The impact of socioeconomic deprivation on total shoulder arthroplasty outcomes

Abstract

Background

Total shoulder arthroplasty (TSA) is a highly effective surgical intervention to alleviate pain and restore function in various shoulder joint pathologies. As the demand for TSA continues to rise, it is important to understand patient factors that affect clinical outcomes. The Area Deprivation Index (ADI) quantifies socioeconomic deprivation, accounting for multiple variables including income, education, employment, and housing quality. The purpose of this study is to examine the effects of socioeconomic deprivation on patient outcomes after TSA.

Methods

A retrospective chart review was performed on patients who received TSA from January 2007 to December 2022. Records were reviewed for demographic factors including age, sex, race, body mass index, and Charlson Comorbidity Index. Postoperative outcomes included active forward flexion and active external rotation along with visual analog scale pain scores and American Shoulder and Elbow Surgeons (ASES) scores. Socioeconomic deprivation was evaluated through ADI scores from the Neighborhood Atlas. Chi-squared and analysis of variance tests were performed for analysis.

Results

In this study, 955 patients were included, with 426 receiving anatomic TSA (aTSA) and 529 receiving reverse TSA (rTSA) cohort. Among aTSA patients, 32% were least, 52% intermediate, and 16% most deprived. For rTSA, 27% of patients were least, 59% intermediate, and 13% most. The aTSA cohort saw significant differences among socioeconomic groups in postoperative active external rotation at 6 months (most deprived [47.0°] >> intermediate [44.6°], least [43.9°]), and postoperative ASES at the most recent follow-up (most deprived [76.5] << least [89.4], intermediate [80.4]). There were also differences in mean pain scores at 3 months (most deprived [2.9] >> least [1.4], intermediate [1.7]), 1 year (most deprived [3.1] >> least [1.3], intermediate [1.7]), and the most recent follow-up (most deprived [2.9] >> intermediate [2.0] >> least [1.2]). The rTSA cohort saw significant differences among socioeconomic groups in preoperative ASES (most deprived [24.3] << intermediate [32.0], least [35.8]). Differences in mean pain score were also significant at preoperative (most deprived [6.6] >> intermediate [5.6] >> least [4.7]) and 1 year time points (most deprived [2.6] >> intermediate [1.6], least [1.4]).

Conclusion

Some interval postoperative disparities highlight the impact of socioeconomic deprivation, with most deprived patients having worse outcomes in TSA cohorts. While outcomes were generally similar at the most recent follow-up, prospective studies are needed to more accurately elucidate differences. Nevertheless, incorporating ADI metrics care could maximize optimal outcomes in shoulder arthroplasty.

Total shoulder arthroplasty (TSA) is a highly effective surgical intervention aimed at alleviating pain and restoring function in patients with various shoulder joint pathologies, such as osteoarthritis, rheumatoid arthritis, and rotator cuff tear arthropathy.²² In the United States, TSA procedures increased by 121.8% between 2013 and 2021, a trend that is expected to persist given the aging population and improved surgical techniques.⁶ As the demand for TSA continues to rise, it becomes increasingly important to examine the myriad of factors influencing patient outcomes, particularly those beyond the clinical and technical aspects of the procedure.^4,5,10

One key area of investigation is the role of social determinants of health (SDOH) in shaping patient outcomes. SDOH refers to the various nonmedical factors, including social, economic, and environmental conditions, that play a critical role in shaping health outcomes³¹ and has been designated as 1 of the 3 priority objectives of the Healthy People 2030 initiative.³ Moreover, the National Academy of Medicine has estimated that 80%-90% of health outcomes are attributable to social determinants while only 10%-20% linked directly to the quality of medical care.¹⁴ Recognizing these disparities, research into the intersection of SDOH and orthopedic outcomes has gained momentum in recent years, yet there remains a substantial gap in understanding their specific impact on shoulder arthroplasty.^7,9,23,26

The Area Deprivation Index (ADI) offers a robust and multifaceted approach to quantifying socioeconomic deprivation by assessing variables such as income, education, employment, and housing quality. Prior studies have demonstrated the utility of ADI as a powerful measure of SDOH, enabling researchers and clinicians to identify communities at higher risk for poor health outcomes.^1,8,12,15 In the context of TSA, evaluating ADI-related discrepancies in postoperative outcomes is essential for addressing inequities and optimizing care delivery. Within orthopedics, ADI has emerged as a useful proxy for SDOH, and several studies have explored its association with outcomes in arthroplasty procedures.

