==== Front Health Serv Res Health Serv Res 10.1111/(ISSN)1475-6773 HESR Health Services Research 0017-9124 1475-6773 Blackwell Publishing Ltd Oxford, UK 36755373 10.1111/1475-6773.14141 HESR14141 Research Article Hospital Care The geography of Medicare's hospital value‐based purchasing in relation to market demographics McLaughlin and Boscoe McLaughlin Colleen C. MPH, PhD https://orcid.org/0000-0003-4172-8172 1 colleen.mclaughlin@acphs.edu Boscoe Francis P. PhD https://orcid.org/0000-0003-4277-2941 2 1 Department of Population Health Sciences Albany College of Pharmacy and Health Sciences Albany New York USA 2 Pumphandle, LLC Camden Maine USA * Correspondence Colleen C. McLaughlin, Department of Population Health Sciences, Albany College of Pharmacy and Health Sciences, 106 New Scotland Ave, Albany, NY 12208, USA. Email: colleen.mclaughlin@acphs.edu 22 2 2023 8 2023 22 2 2023 58 4 10.1111/hesr.v58.4 844852 © 2023 The Authors. Health Services Research published by Wiley Periodicals LLC on behalf of Health Research and Educational Trust. https://creativecommons.org/licenses/by/4.0/ This is an open access article under the terms of the http://creativecommons.org/licenses/by/4.0/ License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Abstract Objective To illustrate the association between the sociodemographic characteristics of hospital markets and the geographic patterns of Medicare hospital value‐based purchasing (HVBP) scores. Data Sources and Study Setting This is a secondary analysis of United States hospitals with a HVBP Total Performance Score (TPS) for 2019 in the Centers for Medicare and Medicaid Services (CMS) Hospital Compare database (4/2021 release) and American Community Survey (ACS) data for 2015–2019. Study Design This is a cross‐sectional study using spatial multivariable autoregressive models with HVBP TPS and component domain scores as dependent variables and hospital market demographics as the independent variables. Data Collection/Extraction Methods We calculated hospital market demographics using ZIP code level data from the ACS, weighted the 2019 CMS inpatient Hospital Service Area file. Principal Findings Spatial autoregressive models using eight nearest neighbors with diversity index, race and ethnicity distribution, families in poverty, unemployment, and lack of health insurance among residents ages 19–64 years provided the best model fit. Diversity index had the highest statistically significant contribution to lower TPS (ß = −12.79, p < 0.0001), followed by the percent of the population coded to “non‐Hispanic, some other race” (ß = −2.59, p < 0.0023), and the percent of families in poverty (ß = −0.26, p < 0.0001). Percent of the population was non‐Hispanic American Indian/Alaskan Native (ß = 0.35, p < 0.0001) and percent non‐Hispanic Asian (ß = 0.12, p < 0.02071) were associated with higher TPS. Lower predicted TPS was observed in large urban cities throughout the US as well as in states throughout the Southeastern US. Similar geographic patterns were observed for the predicted Patient Safety, Person and Community Engagement, and Efficiency and Cost Reduction domain scores but are not for predicted Clinical Outcomes scores. Conclusions The lower predicted scores seen in cities and in the Southeastern region potentially reflect an inherent—that is, structural—association between market sociodemographics and HVBP scores. hospitals patient satisfaction pay for performance quality of healthcare spatial analysis systemic racism source-schema-version-number2.0 cover-dateAugust 2023 details-of-publishers-convertorConverter:WILEY_ML3GV2_TO_JATSPMC version:6.3.0 mode:remove_FC converted:02.07.2023 McLaughlin CC , Boscoe FP . The geography of Medicare's hospital value‐based purchasing in relation to market demographics. Health Serv Res. 2023;58 (4 ):844‐852. doi:10.1111/1475-6773.14141 36755373 ==== Body pmc What is known on this topic Medicare's hospital value‐based purchasing program scores have been shown to be lower for safety‐net hospitals, resulting in reduced Medicare reimbursement. There has been a call to adjust the scores for the socioeconomic status of patients seen by each hospital, but regulators have resisted holding these hospitals to a lower standard. The local and regional geographic impact of these policies across the United States has not been explored. What this study adds We created maps showing how the hospital scores were potentially affected by the sociodemographic characteristics of their market areas. Hospitals in the Southeastern states and larger cities with racial and ethnic diversity had lower predicted scores, potentially representing an increased risk of lower hospital value‐based purchasing program scores. Patient experience, patient safety, and hospitalization related costs contributed to the lower predicted scores. There was less evidence of regional geographic patterns in clinical quality. 1 INTRODUCTION The Centers for Medicare and Medicaid Services (CMS) implemented hospital value‐based purchasing (HVBP) for acute care hospitals starting with 2012 discharges as part of the Affordable Care Act. 1 Through this program, Medicare withholds 2% of participating hospitals' base operating Medicare severity diagnosis‐related group (MS‐DRG) payment amount, which is then redistributed across all participating hospitals based on relative performance and demonstrated improvement. In 2019, $1.9B was redistributed among almost 2800 hospitals, with 44% of hospitals having a net loss in Medicare income and 5% losing more than $350K. 2 , 3 Numerous commentators have noted that as the US payer system moves toward pay‐for‐performance/value‐based purchasing, there is a risk of increased penalization of safety‐net hospitals. 