Abstract
Background
Inequities in healthcare delivery and outcomes remain pervasive. The United States’ Centers for Medicare and Medicare Services (CMS) started confidentially reporting to hospitals data on disparities in readmission rates for Medicare beneficiaries in 2019. Whether hospitals with readmission disparities are more likely to establish equity efforts is an important policy question. We examined relationships between hospitals with readmission disparities for Medicare beneficiaries in 2019, and hospital equity efforts in 2022, measured by (a) the presence of an equity officer and (b) the strength of a hospital’s equity environment.
Methods
We conducted a retrospective study of US hospitals in the American Hospital Association Annual Survey that were eligible for CMS’ Hospital-Wide Readmission and Disparity measures using 2019 Medicare data. Outcomes: hospital equity efforts in 2022 measured by the presence of a hospital equity officer and a hospital’s equity environment composite score. Exposure: Disparities in 2019 Medicare readmissions by insurance (dual-eligible vs. non-dual-eligible patients) and by race (Black vs. White patients). Covariates: hospital characteristics. We used regression analyses to examine relationships between hospitals with and without disparities in readmissions in 2019 (by insurance and race separately) and our two outcomes.
Results
2019 hospital-level disparities by insurance conferred a 1.35 times increased odds (95% CI: 1.17–1.56) of having an equity officer and 0.76 (0.22) point increase in equity environment composite scores in 2022 in unadjusted analyses. These relationships were not significant after adjusting for hospital covariates. There was no relationship between disparities by race and 2022 equity efforts.
Conclusions
Our findings suggest that hospital-level readmission disparities do not necessarily incentivize hospital-level equity efforts. Hospital organizational type, which likely influences not only its patient population but also hospital culture, may be a stronger predictor of hospital equity efforts. This study contributes to the discussion of how we measure, report on and create accountability to equity at the hospital-level.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12913-025-13874-4.
Keywords: Readmissions, Equity officers, Quality measures, Outcome measurement, Disparities
Introduction
Reducing health disparities has been a U.S. priority for decades, yet inequitable care access, chronic disease burden, and premature death remain pervasive [1, 2]. While the root causes of health disparities are complex and inherently structural [3]. disparities can also be traced to differential quality of hospital-based care [4–10].
Hospitals have unique influence and potential to propagate or mitigate health disparities. Recognizing this, regulatory agencies such as the Centers for Medicare and Medicaid Services (CMS) and the National Committee for Quality Assurance have used accreditation and public reporting programs to hold hospitals accountable to reducing disparities [11–14]. For example, the Joint Commission’s equity standards now instruct hospitals to “designate an individual(s) to lead activities to reduce healthcare disparities [14].
Despite these measures hospitals have been slow to adopt robust equity efforts [15, 16]. The Equity Officer National Study (EONS) investigated equity officers nationwide; responsibilities ranged from community benefits/relations, population/community health, and health equity – all patient-facing roles – to workforce – a staff facing role [17]. This study also revealed variation in the strength of hospital equity environments (e.g., reporting structure, budget, strategic planning) and scope of hospital equity efforts (e.g., instituting policies and practices, partnering with community-based organizations) [15, 17–19].
Whether hospitals with clinical disparities are more likely to have equity efforts is an important and relevant policy question. CMS started confidentially reporting data to hospitals on disparities in readmissions for Medicare beneficiaries in 2018 (later added disparities data for mortality measures) [13, 20, 21]. Whether hospitals were motivated or impacted by confidential reporting remains unknown. Knowledge of disparities might have encouraged equity efforts. On the other hand, institutions with disparities may not have had capacity for equity efforts [21]. Now in 2025, with the future of the disparities reporting program and hospital equity efforts uncertain, it is even more important to address these knowledge gaps.
With these questions in mind, we examined relationships between hospitals with disparities in readmission rates for Medicare beneficiaries in 2019, and hospital equity efforts - the presence of an equity officer and the strength of a hospital’s equity environment – in 2022.
Methods
Study overview
We sought to understand the relationship between a 2019 exposure - disparities in Medicare readmission rates – and 2022 outcomes – hospital equity efforts. For this study, we assessed disparities in 2019 readmission rates for two at-risk groups. (1) We first examined disparities by insurance, comparing outcomes for patients dual-eligible for both Medicaid and Medicare (dual-eligible), to patients only insured by Medicare. (2) We then examined disparities by race comparing outcomes for Black patients to White patients. We employed the equitable readmission measure derived from the two CMS Disparity Methods (see Exposure) [21]. We characterized hospital equity efforts by the presence of an equity officer (primary outcome) and a composite score assessing a hospital’s equity environment (secondary outcome).
Data sources
We used 2019 Medicare inpatient claims data to identify hospitals with disparities. Patient-level demographic data were extracted from the Medicare Master Beneficiary Summary File and Enrollment Database. Hospital-level patient demographics (e.g., proportion of patients Medicaid insured, etc.) were extracted from the 2019 Medicare fee-for-service claims’ Provider of Service files.
