Abstract
Background
The aim of this study was to evaluate the association between Area Deprivation Index (ADI) and patient outcomes following reverse total shoulder arthroplasty (rTSA).
Methods
A retrospective analysis of patients who underwent an rTSA at a single institution between 2011 and 2021 with minimum 2-year follow-up. Each patient's home address was mapped to the ADI to determine the level of socioeconomic disadvantage. Patients were categorized into 5 groups based on socioeconomic status (SES): ADI group 1; the least deprived group and ADI group 5; the most deprived group. Bivariate analysis was performed to determine the association between the level of SES and 2-year postoperative American Shoulder and Elbow Surgeons (ASES) score. Multivariable regression analysis was utilized to assess the role of independent variables in achieving minimal clinically important difference (MCID), substantial clinical benefit (SCB), and patient acceptable symptomatic state (PASS) for ASES.
Results
A total of 551 patients, mean age: 71.1 ± 9.1 year/old and overall mean follow-up time of 42.5 ± 29.9 months. The mean ADI value of all cohorts was 49.3 ± 29.4. The mean ADI for groups 1 through 5 were 9.0 ± 4.9, 30.1 ± 7.6, 47.6 ± 4.4, 70.9 ± 6.7, and 89.9 ± 5.2. There were no differences in age, sex, body mass index, or preoperative medical comorbidities. The average preoperative ASES score across ADI subgroups was 30.6 ± 18.0. Preoperative ASES scores were lowest in both ADI group 1:26.5 ± 15.3 and ADI group 5:25.9 ± 16.7. There was no difference in preoperative range of motion (ROM) across all ADI subgroups. The average postoperative ASES score was 74.2 ± 23.7. There was a significant inverse relationship between ADI and postoperative ASES (P = .047). ADI group 1 had the highest postoperative ASES score of 78.6 ± 21.6 compared to 70.0 ± 24.1 in group 5. There was no difference in change preoperative to postoperative ASES scores across ADI subgroups with an average delta ASES score of 42.8 ± 26.2. Like preoperative ROM, there was no difference across ADI subgroups in terms of postoperative ROM. The average percentage of the cohort of patients across ADI subgroups that achieved MCID, SCB, and PASS for ASES was 87.6%, 68.9%, and 57.5%, respectively. There was no difference in terms of achieving MCID, SCB, or PASS for ASES across ADI subgroups.
Conclusion
The current study supports an inverse relationship between ADI and postoperative outcomes in patients undergoing rTSA. Additionally, our study found that a patient's ability to achieve MCID, SCB, or PASS for ASES at a minimum of 2 years after rTSA was not dependent on SES. Lastly, our study demonstrated that the risk of suffering an adverse event or undergoing a revision surgery were not associated with SES.
Keywords: Area Deprivation Index, Socioeconomic status, Reverse total shoulder arthroplasty, Social determinants of health, Health-care disparities, Patient-reported outcome measures
In the United States, well-documented evidence highlights socioeconomic disparities that impede access to sufficient health care.2,5 Despite ongoing efforts, the health-care system continues to grapple with providing universal, affordable care.1,4,5,10,35,37,42 While previous research predominantly focused on simplistic associations between race and ethnicity and health disparities, recent years have witnessed a growing acknowledgment that race and ethnicity are only 2 factors among a broader array of social determinants of health (SDOH).12,38 These include socioeconomic status (SES), geographic location, primary insurance coverage, education, employment and income, all of which significantly contribute to population health disparities.11,12,25,38
Understanding the role of SES in patients undergoing surgical procedures may facilitate the development of treatment plans tailored to at-risk individuals, thereby minimizing barriers in providing health care.5 Consequently, researchers and policymakers are shifting their focus to geographically based SDOH indices. These indices integrate multiple SDOH characteristics into a single score, enabling comprehensive analyses of health disparities across diverse patient populations.5 Among the frequently utilized geography based SDOH indices, the Area Deprivation Index (ADI) stands out for its integration of the highest number of SDOH factors (a total of 17).5,17,23 Employing a geographic model of ZIP code tabulation areas and neighborhood markers, ADI has demonstrated the strongest association with physical function and mental health symptoms in patients undergoing orthopedic surgery, thereby earning recognition as the preferred geography based SDOH index for orthopedic research.5,17
Despite advancements in orthopedic surgery, health disparities persist, with numerous studies indicating a correlation between socioeconomic disparities and poorer clinical outcomes, particularly in fields such as orthopedic trauma, spine surgery, and hip and knee arthroplasties.1,3,28,41 The incidence of shoulder arthroplasty in the US continues to rise, with projections indicating a significant increase (∼235%) by 2025.39 Reverse total shoulder arthroplasty (rTSA) has emerged as a successful and reliable treatment option, dominating the number of shoulder arthroplasty procedures performed each year.7,8,14,18,20,39 However, despite the exponential rise in shoulder arthroplasties, studies exploring the association between SES and outcomes among shoulder arthroplasty patients remain limited. Furthermore, there is a scarcity of research investigating the link between geography based SDOH and outcomes in individuals undergoing shoulder arthroplasty.24,30
Given the paucity of literature regarding the impact of SES and geography-based SDOH on postoperative outcomes after shoulder arthroplasty, this study aims to evaluate the association between ADI and patient outcomes following rTSA. Since rTSA is widely acknowledged as a reliable and effective treatment option for shoulder conditions we hypothesized that clinical outcomes and patient-reported outcome measures (PROMs) after rTSA will not be affected by SES.
Materials and methods
Patient selection
This study was approved by the institutional review board at the facility where all surgeries were performed. We identified all patients who received an rTSA by a single shoulder and elbow fellowship trained orthopedic surgeon between August 2011 and February 2021 with a minimum follow-up of 2 years. Inclusion criteria were any patient receiving an rTSA for osteoarthritis, rheumatoid arthritis, cuff tear arthropathy, osteonecrosis, or fracture. Exclusion criteria were age <18 years, revision cases, and patients undergoing anatomic total shoulder arthroplasty (aTSA). The data utilized for this study were captured prospectively by the senior author (or their surrogate) preoperatively and postoperatively at routine time points on an annual basis. The data were maintained and deidentified on a server.
