Impact of area deprivation index on outcomes and health care utilization following shoulder arthroplasty: a systematic review

Abstract

Background

Socioeconomic factors play a critical role in shaping health care outcomes, yet their impact on surgical recovery and post-operative quality of life is often underexplored. The Area Deprivation Index (ADI) is a validated metric used to quantify socioeconomic disadvantage, providing insight into disparities in health care access and outcomes. Despite its relevance, the extent of ADI score reporting in the context of shoulder arthroplasty (SA) remains unclear. This systematic review assesses the extent to which ADI has been incorporated into SA studies and examines its association with post-operative outcomes and health care utilization.

Methods

A Preferred Reporting Items for Systematic Reviews and Meta-Analyses–compliant search of PubMed, Scopus, and Embase was conducted using the keywords “area deprivation index,” “shoulder arthroplasty,” and “outcomes.” Eligible studies, published in English within the past 10 years, included randomized controlled trials and cohort studies reporting ADI or related socioeconomic metrics in SA populations, focusing on post-operative outcomes, complications, health care utilization, and readmissions. We excluded studies that did not involve SA or failed to report ADI. In addition, case reports, reviews, editorials, conference abstracts, and non-English publications were excluded.

Results

Eight studies met the inclusion criteria, comprising 264,287 patients. Socioeconomic disadvantage, as measured by the ADI, was associated with lower pre-operative and post-operative functional scores, with 2 studies reporting significantly worse ASES scores in the most disadvantaged groups (P < .01). Readmission rates showed mixed findings; one study found higher odds of readmission in the most disadvantaged group (odds ratio = 1.56, P = .001), while others reported no significant association. Two studies identified a modest but significant increase in post-operative complications among disadvantaged patients (odds ratio range: 1.1-1.13, P < .005). Heterogeneity in ADI stratification and outcome reporting limited quantitative synthesis.

Conclusion

Socioeconomic disadvantage, as measured by the ADI, is associated with worse functional outcomes following shoulder arthroplasty. Patients from the most disadvantaged neighborhoods had lower pre-operative and post-operative patient-reported outcome measures, while findings on readmissions and complications were not consistent amongst studies. Integrating ADI into clinical decision-making may help identify at-risk patients and reduce disparities in SA outcomes.

Shoulder arthroplasty (SA) is increasingly performed due to the rising prevalence of shoulder arthritis and advancements in surgical techniques and implant design.²⁸ As the incidence of SA grows, understanding the influence of socioeconomic factors on surgical outcomes becomes critical, especially in a field where disparities in health care access and utilization are well documented.⁶ In the field of orthopedics, such factors have been shown to be correlated with higher incidence of amputation, lower rates of arthroplasty, and higher post-operative complications.^6,12,14 While SA tends to be associated with low complication rates and lengths of stay, lower socioeconomic status (SES) can negatively impact patients' outcomes and lead to measurable differences in complications. While such factors are often overlooked or under-reported in SA, research suggests both modifiable and nonmodifiable patient factors influence outcomes.¹⁷ In understanding the relationship between surgical outcomes and social determinants of health, various methods of assessing a patient's risk have been developed.¹⁰ These screening tools aim to take into account all aspects of a patient's life that may contribute to having poorer health outcomes. Neighborhood-level socioeconomic challenges, such as deprivation, have been shown to impact various health care metrics, yet their specific role in SA outcomes remains underexplored.²⁴

The Area Deprivation Index (ADI) is a validated tool used to quantify social determinants of health through the measurement of neighborhood-level socioeconomic disadvantage. A total of 17 variables from 4 domains (income, education, employment, and housing quality) are used to provide a composite score that reflects the relative deprivation of specific geographic areas.^11,23 A higher ADI score indicates greater levels of socioeconomic disadvantage,²⁷ and worse ADIs have been linked to poorer health outcomes for various medical conditions, including cardiovascular disease, diabetes management, and post-operative recovery after surgery.¹⁵ As health care systems increasingly recognize the importance of addressing social determinants of health, ADI has emerged as a valuable metric for identifying populations at greater risk of adverse outcomes.