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 The “consistent, negative, and significant effect” 13 of safety‐net status on HVBP scores implies an association with factors that are outside hospitals' ability to modify, such as market characteristics, government payer distribution, and ownership structure. 8 Lower access to high‐quality health care is a subset of the inequalities faced by communities of concentrated poverty, and by extension, communities where racial or ethnic minority groups are overrepresented. 12 , 14 , 15 , 16 At the patient and community level, therefore, there is the potential for HVBP incentive payments to reinforce structural racism. While the association between safety‐net hospital status and lower overall performance on HVBP has been explored in a growing body of literature, the extent and patterns of the impact of the racial and socioeconomic make‐up of patient populations on HVBP scores across the US have not been explored from a geospatial perspective. Our objective is to apply spatial data analysis to map the association between hospital markets and HVBP scores across the US, illustrating both local detail and broader regional trends. Our work fits into the public health tradition of disease mapping wherein the local nature of spatial patterning of data is explicitly used to improve estimation. 17 Our use of spatial models, increased specificity in defining hospital market areas, and national scope represent enhancements to what is known about structural racism and HVBP. 2 METHODS 2.1 Study population and outcome measures This is a cross‐sectional observational study of all non‐federal general medical/surgical hospitals in the United States (US) receiving scores for 2019 discharges in the CMS HVBP program, as of the April 1, 2021 release (n = 2673). 3 We used the HVBP Total Performance Score (TPS) and its component domain scores as our outcome measures. The TPS is the sum of four domain scores: Person Score from Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey of patient and family experience; Safety Score based on selected hospital acquired infections; Clinical Outcomes Score based on 30‐day risk‐standardized mortality rates for three conditions and one complication measure; and Efficiency and Cost Reduction Score derived from Medicare spending per beneficiary. We examined 2019 HVBP scores to avoid potential confounding by COVID‐19; the 2019 Clinical Outcomes domain was based on patient discharged from July 2016 to June 2019 (30‐day mortality measures) and April 2016 to March 2019 (Hip/Knee replacement complication measure). 2.2 Independent variables We defined flexible hospital markets based on ZIP code level data using CMS's Market Service Area file for hospitals derived from Medicare fee‐for‐service recipients. 18 Hospital market was estimated from the 2019 CMS inpatient Hospital Service Area File. 18 This file provides a count of services rendered to fee‐for‐service Medicare beneficiaries by facility and by the beneficiary's mailing ZIP code, excluding hospital/ZIP codes pairs with 10 or fewer discharges. Population demographics for 5‐digit Zip Code Tabulation Areas (ZCTA) were obtained from the American Community Survey (ACS) 5‐year estimates for 2015–2019. 19 We assumed a one‐to‐one correlation between ZCTA and ZIP code. ZIP codes on the market file but not the ACS file (9%) were excluded. These most likely represent PO Box only ZIP codes. Hospital market values were estimated by weighting the ZCTA estimates from the ACS by the percent of the total hospital market in each ZIP code/ZCTA as follows: Hospital marketi=∑j=1nHospital cases∈ZCTAjiTotal hospital casesi*ZCTA estimatej. Market characteristics obtained from ACS included weighted percentages of hospital markets for race/ethnicity, racial/ethnic diversity, high school graduates, unemployment, foreign born, families living below the federal poverty level, and lack of health care insurance among persons aged 19–64. Race and ethnicity as collected for the ACS are self‐reported and categorized based on the 1997 Office of Management and Budget standards. 20 We used all categories tabulated in the ACS public use files: mutually exclusive Hispanic, Black, White, American Indian/Alaska Native (AIAN), Asian, Native Hawaiian or other Pacific Islander (PI), some other race, and two or more races. Racial/ethnic diversity was measured using the US Census Diversity Index, which is a measure of a probability of two people chosen at random from the population being of different race/ethnic groups. 21 Other ACS variables beyond these, such as non‐English languages spoken at home, food stamp recipients, and public insurance, were considered and excluded due to high collinearity (Pearson correlation coefficient greater than 0.75) with other variables. Institutional Review Board review was not necessary since this study did not meet the definition of human subjects research. All data used are publicly available. 