Definition of variables studied
Baseline demographics were collected including age, gender, body mass index (BMI), medical comorbidities, tobacco use, history of prior ipsilateral shoulder surgery, and preoperative indication for shoulder arthroplasty procedure. Perioperative and postoperative adverse events (AEs) were documented. These include superficial wound complications or hematoma, deep periprosthetic joint infection requiring surgical treatment, aseptic loosening of humeral and/or glenoid components, postoperative periprosthetic fracture, mechanical failure, dislocation, and medically related morbidity and mortality. Revision surgeries and their circumstances were also documented. American Shoulder and Elbow Surgeons (ASES) score was recorded preoperatively and at each postoperative visit. At latest follow-up, patient satisfaction was assessed as “much better”, “better,”, “unchanged,” and “worse” relative to their preoperative condition. Range of motion was measured preoperatively and postoperatively with a goniometer by the implanting surgeon or their surrogate including: active forward elevation (FE), abduction (AB), and active external rotation (ER).
Area Deprivation Index
The ADI was initially developed by the Health Resources & Services Administration and has since been refined, adapted, and validated at the University of Wisconsin–Madison to the census block group neighborhood level.14,15,33 The ADI enables rankings of neighborhoods by their level of socioeconomic disadvantage within a region of interest. It encompasses 17 factors related to the domains of income, education, employment, health-care access, and housing quality. It can be used to inform health delivery and policy, particularly for the most disadvantaged neighborhood groups. The ADI provides the ability to take a specific address and output a numeric value that is standardized to provide a score between 0 and 100. Scores closer to 0 represent areas that are at lower social deprivation, and scores closer to 100 correspond to higher social deprivation.15 The score is normalized relative to a national mean score of 50.15 Patient addresses were coded into census blocks via R Studio 4.13 (Posit, Boston, MA, USA). After a census block was assigned to each patient, it was mapped over the census block ADI reference provided by Center for Health Disparities at the University of Wisconsin.15 The ADI scores were then assigned to each patient based on their census block group. Patients in a census block that did not contain an ADI value and patients without a 9-digit ZIP code matched to their self-reported address were excluded from analysis. For risk assessment, the study population was divided into 5 quantiles based on ADI value.
Statistical and other analyses
All statistical analysis was performed in R (version 3.6.2; R Foundation for Statistical Computing, Vienna, Austria). Statistical significance was set at a P value of less than .05. For all continuous and categorical variables, descriptive statistics were calculated. Continuous variables were reported as weighted mean and estimated standard deviation, whereas categorical variables were reported as frequencies with percentages. The percentage of each cohort that exceeded the minimal clinically important difference (MCID), substantial clinical benefit (SCB), and patient acceptable symptomatic state (PASS) for ASES was calculated based on previously validated cutoffs published by Simovitch et al.31,32 Multivariate logistic regression analysis was used to assess the role of independent variables in achieving MCID, SCB, and PASS for ASES. Feature importance was determined from the magnitude of value of standardized regression coefficients.
Results
Study population
A total of 551 patients who underwent rTSA were included. The mean age of all groups was 71.1 ± 9.1 years with an overall mean follow-up time of 42.5 ± 29.9 months. The mean ADI value of all cohorts was 49.3 ± 29.4. The mean ADI for each of the 5 cohorts were as follows: ADI group 1: 9.0 ± 4.9, ADI group 2: 30.1 ± 7.6, ADI group 3: 47.6 ± 4.4, ADI group 4: 70.9 ± 6.7, and ADI group 5: 89.9 ± 5.2. As seen on Table I, there were no differences in age, sex, BMI, preoperative medical comorbidities (eg, hypertension, heart disease, diabetes, depression, and anxiety), tobacco use or indication for surgery across all ADI subgroups.
Table I.
Demographic data for patients undergoing reverse total shoulder arthroplasty stratified by ADI quantile subgroups.
| Overall N = 551 |
Group 1 N = 111 |
Group 2 N = 114 |
Group 3 N = 106 |
Group 4 N = 110 |
Group 5 N = 110 |
P value | |
|---|---|---|---|---|---|---|---|
| ADI Score∗ | 49.3 (29.4) | 9.0 (4.9) | 30.1 (7.6) | 47.6 (4.4) | 70.9 (6.7) | 89.9 (5.2) | <.001‡ |
| Age (yr)∗ | 71.1 (9.1) | 71.2 (9.0) | 70.6 (8.0) | 73.0 (10.5) | 70.4 (9.1) | 70.5 (8.6) | .1 |
| Follow-Up (Mo)∗ | 42.5 (29.9) | 39.5 (27.8) | 40.7 (27.0) | 40.6 (29.6) | 43.4 (29.6) | 48.5 (34.6) | .088 |
| Gender = Female† | 341 (64.3%) | 72 (66.1%) | 62 (56.9%) | 62 (60.2%) | 66 (64.7%) | 79 (73.8%) | .1 |
| BMI∗ | 29.2 (6.5) | 29.2 (7.6) | 28.9 (6.5) | 28.1 (5.3) | 29.9 (6.4) | 30.0 (6.3) | .12 |
| Previous Surgery = Yes† | 212 (40.6%) | 48 (46.2%) | 45 (41.7%) | 34 (33.7%) | 44 (42.7%) | 41 (38.7%) | .4 |
