Intersectional disparities in outpatient alcohol treatment completion by gender and race and ethnicity.

BACKGROUND: Untreated alcohol use disorder (AUD) can have negative outcomes, including premature death. Completing specialty treatment for AUD can improve economic and educational outcomes. However, there are large racial and ethnic disparities in treatment completion, and how these disparities vary intersectionally (e.g., by gender and race and ethnicity) is unknown. Recent studies suggest that not using an intersectional approach can mask important disparities. We estimated disparities in AUD nonintensive outpatient treatment completion by gender alone, race and ethnicity alone, and intersectionally in a gender-by-race-and-ethnicity model. Accurately quantifying treatment completion disparities is critical not only for understanding healthcare disparities but reducing them to advance health equity. METHODS: Data are from SAMHSA's 2017 to 2019 Treatment Episode Dataset-Discharges for adults aged 18+ who entered nonintensive outpatient treatment primarily for alcohol (n = 559,447 episodes; 30.3% women; 63.7% White, 18.0% Black, 14.4% Hispanic/Latinx, 2.1% American Indian/Alaska Native [AIAN], 1.0% Asian/Pacific Islander). Using the rank-and-replace method, treatment completion disparities were estimated by gender, race and ethnicity, and gender-by-race-and-ethnicity due to any reason other than differences in need for treatment, consistent with the Institute of Medicine's definition of a healthcare disparity. RESULTS: The intersectional gender-by-race-and-ethnicity model identified the widest range of disparities among all models tested. Using this model, the largest disparities were identified for minoritized women's treatment episodes. Compared to White men whose completion rate was 60.79% (95% confidence interval [CI]: 60.06, 60.98), Black, Hispanic/Latina, AIAN, and Asian-American/Pacific Islander women had treatment episode completion rates that were 12.35 (CI: 12.33, 12.37), 9.08 (CI: 9.06, 9.11), 10.27 (CI: 10.22, 10.32), and 4.87 (CI: 4.78, 4.95) percentage points lower, respectively. CONCLUSIONS: In the United States, treatment completion rates for non-intensive outpatient alcohol treatment episodes are significantly lower for minoritized women than White men. The extent of the disparity is not apparent in univariate models, highlighting the importance of an intersectional approach to understanding disparities in the completion of non-intensive outpatient treatment for AUD.

Patterns of health care interactions of individuals with alcohol use disorder: A latent class analysis.

INTRODUCTION: Given the high rates at which individuals with alcohol use disorder (AUD) utilize health care for co-existing conditions, health systems are promising venues for interventions that will facilitate access to AUD treatment. However, how individuals with AUD interact with such systems and, thus, how systems should intervene is unclear. In this study, we seek to identify patterns in how individuals diagnosed with AUD within an academic health system interacted with the system prior to diagnosis. METHODS: We use electronic health records from a single academic health system in a major US metropolitan area to create a deidentified retrospective cohort including all individuals age 18+ diagnosed with AUD 2010-2019 (n = 26,899). Latent class analysis (LCA) identified subgroups defined by aspects of previous system interaction and health status, including having an in-system primary care provider, previous utilization of primary and specialty care, diagnosis setting, payer, and presence of other chronic conditions. We then assessed subgroup differences in demographics and associations with in-system AUD treatment receipt in the year following diagnosis, adjusting for demographics. RESULTS: The population was on average 38.6 years old (standard deviation = 15.4) and predominantly male (66.1 %), White (64.5 %), and not of Hispanic/Latino ethnicity (87.8 %). Only 4.7 % received in-system treatment following diagnosis. We deemed the four-class model the optimal LCA model. This model identified subgroups that can be described as 1) average utilization (20.7 % of population), 2) low utilization (54.5 %), 3) high health burden and low utilization (14.2 %), and 4) high health burden and high utilization (10.6 %). Predicted membership in the high health burden and high utilization subgroup and low utilization subgroup were associated with higher and lower odds of treatment receipt, respectively, compared with predicted membership in the average utilization subgroup (odds ratio (OR) for high/high subgroup = 1.21, 95 % confidence interval (CI) = 1.01, 1.27; OR for low subgroup = 0.29 95 % CI = 0.24, 0.34). CONCLUSION: Individuals diagnosed with AUD within a health system interact with that system in markedly different ways and are unlikely to benefit uniformly from system-based interventions to facilitate treatment. Group-tailored interventions are more likely to have impact and provide returns on investments for systems.