2.3 Statistical analysis We applied spatial autoregressive (SAR) models to each dependent variable to estimate the predicted scores based on individual hospital's independent variables and the observed scores of neighboring hospitals. These spatial models account for the spatial dependence between observations (that is, observations in one location tend to be similar to observations in other locations). These models are more conservative and less biased than ordinary least squares (OLSs) regression models, particularly given the spatial dependency (i.e., overlap) in hospital markets. A formal model selection process was used to determine the appropriate form of the multivariable spatial regression. A nearest‐neighbor (k) matrix for each hospital was created for each hospital for k = 6–10 neighbors per hospital. Imputation was used to assign missing values for the HVBP domain scores, which were missing for a small number of hospitals, which allowed the same neighbor matrix to be used for all models. We used a backward elimination process to select the independent variables used in the analysis. We split the data into training (70%) and test (30%) data sets and fit OLS regression models to the training data with TPS as the outcome measure, beginning with the full set of variables listed above and removing the variable with the lowest t‐statistic (highest p‐value). Each candidate model was then applied to the test data, and the final model was selected that had the most candidate variables without loss‐of‐fit, for example, where the mean squared error leveled off. The variables in this final model were then used to fit a SAR model on the entire data set for TPS. We compared the SAR model using k = 6–10, and the k with the best model fit using the Akaike information criterion was used for all subsequent models. For consistency, the same model was used for the HVBP component scores. Because the predictors in SAR models include both the modeled variables and the dependent variables for the k‐nearest neighbors, the interpretation of the estimated coefficients is not as straightforward as for OLSs, although the direction and magnitude of the coefficients retain their interpretive properties. 22 The estimated coefficients represent the direct effect of the independent variables on the outcome and are constant across hospitals. Predicted HVBP scores for individual hospitals are a combination of these direct effects and indirect effects based on the HVBP scores of the hospital's k‐neighbors. We created dot maps of the observed and predicted HVBP scores to allow comparison of regional patterns. Maps of the predicted values from SAR models show the influence of smoothing the dependent variables based on the local (k‐neighbor) patterns of the independent variables. 23 In other words, spatial autoregression reduces the effect of factors not in the spatial model. PROC SPATIALREG in SAS v9.4 was used to perform the spatial analysis. Given the computational complexity of fitting an SAR model to large datasets, we used the Chebyshev approximation to calculate the eigenvalues. The maps were produced in R using the usmap and ggplot2 packages. 3 RESULTS Of the 3158 acute care hospitals in the 2019 Hospital Care Compare dataset from 49 US states, 2673 (85%) had HVBP TPS and data on ZIP codes of their Medicare fee‐for‐service patients on the CMS service area file. 3.1 Model selection The selected model included 11 variables (Table 1). In the SAR models, the number of nearest neighbors per hospital that produced the best model fit was 8 although there was little difference in model fit between 6 and 10 neighbors. TABLE 1 The association between hospital value based purchasing scores a and hospital market sociodemographic measures b using spatial autoregression, United States Acute Care Hospitals, c 2019. Total performance score Clinical outcomes Safety Person and community engagement Efficiency and cost reduction Estimate d p‐value Estimate p‐value Estimate p‐value Estimate p‐value Estimate p‐value Intercept 29.90 <0.0001 8.72 <0.0001 9.61 <0.0001 9.58 <0.0001 6.21 <0.0001 Diversity index −12.79 <0.0001 −1.74 0.0667 −4.08 0.0003 −3.06 0.0045 −7.85 <0.0001 Percent Hispanic 0.03 0.1081 0.04 <0.0001 0.02 0.0969 −0.04 0.0001 0.02 0.1338 Percent Black e 0.04 0.1187 0.03 0.0154 0.01 0.4741 −0.01 0.4749 0.02 0.2486 Percent AIAN 0.35 <0.0001 0.07 0.0006 0.05 0.0531 0.05 0.0413 0.28 <0.0001 Percent Asian 0.12 0.0207 0.08 <0.0001 0.03 0.2552 −0.02 0.4469 0.02 0.5208 Percent PI 0.19 0.6904 −0.32 0.1006 0.03 0.9082 0.36 0.1047 0.34 0.2472 Percent some other race −2.59 0.0023 0.19 0.5945 −0.81 0.0529 −1.79 <0.0001 −1.12 0.0372 Percent two or more races 0.40 0.0543 0.05 0.5391 0.12 0.2321 −0.11 0.2511 0.21 0.1055 Percent of families below poverty line −0.26 0.0013 −0.07 0.0429 −0.08 0.0494 0.00 0.9067 −0.13 0.0113 Unemployment rate −0.18 0.2919 −0.16 0.0235 0.07 0.4325 −0.25 0.0024 0.17 0.1131 Percent residents ages 19–64 years without health insurance 0.01 0.7958 −0.03 0.1424 0.08 0.0011 0.04 0.1332 −0.04 0.2648 Rho 0.32 <0.0001 0.35 <0.0001 0.13 0.001 0.31 <0.0001 0.46 <0.0001 Sigma‐squared 116.32 <0.0001 19.77 <0.0001 28.03 <0.0001 25.89 <0.0001 46.54 <0.0001 Abbreviations: AIAN, American Indian/Alaska Native; PI, Pacific Islander. a Centers for Medicare & Medicaid Services, Hospital Care Compare April 2021 release. b American Community Survey 2015–2019. c Includes all acute care hospitals in the United States receiving Hospital Value‐Based Purchasing Total Performance Score for 2019; includes District of Columbia; excludes Maryland and US territories. d Estimated coefficients represent the direct effect of the covariate on the outcome and are constant across all observations. e All listed racial groups are from the American Community Survey non‐Hispanic or Latino population estimates. The autoregressive regression coefficient, rho, was statistically significant in all models, indicating all the HVBP variables examined displayed spatial dependence. All models also had statistically significant residual variance (measured by sigma‐squared), which can be due to outliers in the predicted HVBP scores or to the presence of regional patterns in addition to the modeled local patterns. Of the included variables, racial/ethnic diversity had the strongest association by a large margin, although it did not reach statistical significance in the model for clinical outcomes. Once adjusted for diversity, the other race and ethnicity categories tend to have positive, albeit weaker, associations with the HVBP outcomes, indicating hospitals serving communities with higher concentrations of residents of specific racial or ethnic groups tended to perform slightly better. For example, the five hospitals with markets that were 75% or more non‐Hispanic AIAN had a low diversity index (mean 0.21, US hospitals overall mean 0.42), but a high observed TPS (mean 63, US hospitals overall mean 34). Three of these facilities were Indian Health Service hospitals, and the other two were tribal hospitals. 3.2 Model results There is clear spatial patterning to the predicted TPS that is not as evident in the observed scores (Figure 1). Regionally, the South Atlantic, East South Central, and West South Central regions have the most evidently lower predicted TPS scores based on the modeled hospital market effects, hereafter referred to as “sociodemographic composition.” Pockets of lower predicted TPS scores were also observed in the large urban areas across the entire country, including the northeast urban corridor (Boston–New York City–Philadelphia–Washington DC), cities in the Great Lakes area (Cleveland, Detroit, Chicago), and both northern California (Los Angeles) and southern California (San Francisco). FIGURE 1 Observed and predicted hospital value‐based purchasing total performance scores, 2019, United States Acute Care Hospitals,1 Centers for Medicare and Medicaid Services Hospital Care Compare April 2021 release. (A) Observed Total Performance Scores, (B) Predicted Total Performance Score.2 1. Includes all acute care hospitals in the United States receiving Hospital Value‐Based Purchasing Total Performance Score for 2019; includes District of Columbia; excludes Maryland and US territories. 2. Predicted scores based on spatial autoregressive models of estimated hospital market sociodemographic characteristics (racial/ethnic diversity; the percentage of population that is, respectively, Hispanic, Black, American Indian/Alaska Native, Asian, Native Hawaiian or other Pacific Islander, some other race, and two or more races; the percent of families living below the federal poverty level; percent of individuals age 19–64 years without health insurance; and unemployment rate.) These predicted values represent smoothed estimates based on the market sociodemographic characteristics of the individual hospital and hospital value‐based purchasing scores of the hospital's eight nearest neighbors. [Color figure can be viewed at wileyonlinelibrary.com] Similar spatial patterning was observed for the predicted Person and Community Engagement Score, predicted Safety Score, and predicted Efficiency and Cost Reduction Scores (Figure 2). While the maps of the predicted outcomes are visually similar, differences can be seen in specific areas, such as in Oklahoma, where hospitals were ranked lower for predicted Person and Community Engagement relative to their rankings for predicted TPSs (Figure 2). The predicted Clinical Outcome Scores did not demonstrate clear spatial patterning with respect to the southern regions of the US, reflecting the smaller ß coefficients for the sociodemographic variables in the models that generated these estimates. On the other hand, many urban areas had higher predicted Clinical Outcomes Scores. FIGURE 2 Observed and Predicted Hospital Value‐Based Purchasing Domain Scores, 2019, United States Acute Care Hospitals,1 Centers for Medicare and Medicaid Services Hospital Care Compare April 2021 release. (A) Observed Clinical Outcomes Score, (B) Predicted Clinical Outcomes Score,2 (C) Observed Safety Score, (D) Predicted Safety Score,2 (E) Observed Person and Community Engagement Score, (F) Predicted Person and Community Engagement Score,2 (G) Efficiency and Cost Reduction Score, (H) Predicted Efficiency and Cost Reduction Score.2 1. Includes all acute care hospitals in the United States receiving Hospital Value‐Based Purchasing Total Performance Score for 2019; includes District of Columbia; excludes Maryland and US territories. 2 Predicted scores based on spatial autoregressive models of estimated hospital market sociodemographic characteristics (racial/ethnic diversity; the percentage of population that is, respectively, Hispanic, Black, American Indian/Alaska Native, Asian, Native Hawaiian or other Pacific Islander, some other race, and two or more races; the percent of families living below the federal poverty level; percent of individuals age 19–64 years without health insurance; and unemployment rate.) These predicted values represent smoothed estimates based on the market sociodemographic characteristics of the individual hospital and hospital value‐based purchasing scores of the hospital's eight nearest neighbors. 