| Previous Injection = Yes† | 254 (48.8%) | 43 (41.7%) | 58 (54.2%) | 51 (50.5%) | 48 (46.6%) | 54 (50.9%) | .4 |
| Previous Analgesics = Yes† | 302 (58.3%) | 60 (58.8%) | 62 (57.9%) | 58 (58.0%) | 60 (58.3%) | 62 (58.5%) | >.9 |
| Hypertension† | 308 (59.0%) | 51 (49.0%) | 66 (61.1%) | 63 (61.8%) | 62 (60.8%) | 66 (62.3%) | .2 |
| Heart Disease† | 65 (12.5%) | 5 (4.8%) | 16 (14.8%) | 18 (17.6%) | 14 (13.7%) | 12 (11.3%) | .063 |
| Diabetes† | 68 (13.0%) | 7 (6.7%) | 15 (13.9%) | 17 (16.7%) | 17 (16.7%) | 12 (11.3%) | .2 |
| Depression† | 107 (19.4%) | 16 (14.4%) | 20 (17.5%) | 28 (26.4%) | 26 (23.6%) | 17 (15.5%) | .1 |
| Anxiety† | 84 (15.2%) | 12 (10.8%) | 15 (13.2%) | 24 (22.6%) | 16 (14.5%) | 17 (15.5%) | .2 |
| Tobacco Use† | 49 (9.4%) | 6 (5.8%) | 15 (13.9%) | 9 (8.8%) | 9 (8.8%) | 10 (9.4%) | .4 |
| Indication for Surgery | |||||||
| Diagnosis | |||||||
| Osteoarthritis† | 284 (54.4%) | 60 (57.7%) | 64 (59.8%) | 60 (58.8%) | 49 (47.6%) | 51 (48.1%) | .2 |
| Osteonecrosis† | 5 (1.0%) | 0 (0.0%) | 1 (0.9%) | 2 (2.0%) | 1 (1.0%) | 1 (0.9%) | .7 |
| Rotator Cuff Arthropathy† | 91 (17.4%) | 14 (13.5%) | 20 (18.7%) | 14 (13.7%) | 18 (17.5%) | 25 (23.6%) | .3 |
| Rotator Cuff Tear† | 100 (19.2%) | 17 (16.3%) | 15 (14.0%) | 26 (25.5%) | 21 (20.4%) | 21 (19.8%) | .3 |
| Rheumatoid Arthritis† | 23 (4.4%) | 4 (3.8%) | 1 (0.9%) | 5 (4.9%) | 5 (4.9%) | 8 (7.5%) | .2 |
ADI, Area Deprivation Index; BMI, body mass index.
Mean (Standard Deviation).
Number of patients n (%).
Statistically significant P < .05.
The average preoperative ASES score across ADI subgroups was 30.6 ± 18.0. As seen on Table II, preoperative ASES scores were lowest in both ADI group 1: 26.5 ± 15.3 and ADI group 5: 25.9 ± 16.7. The average preoperative AB, FE, and ER across all ADI subgroups was 61 ± 30°, 76 ± 36°, and 17 ± 17° respectively. There was no difference in preoperative range of motion across all ADI subgroups. The average postoperative ASES score was 74.2 ± 23.7. There was a significant inverse relationship between ADI and postoperative ASES (P = .047). More specifically, ADI group 1 had the highest 2-year postoperative ASES score of 78.6 ± 21.6 compared to 70.0 ± 24.1 in group 5. However, there was no difference in change pre to postoperative ASES scores across all ADI subgroups with an average delta ASES score of 42.8 ± 26.2. Similar to preoperative range of motion, there was no difference across ADI subgroups with an average AB, FE, and ER of 92 ± 19, 130 ± 31, and 32 ± 15, respectively. The average percentage of the cohort of patients across all ADI subgroups that achieved MCID, SCB, and PASS for ASES was 87.6%, 68.9%, and 57.5%, respectively. There was no difference in terms of achieving MCID, SCB, or PASS for ASES across all ADI subgroups.
Table II.
Functional and clinical outcome scores stratified by ADI subgroup.
| Overall N = 551 |
Group 1 N = 111 |
Group 2 N = 114 |
Group 3 N = 106 |
Group 4 N = 110 |
Group 5 N = 110 |
P value | |
|---|---|---|---|---|---|---|---|
| Preoperative ASES∗ | 30.6 (18.0) | 26.5 (15.3) | 33.2 (17.3) | 31.5 (19.5) | 35.1 (19.1) | 25.9 (16.7) | .009‡ |
| Preoperative Active Abduction∗ | 61 (30) | 62 (31) | 67 (32) | 56 (29) | 60 (29) | 60 (31) | .3 |
| Preoperative Active Forward Elevation∗ | 76 (36) | 75 (37) | 77 (34) | 76 (36) | 78 (38) | 74 (37) | >.9 |
| Preoperative Active External Rotation∗ | 17 (17) | 16 (17) | 15 (18) | 17 (16) | 15 (17) | 20 (19) | .2 |
| Postoperative Functional Scores, Range of Motion, and Patient Satisfaction | |||||||
| Postoperative ASES∗ | 74.2 (23.7) | 78.6 (21.6) | 75.8 (23.9) | 74.9 (23.4) | 71.4 (24.9) | 70.0 (24.1) | .047‡ |
| Postoperative Active Abduction∗ | 92 (19) | 92 (16) | 93 (17) | 94 (20) | 92 (22) | 90 (21) | .5 |
| Postoperative Active Forward Elevation∗ | 130 (31) | 135 (30) | 132 (31) | 128.0 (30) | 125 (34) | 129 (33) | .2 |
| Postoperative Active External Rotation∗ | 32 (15) | 33 (15) | 33 (18) | 33 (15) | 30 (14) | 31 (14) | .3 |
| Patient Satisfaction† | .3 | ||||||
| Much better | 378 (68.6%) | 86 (77.5%) | 78 (68.4%) | 77 (72.6%) | 67 (60.9%) | 70 (63.6%) | |
| Better | 121 (22.0%) | 18 (16.2%) | 25 (21.9%) | 18 (17.0%) | 31 (28.2%) | 29 (26.4%) | |
| Unchanged | 26 (4.7%) | 4 (3.6%) | 7 (6.1%) | 3 (2.8%) | 6 (5.5%) | 6 (5.5%) | |
| Worse | 26 (4.7%) | 3 (2.7%) | 4 (3.5%) | 8 (7.5%) | 6 (5.5%) | 5 (4.5%) | |
| Δ Difference | |||||||
| ASES Difference∗ | 42.8 (26.2) | 48.6 (25.3) | 42.0 (26.1) | 44.3 (22.5) | 36.5 (29.1) | 43.1 (26.8) | .2 |
| Patient-Reported Outcome Measures | |||||||
| MCID ASES† | 483 (87.6%) | 102 (90.9%) | 104 (91.4%) | 94 (88.7%) | 88 (79.7%) | 95 (86.7%) | .2 |
| SCB ASES† | 380 (68.9%) | 86 (77.3%) | 78 (68.6%) | 70 (66.1%) | 69 (62.5%) | 77 (70.0%) | .5 |
| PASS ASES† | 317 (57.5%) | 64 (57.4%) | 76 (66.7%) | 55 (51.6%) | 57 (51.5%) | 51 (46.2%) | .14 |
| Adverse Outcomes | |||||||
| Adverse Event† | 54 (9.8%) | 13 (11.7%) | 10 (8.8%) | 6 (5.7%) | 14 (12.7%) | 11 (10.0%) | .4 |
| Revision Rate† | 36 (6.5%) | 10 (9.0%) | 10 (8.8%) | 1 (0.9%) | 9 (8.2%) | 6 (5.5%) | .085 |
ADI, Area Deprivation Index; ASES, American Shoulder and Elbow Surgeons; MCID, minimal clinically important difference; PASS, patient acceptable symptomatic state; SCB, substantial clinical benefit.