Interestingly, the utilization of ADI in the context of SA remains limited. As SA procedures become more widespread, understanding the nuanced relationship between neighborhood-level deprivation scores and surgical outcomes is essential for developing equitable care models. This review aims to assess the reported utilization of ADI scores in SA studies, focusing on their association with post-operative outcomes and health care utilization. By critically evaluating the evidence, this review highlights trends, gaps, and potential opportunities for targeted interventions to address disparities in SA.

Methods

Search strategy

A literature search was reported following the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. In December 2024, 2 independent reviewers (A.M and H.B) searched PubMed, Embase, and Scopus for human clinical studies examining the impact of socioeconomic deprivation on post-operative outcomes and health care utilization after SA. The search string included “area deprivation index,” “shoulder arthroplasty,” and “outcomes,” with relevant Medical Subject Heading terms. Eligible studies were required to focus specifically on patients undergoing SA, excluding studies on rotator cuff repair, nonoperative treatment, or other types of upper-extremity surgeries. In addition, to be included studies had to have been published in English, within the past 10 years, and reported on post-operative outcomes, quality of life, complications, or readmissions. We excluded studies unrelated to SA, lacking ADI categorization data, or categorized as case reports, reviews, editorials, conference abstracts, or non-English publications. Two independent reviewers screened titles and abstracts. All titles and abstracts were reviewed for relevance, with studies meeting the criteria progressing to full-text review. Full-text review followed for studies meeting eligibility criteria. Discrepancies were resolved through discussion with the assistance of a third reviewer (K.S).

Data extraction

Data extraction followed a standardized template to ensure consistency across studies. Key details captured included author, year, study title, journal, study design, and level of evidence. Recorded population characteristics included sample size, mean age, sex distribution, body mass index (BMI), and socioeconomic metrics such as ADI stratification by quartile or tertile depending on the study. Extracted outcomes included post-operative patient-reported outcome measures (PROMS), readmission rates, complications, inpatient hospitalization costs, and post-operative emergency department (ED) visits.

Data synthesis

The primary aim of this study was to qualitatively assess the range of evidence relevant to our research question. We used descriptive statistics to summarize study characteristics and used a narrative approach to summarize relevant findings. Given the heterogeneity in ADI stratification (quartiles vs. tertiles) available and outcomes reported, a meta-analysis was not performed, and results are presented descriptively.

Results

Selection of sources

The initial search yielded 565 articles. After removal of duplicates (n = 225) and title/abstract screening (n = 340), 24 full-text articles were assessed, and 8 studies met the inclusion criteria. Figure 1 presents the Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram detailing study selection.

Figure 1.

PRISMA flow diagram depicting the article selection process. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses; ADI, Area Deprivation Index.

Characteristics of the studies included

All included studies were retrospective cohorts or observational designs with Level of Evidence III. Combined, these studies analyzed a total of 264,287 patients, with sample sizes ranging from 170 to 145,435. Four studies used ADI quartiles, while 3 used ADI tertiles, and 1 classified high ADI as > 95%. The mean age of the general cohort ranged from 67.1 to 74.5 years when reported. Female representation varied between 40% and 60.6%. Age data was not reported in 2 studies, and sex distribution was missing in one study. Demographic information for each study is presented in Table I.

Table I.

Overall characteristics of included studies.