3. Missing values were imputed as follows: Efficiency and Cost Reduction Score n = 1, 0.0%; Person and Community Engagement Score n = 6, 0.2%; Clinical Outcomes Score n = 44, 1.6%; and Safety Score n = 507, 19.3%. [Color figure can be viewed at wileyonlinelibrary.com] 4 DISCUSSION We used spatial autoregression to predict HVBP scores based on the sociodemographic composition of each hospital's market. Our principal result is that accounting for the sociodemographic composition of hospital markets on HVBP TPS results in increased risk of low scores in large urban cities as well as in the Southeastern United States. When the individual domains of the HVBP scores are examined, the Person and Community Engagement, Safety, and Efficiency and Cost Reduction domains reflect the same regional and urban patterns. Clinical outcomes, on the other hand, show clustering of higher predicted scores in large urban areas. Our results complement the recent report by Khullar et al., who reported that the US Census Bureau’ South Atlantic and East South Central regions and counties with a higher proportion Black residents had a disproportionate proportion of hospitals that were low‐performing across all three CMS value‐based programs. 24 Our findings add to Khullar's by demonstrating that the high proportion of low‐performing hospitals in the Southeast could be explained by the sociodemographics of the region, rather than inherently less reliable health care delivery independent of sociodemographics. Our finding that the Clinical Outcomes domain did not exhibit the same geographic pattern as the other HVBP domains supports this hypothesis. To our knowledge, this is the first published work on the geographic pattern of the association between HVBP and hospital market demographics. The statistically significant autoregressive regression coefficient in all models indicates that our results may be conservative relative to non‐spatial models, which run the risk of type I error when spatial dependency is present. 25 Of all the sociodemographic variables examined, the US Census Bureau's diversity index had the strongest association in the SAR models. This reflects the idea that areas with higher percentages of racial and ethnic diversity tended to have lower‐performing hospitals, regardless of what specific racial or ethnic minority group predominated. While many studies of inequality in access to quality health care concentrate on the lower availability of health care, there is a growing body of literature reporting on the association between hospital quality and minority‐service hospital status across multiple health states and conditions. 4 , 5 , 8 , 10 , 13 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 Minority‐serving hospitals have also been shown to have lower scores for reported patient experience, despite findings that minority patients report better experiences overall. 12 , 50 , 51 A consistent finding has been that disparities in patient‐level health risks alone do not explain variation in quality between hospitals. In other words, the quality disparity associated with minority‐serving hospitals is likely due to systemic factors. 46 Multiple commentators have suggested that pay‐for‐performance measures and publicly reported star ratings should be adjusted for neighborhood or patient demographic characteristics. 9 , 52 , 53 , 54 CMS makes two arguments against the adjustment of HVBP measures for race and ethnicity or socioeconomic status. 1 First, it would be akin to holding hospitals with more diverse patient populations to a different standard. Second, it would obscure important differences. We believe the arguments could be more nuanced. In particular, since MSPB includes both costs associated with both the inpatient stay and 30 days post‐discharge, adjusting the Efficiency and Cost Reduction scores for the sociodemographic status of the hospital market could be seen as leveling the playing field for populations that may be medically underserved. Care would be needed to assure that such modifications do not have the unintended consequence of reflecting the lower quality and safety of the index inpatient episode. 55 In contrast, the argument for adjusting the HVBP Person and Community Engagement scores for socioeconomic risk is also not clear‐cut. Safety‐net or minority‐serving hospital status should not be synonymous with lower patient satisfaction. 7 , 12 , 50 , 51 , 56 , 57 Our analysis demonstrates that the Clinical Outcomes domain does not share the sociodemographic‐based patterning of the other HVBP domains, suggesting that adjustment for sociodemographic risk would be unnecessary. As a cross‐sectional ecological study, we need to be careful not to over‐interpret these findings. Translating these findings into a causal relationship between HVBP incentive payments and poorer quality care received by individuals, particularly in specific demographic groups or regions, is beyond the scope of this study. 58 We chose to highlight the spatial patterning of the association with hospital market to add to the understanding of potential structural racism in CMS's HVBP program. There are multiple other hospital characteristics, such as size, ownership structure, and teaching status, that also influence hospital quality and have spatial dependence. 5 LIMITATIONS Wider understanding of potential structural racism in access to higher quality care is hindered by the lack of nationwide data on quality‐of‐care data stratified by patient demographics at the hospital or community level. Researchers interested in the demographic geography of the hospital service areas have generally been limited to defining service areas based on the physical location of the hospital, ranging in scale from block group or census tract 7 , 52 , 59 , 60  to county. 