We used the Equity Officer National Survey (EONS), which we conducted in collaboration with the American Hospital Association (AHA) in 2022–2023, and the 2022 AHA annual survey to measure our two outcomes (presence of equity officer and equity environment composite score), and additional hospital characteristics [22, 23].
Cohort
Of the 6,193 hospitals that completed the 2022 AHA Survey (Fig. 1), we excluded hospital types (see Fig. 1) in line with criteria outlined by the EONS as well as hospitals without sufficient Medicare claims data to calculate readmission rates [21]. We further excluded hospitals caring for insufficient patients to measure disparities by insurance and by race (see “Primary Exposure”), creating two cohorts of hospitals: an insurance disparities cohort and a race disparities cohort. We additionally excluded hospitals with insufficient data to determine the presence of an equity officer (see “Primary Outcome”), and the Equity Environment Composite Score (see “Secondary Outcome”).
Fig. 1.
Flowchart – development of study cohorts for primary and secondary analyses. Legend: a These hospitals did not have associated Medicare fee-for-service claims data to calculate CMS’s Hospital-Wide Readmission measure. b The Centers for Medicare and Medicaid (CMS)’s Disparity Methods cannot be used to calculate disparities if hospitals did not serve at least 25 beneficiaries in the group at-risk for disparities in care during the study period (insurance disparities analysis: beneficiaries dual-eligible for Medicare and Medicaid; race disparities analysis: Black identifying beneficiaries). c Hospitals were excluded from analyses if we could not determine whether or not they had an equity officer in 2022 – i.e. they did not have responses to the American Hospital Association (AHA) Annual survey questions F6 or did not have response to the Equity Officer National Survey (EONS) triage questions or full survey. d Hospitals were excluded from analyses if they did not have responses to each of the five AHA annual survey questions/domains included in our composite score of hospitals’ equity environment
Primary exposure: Hospital-level disparities
We defined hospital-level disparities by the previously defined equitable readmissions measure, identifying hospitals with equitable readmissions and those without equitable readmissions (i.e., disparities in readmissions) [21]. To be categorized as having equitable readmissions a hospital must meet two criteria corresponding to two CMS’s Disparity Methods (Supplemental Methods Fig. 1a/b): (1) The Across-Hospital Method: a hospital’s risk-adjusted readmission rate for the at-risk group must be lower (better) than the median hospital-level risk-adjusted readmission rate for all eligible hospitals nationally, for the same at-risk population; and, (2) The Within-A-Single-Hospital Method: the difference in risk-adjusted readmission rates between the populations at-risk and not at-risk for disparities within an individual hospital must be less than 1%. Therefore, hospitals with equitable readmissions must be better at treating the at-risk group compared to peer hospitals and only have small differences in outcomes between the at-risk population and other patients within their hospital [21].
We note two key considerations. The first is that measuring disparities for different at-risk groups likely impacts disparities outcomes [21]. Therefore, based on CMS standards and prior work [21], we measured equitable readmissions (i.e. disparities) by insurance and race separately. (1) We measured insurance disparities by comparing outcomes for dual-eligible patients to non–dual-eligible. (2) We measured race disparities by comparing non-Hispanic Black patients to non-Hispanic White patients [24]. We focused on the latter for two reasons: (a) literature demonstrates inequities in readmissions between Black and White patients [25] and (b) while Medicare’s race data are not fully reliable, Black race specifically is more consistent with self-report [26, 27]. Our second consideration is recognizing that many US hospitals do not have sufficient patients to calculate disparities. By CMS Disparity Methods specifications, to measure outcome disparities hospitals must care for ≥25 patients from the population at-risk (Fig. 1). As a result, many hospitals were excluded from the measure and the hospital cohorts used to analyze insurance and race disparities were different.
Primary outcome: presence of an equity officer in 2022
We classified hospitals as having an equity officer if they met any of the following criteria: (1) senior executive or middle management responsible for equity goals (AHA survey); (2) “yes” to the EONS survey triage question “Does your hospital have an equity officer?”; or (3) completed the full EONS survey. We excluded hospitals that did not have data for any of these three sources. Our goal was to create a sensitive definition and not omit hospitals with an equity officer.
We explored a more specific definition for presence of an equity officer (classifying hospitals with an equity officer only if data from all three criteria (listed above) were consistent, with no conflicts or missing data. However, with this definition, in combination with exclusions due to Disparity Methods specifications, we did not have sufficient hospitals to conduct analyses.
Secondary outcome: 2022 equity environment composite score
We created an Equity Environment Composite Score (EECS) based on five domains, corresponding to pertinent questions in the AHA Survey Section F: “Addressing patient social needs and community social determinants of health (SDOH),” and in alignment with previously published methods [28]. The five domains included: (1) SDOH screening, (2) SDOH programs, (3) Community engagement, (4) Diversity, Equity, and Inclusion Data, and (5) Strategic plans. Including only hospitals with answers to each of the five questions included in the EECS (see Fig. 1; Supplemental Methods), we first summed scores for all questions within each domain and then transformed each domain score into a tercile score (i.e., 0 and 1, 2, 3). We then added tercile scores, weighting each domain equally, to create a composite score (0–15). Hospitals with a higher composite score were presumed to have stronger equity environments (See Supplemental Methods).