Mean (Standard Deviation).
Number of patients n (%).
Statistically significant P < .05.
As seen in Table III, a higher deprivation index was marginally predictive of lower pre to postoperative changes in ASES scores when controlling for age, gender, BMI and ethnic origin. Additionally, Table IV, Table V, Table VI demonstrate that there was no association between ADI level and achieving MCID, SCB, and PASS for ASES. Additionally, the only factors predictive of achieving 2-year postoperative SCB and PASS for ASES were age and gender. Lastly, as seen in Tables VII and VIII, there was no association between ADI level and suffering an AE or undergoing revision surgery in our cohort of patients.
Table III.
Multivariate linear regression looking at whether the changes in ASES are dependent on ADI group when controlled for age, gender, BMI, and ethnic origin.
| Characteristic | Beta | 95% CI | P value |
|---|---|---|---|
| ADI Quartiles | |||
| 1 | — | — | |
| 2 | −2.5 | −12 to 6.4 | .6 |
| 3 | −4.5 | −14 to 4.8 | .3 |
| 4 | −11 | −20 to −1.9 | .018 |
| 5 | −2.5 | −12 to 6.8 | .6 |
| Age (yr) | 0.4 | 0.07 to 0.73 | .017 |
| Gender = Female | 6.2 | 0.00 to 12 | .05 |
| BMI | 0.27 | −0.18 to 0.73 | .2 |
| Ethnic Origin | |||
| Asian | — | — | |
| Black or African American | −3.8 | −41 to 33 | .8 |
| Caucasian | −0.94 | −37 to 35 | >.9 |
| Hispanic | 3.1 | −35 to 41 | .9 |
| Other | 14 | −30 to 59 | .5 |
ADI, Area Deprivation Index; BMI, body mass index; ASES, American Shoulder and Elbow Surgeons; CI, confidence interval.
Statistically significant P < .05.
Table IV.
Multivariable logistic regression looking at whether achieving MCID for ASES was dependent on ADI quartiles.
| Characteristic | OR | 95% CI | P value |
|---|---|---|---|
| ADI Quartiles | |||
| 1 | — | — | |
| 2 | 0.64 | 0.16-2.24 | .5 |
| 3 | 0.65 | 0.16-2.34 | .5 |
| 4 | 0.48 | 0.12-1.58 | .2 |
| 5 | 0.44 | 0.11-1.48 | .2 |
| Age (yr) | 1.03 | 0.99-1.07 | .2 |
| Gender = Female | 1.44 | 0.69-2.97 | .3 |
| BMI | 1 | 0.95-1.06 | .9 |
| Ethnic Origin | |||
| Asian | — | — | |
| Black or African American | 0 | >.9 | |
| Caucasian | 0 | >.9 | |
| Hispanic | 0 | >.9 |
ADI, Area Deprivation Index; ASES, American Shoulder and Elbow Surgeons; BMI, body mass index; MCID, minimal clinically important difference; CI, confidence interval; OR, odds ratio.
Statistically significant P < .05.
Table V.
Multivariable logistic regression looking at whether achieving SCB for ASES was dependent on ADI quartiles.
| Characteristic | OR | 95% CI | P value |
|---|---|---|---|
| ADI Quartiles | |||
| 1 | — | — | |
| 2 | 0.72 | 0.32-1.59 | .4 |
| 3 | 0.63 | 0.27-1.43 | .3 |
| 4 | 0.59 | 0.26-1.28 | .2 |
| 5 | 0.75 | 0.32-1.75 | .5 |
| Age (yr) | 1.03 | 1.01-1.06 | .017 |
| Gender = Female | 1.81 | 1.07-3.08 | .027 |
| BMI | 1.04 | 1.00-1.08 | .087 |
| Ethnic Origin | |||
| Asian | — | — | |
| Black or African American | 1.08 | 0.04-30.7 | >.9 |
| Caucasian | 2.17 | 0.08-59.8 | .6 |
| Hispanic | 2.43 | 0.08-72.9 | .6 |
ADI, Area Deprivation Index; ASES, American Shoulder and Elbow Surgeons; BMI, body mass index; SCB, substantial clinical benefit; CI, confidence interval; OR, odds ratio.
Statistically significant P < .05.
Table VI.