Residence in a Medicaid-expansion state and receipt of alcohol screening and brief counseling by adults with lower incomes: Is increased access to primary care enough?

BACKGROUND: We investigate whether living in a state that expanded Medicaid eligibility is associated with receiving alcohol screening and brief counseling among nonelderly, low-income adults and a subgroup with chronic health conditions caused or exacerbated by alcohol use. METHOD: Data are from the 2017 and 2019 Behavioral Risk Factor Surveillance System (N = 15,743 low-income adults; n = 7062 with a chronic condition). We used propensity score-weighted, covariate-adjusted, modified Poisson regression to estimate associations between residence in a Medicaid-expansion state and receipt of alcohol screening and brief counseling. Models estimated associations in the overall sample and chronic conditions subsample, as well as differential associations across sex, race, and ethnicity using interaction terms. RESULTS: Living in a state that expanded Medicaid eligibility was associated with being asked whether one drank (prevalence ratio (PR) = 1.15, 95% confidence interval (CI) = 1.08, 1.22), but not with further alcohol screening, guidance about harmful drinking, or advice to reduce drinking. Among individuals with alcohol-related chronic conditions, expansion state residence was associated with being asked about drinking (PR = 1.13, 95% CI = 1.05, 1.20) and, among past 30-day drinkers with chronic conditions, being asked how much one drank (PR = 1.28, 95% CI = 1.04, 1.59) and about binge drinking (PR = 1.43, 95% CI = 1.03, 1.99). Interaction terms suggest that some associations differ by race and ethnicity. CONCLUSIONS: Living in a state that expanded Medicaid is associated with a higher prevalence of receiving some alcohol screening at a check-up in the past 2 years among low-income residents, particularly among individuals with alcohol-related chronic conditions, but not with the receipt of high-quality screening and brief counseling. Policies may have to address provider barriers to delivery of these services in addition to access to care.

Hazardous drinking by older adults with chronic conditions during the COVID-19 pandemic: Evidence from a Chicago-based cohort.

BACKGROUND: It is unclear how older adults with chronic conditions, who have greater risk of alcohol-related adverse outcomes, used alcohol throughout the COVID-19 pandemic. We assess changes in hazardous drinking prevalence May 2020-December 2021 and factors associated with hazardous drinking. METHODS: Data are from structured phone interviews of older adults (age 60+) with chronic conditions (e.g., hypertension, diabetes, pulmonary disease, heart disease) in a Chicago-based longitudinal cohort (Chicago COVID-19 Comorbidities survey, Waves 3-7, n = 247). We tested differences in the prevalence of hazardous drinking (defined as AUDIT-C score of 3+ for women and 4+ for men) across waves for the full sample, by demographic group (sex, race, and ethnicity), and by chronic condition burden (<3 conditions, 3+ conditions). Generalized estimating equations investigated associations of hazardous drinking with sociodemographic and pandemic coping-related factors (stress, loneliness, outside contacts, depression, anxiety). RESULTS: Participants were 66.8% female; 27.9% non-Hispanic Black, 14.2% Hispanic, 4.9% other race. Hazardous drinking was reported by 44.9% of participants in May 2020, but declined to 23.1% by July-August 2020 and continued to slowly decline to 19.4% by September-December 2021. Differences from May 2020 were significant at the 0.05 level. Subgroups followed similar trajectories. Hazardous drinking prevalence was initially higher but declined more among men than women, consistently higher among non-Hispanic White respondents than among Hispanic and non-Hispanic Black respondents, and declined more rapidly among adults with 3+ chronic conditions. In adjusted models, race and ethnicity were associated with lower prevalence of hazardous drinking (non-Hispanic Black: adjusted prevalence ratio [aPR] = 0.50, 95% confidence interval [CI] = 0.33, 0.74; other race: aPR = 0.26, 95% CI = 0.09, 0.81, compared with non-Hispanic White). No coping-related factors were significantly associated with hazardous drinking. CONCLUSION: Among a cohort of older adults with chronic conditions, almost half engaged in hazardous drinking in early summer of the COVID-19 pandemic. While prevalence fell, these rates reinforce the need for alcohol screening and intervention in clinical settings among this population.