Covariates
We included hospital covariates in our analysis based on a conceptual model [21] and prior literature [29–31]. These hospital characteristics, measured from the AHA Survey, included ownership status, teaching status, urbanicity, geographic location, number of staffed beds, and nurse to bed ratio. We also included hospital-level patient demographics (mean percentage of patients insured by Medicare, dual-eligible, and who identify as Black and White) [31].
Analyses
We conducted each of the following analyses separately for the cohort of hospitals included in the insurance disparities analysis and the cohort of hospitals included in the race disparities analysis. We first compared hospital characteristics and patient demographics of hospitals with and without disparities in readmissions using descriptive statistics. Next, we used logistic and linear regression to examine relationships between hospitals with and without disparities in readmissions in 2019 and our primary and secondary outcomes. Adjusted models included all hospital covariates.
To investigate how hospital covariates moderated relationships between 2019 disparities and 2022 hospital equity efforts, we conducted two additional analyses. We first examined simple bivariate associations between hospital covariates and our primary and secondary outcomes. We then stratified these bivariate results by 2019 readmission disparities to see whether relationships between hospital covariates and equity outcomes differed for hospitals with and without disparities in 2019.
Results
Of the 6,193 hospitals completing the AHA annual survey, we included 4,324 hospitals after excluding selected hospital types (Fig. 1). Examining insurance disparities, 2,047 hospitals had sufficient data to measure our primary outcome - presence of an equity officer - and 1,691 hospitals for our secondary outcome - EECS. To examine racial disparities, 1,195 hospitals had sufficient data for our primary outcome and 1,026 hospitals for our secondary outcome. Characteristics of hospitals included and excluded (due to missing AHA data) are found in Supplemental Tables 1a/b.
Of the 2,047 hospitals included in our insurance disparities analysis, 1,722 (84.1%) had disparities in readmissions for dual-eligible patients. Of the 1,195 hospitals included in our racial disparities analysis, 845 hospitals (70.7%) had disparities in readmissions. As published previously, [21]hospitals with disparities, by either insurance or race, were more likely to be teaching hospitals, urban, larger in size, and have a higher proportion of Black patients (Table 1).
Table 1.
Characteristics of hospitals with and without readmission disparities, 2019
| 2019 Insurance Disparities Comparing dual-eligible vs. non-dual-eligible patients |
2019 Race Disparities Comparing Black vs. White patients |
|||||||
|---|---|---|---|---|---|---|---|---|
| Hospital Characteristic | Total = 2047 | Hospitals WITH disparities N = 1722 | Hospitals WITHOUT disparities N = 325 |
Total = 1195 | Hospitals WITH disparities N = 845 | Hospitals WITHOUT disparities N = 350 | ||
| Categorical Variables | N (Col %) | N (Col %) | N (Col %) | p-value | N (Col %) | N (Col %) | N (Col %) | p-value |
| Ownership | 0.002 | 0.957 | ||||||
| Government, nonfederal | 327 (16.0%) | 254 (14.8%) | 73 (22.5%) | 168 (14.1%) | 119 (14.1%) | 49 (14.0%) | ||
| Nongovernment, not-for-profit | 1604 (78.4%) | 1368 (79.4%) | 236 (72.6%) | 949 (79.4%) | 672 (79.5%) | 277 (79.1%) | ||
| Investor-owned, for-profit | 116 (5.7%) | 100 (5.8%) | 16 (4.9%) | 78 (6.5%) | 54 (6.4%) | 24 (6.9%) | ||
| Teaching Status | < 0.001 | 0.002 | ||||||