Recent literature examining the impact of socioeconomic deprivation using ADI on shoulder arthroplasty outcomes has yielded mixed results. For instance, Morgan et al. reported that patients from more socioeconomically disadvantaged neighborhoods experienced higher levels of pain and decreased functionality in both the preoperative and postoperative periods.¹⁷ However, the degree of improvement from preoperative to postoperative outcomes was comparable to that of patients from less deprived backgrounds, suggesting similar relative gains despite baseline disparities. In contrast, Moverman et al. found no significant differences in outcomes or costs following elective shoulder arthroplasty when stratified by ADI, concluding that neighborhood-level socioeconomic disadvantages did not predict poorer results.¹⁸

Despite these contributions, the literature remains limited and somewhat inconsistent particularly regarding how ADI influences specific postoperative outcomes such as pain scores, range of motion (ROM), functional scores, and complication rates. Moreover, many existing studies have been constrained by small sample sizes, a lack of implant specific analysis, or relatively short follow-up periods, underscoring the need for further investigation into the role of socioeconomic deprivation in shoulder arthroplasty outcomes.

The purpose of this study is to examine the effects of socioeconomic deprivation as measured by ADI, on patient outcomes after TSA in hopes of providing information that could be used to improve surgical techniques, patient education, and postoperative management to achieve equitable health outcomes for all patients undergoing TSA. We hypothesized that patients from more socioeconomically deprived backgrounds, as measured by the ADI, would experience worse postoperative pain, functional scores, and complication rates compared to lesser deprived counterparts.

Methods and materials

After institutional review board approval, a retrospective chart review was performed on patients who received all-cause TSA by 3 fellowship-trained shoulder and elbow surgeons at a single level 1 trauma center from January 2007 to December 2022. Exclusion criteria included patients with less than 2 year follow-up, patients younger than 18 years old at the time of surgery, revision arthroplasty, and patients undergoing TSA for fracture. Electronic records were reviewed for demographic information such as age, sex, race, and home address; body mass index; Charlson Comorbidity Index; length of follow-up; length of surgery; and postoperative outcomes.

Postoperative outcomes included active forward flexion (aFF) and active external rotation (aER) as measured with a goniometer by the treating surgeon preoperatively, at 2 months, 6 months, 1 year, 2 years, and at most recent follow-up (MRF) times. Visual analog scale (VAS) pain scores and American Shoulder and Elbow Surgeons (ASES) scores were collected during similar time points. A delta value equal to the percent difference between preoperative and MRF values was calculated for each outcome measure. Lastly, clinical outcomes including readmission, mortality, and revision rates were evaluated at 1 month, 3 month, and 1 year times, with the etiologies for each instance also collected.

Socioeconomic deprivation of each patient was evaluated through ADI scores from the Neighborhood Atlas, a public resource from the University of Wisconsin.^11,20 The database employs census data from 2020 to assign a level of socioeconomic deprivation to different Census Block Groups based on 17 factors including education, employment, and quality of housing.^11,20 The calculated score is represented as a national percentile from 1 to 100, with higher levels correlating with higher levels of socioeconomic deprivation. Each patient's full home address was thus entered, with their national percentile assigning them to tertiles of least deprived (1-33), intermediate deprived (34-66), and most deprived (67-100).

Percentages were reported for categorical variables while means were calculated for continuous variables for statistical analysis. After separating results into anatomic TSA (aTSA) and reverse TSA (rTSA) cohorts, descriptive and comparative statistics between tertiles of patients were completed using pairwise Chi-squared tests with post-hoc Benjamini-Hochberg corrections for percentages while analysis of variance tests with post-hoc Tukey's Honestly Significant Difference tests and Benjamini-Hochberg corrections were calculated for means. Similar analysis of variance and post-hoc testing was performed for aFF, aER, pain score, ASES score, readmission rate, revision rate, and mortality rate for each preoperative, postoperative, and delta value. All statistical analysis was performed using R Statistical Software (v4.2.2; R Foundation for Statistical Computing, Vienna, Austria).²¹ Statistical significance was set at P values < .05.

Results

Demographics

A total of 1,262 patients underwent TSA between January 2007 and December 2022. Following exclusions, a total of 955 patients were analyzed with a mean follow-up of 4.6 years (range 2.0-16.4) and most common preoperative diagnoses of glenohumeral arthritis (52%) and rotator cuff arthropathy (43%). Of these patients, 426 underwent aTSA with a mean follow-up of 5.8 years (range 2.0-16.4) and had glenohumeral arthritis (97%) as the most common preoperative diagnosis. By socioeconomic deprivation, 138 (32%) were considered least deprived, 220 (52%) were intermediate deprived, and 68 (16%) were most deprived. A total of 529 patients underwent rTSA with a mean follow-up of 3.6 years (range 2.0-13.9) and had rotator cuff arthropathy (78%) and glenohumeral arthritis (15%) representing the most common preoperative diagnoses. By socioeconomic deprivation, 145 (27%) were considered least deprived, 314 (59%) were intermediate deprived, and 70 (13%) were most deprived.