Author	Yr	Journal	LOE	ADI categorization	Sample size	Mean age (SD)	Sex distribution; female (%)
Castle5	2024	SIA: JSES	III	Quartiles	248	67.9 (8.3)	58.90%
Sheth22	2020	JAAOS	III	Quartiles	982	67.1	40%
Schell20	2024	SIA: JSES	III	Quartiles	115,852	NR	56.10%
VanBoxtel26	2024	SIA: JSES	III	Tertiles	780	NR	60.60%
Moverman16	2022	JSES	III	Tertiles	380	69 (7.8)	59.20%
Morgan15	2024	JSES International	III	Tertiles	170	69.89 (6.95)	54.70%
Bethell3	2024	JSES	III	Tertiles	145,435	74.5 (5.8)	59.50%
Gordon7	2023	HSS Journal	III	>95% ADI = high	49,440	70	NR

LOE, level of evidence; ADI, Area Deprivation Index; SD, standard deviation; SIA: JSES, Seminars in Arthroplasty: Journal of Shoulder and Elbow Surgery; JAAOS, Journal of the American Academy of Orthopaedic Surgeons; JSES, Journal of Shoulder and Elbow Surgery; HSS Journal, The Musculoskeletal Journal of Hospital for Special Surgery; NR, not reported.

A comparative analysis of sample size, age, and BMI between the least and most disadvantaged ADI cohorts was conducted. Age comparisons indicated generally similar distributions between the least and most disadvantaged cohorts, with mean ages ranging from 64.9 to 74.9 years. BMI data, where available, showed a trend for higher BMI in the most disadvantaged groups compared to the least disadvantaged groups (Table II).

Table II.

Sample size, age, and BMI comparisons between lowest ADI and highest ADI cohorts.

Author	Yr	Lowest ADI sample size	Highest ADI sample size	Lowest ADI mean age (SD)	Highest ADI mean age (SD)	Lowest ADI BMI (SD)	Highest ADI BMI (SD)
Castle5	2024	31	65	67.0 (8.6)	68.8 (9.6)	30.3 (5.4)	32.5 (8.0)
Sheth22	2020	247	249	67.1 (9.2)	67.1 (10.4)	28.7 (5.4)	31.6 (6.5)
Schell20	2024	23,960	21,745	70.2 (8.98)	69.4 (9.30)	NR	NR
VanBoxtel26	2024	195	214	70.0 (10.3)	64.9 (11.3)	NR	NR
Moverman16	2022	179	72	70.5 (7.6)	68.4 (7.7)	29.9 (6.0)	30.3 (5.7)
Morgan15	2024	46	34	71.32 (6.38)	69.37 (7.36)	29.06 (5.09)	29.56 (5.85)
Bethell13	2024	20,147	10,942	74.9 (5.7)	74.3 (6.0)	NR	NR
Gordon7	2023	24,720	24,720	NR	NR	NR	NR

BMI, body mass index; ADI, Area Deprivation Index; SD, standard deviation; NR, not reported.

Patient-reported outcome measures

Four studies analyzed PROMs, with most reporting that patients in disadvantaged ADI groups had lower baseline and post-operative scores.^3,5,16,27 Pre-operatively, Sheth et al and Morgan et al found significantly lower ASES scores in the most disadvantaged group (P < .001), while Moverman et al reported no significant difference (P = .72) (Table III).5, 6, 7^,22,27 Morgan et al found that disparities persisted post-operatively, with significantly lower ASES scores in the most disadvantaged quartile at 6 months (75.32 vs. 84.12, P = .01) and 2 years (80.51 vs. 88.87, P < .01). Castle et al³ similarly found lower Patient-Reported Outcomes Measurement Information System Upper Extremity (PROMIS-UE) scores at 12 months in the most disadvantaged highest ADI quartile (33.9 ± 8.9 vs. 41.7 ± 9.2; P = .004). In contrast, Moverman et al⁵ found no significant differences in PROMs across ADI groups, suggesting potential variability in socioeconomic disparities at the individual patient level.

Table III.

Pre-operative and 2-year post-operative ASES scores by socioeconomic disadvantage.

Author	Yr	Lowest ADI pre-operative ASES (SD)	Highest ADI pre-operative ASES (SD)	P value	2 yr lowest ADI post-operative ASES (SD)	2 yr highest ADI post-operative ASES (SD)	P value
Sheth22	2020	31.7 (18.1)	39.8 (19.3)	<.001	NR	NR	NR
Moverman16	2022	35.7 (16.4)	37.1 (16.9)	.72	84.3	84	.56
Morgan15	2024	49.24 (17.84)	40.83 (16.70)	<.01	88.87 (13.30)	80.51 (16.86)	<.01

ASES, American Shoulder and Elbow Surgeons; ADI, Area Deprivation Index; SD, standard deviation; NR, not reported.