| Non-teaching Hospital | 783 (38.3%) | 578 (33.6%) | 205 (63.1%) | 256 (21.4%) | 161 (19.1%) | 95 (27.1%) | ||
| Teaching Hospital | 1264 (61.7%) | 1144 (66.4%) | 120 (36.9%) | 939 (78.6%) | 684 (80.9%) | 255 (72.9%) | ||
| Urbanity | < 0.001 | < 0.001 | ||||||
| Rural | 658 (32.1%) | 487 (28.3%) | 171 (52.6%) | 138 (11.5%) | 78 (9.2%) | 60 (17.1%) | ||
| Urban | 1389 (67.9%) | 1235 (71.7%) | 154 (47.4%) | 1057 (88.5%) | 767 (90.8%) | 290 (82.9%) | ||
| Region | 0.004 | 0.549 | ||||||
| West | 419 (20.5%) | 357 (20.7%) | 62 (19.1%) | 203 (17.0%) | 152 (18.0%) | 51 (14.6%) | ||
| Midwest | 370 (18.1%) | 297 (17.2%) | 73 (22.5%) | 168 (14.1%) | 117 (13.8%) | 51 (14.6%) | ||
| Northeast | 377 (18.4%) | 337 (19.6%) | 40 (12.3%) | 241 (20.2%) | 167 (19.8%) | 74 (21.1%) | ||
| South | 881 (43.0%) | 731 (42.5%) | 150 (46.2%) | 583 (48.8%) | 409 (48.4%) | 174 (49.7%) | ||
| Number of beds | < 0.001 | < 0.001 | ||||||
| 6–99 beds | 772 (37.7%) | 561 (32.6%) | 211 (64.9%) | 178 (14.9%) | 104 (12.3%) | 74 (21.1%) | ||
| 100–199 beds | 451 (22.0%) | 388 (22.5%) | 63 (19.4%) | 280 (23.4%) | 168 (19.9%) | 112 (32.0%) | ||
| 200–299 beds | 275 (13.4%) | 250 (14.5%) | 25 (7.7%) | 226 (18.9%) | 166 (19.6%) | 60 (17.1%) | ||
| 300–399 beds | 190 (9.3%) | 180 (10.5%) | 10 (3.1%) | 168 (14.1%) | 126 (14.9%) | 42 (12.0%) | ||
| 400 or more beds | 359 (17.5%) | 343 (19.9%) | 16 (4.9%) | 343 (28.7%) | 281 (33.3%) | 62 (17.7%) | ||
| Continuous variables | Total N | Median (IQR) | Median (IQR) | p-value | Total N | Median (IQR) | Median (IQR) | p-value |
| Nurse-to-bed Ratio | 2047 | 1.6 (1.1–2.1) | 1.4 (0.9–1.9) | < 0.001 | 1195 | 1.6 (1.2-2.0) | 1.6 (1.2-2.0) | 0.985 |
| Medicare population % | 2047 | 50.0 (43.0-57.3) | 54.3 (46.7–64.3) | < 0.001 | 1195 | 48.1 (41.3–55.3) | 50.1 (43.1–56.2) | 0.003 |
| Dual-eligible population % | 2047 | 19.5 (14.0–26.0) | 17.5 (9.6–25.1) | < 0.001 | 1195 | 20.7 (15.2–27.7) | 19.4 (13.3–26.0) | 0.004 |
| Black population % | 2047 | 3.1 (0.6–9.9) | 1.5 (0.0-8.4) | < 0.001 | 1195 | 8.2 (3.6–18.0) | 6.6 (2.9–15.5) | 0.023 |
| White population % | 2047 | 90.2 (78.0-95.3) | 92.6 (80.6–97.5) | 0.01 | 1195 | 82.7 (67.7–90.8) | 84.4 (74.1–91.5) | 0.095 |
Readmission disparities in 2019 and hospital equity officers in 2022
For our insurance disparities analysis, 1,706 of 2,047 total hospitals (83.3%) had an equity officer in 2022 (Table 2a). Of the 1,722 hospitals with insurance disparities, 1,461 (84.8%) had equity officers, compared to 245/325 (75.4%) hospitals without disparities. Therefore, disparities for dual-eligible patients in 2019 conferred a 1.35 times increased odds (95% CI: 1.17–1.56) of having an equity officer in 2022 in unadjusted analyses. This relationship was no longer statistically significant after adjusting for hospital covariates.
Table 2a.
The relationship between hospital disparities in 2019 with equity efforts in 2022
| Hospitals WITH and WITHOUT an Equity Officer in 2022 | Likelihood of having a hospital equity officer in 2022 | |||||
|---|---|---|---|---|---|---|
| WITH | WITHOUT | Unadjusted OR | Adjusted a OR | |||
| N(%) | N(%) | OR (95% CI) | p | OR (95% CI) | p | |
| Disparities by Insurance in 2019 (Total N = 2047) | 1706 (83.3%) | 341 (16.7%) | ||||
| Hospitals WITH disparities (N = 1722) | 1461 (84.8%) | 261 (15.2%) | 1.35 (1.17–1.56) | < 0.0001 | 1.02 (0.87–1.20) | 0.770 |
| Hospitals WITHOUT disparities (N = 325) | 245 (75.4%) | 80 (24.6%) | ref | ref | ||
| Disparities by Race in 2019 (Total N = 1195) | 1051 (87.9%) | 144 (12.1%) | ||||
| Hospitals WITH disparities (N = 845) | 748 (88.5%) | 97 (11.5%) | 1.09 (0.91–1.32) | 0.347 | 1.03 (0.83–1.27) | 0.812 |
| Hospitals WITHOUT disparities (N = 350) | 303 (86.6%) | 47 (13.4%) | ref | ref | ||
Table 2b.