In the aTSA cohort, the mean age was 63.5 years old and the majority of patients were female (50.5%) and Caucasian (82.9%). The indications for surgery did not distinctly differ between the socioeconomic groups, with glenohumeral arthritis representing the majority of cases. Further details on aTSA surgical indications can be found in Table I. On the other hand, there were significant differences in demographic characteristics among the socioeconomic groups, particularly in both age and race. Most deprived patients had a mean age of 60.5 years old, significantly younger than both least (65.7 years old, P ≤ .001) and intermediate deprived patients (60.5 year old, P = .014). With regard to racial characteristics, 23.5% of most deprived patients identified as Black, significantly greater than least (2.9%, P ≤ .001) and intermediate deprived patients (6.8%, P ≤ .001). Similarly, 16.2% of most deprived patients identified as Hispanic, greater than least (1.5%, P ≤ .001) and intermediate deprived patients (6.8%, P = .020) with significant difference between least and intermediate patients as well (P = .020). Furthermore, 92.8% of least deprived patients identified as Caucasian/White, which was significantly greater than intermediate (83.6%, P = .012) and most deprived patients (60.3%, P ≤ .001) with significant difference between intermediate and most (P ≤ .001). Further information regarding demographics in the aTSA cohort can be found in Table II.

Table I.

Indications for anatomic total shoulder arthroplasty operation.

aTSA reason for operation
Socioeconomic deprivation category	Rank of commonality (percent of group)
	Most common	Second most common	Third most common
All aTSA patients	Glenohumeral arthritis (96%)	Inflammatory arthritis∗ (1%)	Avascular necrosis (1%)
Least deprived	Glenohumeral arthritis (97%)	Inflammatory arthritis∗ (2%)	Avascular necrosis (1%)
Intermediate deprived	Glenohumeral arthritis (95%)	Avascular necrosis (3%)	Inflammatory arthritis∗ (1%)
Most deprived	Glenohumeral arthritis (99%)	Inflammatory arthritis∗ (1%)	Null

aTSA, anatomic total shoulder arthroplasty.

^∗Etiologies including rheumatoid arthritis and ankylosing spondylitis.

Table II.

aTSA cohort demographics.

Parameter	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
Sex: Female (%)	49.3	51.8	48.5	.920	.920	.920
Average age (yr)	65.7	63.0	60.5	.057	<.001	.014
Race: Asian (%)	1.5	0.0	0.0	.573	.999	.573
Race: African American/Black (%)	2.9	6.8	23.5	.107	<.001	<.001
Race: Hispanic (%)	1.5	6.8	16.2	.020	<.001	.020
Race: Other (%)	1.5	2.7	0.0	.819	.999	.819
Race: Caucasian/White (%)	92.8	83.6	60.3	.012	<.001	<.001
Average BMI (kg/m2)	31.9	32.6	33.3	.500	.500	.500
Average CCI	3.1	3.0	2.6	.750	.470	.470
Average Follow-up (yr)	5.6	6.1	5.0	.668	.668	.417

aTSA, anatomic total shoulder arthroplasty; BMI, body mass index; CCI, Charlson Comorbidity Index.

Bold values indicate statistical significance (P < .05).

In the rTSA cohort, the mean age was 70.0 year old and the majority of patients were female (58.4%) and Caucasian (75.8%). The indications for surgery did not distinctly differ between the socioeconomic groups, with rotator cuff arthropathy representing the majority of cases. Further details on rTSA surgical indications can be found in Table III. Regarding demographic characteristics, there were significant differences across sex, age, and racial groups. Least deprived patients had a significantly lower ratio of females (45.5%) compared to intermediate (61.1%, P ≤ .001) and most deprived patients (68.6%, P ≤ .001). Additionally, the least deprived patients had a significantly higher mean age (71.7) compared to intermediate deprived (69.6, P = .042) and most deprived (68.7, P = .042) patients. Regarding racial characteristics, a significant proportion of patients identified as Black, comprising 25.7% of most deprived patients compared to least (4.1%, P ≤ .001) and intermediate deprived (8.9%, P ≤ .001) patients. Additionally, 2.8% of least deprived patients identified as Hispanic, which was significantly less than intermediate (16.2%, P ≤ .001) and most deprived (12.9%, P ≤ .001) patients. Finally, significant differences occurred with patients identified as Caucasian/White, making up 58.6% of the most deprived patients compared to least deprived (93.1%, P ≤ .001) and intermediate deprived (71.7%, P = .032) patients. The difference between least and intermediate groups was also significant (P < .001). Further details regarding demographics in the rTSA cohort can be found in Table IV. Post-operative follow-up was similar among all groups in both aTSA and rTSA cohorts, as depicted in Table II. and IV, respectively.

Table III.