Multivariable logistic regression looking at patients achieving PASS for ASES.
| Characteristic | OR | 95% CI | P value |
|---|---|---|---|
| ADI Quartiles | |||
| 1 | — | — | |
| 2 | 1.07 | 0.59-1.95 | .8 |
| 3 | 0.8 | 0.44-1.46 | .5 |
| 4 | 0.84 | 0.45-1.53 | .6 |
| 5 | 0.72 | 0.39-1.31 | .3 |
| Age (yr) | 1.03 | 1.00-1.05 | .02 |
| Gender = Female | 0.59 | 0.39-0.89 | .012 |
| BMI | 1 | 0.97-1.03 | .9 |
| Ethnic Origin | |||
| Asian | — | — | |
| Black or African American | 906,820 | 0.00, NA | >.9 |
| Caucasian | 2,146,228 | 0.00, NA | >.9 |
| Hispanic | 1,248,810 | 0.00, NA | >.9 |
ADI, Area Deprivation Index; ASES, American Shoulder and Elbow Surgeons; BMI, body mass index; PASS, patient acceptable symptomatic state; CI, confidence interval; OR, odds ratio.
Statistically significant P < .05.
Table VII.
Multivariable logistic regression looking at the association between ADI and suffering from an adverse event.
| Characteristic | OR | 95% CI | P value |
|---|---|---|---|
| ADI Quartiles | |||
| 1 | — | — | |
| 2 | 0.64 | 0.25-1.57 | .3 |
| 3 | 0.43 | 0.15-1.14 | .1 |
| 4 | 1.14 | 0.50-2.59 | .8 |
| 5 | 0.82 | 0.34-1.92 | .6 |
| Age (yr) | 1 | 0.97-1.03 | >.9 |
| Gender = Female | 0.94 | 0.52-1.74 | .8 |
| BMI | 0.99 | 0.94-1.03 | .6 |
ADI, Area Deprivation Index; BMI, body mass index; CI, confidence interval; OR, odds ratio.
Statistically significant P < .05.
Table VIII.
Multivariable logistic regression looking at the association between ADI and undergoing a revision surgery.
| Characteristic | OR | 95% CI | P value |
|---|---|---|---|
| ADI Quartiles | |||
| 1 | — | — | |
| 2 | 0.85 | 0.32-2.22 | .7 |
| 3 | 0.09 | 0.01-0.51 | .026 |
| 4 | 0.9 | 0.34-2.34 | .8 |
| 5 | 0.55 | 0.18-1.55 | .3 |
| Age (yr) | 0.99 | 0.95-1.03 | .5 |
| Gender = Female | 0.92 | 0.45-1.95 | .8 |
| BMI | 1 | 0.95-1.05 | >.9 |
ADI, Area Deprivation Index; BMI, body mass index; CI, confidence interval; OR, odds ratio.
Statistically significant P < .05.
Discussion
In the current US health-care environment, which is moving toward emphasis on pay for performance, it is important to understand how PROMs are influenced by SDOH.41 Social circumstances have been recognized as factors that directly impact physical and mental health—with studies linking SDOH to cancer incidence, cardiovascular and pulmonary health, and even depression.16,21,26,34,36 Similarly, in recent literature, social disparities have been shown to directly impact patient outcomes following a variety of orthopedic surgical procedures.4,5,10,11 However, the effect of social deprivation on patients undergoing rTSA has not been well defined. Our study found that SES, as defined by ADI, at a minimum of 24 months postoperatively was not a predictor of clinical or functional outcome measures in patients undergoing rTSA. Furthermore, our study demonstrates that irrespective of ADI and preoperative and postoperative PROMs, patients achieved similar rates of MCID, SCB, and PASS for ASES.
To date, the majority of studies looking at the connection between socioeconomic disparities and clinical outcomes in patients undergoing total shoulder arthroplasty are primarily based on large national database registries with follow-up time limited to 90 days.23,27,29 Utilizing insurance type as a proxy for SES, Sharma et al found increased rates of revision and lower patient-reported outcome scores in patients undergoing shoulder arthroplasty who have Medicare and Medicaid compared with commercial insurance.29 Similarly, Waldrop et al found that patients with Medicare and Medicaid insurance undergoing primary shoulder arthroplasty demonstrated worse preoperative and postoperative patient-reported scores (though similar improvement) than those with commercial insurance.40 As our understanding of SDOH has continued to evolve, we now understand it is multifactorial and assigning patients a SES based on individual factors like insurance type, income, or ZIP code fails to capture the full complexity of a patient's situation.5,9,19 We therefore have transitioned to utilizing ADI. ADI has proven to be the most comprehensive and generalizability as it quantifies a patient's level of social deprivation relative to others at both the state and national level.5 As such, it provides a value that can be reliably compared across studies.6
In a recent cross-sectional study utilizing ADI as a surrogate of SES, Sheth et al found that ADI was inversely proportional to preoperative ASES scores.30 However, the shortcoming of this study was that only preoperative outcome scores were analyzed. In contrast, the study by Moverman et al which included a total of 380 patients (115 aTSA and 265 rTSA) found no difference in preoperative or 2-year postoperative ASES, Single Assessment Numeric Evaluation, or Visual Analogue Scale pain scores when patients were stratified by ADI.24 Interestingly, in our cohort of patients we found that patients in the highest deprivation cohort (ADI group 5) and lowest deprivation cohort (ADI group 1) had the lowest preoperative ASES scores compared to groups 2 through 4. However, unlike the study by Moverman et al, our study found a linear increase in 2-year postoperative ASES scores as ADI decreased.24 More specifically, our study demonstrated an 8.6 ± 2.5 difference in 2-year postoperative ASES scores between ADI group 1 and ADI group 5. Similarly, while not statistically significant, ADI group 1 (48.6 ± 25.3) had the largest change preoperatively to postoperatively in ASES scores compared to both ADI group 4 (36.5 ± 29.1) and ADI group 5 (43.1 ± 26.8). These findings were further validated by a multivariate linear regression model which demonstrated that a higher deprivation index was marginally predictive of lower pre to postoperative changes in ASES scores when controlling for age, gender, BMI, and ethnic origin.