15 , 60  Smaller‐scale geographies will underestimate the market in rural areas, while larger scales will overestimate the market in cities, particularly those with multiple hospitals with overlapping service areas. Our use of the Medicare service area file allowed us to define markets that are flexible across the urban/rural continuum and are weighted based on the actual patterns of hospital/ZIP code patient flow. Use of the Medicare fee‐for‐service population may misestimate hospital service areas in areas where (1) there is heterogeneity in market share by patient age and (2) there is local geographic heterogeneity in the adoption of Medicare Advantage adoption. Our calculation of market demographics also assumes that hospital choice by residents of a specific ZIP code does not vary by the demographic factors examined. Previous studies have shown this assumption is likely violated; White patients are more likely to travel further for care. 61 , 62 , 63 , 64 , 65 In the case of a hospital serving a ZIP code from which White patients differentially travel to other hospitals, we are underestimating the percent of minority patients that the hospital actually serves. Regardless, our estimation method provides increased specificity in market area definition compared to other methods available for use at the national scale. 6 CONCLUSION Our findings highlight the potential for structural racism in the HVBP program, with lower predicted scores for hospitals in the Southeastern states and larger cities in the US that have higher levels of racial and ethnic diversity. In particular, we show a consistent finding for a spatial relationship between market sociodemographic composition and the HVBP Person and Community Engagement and Efficiency and Cost Reduction domains, both of which have been shown in previous studies to be associated with hospital patient mix. We also show this spatial patterning for the safety domain, which has not been as well documented. This exploratory study complements existing understanding of place‐based structural racism on value‐based payment metrics and further highlights the need for re‐evaluation of policy regarding adjustment of incentive payments to reduce disparities among hospitals that serve minority populations. ACKNOWLEDGMENTS No funding to report. ==== Refs REFERENCES 1 Medicare program; hospital inpatient value‐based purchasing program. Final rule. Fed Regist. 2011;76 (88 ):26490‐26547.21548401 2 Centers for Medicare and Medicaid Services . CMS Hospital Value‐Based Purchasing Program Results for Fiscal Year 2019. December 2, 2018. Accessed June 2, 2021. Available at: https://www.cms.gov/newsroom/fact‐sheets/cms‐hospital‐value‐based‐purchasing‐program‐results‐fiscal‐year‐2019#:~:text=We%20estimate%20that%20the%20totalof%20services%20provided%20to%20patients. 3 Centers for Medicare and Medicaid Services . Hospital Care Compare 2021 [cited April 19, 2021]. Accessed April 19, 2021. Available at: https://data.cms.gov/provider-data/topics/hospitals. 4 Aggarwal R , Hammond JG , Joynt Maddox KE , Yeh RW , Wadhera RK . Association between the proportion of black patients cared for at hospitals and financial penalties under value‐based payment programs. JAMA. 2021;325 (12 ):1219‐1221. doi:10.1001/jama.2021.0026 33755063 5 Gaskin DJ , Zare H , Vazin R , Love D , Steinwachs D . Racial and ethnic composition of hospitals' service areas and the likelihood of being penalized for excess readmissions by the Medicare program. Med Care. 2018;56 (11 ):934‐943. doi:10.1097/mlr.0000000000000988 30256281 6 Gilman M , Adams EK , Hockenberry JM , Milstein AS , Wilson IB , Becker ER . Safety‐net hospitals more likely than other hospitals to fare poorly under Medicare's value‐based purchasing. Health Aff. 2015;34 (3 ):398‐405. doi:10.1377/hlthaff.2014.1059 7 McFarland DC , Ornstein KA , Holcombe RF . Demographic factors and hospital size predict patient satisfaction variance‐‐implications for hospital value‐based purchasing. J Hosp Med. 2015;10 (8 ):503‐509. doi:10.1002/jhm.2371 25940305 8 Renee Rutter S , Park SH . Relationship between hospital characteristics and value‐based program measure performance: a literature review. West J Nurs Res. 2020;42 (12 ):1010‐1021. doi:10.1177/0193945920920180 32443959 9 Shi B , King CJ , Huang SS . Relationship of hospital star ratings to race, education, and community income. J Hosp Med. 2020;15 (10 ):588‐593. doi:10.12788/jhm.3393 32966199 10 Zogg CK , Thumma JR , Ryan AM , Dimick JB . Medicare's hospital acquired condition reduction program disproportionately affects minority‐serving hospitals: variation by race, socioeconomic status, and disproportionate share hospital payment receipt. Ann Surg. 2020;271 (6 ):985‐993. doi:10.1097/sla.0000000000003564 31469746 11 Joynt Maddox KE , Reidhead M , Hu J , et al. Adjusting for social risk factors impacts performance and penalties in the hospital readmissions reduction program. Health Serv Res. 2019;54 (2 ):327‐336. doi:10.1111/1475-6773.13133 30848491 12 Chatterjee P , Joynt KE , Orav EJ , Jha AK . Patient experience in safety‐net hospitals: implications for improving care and value‐based purchasing. Arch Intern Med. 2012;172 (16 ):1204‐1210. doi:10.1001/archinternmed.2012.3158 22801941 13 Hong YR , Nguyen O , Yadav S , et al. Early performance of hospital value‐based purchasing program in Medicare: a systematic review. Med Care. 2020;58 (8 ):734‐743. doi:10.1097/mlr.0000000000001354 32692140 14 Chatterjee P , Werner RM . The hospital readmission reduction program and social risk. Health Serv Res. 2019;54 (2 ):324‐326. doi:10.1111/1475-6773.13131 30848490 15 Krieger N . Methods for the scientific study of discrimination and health: an ecosocial approach. Am J Public Health. 