Indications for reverse total shoulder arthroplasty operation.

rTSA reason for operation
Socioeconomic deprivation category	Rank of commonality (percent of group)
	Most common	Second most common	Third most common
All rTSA patients	Rotator cuff arthropathy (77%)	Glenohumeral arthritis (16%)	Inflammatory arthritis∗ (3%)
Least deprived	Rotator cuff arthropathy (76%)	Glenohumeral arthritis (19%)	Dislocation (2%)
Intermediate deprived	Rotator cuff arthropathy (77%)	Glenohumeral arthritis (16%)	Inflammatory arthritis∗ (4%)
Most deprived	Rotator cuff arthropathy (81%)	Glenohumeral arthritis (13%)	Dislocation (4%)

rTSA, anatomic total shoulder arthroplasty.

^∗Etiologies including rheumatoid arthritis and ankylosing spondylitis.

Table IV.

rTSA cohort demographics.

Parameter	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
Sex: Female (%)	45.5	62.10	68.576	<.001	<.001	.309
Average age (yr)	71.7	69.6	68.7	.042	.042	.493
Race: Asian (%)	0.0	0.6	0.0	.952	.601	.504
Race: African American/Black (%)	4.1	8.9	25.7	.069	<.001	<.001
Race: Hispanic (%)	2.8	16.2	12.9	<.001	<.001	.481
Race: Other (%)	0.0	2.6	2.9	.179	.883	.202
Race: Caucasian/White (%)	93.1	71.7	58.6	<.001	<.001	.032
Average BMI (kg/m2)	30.6	31.2	31.8	.550	.550	.550
Average CCI	4.9	4.2	3.9	.094	.094	.496
Average Follow-up (yr)	3.7	3.6	3.3	.981	.981	.981

BMI, body mass index; rTSA, reverse total shoulder arthroplasty; CCI, Charlson Comorbidity Index.

Bold values indicate statistical significance (P < .05).

Range of motion outcomes

Regarding aFF, no significant differences were seen at any time point among all socioeconomic groups for both aTSA and rTSA cohorts. The same was observed for percent change from preoperative to MRF. Further details regarding aFF outcomes can be found in Tables V and VI.

Table V.

aTSA active forward flexion.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
Preoperative (°)	111.4	107.9	110.0	.510	.510	.510
2 mo (°)	127.2	127.1	128.5	.830	.170	.170
6 mo (°)	142.6	141.0	144.9	.660	.280	.280
1 yr (°)	150.2	146.6	145.6	.280	.230	.480
2 yr (°)	153.2	143.2	149.6	.620	.620	.620
MRF (°)	144.0	141.2	140.2	.570	.570	.570
Δ (%)	156.1	139.9	472.8	.494	.106	.085

aTSA, anatomic total shoulder arthroplasty; MRF, most recent follow-up; Δ, percent change from preoperation to most recent follow-up.

Table VI.

rTSA active forward flexion.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. Most	Intermediate vs. most
Preoperative (°)	93.0	84.3	83.0	.130	.170	.820
2 mo (°)	117.2	118.1	116.7	.920	.920	.920
6 mo (°)	128.1	132.0	134.7	.320	.320	.520
1 yr (°)	137.3	138.0	139.4	.830	.830	.830
2 yr (°)	143.0	136.6	140.0	.750	.750	.750
MRF (°)	132.2	133.9	131.7	.850	.910	.850
Δ (%)	501.5	1,002.5	1,043.2	.340	.380	.920

rTSA, reverse total shoulder arthroplasty; MRF, most recent follow-up; Δ, percent change from preoperation to most recent follow-up.

For aER, the only statistically significant difference within the aTSA cohort was seen at the 6 month postoperative period. The most deprived patients had a significantly greater mean aER (47.0°) compared to the least deprived (43.9°, P = .049) and the intermediate deprived (44.6°, P = .049) groups. However, while statistically significant, the absolute differences in aER between groups fell below the minimal clinically important difference threshold of 14.5 established for external rotation following aTSA, as reported by Simovitch et al.²⁵ This suggests that although the difference reached statistical significance, it may not be clinically meaningful. Once again the rTSA cohort did not demonstrate any significant differences at any time point across socioeconomic groups. Further details regarding aER outcomes can be found in Tables VII and VIII.

Table VII.

aTSA active external rotation.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
Preoperative	24.8	27.3	29.8	.290	.290	.080
2 mo (°)	39.3	40.1	38.0	.870	.870	.870
6 mo (°)	43.9	44.6	47.0	.507	.049	.049
1 yr (°)	46.8	47.2	45.9	.780	.430	.430
2 yr (°)	51.0	49.3	48.1	.600	.600	.860
MRF (°)	46.3	46.7	43.4	.380	.980	.380
Δ (%)	1,126.2	1,295.6	867.2	.660	.660	.660

aTSA, anatomic total shoulder arthroplasty; MRF, most recent follow-up; Δ, percent change from preoperation to most recent follow-up.