The relationship between hospital disparities in 2019 with equity efforts in 2022
| Effect on Equity Environment Composite Score in 2022 | |||||
|---|---|---|---|---|---|
| Median Equity Environment Composite Score in 2022 | Unadjusted Estimate (S.E.) | Adjusted Estimate (S.E.) | |||
| Median (IQR) | (95% CI) | p | (95% CI) | p | |
| Disparities by Insurance in 2019 (Total N = 1691) | 10.0 (8.0–13.0) | ||||
| Hospitals WITH disparities (N = 1449) | 10.0 (8.0–13.0) | 0.76 (0.22) | < 0.001 | -0.13 (0.21) | 0.551 |
| Hospitals WITHOUT disparities (n-242) | 9.5 (7.0–13.0) | ref | ref | ||
| Disparities by Race in 2019 (Total N = 1026) | 11.0 (9.0–13.0) | ||||
| Hospitals WITH disparities (N = 736) | 11.0 (9.0–13.0) | 0.29 (0.20) | 0.152 | -0.20 (0.20) | 0.316 |
| Hospitals WITHOUT disparities (n-290) | 10.5 (8.0–13.0) | ref | ref | ||
a Adjusted for hospital characteristics listed in Table 1: Ownership, teaching status, urbanity, region, number of beds, nurse to bed ratio, percent of Medicare patients, percent of senior patients aged 65 + dual-eligible for Medicare and Medicaid
For our race disparities analysis, 1,051 of 1,195 total hospitals (87.9%) had an equity officer in 2022. Of the 845 hospitals with racial disparities in 2019, 748 (88.5%) had equity officers, compared to 303/350 (86.6%) hospitals without disparities. Relationships between disparities by race and the presence of an equity officer were not statistically significant in unadjusted or adjusted analyses (Table 2a).
Readmission disparities in 2019 and hospitals’ equity environment composite score in 2022
Among the 1,691 hospitals in the insurance disparities analysis, the median 2022 EECS was 10.0 (IQR 8.0–13.0) (Table 2b). Hospitals with 2019 insurance disparities had a higher median score (10.0, IQR: 8.0–13.0) than hospitals without disparities (9.5, IQR: 7.0–13.0) in unadjusted analyses (p < 0.001), however these differences were not statistically significant in adjusted analyses (Table 2b).
Among the 1026 hospitals in the racial disparities analysis, the median 2022 EECS was 11.0 (IQR: 9.0–13.0). Hospitals with disparities in readmissions for Black patients in 2019 had higher median scores (11.0, IQR: 9.0–13.0) than hospitals without disparities (10.5, IQR: 8.0–13.0), however these differences were not statistically significant in unadjusted nor adjusted analyses (Table 2b).
Hospitals with equity officers by hospital characteristics
Hospitals with and without equity officers differed by hospital characteristics (Fig. 2). By ownership type, a higher proportion of non-profit hospitals had an equity officer (insurance disparities: 86.6%; racial disparities: 90.6%) than government owned (insurance: 73.4%, race: 83.3%), and investor owned/for-profit hospitals (insurance: 66.4%, race: 65.4%). A higher proportion of teaching hospitals had equity officers (insurance: 88.4%, race: 91.1%) compared to non-teaching hospitals (insurance: 75.1, race: 76.6%). A higher proportion of urban (insurance: 88.8%; race: 90.0%) vs. rural (insurance 71.9%; race: 72.5%) hospitals had equity officers. By region, a higher proportion of hospitals in the South and Northeast had equity officers compared to the Midwest and West. A lower proportion of small hospitals (< 100 beds) had equity officers (insurance: 73.7%; race: 74.2%) compared to larger hospitals (> 400 beds) (insurance: 92.8%; race: 92.4%). A higher proportion of hospitals with high nurse-to-bed ratios had an equity officer (insurance: 88.6%; race: 93.8%) compared to hospitals with lower nurse-to-bed ratios (insurance: 75.7%; race: 74.7%). All p-values < 0.001 (Fig. 2).
Fig. 2.
Proportion of hospitals with an equity officer (1o outcome) by hospital covariates a P-values for these variables indicate, in chi-square tests, there was a significant difference by hospital covariates across all four quartiles (not only for the 1st and 4th quartile shown in the table) b All hospitals with <25 Black patients in 2019 were excluded from the analysis of race disparities
By hospital patient demographics, a higher proportion of hospitals with fewer dual-eligible patients (0–10% dual-eligible patients) had equity officers (insurance: 89.3%, race: 90.9%) compared to hospitals with more (23.7–100%) dual-eligible patients (insurance: 74.2%; race: 77.8%), p-value < 0.001. A higher proportion of hospitals with larger Black populations (8.7–97.3% Black patients) had equity officers (insurance: 85.7%, race: 86.3%) compared to hospitals with fewer (< 1%) Black patients (insurance: 68.0%, race: not-reportable), p-value 0.02 (Fig. 2).
Equity environment composite score by hospital characteristics
Patterns across hospital covariates for the EECS mirrored patterns for equity officers (Fig. 3). Non-for-profit, teaching, urban, Northeast/Southern, larger hospitals, and hospitals with a higher nurse-to bed ratios had higher EECS. In our insurance disparities analysis, EECS’s were higher for hospitals with higher proportions of Black and Medicaid-insured patients and lower proportions of Medicare-insured patients. These differences were not statistically different in the racial disparities cohort.
Fig. 3.