Over recent years, more and more literature has focused on translating PROM scores into predictors of clinical success at the individual level.13,31,32 The benefit of MCID, SCB, and PASS thresholds is that they represent tangible and clinically relevant treatment targets that align clinical outcomes with a patient's perception of the experience. In our study, the overall percentage of patients that achieved MCID, SCB, and PASS for ASES were 87.6%, 68.9%, and 57.5%, respectively. It is important to note that while not significant, ADI group 1 exceeded the mean MCID, SCB, and PASS by an average percentage of 6.97%. In contrast, ADI group 5 fell short of achieving the overall average MCID, SCB, and PASS percentage by 5.03%. More specifically, we found a significantly higher rate of patients in ADI group 1 who achieved PASS for ASES when compared to ADI group 5 at an average follow-up of 42.5 months (63.1% vs. 47.6%, P = .048). Similarly, when we looked at patient satisfaction across all cohorts at 2 years postoperatively we found a statistically significant difference in the rate of patients reporting their shoulder to be “much better” when comparing ADI groups 1 and 5 (77.5% vs. 63.6%, P = .02). However, when we compared MCID, SCB, PASS, and patient satisfaction across all groups there was no statistical difference. When performing a multivariable linear regression, we found that achieving MCID, SCB or PASS for ASES was not dependent on ADI. In fact, the only factors predictive of achieving these 3 outcome score thresholds were age and gender.
A comprehensive understanding of patients at higher risk of complications following shoulder arthroplasty is essential for patient safety and surgical optimization.22 In a large national database study looking at both aTSA and rTSA patients Raso et al found that patients with any degree of SDOH compromise were associated with an increased risk of 90-day medical complications, readmission, and emergency department visits compared with matched controls without SDOH compromise.27 Additionally, Raso et al found that SDOHs were associated with increased rates of 1-year prosthetic instability, aseptic loosening, and ipsilateral revision compared with controls.27 Unlike the study by Raso et al, our study reduced heterogeneity by limiting our analysis to only patients undergoing rTSA. More specifically, our study found no difference across groups in terms of preoperative risk factors, comorbidities, and indication for surgery. Additionally, our study demonstrates similar rates of AEs and revision rate across all ADI cohorts. Lastly, unlike the study by Raso et al our study found no association between ADI and AEs or revision rates.
Limitations
The present study had several limitations. First, this study was a retrospective review of a prospectively collected database and thus was subject to the typical biases of retrospective studies. Another limitation of the study design was that we used demographics and ZIP codes for the ADI, and these variables represent an individual at a single time point. Although the ADI is validated and well-studied, utilizing ZIP code to identify and classify an individual may introduce bias into the model, especially in areas that have heterogeneous or changing demographics that do not accurately represent the SES of all individuals living in that area. Although we controlled for a large number of potential confounding variables, including age, gender, BMI, and ethnic origin in the multivariable regression models, it is possible the influence of ADI might still be partly explained by other confounding factors. Patients in this study are part of a large single institutional database and as such, there may be concern for external generalizability. However, the benefit of our institutional shoulder database is that patients were treated by a single surgeon and followed longitudinally, which helps reduce treatment effect as a confounder while at the same time reduces the interval between patient follow-up.
Conclusion
The current study supports an inverse relationship between ADI and postoperative outcomes in patients undergoing rTSA. Additionally, our study found that a patient's ability to achieve MCID, SCB, or PASS for ASES at a minimum of 2 years after rTSA was not dependent on SES. Lastly, our study demonstrated that the risk of suffering an AE or undergoing a revision surgery were not associated with SES. Future long-term studies are needed to assess the durability of the results of this study.
Disclaimers:
Funding: No funding was disclosed by the authors.
Conflicts of interest: Joseph D. Zuckerman is a design surgeon for Exactech, Inc. shoulder arthroplasty system. The other authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.
Footnotes
The NYU Office of Science and Research Institutional Review Board approved this study; Study #: i05-144 CR20.
References
- 1.Amen T.B., Varady N.H., Shannon E.M., Chopra A., Rajaee S., Chen A.F. Racial and ethnic disparities in hip fracture surgery care in the United States from 2006 to 2015: a nationwide trends study. J Am Acad Orthop Surg. 2022;30:e182–e190. doi: 10.5435/JAAOS-D-21-00137. [DOI] [PubMed] [Google Scholar]
- 2.National Academies of Sciences, Engineering, and Medicine; Health and Medicine Division; Board on Population Health and Public Health Practice; Committee on Community-Based Solutions to Promote Health Equity in the United States . National Academies Press (US); Washington (DC): 2017. Communities in Action: Pathways to Health Equity. Baciu A, Negussie Y, Geller A, Weinstein JN, editors. [PubMed] [Google Scholar]
- 3.Bass A.R., McHugh K., Fields K., Goto R., Parks M.L., Goodman S.M. Higher total knee arthroplasty revision rates among United States blacks than whites: a systematic literature review and meta-analysis. J Bone Joint Surg Am. 2016;98:2103–2108. doi: 10.2106/JBJS.15.00976. [DOI] [PubMed] [Google Scholar]