Bold values indicate statistical significance (P < .05).

Table VIII.

rTSA active external rotation.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
Preoperative (°)	21.4	19.6	21.5	.710	.990	.710
2 mo (°)	27.8	30.2	30.5	.310	.310	.870
6 mo (°)	33.3	36.3	34.2	.210	.720	.550
1 yr (°)	37.9	36.0	36.4	.880	.880	.890
2 yr (°)	41.5	35.3	32.2	.210	.210	.590
MRF (°)	37.2	35.6	34.7	.430	.430	.670
Δ (%)	946.4	1,096.9	855.9	.620	.740	.620

rTSA, reverse total shoulder arthroplasty; MRF, most recent follow-up; Δ, percent change from preoperation to most recent follow-up.

American Shoulder and Elbow Surgeons functional outcomes and visual analog scale pain outcomes

Regarding ASES scores, the only statistically significant difference within the aTSA cohort was seen at the MRF. The most deprived patients had a significantly lower average score (76.5) compared to the least deprived (89.4, P = .003) and the intermediate deprived (80.4, P = .003) groups. Within the rTSA patients, statistically significant differences were observed preoperatively, with most deprived patients having significantly lower mean ASES (24.3) compared to least (35.8, P = .002) and intermediate deprived (32.0, P = .017) patients. No significant postoperative differences were present at any time point or in percent change by MRF. While these differences were statistically significant, the absolute difference between the most and least deprived groups also exceeds the reported minimal clinically important difference for ASES score following shoulder arthroplasty. This has been estimated to be between 6.5-13.5 by Werner et al., suggesting these disparities may be both statistically and clinically meaningful.³⁰ Further details regarding differences in ASES score can be found in Tables IX and X.

Table IX.

aTSA ASES scores.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
Preoperative	38.3	32.8	34.6	.610	.980	.610
6 Weeks	59.9	49.9	61.7	.700	.630	.630
3 mo	68.4	67.2	52.5	.880	.190	.190
6 mo	85.8	75.9	67.3	.200	.200	.430
1 yr	90.0	82.6	78.4	.190	.190	.510
2 yr	97.6	78.2	91.9	.110	.590	.300
MRF	89.4	80.4	76.5	.285	.003	.003
Δ (%)	689.3	1,063.3	783.0	.259	.259	.065

aTSA, anatomic total shoulder arthroplasty; ASES, American Shoulder and Elbow Surgeons score; MRF, most recent follow-up; Δ, percent change from preoperation to most recent follow-up.

Bold values indicate statistical significance (P < .05).

Table X.

rTSA ASES scores.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
Preoperative	35.8	32.0	24.3	.092	.002	.017
6 Weeks	53.6	55.1	45.7	.730	.350	.350
3 mo	62.2	62.4	51.7	.970	.200	.200
6 mo	68.4	70.4	60.9	.650	.450	.450
1 yr	82.4	78.6	71.4	.320	.230	.320
2 yr	83.8	79.2	86.0	.680	.810	.680
MRF	79.7	78.8	71.8	.749	.088	.088
Δ (%)	1,540.6	1,583.2	2,040.9	.910	.580	.580

rTSA, reverse total shoulder arthroplasty; ASES, American Shoulder and Elbow Surgeons score; MRF, most recent follow-up; Δ, percent change from preoperation to most recent follow-up.

Bold values indicate statistical significance (P < .05).

Regarding VAS pain scores, multiple statistically significant differences were seen within the aTSA cohort. Most deprived patients had a mean pain score of 2.9, which was significantly greater than least (1.4, P = .002) and intermediate deprived patients (1.7, P = .013). This was continued at the 1-year follow-up, with most deprived patients having a mean score of 3.1 compared to least (1.3, P = .003) and intermediate (1.7, P = .004) deprived patients. This was again observed at the MRF, with most deprived patients having a mean pain score of 2.9 vs. least (1.2, P ≤ .001) and intermediate (2.0, P = .031), with an additional difference noted between least and intermediate groups at this time (P = .007). Within the rTSA cohort, statistically significant differences were seen at the preoperative time period with the most deprived group having a significantly higher average pain score (6.6) compared to the least (4.7, P = .002) and intermediate deprived (5.6, P = .021) groups. There was also a significant difference between least and intermediate deprived (P ≤ .001) groups at this time point. Additionally, the most deprived patient group had a mean pain score of 2.6 by 1 year, which was significantly greater compared to patients within the least (1.4, P = .029) and intermediate (1.6, P = .029) groups. Many of the statistically significant differences in VAS scores exceed the minimal clinically important difference of 1.6, as reported by Simovitch et al., indicating clinical as well as statistical relevance.²⁵ Further details regarding differences in VAS pain scores can be found in Tables XI and XII.