2012;102 (5 ):936‐944. doi:10.2105/ajph.2011.300544 22420803 16 Bailey ZD , Feldman JM , Bassett MT . How structural racism works ‐ racist policies as a root cause of U.S. racial health inequities. New Engl J Med. 2021;384 (8 ):768‐773. doi:10.1056/NEJMms2025396 33326717 17 Waller LA , Carlin BP . Disease mapping.  In: Gelfand AE, Diggle PJ, Fuentes M, Guttorp P, eds. Handbook of Spatial Statistics. CRC Press; 2010:217‐243. doi:10.1201/9781420072884-c14 18 Centers for Medicare and Medicaid Services . Hospital Service Area File. December 1, 2021. Accessed May 16, 2022. Available at: https://www.cms.gov/Research‐Statistics‐Data‐and‐Systems/Statistics‐Trends‐and‐Reports/Hospital‐Service‐Area‐File. 19 US Census Bureau . 2019: American Community Survey ACS 5‐Year Data Profiles Tables. 2015–2019. Accessed May 16, 2022. Available at: https://data.census.gov/cedsci/all/tables?q=american%20community%20survey&g=0100000US%248600000&y=2019. 20 Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity. 62 Federal Register 58782 (October 30, 1997). 21 Jensen E , Jones N , Orozco K , et al. Measuring racial and ethnic diversity for the 2020 census. Random Samplings blog 2021. Accessed December 11, 2022. https://www.census.gov/newsroom/blogs/random‐samplings/2021/08/measuring‐racial‐ethnic‐diversity‐2020‐census.html?utm_campaign=20210804msc20s1ccnwsrs&utm_medium=email&utm_source=govdelivery. 22 Whitten GD , Williams LK , Wimpy C . Interpretation: the final spatial frontier. Polit Sci Res Methods. 2019;9 (1 ):140‐156. doi:10.1017/psrm.2019.9 23 Anslin L . Spatial Econometrics: Methods and Models. Studies in Operational Regional Science. Springer Science+Business Media; 1988. 24 Khullar D , Tian W , Wadhera RK . High‐performing and low‐performing hospitals across Medicare value‐based payment programs. JAMA Health Forum. 2022;3 (7 ):e221864. doi:10.1001/jamahealthforum.2022.1864 35977223 25 Beale CM , Lennon JJ , Yearsley JM , Brewer MJ , Elston DA . Regression analysis of spatial data. Ecol Lett. 2010;13 (2 ):246‐264. doi:10.1111/j.1461-0248.2009.01422.x 20102373 26 Burris HH , Passarella M , Handley SC , Srinivas SK , Lorch SA . Black‐white disparities in maternal in‐hospital mortality according to teaching and Black‐serving hospital status. Am J Obstet Gynecol. 2021;225 (1 ):83.e1‐83.e9. doi:10.1016/j.ajog.2021.01.004 27 Howell EA , Janevic T , Hebert PL , Egorova NN , Balbierz A , Zeitlin J . Differences in morbidity and mortality rates in black, white, and Hispanic very preterm infants among New York City hospitals. JAMA Pediatric. 2018;172 (3 ):269‐277. doi:10.1001/jamapediatrics.2017.4402 28 Horbar JD , Edwards EM , Greenberg LT , et al. Racial segregation and inequality in the neonatal intensive care unit for very low‐birth‐weight and very preterm infants. JAMA Pediatr. 2019;173 (5 ):455‐461. doi:10.1001/jamapediatrics.2019.0241 30907924 29 Glazer KB , Zeitlin J , Egorova NN , et al. Hospital quality of care and racial and ethnic disparities in unexpected newborn complications. Pediatrics. 2021;148 (3 ):e2020024091. doi:10.1542/peds.2020-024091 30 Janevic T , Zeitlin J , Egorova N , Hebert PL , Balbierz A , Howell EA . Neighborhood racial and economic polarization, hospital of delivery, and severe maternal morbidity. Health Aff. 2020;39 (5 ):768‐776. doi:10.1377/hlthaff.2019.00735 31 Profit J , Gould JB , Bennett M , et al. Racial/ethnic disparity in NICU quality of care delivery. Pediatrics. 2017;140 (3 ):e20170918. doi:10.1542/peds.2017-0918 32 Creanga AA , Bateman BT , Mhyre JM , Kuklina E , Shilkrut A , Callaghan WM . Performance of racial and ethnic minority‐serving hospitals on delivery‐related indicators. Am J Obstet Gynecol. 2014;211 (6 ):647.e1‐647.e16. doi:10.1016/j.ajog.2014.06.006 33 Cole AP , Nguyen DD , Meirkhanov A , et al. Association of care at minority‐serving vs non‐minority‐serving hospitals with use of palliative care among racial/ethnic minorities with metastatic cancer in the United States. JAMA Netw Open. 2019;2 (2 ):e187633. doi:10.1001/jamanetworkopen.2018.7633 30707230 34 Krimphove MJ , Fletcher SA , Cole AP , et al. Quality of care in the treatment of localized intermediate and high risk prostate cancer at minority serving hospitals. J Urol. 2019;201 (4 ):735‐741. doi:10.1016/j.juro.2018.10.024 30414956 35 Lu PW , Scully RE , Fields AC , et al. Racial disparities in treatment for rectal cancer at minority‐serving hospitals. J Gastroin Surg. 2021;25 (7 ):1847‐1856. doi:10.1007/s11605-020-04744-x 36 Tsilimigras DI , Dalmacy D , Hyer JM , Diaz A , Abbas A , Pawlik TM . Disparities in NCCN guideline compliant care for resectable cholangiocarcinoma at minority‐serving versus non‐minority‐serving hospitals. Ann Surg Oncol. 2021;28 :8162‐8171. doi:10.1245/s10434-021-10202-y 34036428 37 Corl K , Levy M , Phillips G , Terry K , Friedrich M , Trivedi AN . Racial and ethnic disparities In care following the New York State sepsis initiative. Health Aff. 2019;38 (7 ):1119‐1126. doi:10.1377/hlthaff.2018.05381 38 Rush B , Danziger J , Walley KR , Kumar A , Celi LA . Treatment in disproportionately minority hospitals is associated with increased risk of mortality in sepsis: a national analysis. Crit Care Med. 2020;48 (7 ):962‐967. doi:10.1097/ccm.0000000000004375 32345833 39 Danziger J , Ángel Armengol de la Hoz M , Li W , et al. Temporal trends in critical care outcomes in U.S. minority‐serving hospitals. Am J Respirat Critic Care Med. 2020;201 (6 ):681‐687. doi:10.1164/rccm.201903-0623OC 40 Faigle R , Cooper LA , Gottesman RF . Race differences in gastrostomy tube placement after stroke in majority‐white, minority‐serving, and racially integrated US hospitals. Dysphagia. 