Median equity environment composite score (2o outcome) by hospital covariates. a P-values for these variables indicate, in chi-square tests, there was a significant difference by hospital covariates across all four quartiles (not only for the 1st and 4th quartile shown in the table). b All hospitals with <25 Black patients in 2019 were excluded from the analysis of race disparities
Outcomes stratified by hospital covariates and readmission disparities
Supplemental Tables 2–3 presents the proportion of hospitals with equity officers and the median EECS by both hospital type and disparities status. Generally, hospitals with disparities were more likely to have equity officers and higher median EECS. For a few categories, this relationship reversed (insurance disparities analysis: Northeast hospitals and hospitals with 200–299 beds; racial disparities analysis: urban hospitals, some regions, and bed size).
Discussion
This study investigated relationships between hospital-level disparities in 2019 readmission rates and hospital-level equity efforts in 2022. We hypothesized that hospitals with disparities in readmission rates, reported confidentially by CMS starting in 2019, would be more likely to have equity efforts to address these disparities. However, our adjusted results did not support this hypothesis. We found that hospitals with insurance disparities in 2019 were more likely to have an equity officer and a stronger equity environment score in 2022 compared to hospitals without disparities. However, these associations were not statistically significant after adjusting for hospital characteristics. In examining disparities by race, relationships with equity efforts were not significant in unadjusted nor adjusted analyses. These results suggest that hospital type (e.g., location, teaching status, and demographics), potentially as a reflection of hospital culture, may be more predictive of equity efforts than disparities in readmissions.
We chose to measure disparities using a previously defined standard for equitable readmissions. We focused on this measure for its policy relevance - outcomes for the Disparity Methods that define this measure were confidentially reported by CMS to all US hospitals accepting Medicare patients until this year [13]. Whether hospitals changed behavior in response to these confidential reports is unknown. Examining relationships between hospital equity efforts and readmission disparities begins to address this question and lays key groundwork for future causal analyses.
Several factors may explain why we did not find relationships between disparities and equity efforts. First, the time horizon may have been too short; equity efforts and cultural change take time [15, 19]. Even hospitals that intended to address readmission disparities reported in 2019 may not have made changes detected on the 2022 AHA survey. Second, the AHA survey does not indicate tenure of equity efforts; hospitals with longstanding equity efforts cannot be differentiated from hospitals that initiated efforts in 2022. Third, our study outcomes were measured in the wake of the COVID-19 pandemic and racial justice movement of 2020; likely a key unmeasured confounder. Fourth, we measured disparities using Medicare fee-for-service claims data, excluding the population of Medicare beneficiaries insured by Medicare Advantage plans (disproportionately dual-eligible and minoritized populations) [32].
Lastly, our results suggest the importance of hospital type/culture. Our prior work demonstrated that hospitals with readmission disparities were more likely to be teaching, non-for profit, urban, and large hospitals (>400 beds), and hospitals with larger populations of dual-eligible and Black patients [21]. In the present analysis, we found that hospitals with equity officers and stronger equity environments were similar. These teaching, non-profit, large, urban hospitals, that are more likely to serve a more medically and socially complex patient population at higher risk for poor outcomes (including disparities), [21, 33] may also be more likely, culturally, to have equity efforts that directly address the needs of their patient population [34, 35].
Despite the equitable readmission measure’s policy relevance, it has limitations (like all measures). Using a single measure, that only includes the Medicare population, to assess disparities in quality may underestimate relationships between disparities and equity efforts (e.g., among other payers or other indicators of quality), and may limit study generalizability [36]. Hospitals without disparities in readmissions may have disparities in other measures of quality and/or may have readmission disparities for non-Medicare populations. Many hospitals with robust equity efforts, for example, have chosen to measure, track and work to improve disparities in other metrics [37]. Furthermore, measure specifications exclude the many US hospitals that do not serve sufficient patients to calculate disparities; more than half of hospitals were excluded from our analysis of race disparities. This exclusion is not a data limitation, but reflects small-volume rural hospitals and geographic/care segregation in the U.S.
This study has several data limitations. AHA survey data may be influenced by social desirability bias and. Large survey missingness may impact study results (see Supplemental Table 1). Nonetheless, the AHA survey is the best source of national hospital data currently available. The AHA survey only started to include all study questions in 2022 and does not assess equity officers’ tenure, impacting interpretation. Future work with additional years of data may be used to examine temporal trends and/or how discontinuing confidential reporting of disparities in 2025 impacts hospital equity efforts in future years.
This study contributes to the discussion of how we measure, report on and create accountability to equity at the hospital-level. Whether confidential reporting of disparities impacts hospital behavior remains uncertain, and the changing political climate may prompt hospitals to lessen equity efforts. However, understanding what drives hospitals to pursue equity, as well as continued discussion around what it means to be an equitable hospital is crucial for developing effective accountability metrics (e.g., outcome reporting, attestation, or cultural assessments) now and in the future. Hospitals without measured disparities are not necessarily working towards equity and justice – prompting questions about what outcomes or efforts we should reward and celebrate.