Table XI.

aTSA pain (VAS) scores.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
Preoperative	5.5	6.0	6.5	.760	.760	.760
6 Weeks	1.7	2.3	2.7	.980	.190	.190
3 mo	1.4	1.7	2.9	.199	.002	.013
6 mo	1.5	1.5	2.2	.720	.300	.300
1 yr	1.3	1.7	3.1	.473	.003	.004
2 yr	0.6	1.8	2.1	.053	.053	.523
MRF	1.2	2.0	2.9	.007	.0002	.031
Δ (%)	−22.1	−3.0	75.2	.870	.600	.600

aTSA, anatomic total shoulder arthroplasty; VAS, visual analog scale; MRF, most recent follow-up; Δ, percent change from preoperation to most recent follow-up.

Bold values indicate statistical significance (P < .05).

Table XII.

rTSA pain (VAS) scores.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
Preoperative	4.7	5.6	6.6	.0009	.0002	.021
6 Weeks	2.0	2.5	2.8	.071	.071	.467
3 mo	1.6	1.9	2.1	.590	.590	.590
6 mo	1.7	1.6	2.0	.660	.660	.660
1 yr	1.4	1.6	2.6	.618	.029	.029
2 yr	1.6	1.6	2.8	.900	.260	.260
MRF	1.5	1.8	2.2	.380	.210	.280
Δ (%)	−44.9	−60.5	−55.6	.440	.730	.730

rTSA, reverse total shoulder arthroplasty; VAS, visual analog scale; MRF, most recent follow-up; Δ, percent change from preoperation to most recent follow-up.

Bold values indicate statistical significance (P < .05).

Surgery duration and complications

The overall surgery duration across all socioeconomic groups was approximately 2 hours. There were no significant differences found within the aTSA and rTSA cohorts. Further details on surgery duration can be found in Tables XIII and XIV.

Table XIII.

aTSA surgery duration.

Socioeconomic deprivation category			P value
Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
2.1 h	2.1 h	2.1 h	.880	.880	.880

aTSA, anatomic total shoulder arthroplasty.

Table XIV.

rTSA surgery duration.

Socioeconomic deprivation category			P value
Least deprived	Moderate deprived	Most deprived	Least vs. moderate	Least vs. most	Moderate vs. most
1.8 h	1.8 h	1.8 h	.890	.890	.890

rTSA, reverse total shoulder arthroplasty.

There were no significant differences in mortality rates at any time point among socioeconomic groups for both arthroplasty cohorts. Further details on mortality rates can be found in Tables XV and XVI.

Table XV.

aTSA mortality rates.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
1 mo	0.00	0.00	0.00	N/A
3 mo	0.00	0.00	0.00	N/A
1 yr	0.00	0.00	0.00	N/A

aTSA, anatomic total shoulder arthroplasty.

Table XVI.

rTSA mortality rates.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
1 mo	0.00	0.00	0.00	N/A
3 mo	0.01	0.01	0.00	.940	.790	.790
1 yr	0.02	0.01	0.00	.440	.440	.500

rTSA, reverse total shoulder arthroplasty.

The most common causes for hospital readmission within the first year of aTSA were non–orthopedics-related (56%), operation of the other shoulder (23%), and revision of the operative shoulder (18%). The percentages were similar across all socioeconomic groups. There were no significant differences among the socioeconomic groups in readmission rates at any time point within 1 year. Further details on aTSA readmission can be found in Tables XVII and XVIII. For the rTSA cohort, the most common causes for readmission were non–orthopedics-related (55%), revision of the operative shoulder (19%), and operation of the other shoulder (15%). The percentages were relatively similar across all socioeconomic groups and no significant differences were found among socioeconomic groups in readmission rates at any time point within 1 year. Further details on rTSA readmission can be found in Tables XIX and XX.

Table XVII.

aTSA indications for readmission within 1 yr.

Socioeconomic deprivation category	Rank of commonality (percent of group)
	Most common	Second most common	Third most common
All aTSA patients	Nonorthopedics (56%)	Other shoulder (23%)	Revision (18%)
Least deprived	Nonorthopedics (67%)	Other shoulder (24%)	Revision (10%)
Intermediate deprived	Nonorthopedics (50%)	Other shoulder (24%)	Revision (24%)
Most deprived	Nonorthopedics (60%)	Other shoulder (20%)	Revision (10%)

aTSA, anatomic total shoulder arthroplasty.

Table XVIII.

aTSA readmission rates.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprive	Most disadvantage	Least vs. intermediate	Least vs. most	Intermediate vs. most
1 mo	0.00	0.02	0.04	.184	.085	.255
3 mo	0.04	0.07	0.06	.670	.780	.780
1 yr	0.16	0.25	0.16	.330	.970	.340

aTSA, anatomic total shoulder arthroplasty.

Table XIX.

rTSA indications for readmission within 1 yr.