- 4.Bernstein D.N., Kurucan E., Fear K., Hammert W.C. Evaluating the impact of patient social deprivation on the level of symptom severity at carpal tunnel syndrome presentation. Hand (N Y) 2022;17:339–345. doi: 10.1177/1558944720928487. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bernstein D.N., Shin D., Poolman R.W., Schwab J.H., Tobert D.G. Are commonly used geographically based social determinant of health indices in orthopaedic surgery research correlated with each other and with PROMIS global-10 physical and mental health scores? Clin Orthop Relat Res. 2024;482:604–614. doi: 10.1097/CORR.0000000000002896. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Cheng A.L., McDuffie J.V., Schuelke M.J., Calfee R.P., Prather H., Colditz G.A. How should we measure social deprivation in orthopaedic patients? Clin Orthop Relat Res. 2022;480:325–339. doi: 10.1097/CORR.0000000000002044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Collin P., Herve A., Walch G., Boileau P., Muniandy M., Chelli M. Mid-term results of reverse shoulder arthroplasty for glenohumeral osteoarthritis with posterior glenoid deficiency and humeral subluxation. J Shoulder Elbow Surg. 2019;28:2023–2030. doi: 10.1016/j.jse.2019.03.002. [DOI] [PubMed] [Google Scholar]
- 8.Cuff D.J., Pupello D.R., Santoni B.G., Clark R.E., Frankle M.A. Reverse shoulder arthroplasty for the treatment of rotator cuff deficiency: a concise follow-up, at a minimum of 10 years, of previous reports. J Bone Joint Surg Am. 2017;99:1895–1899. doi: 10.2106/JBJS.17.00175. [DOI] [PubMed] [Google Scholar]
- 9.D'Apuzzo M.R., Villa J.M., Alcerro J.C., Rossi M.D., Lavernia C.J. Total joint arthroplasty: a granular analysis of outcomes in the economically disadvantaged patient. J Arthroplasty. 2016;31(9 Suppl):41–44. doi: 10.1016/j.arth.2016.02.066. [DOI] [PubMed] [Google Scholar]
- 10.Delanois R.E., Tarazi J.M., Wilkie W.A., Remily E., Salem H.S., Mohamed N.S., et al. 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;103-B(6 Supple A):113–118. doi: 10.1302/0301-620X.103B6.BJJ-2020-2430.R1. [DOI] [PubMed] [Google Scholar]
- 11.Faison W.E., Harrell P.G., Semel D. Disparities across diverse populations in the health and treatment of patients with osteoarthritis. Healthcare (Basel) 2021;9:1421. doi: 10.3390/healthcare9111421. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hadad M.J., Rullan-Oliver P., Grits D., Zhang C., Emara A.K., Molloy R.M., et al. Racial disparities in outcomes after THA and TKA are substantially mediated by socioeconomic disadvantage both in black and white patients. Clin Orthop Relat Res. 2023;481:254–264. doi: 10.1097/CORR.0000000000002392. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hao K.A., Hones K.M., O'Keefe D.S., Saengchote S.A., Burns M.Q., Wright J.O., et al. Quantifying success after first revision reverse total shoulder arthroplasty: the minimal clinically important difference, substantial clinical benefit, and patient acceptable symptomatic state. J Shoulder Elbow Surg. 2023;32:e516–e527. doi: 10.1016/j.jse.2023.03.032. [DOI] [PubMed] [Google Scholar]
- 14.Kang J.R., Dubiel M.J., Cofield R.H., Steinmann S.P., Elhassan B.T., Morrey M.E., et al. Primary reverse shoulder arthroplasty using contemporary implants is associated with very low reoperation rates. J Shoulder Elbow Surg. 2019;28(6S):S175–S180. doi: 10.1016/j.jse.2019.01.026. [DOI] [PubMed] [Google Scholar]
- 15.Kind A.J.H., Buckingham W.R. Making neighborhood-disadvantage metrics accessible - the neighborhood atlas. N Engl J Med. 2018;378:2456–2458. doi: 10.1056/NEJMp1802313. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kind A.J., Jencks S., Brock J., Yu M., Bartels C., Ehlenbach W., et al. Neighborhood socioeconomic disadvantage and 30-day rehospitalization: a retrospective cohort study. Ann Intern Med. 2014;161:765–774. doi: 10.7326/M13-2946. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Knighton A.J., Savitz L., Belnap T., Stephenson B., VanDerslice J. Introduction of an area deprivation index measuring patient socioeconomic status in an integrated health system: implications for population health. EGEMS (Wash DC) 2016;4:1238. doi: 10.13063/2327-9214.1238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Larose G., Fisher N.D., Gambhir N., Alben M.G., Zuckerman J.D., Virk M.S., et al. Inlay versus onlay humeral design for reverse shoulder arthroplasty: a systematic review and meta-analysis. J Shoulder Elbow Surg. 2022;31:2410–2420. doi: 10.1016/j.jse.2022.05.002. [DOI] [PubMed] [Google Scholar]
- 19.Lasater K.B., McHugh M.D. Reducing hospital readmission disparities of older black and white adults after elective joint replacement: the role of nurse staffing. J Am Geriatr Soc. 2016;64:2593–2598. doi: 10.1111/jgs.14367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Levin J.M., Bokshan S., Roche C.P., Zuckerman J.D., Wright T., Flurin P.H., et al. Reverse shoulder arthroplasty with and without baseplate wedge augmentation in the setting of glenoid deformity and rotator cuff deficiency-a multicenter investigation. J Shoulder Elbow Surg. 2022;31:2488–2496. doi: 10.1016/j.jse.2022.04.025. [DOI] [PubMed] [Google Scholar]
- 21.Lian M., Schootman M., Doubeni C.A., Park Y., Major J.M., Stone R.A., et al. Geographic variation in colorectal cancer survival and the role of small-area socioeconomic deprivation: a multilevel survival analysis of the NIH-AARP Diet and Health Study Cohort. Am J Epidemiol. 2011;174:828–838. doi: 10.1093/aje/kwr162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ling K., Fassler R.P., Nicholson A.J., Komatsu D.E., Wang E.D. A comprehensive analysis of age and 30-day complications following total shoulder arthroplasty: nonagenarians, octogenarians, and septuagenarians. JSES Int. 2024;8:176–184. doi: 10.1016/j.jseint.2023.08.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Mandalia K.P., Brodeur P.G., Li L.T., Ives K., Cruz A.I., Jr., Shah S.S. Higher complication rates following primary total shoulder arthroplasty in patients presenting from areas of higher social deprivation. Bone Joint J. 2024;106-B:174–181. doi: 10.1302/0301-620X.106B2.BJJ-2023-0785.R1. [DOI] [PubMed] [Google Scholar]
