA, Crego D, Russell LA. Disparities in total hip arthroplasty outcomes: census tract data show interactions between race and community deprivation. J Am Acad Orthop Surg. 2018;26(21):e457-64. [DOI] [PubMed] [Google Scholar]
- 16.Goodman SM, Mandl LA, Mehta B, Navarro-Millan I, Russell LA, Parks ML, Dey SA, Crego D, Figgie MP, Nguyen JT, Szymonifka J, Zhang M, Bass AR. Does education level mitigate the effect of poverty on total knee arthroplasty outcomes? Arthritis Care Res (Hoboken). 2018. Jun;70(6):884-91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Carlson LC, Kim J, Samuels-Kalow ME, Yun BJ, Terry DF, Weilburg JB, Lee J. Comparing neighborhood-based indices of socioeconomic risk factors and potentially preventable emergency department utilization. Am J Emerg Med. 2021. Jun;44:213-9. [DOI] [PubMed] [Google Scholar]
- 18.Grits D, Emara AK, Klika AK, Murray TG, McLaughlin JP, Piuzzi NS. Neighborhood socioeconomic disadvantage associated with increased healthcare utilization after total hip arthroplasty. J Arthroplasty. 2022. Oct;37(10):1980-1986.e2. [DOI] [PubMed] [Google Scholar]
- 19.Khlopas A, Grits D, Sax OC, Chen Z, Orr MN, Klika AK, Mont MA, Piuzzi NS. Neighborhood socioeconomic disadvantages associated with prolonged lengths of stay, nonhome discharges, and 90-day readmissions after total knee arthroplasty. J Arthroplasty. 2022. Jun;37(6S):S37-43, 43.e1. [DOI] [PubMed] [Google Scholar]
- 20.Mehta B, Goodman S, Ho K, Parks M, Ibrahim SA. Community deprivation index and discharge destination after elective hip replacement. Arthritis Care Res (Hoboken). 2021. Apr;73(4):531-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Shaw JH, Wesemann LD, Ayoola AS, Les CM, Charters MA, North WT. Comparison of area deprivation index, socioeconomic parameters, and preoperative demographics with postoperative emergency department visits after total knee arthroplasty. J Arthroplasty. 2021. Aug;36(8):2788-94. [DOI] [PubMed] [Google Scholar]
- 22.Havranek EP, Mujahid MS, Barr DA, Blair IV, Cohen MS, Cruz-Flores S, Davey-Smith G, Dennison-Himmelfarb CR, Lauer MS, Lockwood DW, Rosal M, Yancy CW; American Heart Association Council on Quality of Care and Outcomes Research, Council on Epidemiology and Prevention, Council on Cardiovascular and Stroke Nursing, Council on Lifestyle and Cardiometabolic Health, and Stroke Council. Social determinants of risk and outcomes for cardiovascular disease: a scientific statement from the American Heart Association. Circulation. 2015. Sep 1;132(9):873-98. [DOI] [PubMed] [Google Scholar]
- 23.Diez Roux AV, Mujahid MS, Hirsch JA, Moore K, Moore LV. The impact of neighborhoods on CV risk. Glob Heart. 2016. Sep;11(3):353-63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Carmichael H, Moore A, Steward L, Velopulos CG. Disparities in emergency versus elective surgery: comparing measures of neighborhood social vulnerability. J Surg Res. 2020. Dec;256:397-403. [DOI] [PubMed] [Google Scholar]
- 25.Hall YN. Social determinants of health: addressing unmet needs in nephrology. Am J Kidney Dis. 2018. Oct;72(4):582-91. [DOI] [PubMed] [Google Scholar]
- 26.Mesterton J, Willers C, Dahlström T, Rolfson O. Comparison of individual and neighbourhood socioeconomic status in case mix adjustment of hospital performance in primary total hip replacement in Sweden: a register-based study. BMC Health Serv Res. 2020. Jul 10;20(1):645. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kamath CC, O’Byrne TJ, Lewallen DG, Berry DJ, Maradit Kremers H. Association of rurality and neighborhood level socioeconomic deprivation with perioperative health status in total joint arthroplasty patients: analysis from a large, tertiary care hospital. J Arthroplasty. 2022. Aug;37(8):1505-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.George J, Piuzzi NS, Ng M, Sodhi N, Khlopas AA, Mont MA. Association between body mass index and thirty-day complications after total knee arthroplasty. J Arthroplasty. 2018. Mar;33(3):865-71. [DOI] [PubMed] [Google Scholar]
- 29.Davis ET, Lingard EA, Schemitsch EH, Waddell JP. Effects of socioeconomic status on patients’ outcome after total knee arthroplasty. Int J Qual Health Care. 2008. Feb;20(1):40-6. [DOI] [PubMed] [Google Scholar]
- 30.Keeney BJ, Koenig KM, Paddock NG, Moschetti WE, Sparks MB, Jevsevar DS. Do aggregate socioeconomic status factors predict outcomes for total knee arthroplasty in a rural population? J Arthroplasty. 2017. Dec;32(12):3583-90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Barrack RL, Ruh EL, Chen J, Lombardi AV, Jr, Berend KR, Parvizi J, Della Valle CJ, Hamilton WG, Nunley RM. Impact of socioeconomic factors on outcome of total knee arthroplasty. Clin Orthop Relat Res. 2014. Jan;472(1):86-97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Taylor LA, Tan AX, Coyle CE, Ndumele C, Rogan E, Canavan M, Curry LA, Bradley EH. Leveraging the social determinants of health: what works? PLoS One. 2016. Aug 17;11(8):e0160217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Dlott CC, Wiznia DH. CORR synthesis: how might the preoperative management of risk factors influence healthcare disparities in total joint arthroplasty? Clin Orthop Relat Res. 2022. May 1;480(5):872-90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Aseltine RH, Jr, Wang W, Benthien RA, Katz M, Wagner C, Yan J, Lewis CG. Reductions in race and ethnic disparities in hospital readmissions following total joint arthroplasty from 2005 to 2015. J Bone Joint Surg Am. 2019;101(22):2044-50. [DOI] [PubMed] [Google Scholar]
- 35.Prouty A, Cooper M, Thomas P, Christensen J, Strong C, Bowie L, Oermann MH. Multidisciplinary patient education for total joint replacement surgery patients. Orthop Nurs. 2006. Jul-Aug;25(4):257-61. [DOI] [PubMed] [Google Scholar]
- 36.Thomas KM, Sethares KA. An investigation of the effects of preoperative interdisciplinary patient education on understanding postoperative expectations following a total joint arthroplasty. Orthop Nurs. 2008. Nov-Dec;27(6):374-81. [DOI] [PubMed] [Google Scholar]
- 37.Pelt CE, Gililland JM, Erickson JA, Trimble DE, Anderson MB, Peters CL. Improving value in total joint arthroplasty: a comprehensive patient education and management program decreases discharge to post-acute care facilities and post-operative complications. J Arthroplasty. 2018. Jan;33(1):14-8. [DOI] [PubMed] [Google Scholar]