Conclusions
Our findings suggest that hospital-level readmission disparities are not necessarily associated with greater hospital-level equity efforts. Most likely, hospital organizational type, which dictates not only its patient population but also likely its culture, is a stronger predictor of hospital equity efforts. Future work is needed to understand how disparities in clinical outcomes do or do not drive change in hospital culture and equity efforts.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We would like to thank Dr. Julia Iyasere, Dr. Ndidi Unaka and Dr. Lou Hart for providing their reflections and perspectives of the study results and interpretation as well as Dr. Mark Olfson for his feedback on the manuscript.
Abbreviations
- CMS
The Centers for Medicare and Medicaid Services
- AHA
American Hospital Association Annual Survey
- EONS
Equity Officer National Survey
- SDOH
Social determinants of health
- EECS
Equity environment composite score
Author contributions
KN conceptualized the project and study design, drafted the initial manuscript, and critically reviewed and approved the final manuscript. HY performed the data analysis, contributed to the study design, and critically reviewed and approved the final manuscript. RA contributed to the study design, helped to draft the initial manuscript, and critically reviewed and approved the final manuscript. JH, HW, LH contributed to the project and study design, interpretation of data and analysis, and critically reviewed and approved the final manuscript.
Funding
This work was supported by The Commonwealth Fund, Grant # 23-23844, and the Agency for Health Care Quality through grant number 5R01HS022882. Dr. Nash is supported by the National Center for Advancing Translational Sciences, National Institute of Health, through grant number KL2TR001874.
Data availability
Data used in this manuscript (Medicare claims data and American Hospital Association Data) are both under data use agreements and therefore cannot be shared.
Declarations
Ethics approval
This study was approved by the Yale University Medical School Institutional Review Board and in compliance with the Declaration of Helsinki. Given that this publication only uses secondary data sets, the need for participant consent was waived by the Yale University School of Medicine Institutional Review Board via federal regulation 45CFR46.116.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Institute of Medicine Committee on U, Eliminating R. Ethnic Disparities in Health C. In: Smedley BD, Stith AY, Nelson AR, editors. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington (DC): National Academies Press (US) Copyright 2002 by the National Academy of Sciences. All rights reserved.; 2003. [PubMed]
- 2.Healthy People 2020. United States Department of Health and Human Services, Office of Disease Prevention and Health Promotion. https://www.healthypeople.gov/2020/about/foundation-health-measures/Disparities. Updated February 6th, 2022. Accessed April 10,2022.
- 3.Solar O, Irwin A. A conceptual framework for action on the social determinants of health. Social Determinants of Health Discussion Paper 2 (Policy and Practice). Geneva: World Health Organization;2010.
- 4.McGuire TG, Alegria M, Cook BL, Wells KB, Zaslavsky AM. Implementing the Institute of medicine definition of disparities: an application to mental health care. Health Serv Res. 2006;41(5):1979–2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Downing NS, Wang C, Gupta A, et al. Association of Racial and socioeconomic disparities with outcomes among patients hospitalized with acute myocardial Infarction, heart Failure, and pneumonia: an analysis of within- and Between-Hospital variation. JAMA Netw Open. 2018;1(5):e182044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bensken WP, Alberti PM, Koroukian SM. Health-Related social needs and increased readmission rates: findings from the nationwide readmissions database. J Gen Intern Med. 2021;36(5):1173–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Silvestri D, Goutos D, Lloren A et al. Factors associated with disparities in hospital readmission rates among US adults dually eligible for Medicare and Medicaid. Paper presented at: JAMA Health Forum2022. [DOI] [PMC free article] [PubMed]
- 8.Shannon EM, Zheng J, Orav EJ, Schnipper JL, Mueller SK. Racial/Ethnic disparities in interhospital transfer for conditions with a mortality benefit to transfer among patients with medicare. JAMA Netw Open. 2021;4(3):e213474–213474. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Asch DA, Islam MN, Sheils NE, et al. Patient and hospital factors associated with differences in mortality rates among black and white US medicare beneficiaries hospitalized with COVID-19 infection. JAMA Netw Open. 2021;4(6):e2112842–2112842. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Eberly LA, Richterman A, Beckett AG, et al. Identification of Racial inequities in access to specialized inpatient heart failure care at an academic medical center. Circulation: Heart Fail. 2019;12(11):e006214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Data M. and Equity. https://www.ncqa.org/about-ncqa/health-equity/data-and-measurement/. Accessed February 4th, 2022.
- 12.Binger T, Chen H, Harder B. Hospital rankings and health equity. JAMA 2022. [DOI] [PubMed]
- 13.Disparity Methods Confidential Reporting. The Center for Medicare and Medicaid Services. Quality Net Web site. https://qualitynet.cms.gov/inpatient/measures/disparity-methods. Accessed September 30, 2022.
- 14.Health Care Equity Certification. https://www.jointcommission.org/our-priorities/health-care-equity/health-care-equity-prepublication/. Published 2022. Accessed February 1st, 2023.