Readmission rates

Four studies evaluated the association between ADI and readmission rates, with mixed results.^3,8,20,26 Bethell et al reported a significantly higher odds ratio (OR) of readmission in the most disadvantaged group (OR = 1.56, P = .001). Schell et al found higher readmission rates in the most disadvantaged cohort (P = .001), though the OR was not reported. VanBoxtel et al observed no significant association between ADI and readmission risk (OR = 0.87, P = .044).^20,26 Gordon et al⁸ found no difference in readmission rates between the >95% ADI cohort and their control groups (OR = 1.03, P = .633) (Table IV). These findings suggest that while socioeconomic disadvantage may contribute to readmission risk, its reported impact has varied across studies.

Table IV.

Readmission rates by socioeconomic disadvantage.

Author	Comparison	Readmission rate (%)	Odds ratio (95% CI)	P value
VanBoxtel et al26	Most vs. least disadvantaged	11.2% vs. 8.7%	0.87 (0.4-1.87)	.044
Schell et al20	Most vs. least disadvantaged	10.3% vs. 8.1%	NR	.001
Bethell et al13	Most vs. least disadvantaged	NR	1.56 (1.37-1.78)	.001
Gordon et al7	Most (>95% ADI) vs. control	2.66% vs. 2.71%	1.03 (0.92-1.14)	.633

ADI, Area Deprivation Index; CI, confidence interval; NR, not reported.

Post-operative emergency department visits

VanBoxtel et al and Gordon et al examined the association between socioeconomic disadvantage and post-operative ED visits. VanBoxtel et al⁵ reported comparable ED visits between the least disadvantaged and most disadvantaged ADI cohorts in their study with ORs ranging from 0.28 to 0.73 (P > .36). Similarly, Gordon et al⁸ found no significant difference in ED visit rates between the most and least deprived groups. These findings demonstrate that ADI may not be a strong predictor of post-operative ED utilization.

Post-operative complications

Two studies assessed complication rates in the most disadvantaged ADI cohorts. Gordon et al⁸ reported a higher overall complication rate in the most disadvantaged group (10.84% vs. 9.45%) with a significant increase in odds (OR = 1.1, 95% confidence interval: 1.03-1.18, P = .005) (Table V). Similarly, Schell et al²⁰ found an increased risk of complications in the most disadvantaged group (OR = 1.13, 95% confidence interval: 1.051.22, P = .001). These findings suggest that socioeconomic disadvantage is associated with a modest but significant increase in post-operative complications.

Table V.

Summary of included articles.

Study	Sample size	ADI stratification	Primary outcome	Key findings
Morgan et al15	11,520	Quartiles	PROMs (ASES)	Lower ASES scores at 6 mo (75.32 ± 16.06 vs. 84.12 ± 13.61, P = .01) and 2 yr (80.51 ± 16.86 vs. 88.87 ± 13.30, P < .01) for the most disadvantaged group
Castle et al5	8,215	Tertiles	PROMIS-UE	Lower PROMIS-UE scores at 12 mo in the most disadvantaged quartile (33.9 ± 8.9 vs. 41.7 ± 9.2; P = .004)
Moverman et al16	6,740	Quartiles	PROMs	No significant difference in PROMs across ADI groups
Bethell et al3	9,476	Quartiles	Readmission	Dose-dependent increase in readmission with higher ADI quartiles
Van Boxtel et al26	7,924	Quartiles	Health care utilization	No significant association between ADI and hospital resource utilization
Mandalia et al13	12,320	Tertiles	Complications	Higher fracture and infection rates in highest ADI tertile
Gordon et al7	13,627	Quartiles	Complications	No significant relationship between ADI and post-operative complications
Schell et al20	128,376	Quartiles	Health care disparities	Medicaid/Medicare patients had increased complication, readmission, and mortality risks

ADI, Area Deprivation Index; ASES, American Shoulder and Elbow Surgeons; PROMs, patient-reported outcome measures; PROMIS-UE, Patient-Reported Outcomes Measurement Information System Upper Extremity.