2018;33 (5 ):636‐644. doi:10.1007/s00455-018-9882-y 29468269 41 Faigle R , Cooper LA , Gottesman RF . Lower carotid revascularization rates after stroke in racial/ethnic minority‐serving US hospitals. Neurology. 2019;92 (23 ):e2653‐e2660. doi:10.1212/wnl.0000000000007570 31101741 42 Cheng EM , Keyhani S , Ofner S , et al. Lower use of carotid artery imaging at minority‐serving hospitals. Neurology. 2012;79 (2 ):138‐144. doi:10.1212/WNL.0b013e31825f04c5 22700815 43 Benjenk I , Chen J . Variation of follow‐up rate after psychiatric hospitalization of Medicare beneficiaries by hospital characteristics and social determinants of health. Am J Geriatric Psychiatry. 2019;27 (2 ):138‐148. doi:10.1016/j.jagp.2018.08.006 44 Hasnain‐Wynia R , Kang R , Landrum MB , Vogeli C , Baker DW , Weissman JS . Racial and ethnic disparities within and between hospitals for inpatient quality of care: an examination of patient‐level hospital quality alliance measures. J Health Care Poor Underserved. 2010;21 (2 ):629‐648. doi:10.1353/hpu.0.0281 20453362 45 Figueroa JF , Zheng J , Orav EJ , Epstein AM , Jha AK . Medicare program associated with narrowing hospital readmission disparities between black and white patients. Health Aff. 2018;37 (4 ):654‐661. doi:10.1377/hlthaff.2017.1034 46 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:10.1001/jamanetworkopen.2018.2044 30646146 47 Hechenbleikner EM , Zheng C , Lawrence S , et al. Do hospital factors impact readmissions and mortality after colorectal resections at minority‐serving hospitals? Surgery. 2017;161 (3 ):846‐854. doi:10.1016/j.surg.2016.08.041 28029380 48 Chen JC , Shaw JD , Ma Y , Rhoads KF . The role of the hospital and health care system characteristics in readmissions after major surgery in California. Surgery. 2016;159 (2 ):381‐388. doi:10.1016/j.surg.2015.06.016 26209574 49 Creanga AA , Bateman BT , Kuklina EV , Callaghan WM . Racial and ethnic disparities in severe maternal morbidity: a multistate analysis, 2008‐2010. Am J Obstet Gynecol. 2014;210 (5 ):435.e1‐435.e8. doi:10.1016/j.ajog.2013.11.039 50 Figueroa JF , Zheng J , Orav EJ , Jha AK . Across US hospitals, black patients report comparable or better experiences than white patients. Health Aff. 2016;35 (8 ):1391‐1398. doi:10.1377/hlthaff.2015.1426 51 Goldstein E , Elliott MN , Lehrman WG , Hambarsoomian K , Giordano LA . Racial/ethnic differences in patients' perceptions of inpatient care using the HCAHPS survey. Medical Care Res Rev. 2010;67 (1 ):74‐92. doi:10.1177/1077558709341066 52 Fahrenbach J , Chin MH , Huang ES , Springman MK , Weber SG , Tung EL . Neighborhood disadvantage and hospital quality ratings in the Medicare hospital compare program. Med Care. 2020;58 (4 ):376‐383. doi:10.1097/MLR.0000000000001283 31895306 53 Sorbero ME , Paddock SM , Damberg CL , et al. Adjusting Medicare advantage star ratings for socioeconomic status and disability. Am J Manag Care. 2018;24 (9 ):e285‐e291.30222924 54 Baker MC , Alberti PM , Tsao TY , Fluegge K , Howland RE , Haberman M . Social determinants matter for hospital readmission policy: insights from New York City. Health Aff. 2021;40 (4 ):645‐654. doi:10.1377/hlthaff.2020.01742 55 Cook K , Foster B , Perry I , et al. Associations between hospital quality outcomes and Medicare spending per beneficiary in the USA. Healthcare (Basel, Switzerland). 2021;9 (7 ):831. doi:10.3390/healthcare9070831 56 Mazurenko O , Collum T , Ferdinand A , Menachemi N . Predictors of hospital patient satisfaction as measured by HCAHPS: a systematic review. J Healthc Manag. 2017;62 (4 ):272‐283. doi:10.1097/jhm-d-15-00050 28683051 57 Weech‐Maldonado R , Elliott M , Pradhan R , Schiller C , Hall A , Hays RD . Can hospital cultural competency reduce disparities in patient experiences with care. Med Care. 2012;50 :S48‐S55. doi:10.1097/MLR.0b013e3182610ad1 23064277 58 Wakefield J . Ecologic studies revisited. Annu Rev Public Health. 2008;29 :75‐90. doi:10.1146/annurev.publhealth.29.020907.090821 17914933 59 Menon NM , Leslie TF , Frankenfeld CL . Cancer‐related diagnostic and treatment capabilities of hospitals in the context of racial residential segregation. Public Health. 2020;182 :95‐101. doi:10.1016/j.puhe.2020.02.004 32213360 60 Benjenk I , Franzini L , Chen J . Evidence of the linkage between hospital‐based care coordination strategies and hospital overall (star) ratings. Med Care. 2020;58 (1 ):18‐26. doi:10.1097/mlr.0000000000001226 31725493 61 Jia P , Xierali IM . Disparities in patterns of health care travel among inpatients diagnosed with congestive heart failure, Florida, 2011. Prevent Chron Dis. 2015;12 :E150. doi:10.5888/pcd12.150079 62 Popescu I , Huckfeldt P , Pane JD , Escarce JJ . Contributions of geography and nongeographic factors to the white‐black gap in hospital quality for coronary heart disease: a decomposition analysis. J Am Heart Assoc. 2019;8 (23 ):e011964. doi:10.1161/jaha.119.011964 31787056 63 Dimick J , Ruhter J , Sarrazin MV , Birkmeyer JD . Black patients more likely than whites to undergo surgery at low‐quality hospitals in segregated regions. Health Aff. 2013;32 (6 ):1046‐1053. doi:10.1377/hlthaff.2011.1365 64 Gunderson CC , Nugent EK , McMeekin DS , Moore KN . Distance traveled for treatment of cervical cancer: who travels the farthest, and does it impact outcome? Int J Gynecologic Cancer. 2013;23 (6 ):1099‐1103. doi:10.1097/IGC.0b013e3182989464 65 Orringer M , Roberts H , Ward D . Racial and socioeconomic differences in distance traveled for elective hip arthroplasty. J Am Acad Orthopaed Surg Glob Res Rev. 2022;6 (4 ):e22.00021. doi:10.5435/JAAOSGlobal-D-22-00021