- 15.Weissman JS, Adler RR, Ilkhani S et al. The rise of the hospital chief equity Officer—A National survey of early experiences and attributes. J Gen Intern Med 2025. [DOI] [PMC free article] [PubMed]
- 16.Herrin J, Harris KG, Spatz E, Cobbs-Lomax D, Allen S, León T. Hospital leadership diversity and strategies to advance health equity. Jt Comm J Qual Patient Saf. 2018;44(9):545–51. [DOI] [PubMed] [Google Scholar]
- 17.Adler RR, Pomer A, Diviti S, Lewis J, Tan-McGrory A, Weissman JS. Understanding the Roles, Responsibilities, and factors for success of health equity officers in health care settings: A qualitative study. J Public Health Manag Pract. 2024;30(4):E165–73. [DOI] [PubMed] [Google Scholar]
- 18.Community health needs assessment for charitable hospital organizations. - Sect. 501(r)(3). International Revenue Service https://www.irs.gov/charities-non-profits/community-health-needs-assessment-for-charitable-hospital-organizations-section-501r3. Updated August 20th, 2024. Accessed March 19, 2025.
- 19.Weissman JS, Adler RR, Betancourt J, et al. How hospitals are addressing the effects of racism: A Mixed-Methods study of hospital equity officers. Health Aff (Millwood). 2023;42(10):1402–10. [DOI] [PubMed] [Google Scholar]
- 20.2022 CMS Disparity Methods Updates and Specifications Report. CMS.gov: The Centers for Medicaid and Medicare Services; October 2022 2022.
- 21.Nash KA, Weerahandi H, Yu H, et al. Measuring equity in readmission as a distinct assessment of hospital performance. JAMA. 2024;331(2):111–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.How Hospitals Are Addressing The Effects Of Racism. A Mixed-Methods study of hospital equity officers. Health Aff (Millwood). 2023;42(10):1402–10. [DOI] [PubMed] [Google Scholar]
- 23.AHA Annual Survey Database. AHA Data and Insights Web site. https://www.ahadata.com/aha-annual-survey-database. Accessed May 10th, 2025.
- 24.Lloren A, Liu S, Herrin J, et al. Measuring hospital-specific disparities by dual eligibility and race to reduce health inequities. Health Serv Res. 2019;54(Suppl 1Suppl 1):243–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Joynt KE, Orav EJ, Jha AK. Thirty-day readmission rates for medicare beneficiaries by race and site of care. JAMA. 2011;305(7):675–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Zaslavsky AM, Ayanian JZ, Zaborski LB. The validity of race and ethnicity in enrollment data for medicare beneficiaries. Health Serv Res. 2012;47(3 Pt 2):1300–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Jarrín OF, Nyandege AN, Grafova IB, Dong X, Lin H. Validity of race and ethnicity codes in medicare administrative data compared with Gold-standard Self-reported race collected during routine home health care visits. Med Care. 2020;58(1):e1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Figueroa JF, Duggan C, Toledo-Cornell C, Zheng J, Orav EJ, Tsai TC. Assessment of strategies used in US hospitals to address social needs during the COVID-19 pandemic. JAMA Health Forum. 2022;3(10):e223764–223764. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Horwitz LI, Bernheim SM, Ross JS, et al. Hospital characteristics associated with Risk-standardized readmission rates. Med Care. 2017;55(5):528–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Silvestri D, Goutos D, Lloren A, et al. Factors associated with disparities in hospital readmission rates among US adults dually eligible for medicare and medicaid. JAMA Health Forum. 2022;3(1):e214611. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Herrin J, Yu H, Venkatesh AK, et al. Identifying high-value care for medicare beneficiaries: a cross-sectional study of acute care hospitals in the USA. BMJ Open. 2022;12(3):e053629. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Nancy Ochieng MF. Jeannie Fuglesten Biniek, Anthony Damico, and Tricia Neuman. Medicare Advantage in 2025: Enrollment Update and Key Trends. https://www.kff.org/medicare/medicare-advantage-enrollment-update-and-key-trends/. Published July 28th, 2025. Accessed November 16, 2025.
- 33.Webb Hooper M, Pérez-Stable EJ. Health equity is not possible without addressing disparities. Health Psychol. 2023;42(9):625–7. [DOI] [PubMed] [Google Scholar]
- 34.Chin MH, Pace-Moody A, Vela MB, et al. Theatre of the oppressed to teach medical students about Power, lived Experience, and health equity. J Gen Intern Med. 2025;40(2):330–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Todic J, Cook SC, Spitzer-Shohat S, et al. Critical Theory, culture Change, and achieving health equity in health care settings. Acad Med. 2022;97(7):977–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Horwitz LI, Partovian C, Lin Z, et al. Development and use of an administrative claims measure for profiling hospital-wide performance on 30-day unplanned readmission. Ann Intern Med. 2014;161(10 Suppl):S66–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Golden SH, Powe NR. Hospital equity rating Metrics—Promise, Pitfalls, and perils. JAMA Health Forum. 2023;4(10):e233188–233188. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data used in this manuscript (Medicare claims data and American Hospital Association Data) are both under data use agreements and therefore cannot be shared.