Socioeconomic deprivation category	Rank of commonality (percent of group)
	Most common	Second most common	Third most common
All rTSA patients	Nonorthopedics (55%)	Revision (19%)	Other shoulder (15%)
Least deprived	Nonorthopedics (66%)	Revision (17%)	Other shoulder (10%)
Intermediate deprived	Nonorthopedics (52%)	Other shoulder (20%)	Revision (16%)
Most deprived	Nonorthopedics (45%)	Revision (36%)	Other shoulder (18%)

rTSA, reverse total shoulder arthroplasty.

Table XX.

rTSA readmission rates.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. Intermediate	Least vs. Most	Intermediate vs. Most
1 mo	0.06	0.04	0.07	.660	.810	.660
3 mo	0.09	0.07	0.1	.820	.830	.820
1 yr	0.25	0.24	0.21	.920	.920	.920

rTSA, reverse total shoulder arthroplasty.

The most common causes for revision surgery within the first year of aTSA surgery were tear, arthropathy, or failure of the rotator cuff (57%), infection of the surgical site (29%), and instability of arthroplasty components (7%). Instability was considered as subluxation or any form of prosthetic weakness up to and excluding complete dislocation requiring reduction, which was classified separately. Interestingly, all of the patients from the least and most deprived groups underwent revision surgery due to rotator cuff arthropathy. There were no significant differences among the socioeconomic groups in revision rates at any time point within 1 year. Further details on aTSA revision can be found in Tables XXI and XXII. For rTSA patients, the most common causes of revision surgery were dislocation (42%), instability of arthroplasty components (37%), and infection of the surgical site (16%). The percentages were relatively similar across all socioeconomic groups, though more intermediate deprived patients had revision surgery for instability (40%) than dislocation (30%). There were no significant differences among the socioeconomic groups in revision rates at any time point within 1 year. Further details on rTSA revision can be found in Tables XXIII and XXIV.

Table XXI.

aTSA indications for revision within 1 yr.

Socioeconomic deprivation category	Rank of commonality (percent of group)
	Most common	Second most common	Third most common
All aTSA patients	Rotator cuff arthropathy (57%)	Infection (29%)	Instability (7%)
Least deprived	Rotator cuff arthropathy (100%)	Null	Null
Intermediate deprived	Rotator cuff arthropathy (45%)	Infection (36%)	Instability (9%)
Most deprived	Rotator cuff arthropathy (100%)	Null	Null

aTSA, anatomic total shoulder arthroplasty.

Table XXII.

aTSA revision rates.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
1 mo	0.00	0.01	0.00	.360	.999	.360
3 mo	0.01	0.02	0.00	.340	.680	.340
1 yr	0.01	0.05	0.01	.180	.999	.220

aTSA, anatomic total shoulder arthroplasty.

Table XXIII.

rTSA indications for revision within 1 yr.

Socioeconomic deprivation category	Rank of commonality (percent of group)
	Most common	Second most common	Third most common
All rTSA patients	Dislocation (42%)	Instability (37%)	Infection (16%)
Least deprived	Dislocation (60%)	Instability (40%)	Null
Intermediate deprived	Instability (40%)	Dislocation (30%)	Infection (20%)
Most deprived	Dislocation (50%)	Instability (25%)	Infection (25%)

rTSA, reverse total shoulder arthroplasty.

Table XXIV.

rTSA revision rates.

Time point	Socioeconomic deprivation category			P value
	Least deprived	Intermediate deprived	Most deprived	Least vs. intermediate	Least vs. most	Intermediate vs. most
1 mo	0.02	0.02	0.06	.930	.290	.290
3 mo	0.03	0.03	0.09	.820	.240	.240
1 yr	0.05	0.03	0.11	.560	.160	.080

rTSA, reverse total shoulder arthroplasty.

Discussion

The main findings of this study reveal that socioeconomic deprivation was associated with significant differences in postoperative outcomes within the aTSA cohort, including lower ASES scores and higher pain scores at multiple postoperative intervals for the most deprived group. Similarly, the rTSA cohort demonstrated worse preoperative function and higher pain scores among the most deprived patients with some differences persisting at 1 year. Despite these disparities at intermediate time points, outcomes were generally comparable across socioeconomic groups at the MRF.

When comparing demographic characteristics across the socioeconomic groups, our findings indicate that patients in the least deprived group were older compared to the intermediate and least deprived groups. This aligns with previous research in multiple fields demonstrating that socioeconomically deprived populations often present with more severe disease progression by the time of diagnosis, potentially due to delayed access to preventative care.^16,19,29 While not investigated in this study, lower SDOH status has been associated with higher disease burden and comorbidity indices.^2,13 Further research is necessary to elucidate the relationship between socioeconomic deprivation, age, and health status in patients undergoing TSA.