Discussion

This review identified 8 retrospective studies that met our inclusion criteria, with a total of 264,287 patients and a mean age ranging from 67.1 to 74.5 years with female representation accounting for 40%-60.6%. Higher BMI was generally observed in more disadvantaged ADI groups. Most studies found lower baseline and post-operative PROMs in high-ADI (disadvantaged) groups. Significant disparities in American Shoulder and Elbow Surgeons (ASES) and PROMIS-UE scores persisted at 6 months and 2 years post-operatively. Morgan et al found significantly lower ASES scores at 6 months (75.32 vs. 84.12, P = .01) and at 2 years (80.51 vs. 88.87, P < .01). Castle et al also reported lower PROMIS-UE scores at 12 months in the highest ADI quartile (33.9 ± 8.9 vs. 41.7 ± 9.2; P = .004). Regarding complications, disadvantaged patients had higher rates, with Gordon et al reporting 10.84% vs. 9.45% (OR = 1.1, P = .005) and Schell et al reporting OR = 1.13 (P = .001). Readmission rates were mixed; Bethell et al found significantly increased odds (OR = 1.56, P = .001), while other studies showed no difference. No significant association was found between ADI and post-operative ED visits.

Interestingly, while several studies have reported increased readmission rates among socioeconomically disadvantaged patients undergoing SA, findings remain inconsistent across the literature. A recent single-institution study by Saunders et al¹⁹ evaluated the association between ADI and post-operative outcomes in a matched cohort of 374 patients undergoing SA and found that patients in the highest ADI group had significantly lower 30- and 90-day readmission rates compared to those from less disadvantaged areas (0.5% vs. 5.3%, P = .01), despite a nonsignificant trend toward longer length of stay. This counterintuitive result contrasts with prior findings, including Bethell et al and Schell et al, who reported increased readmissions in high-ADI groups. One possible explanation, as noted by Saunders et al, is that patients from disadvantaged neighborhoods may experience greater barriers to health care access, leading to underutilization of post-operative care services even when medically necessary. These findings underscore the complexity of interpreting readmission as a quality metric and suggest that lower readmission rates in high-ADI groups may reflect unmet needs rather than better outcomes.

These findings underscore the likelihood that socioeconomic deprivation, as measured by the ADI, translates into tangible disparities in surgical outcomes following SA. Patients from the most disadvantaged neighborhoods consistently reported lower pre-operative and post-operative functional scores, suggesting that deprivation affects not only access to surgery but also post-operative recovery trajectories.^3,5,15,16 The persistent gap in PROMs raises important questions about the mechanisms driving these disparities.^4,9 One explanation may be that it is more difficult for deprived patients to access health care, or that they may have a worse overall health status for the same degree of disease severity.²⁹ Deprived patients may present for elective surgery with more advanced disease and higher burdens of chronic diseases secondary to socioeconomic factors, which can hinder optimal recovery, particularly when compounded by disparities in access to postoperative rehabilitation.^1,25,29 When patients have more consistent access to post-operative therapy and rehabilitation services, it can result in faster recovery, better health outcomes, and a reduced risk of complications.¹⁸ Interestingly, while lower ADI was associated with worse PROMs and functional recovery, it did not consistently predict higher rates of ED visits or readmissions across studies.⁸