- 24.Moverman M.A., Sudah S.Y., Puzzitiello R.N., Pagani N.R., Hart P.A., Swanson D., et al. Neighborhood socioeconomic disadvantage does not predict outcomes or cost after elective shoulder arthroplasty. J Shoulder Elbow Surg. 2022;31:2465–2472. doi: 10.1016/j.jse.2022.04.023. [DOI] [PubMed] [Google Scholar]
- 25.Pandya N.K., Wustrack R., Metz L., Ward D. Current concepts in orthopaedic care disparities. J Am Acad Orthop Surg. 2018;26:823–832. doi: 10.5435/JAAOS-D-17-00410. [DOI] [PubMed] [Google Scholar]
- 26.Powell-Wiley T.M., Ayers C., Agyemang P., Leonard T., Berrigan D., Ballard-Barbash R., et al. Neighborhood-level socioeconomic deprivation predicts weight gain in a multi-ethnic population: longitudinal data from the Dallas Heart Study. Prev Med. 2014;66:22–27. doi: 10.1016/j.ypmed.2014.05.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Raso J., Kamalapathy P.N., Sumpter A., Ramamurti P., Werner B.C. Economic and educational disparities are associated with an increased risk of revision surgery following shoulder arthroplasty. J Shoulder Elbow Surg. 2023;32:589–596. doi: 10.1016/j.jse.2022.08.008. [DOI] [PubMed] [Google Scholar]
- 28.Schoenfeld A.J., Sieg R.N., Li G., Bader J.O., Belmont P.J., Jr., Bono C.M. Outcomes after spine surgery among racial/ethnic minorities: a meta-analysis of the literature. Spine J. 2011;11:381–388. doi: 10.1016/j.spinee.2011.03.013. [DOI] [PubMed] [Google Scholar]
- 29.Sharma S., Miller A.S., Pearson Z., Tran A., Bahoravitch T.J., Stadecker M., et al. Social determinants of health disparities impact postoperative complications in patients undergoing total shoulder arthroplasty. J Shoulder Elbow Surg. 2024;33:640–647. doi: 10.1016/j.jse.2023.07.006. [DOI] [PubMed] [Google Scholar]
- 30.Sheth M.M., Morris B.J., Laughlin M.S., Elkousy H.A., Edwards T.B. Lower socioeconomic status is associated with worse preoperative function, pain, and increased opioid use in patients with primary glenohumeral osteoarthritis. J Am Acad Orthop Surg. 2020;28:287–292. doi: 10.5435/JAAOS-D-19-00490. [DOI] [PubMed] [Google Scholar]
- 31.Simovitch R., Flurin P.H., Wright T., Zuckerman J.D., Roche C.P. Quantifying success after total shoulder arthroplasty: the minimal clinically important difference. J Shoulder Elbow Surg. 2018;27:298–305. doi: 10.1016/j.jse.2017.09.013. [DOI] [PubMed] [Google Scholar]
- 32.Simovitch R., Flurin P.H., Wright T., Zuckerman J.D., Roche C.P. Quantifying success after total shoulder arthroplasty: the substantial clinical benefit. J Shoulder Elbow Surg. 2018;27:903–911. doi: 10.1016/j.jse.2017.12.014. [DOI] [PubMed] [Google Scholar]
- 33.Singh G.K. Area deprivation and widening inequalities in US mortality, 1969-1998. Am J Public Health. 2003;93:1137–1143. doi: 10.2105/ajph.93.7.1137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Singh G.K., Kogan M.D. Widening socioeconomic disparities in US childhood mortality, 1969 2000. Am J Public Health. 2007;97:1658–1665. doi: 10.2105/AJPH.2006.087320. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Suleiman L.I., Tucker K., Ihekweazu U., Huddleston J.I., Cohen-Rosenblum A.R. Caring for diverse and high-risk patients: surgeon, health system, and patient integration. J Arthroplasty. 2022;37:1421–1425. doi: 10.1016/j.arth.2022.02.017. [DOI] [PubMed] [Google Scholar]
- 36.Urwin M., Symmons D., Allison T., Brammah T., Busby H., Roxby M., et al. Estimating the burden of musculoskeletal disorders in the community: the comparative prevalence of symptoms at different anatomical sites, and the relation to social deprivation. Ann Rheum Dis. 1998;57:649–655. doi: 10.1136/ard.57.11.649. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Varady N.H., d'Amonville S., Chen A.F. Electronic patient portal use in orthopaedic surgery is associated with disparities, improved satisfaction, and lower No-show rates. J Bone Joint Surg Am. 2020;102:1336–1343. doi: 10.2106/JBJS.19.01080. [DOI] [PubMed] [Google Scholar]
- 38.Virk M.S. CORR insights(R): racial disparities in outcomes after THA and TKA are substantially mediated by socioeconomic disadvantage both in black and white patients. Clin Orthop Relat Res. 2023;481:265–267. doi: 10.1097/CORR.0000000000002452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Wagner E.R., Farley K.X., Higgins I., Wilson J.M., Daly C.A., Gottschalk M.B. The incidence of shoulder arthroplasty: rise and future projections compared with hip and knee arthroplasty. J Shoulder Elbow Surg. 2020;29:2601–2609. doi: 10.1016/j.jse.2020.03.049. [DOI] [PubMed] [Google Scholar]
- 40.Waldrop L.D., 2nd, King J.J., 3rd, Mayfield J., Farmer K.W., Struk A.M., Wright T.W., et al. The effect of lower socioeconomic status insurance on outcomes after primary shoulder arthroplasty. J Shoulder Elbow Surg. 2018;27(6S):S35–S42. doi: 10.1016/j.jse.2018.01.002. [DOI] [PubMed] [Google Scholar]
- 41.Wright M.A., Murthi A.M., Aleem A., Zmistowski B. Patient disparities and provider diversity in orthopaedic surgery: a complex relationship. J Am Acad Orthop Surg. 2023;31:132–139. doi: 10.5435/JAAOS-D-22-00410. [DOI] [PubMed] [Google Scholar]
- 42.Ziedas A., Abed V., Swantek A., Cross A., Chaides S., Rahman T., et al. Social determinants of health influence access to care and outcomes in patients undergoing anterior cruciate ligament reconstruction: a systematic review. Arthroscopy. 2022;38:583–594.e4. doi: 10.1016/j.arthro.2021.06.031. [DOI] [PubMed] [Google Scholar]