Significant differences in demographics also appeared in the racial constitution of both TSA cohorts. Greater proportions of most deprived patients identified as Black or Hispanic, with greater portions of least deprived patients identifying as Caucasian/White. While ascertaining the nature of this connection between race and SDOH lies beyond the scope of our investigation, its potential influence on the availability of TSA cannot be ignored. As encountered in the literature, differences in rates of TSA utilization exist between racial identities that may, in turn, ultimately impact postoperative outcomes.^5,28 Applying knowledge of these documented discrepancies remains key to equitable access to treatment resources.

To minimize potential confounding from differing surgical indications, we analyzed aTSA and rTSA cohorts separately. Nearly all aTSA procedures were performed for glenohumeral osteoarthritis, while the rTSA group included a broader mix of indications, most commonly cuff tear arthropathy. Given the known differences in postoperative expectations and trajectories between these diagnoses, this stratified approach helps ensure that observed differences are more likely attributable to socioeconomic factors rather than underlying pathology.

Statistically significant differences in functional and reported outcomes were observed postoperatively at select time points. Interestingly, the most socioeconomically deprived group in the aTSA cohort experienced the highest range of aER postoperatively at 6 months and the worst ASES score at the MRF. The most socioeconomically deprived group in the rTSA cohort experienced the worst ASES score preoperatively, though with no significant postoperative differences were found across groups. This suggests that socioeconomic deprivation may negatively affect long-term shoulder function particularly after aTSA. In orthopedics, some studies have indicated worse patient reported outcomes in patients with a lower socioeconomic status.^24,26 Most deprived aTSA patients also reported greater mean pain scores postoperatively at 3 months, 1 year, and MRF appointments compared to their least deprived counterparts. More deprived rTSA patients, in parallel, experienced greater pain by 1 year follow-up than lesser deprived cohorts, though these differences resolved by MRF. This aligns with other studies that have published that patients with lower socioeconomic status may have greater difficulty with postoperative pain control.²⁷

Our findings contribute to a growing but mixed body of literature on the relationship between neighborhood-level socioeconomic status and TSA outcomes. Morgan et al. recently demonstrated the patients from socioeconomically disadvantaged neighborhoods experienced reduced ASES scores after TSA.¹⁷ These findings are consistent with our aTSA cohort, in which the most deprived patients had significantly lower ASES scores at MRF. In contrast, Moverman et al. found that neighborhood socioeconomic disadvantage did not predict outcomes or cost following elective shoulder arthroplasty.¹⁸ However, their analysis did not stratify by implant type or explore differences at multiple postoperative intervals. Our study adds to this discussion by incorporating both aTSA and rTSA cohorts, and evaluating outcomes at several time points, including longer-term follow-up.

Our collective findings demonstrate that TSA is able to deliver consistent long-term functional improvements across all socioeconomic groups. However, our findings of discordant outcomes among socioeconomically deprived cohorts hint that there is a need to incorporate socioeconomic metrics into clinical workflows for risk stratification and tailored patient care. Targeted interventions, including enhanced postoperative care coordination, patient education, and social support services, could help mitigate disparities. Moreover, policies aimed at improving healthcare access in underserved communities are vital to achieving equitable outcomes.

This study has several limitations. Its retrospective design and single-institution setting may restrict generalizability, particularly across diverse geographic and healthcare environments. While the ADI is a robust measure of community-level socioeconomic factors, it does not capture individual-level variables such as income, education, or insurance status, nor does it address factors like access to transportation or social support, which may influence recovery and adherence to care. The cohorts' exclusion criteria, including a minimum 2-year follow-up, may have introduced selection bias by excluding patients with shorter follow-up durations. These patients may differ demographically or socioeconomically from those included. Furthermore, the dataset did not account for granular factors like postoperative rehabilitation adherence, specific comorbidities not captured by the Charlson Comorbidity Index, or other potential confounders such as insurance status. Additionally, covariates including age and race were not controlled for in the statistical analysis, potentially confounding the relationship between ADI and outcomes. Finally, as an observational study, causality cannot be established. Prospective, multicenter research that incorporates both individual and community-level SDOH is needed to validate these findings and guide equitable care strategies in shoulder arthroplasty.

Conclusion

Despite differences in demographic and socioeconomic characteristics, TSA offers substantial pain relief and functional improvement across all ADI groups. Numerous disparities in postoperative outcomes appear and highlight the impact of socioeconomic deprivation on functional outcomes and pain relief across both cohorts. Future studies incorporating prospective protocols are needed to accurately elucidate differences among socioeconomic groups. Nevertheless, incorporating ADI into preoperative planning and postoperative care protocols presents an opportunity to optimize equitable outcomes in shoulder arthroplasty.

Disclaimers

Funding: No funding was disclosed by the authors.

Conflicts of interest: The 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.