Several other studies also suggest that socioeconomic deprivation may have a more pronounced effect on long-term recovery trajectories rather than acute post-operative health care utilization.^2,13 In a retrospective study of 463 patients that underwent total knee arthroplasty, ADI was not a predictor for 90-day post-operative ED visit.²¹ In contrast, Bethell et al³ did report an increase in readmission rates with higher ADI quartiles, highlighting the variability in how deprivation influences post-operative care-seeking behavior. Similarly, a nationwide study assessing health care utilization after patients undergo 1- to 2- level lumbar fusion demonstrated that patients with a high ADI incurred higher rates and odds of ED visits within 90 days.⁷ In a study assessing factors that impact healt hcare utilization following SA, Schell et al²⁰ found that insurance status was a significant factor, with Medicaid patients having higher rates of readmission, revision, and nonhome discharge. These inconsistencies suggest that SES alone does not fully explain variations in readmission rates, and that additional factors, such as insurance status, hospital system differences, and inaccessible health care services, play a role as well. The lack of a consistent relationship between ADI and ED visits suggests that patients from disadvantaged neighborhoods may either avoid seeking emergency care due to financial barriers or rely on alternative health care pathways that are not captured in readmission data.

These findings raise broader concerns about whether current health care models adequately address the needs of socioeconomically disadvantaged SA patients. The data suggest that targeted interventions may be necessary to bridge these disparities, including structured pre-operative optimization programs for high-ADI patients, financial assistance for post-operative rehabilitation, and enhanced perioperative education to improve patient engagement in recovery. Without such measures, disparities in functional outcomes are likely to persist despite advances in surgical techniques and perioperative care. While this review highlights clear socioeconomic disparities, several critical knowledge gaps remain. The interaction between ADI and hospital-level factors, such as surgeon volume and institutional resources, remains unclear. Future research should also explore whether integrating ADI into clinical decision-making, alongside traditional surgical risk factors, could enhance personalized treatment planning.

The present study is not without its limitations. Our inclusion criteria focused on studies utilizing ADI as a validated measure of socioeconomic disadvantage and those reporting data disaggregated by ADI category. As a result, studies using alternative socioeconomic measures were not included, since our aim was specifically centered on the application of ADI and its findings in the literature. In addition, the variability in ADI stratification methods and reported outcomes limited our ability to conduct a meta-analysis. However, to address this limitation, we provided a descriptive analysis comparing the least and most disadvantaged cohorts as defined by each of the included studies. Lastly, the risk of bias and methodological rigor of included studies were not assessed, resulting in a descriptive rather than strictly analytical approach.

This review reinforces the notion that SES is a critical but often overlooked determinant of SA outcomes. While disparities in functional recovery persist, their impact on health care utilization remains less clear, underscoring the need for further research and targeted interventions. As SA rates continue to rise, orthopedic care must evolve beyond a purely surgical focus. Integrating ADI into clinical decision-making could help identify at-risk patients and guide the development of interventions to mitigate disparities in surgical outcomes.

Conclusion

Socioeconomic disadvantage, as measured by the ADI, is associated with worse functional outcomes following SA. While PROMs consistently reflected disparities, associations with readmission and complications were mixed. Notably, recent data from Saunders et al¹⁹ challenge the prevailing assumption that greater deprivation always correlates with higher health care utilization, reporting lower readmission rates in high-ADI populations. This paradox may reflect barriers to care rather than better recovery. As such, integrating ADI into clinical decision-making could help identify patients at risk for poor outcomes or inadequate follow-up, enabling tailored interventions to mitigate disparities in orthopedic surgical care.

Disclaimers:

Funding: There are no sources of funding for this study.

Conflicts of interest: Erick M. Marigi has received financial consulting payments from Zimmer Biomet which are not related to the subject of this article.

John W. Sperling, M.D. has received financial payments for consulting, research, and royalties from Zimmer Bioment, Pacira, and Medacta, which are outside the submitted work.

Joaquin Sanchez-Sotelo, M.D. has received financial payments for consulting, research, and royalties from Stryker, Exactech, Synthes, and Acumed, which are outside the submitted work.

Any additional 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.

Given his role as Editor in Chief, Dr. John W. Sperling had no involvement in the peer-review of this article and has no access to information regarding its peer-review. Full responsibility for the editorial process for this article was delegated to Dr. Peter N. Chalmers.