Prevalence of Alcohol in Unintentional Opioid Overdose Deaths, 2017-2020.

This cross-sectional study investigates the prevalence of alcohol in unintentional opioid overdose deaths in Illinois from 2017 through 2020.

Does the relationship between alcohol retail environment and alcohol outcomes vary by depressive symptoms? Findings from a U.S. Survey of Black, Hispanic and White drinkers.

AIMS: To assess whether associations between alcohol availability and consumption, drinking to drunkenness, and negative drinking consequences vary among individuals with elevated depressive symptoms. METHODS: 10,482 current drinkers in 2005-2015 National Alcohol Surveys (50.0% female; 74.4% White, 8.7% Black, 11.1% Hispanic). Elevated depressive symptoms was defined as having symptoms suggestive of major depressive disorder (above CES-D8/PHQ-2 cut-offs) versus no/sub-threshold symptoms (below cut-offs). Inverse probability of treatment weighted and covariate adjusted Poisson models with robust standard errors estimated associations of ZIP-code bar density and off-premise outlet density (locations/1,000 residents), elevated depressive symptoms, and their interaction with past-year volume consumed, monthly drinking to drunkenness, and negative drinking consequences. Models were then stratified by sex and race and ethnicity. RESULTS: Overall, 13.7% of respondents had elevated depressive symptoms. Regarding density, the only statistically significant association observed was between off-premise density and volume consumed (rate ratio = 1.3, 95% confidence interval = 1.0, 1.7). Elevated depressive symptoms were associated with higher volume consumed, prevalence of drinking to drunkenness, and prevalence of negative consequences when controlling for off-premise density or bar density. However, there was no evidence of interaction between symptoms and density in the full sample nor among subgroups. CONCLUSION: This study suggests that, while elevated depressive symptoms do not alter associations between alcohol availability and alcohol use and problems, they remain associated with these outcomes among past-year drinkers in a U.S. general population sample even when accounting for differential availability. Addressing depressive symptoms should be considered along with other policies to reduce population-level drinking and alcohol problems.

Effect of State-Led Delivery System and Payment Reforms on Population-Level Detection and Management of Diabetes.

OBJECTIVE: The Centers for Medicare and Medicaid Services State Innovation Models (SIM) initiative has invested more than $1 billion to test state-led delivery system and payment reforms that can affect diabetes care management. We examined whether SIM implementation between 2013 and 2017 was associated with diagnosed diabetes prevalence or with hospitalization or 30-day readmission rate among diagnosed adults. RESEARCH DESIGN AND METHODS: The quasiexperimental design compared study outcomes before and after the SIM initiative in 12 SIM states versus five comparison states using difference-in-differences (DiD) regression models of 21,055,714 hospitalizations for adults age ≥18 years diagnosed with diabetes in 889 counties from 2010 to 2017 across the 17 states. For readmission analyses, comparative interrupted time series (CITS) models included 11,812,993 hospitalizations from a subset of nine states. RESULTS: Diagnosed diabetes prevalence changes were not significantly different between SIM states and comparison states. Hospitalization rates were inconsistent across models, with DiD estimates ranging from -5.34 to -0.37 and from -13.16 to 0.92, respectively. CITS results indicate that SIM states had greater increases in odds of 30-day readmission during SIM implementation compared with comparison states (round 1: adjusted odds ratio [AOR] 1.07; 95% CI 1.04, 1.11; P < 0.001; round 2: AOR 1.06; 95% CI 1.03, 1.10; P = 0.001). CONCLUSIONS: The SIM initiative was not sufficiently focused to have a population-level effect on diabetes detection or management. SIM states had greater increases in 30-day readmission for adults with diabetes than comparison states, highlighting potential unintended effects of engaging in the multipayer alignment efforts required of state-led delivery system and payment reforms.

Alcohol Use and Blood Pressure Among Adults with Hypertension: the Mediating Roles of Health Behaviors.

BACKGROUND: Alcohol use is associated with increased blood pressure among adults with hypertension, but it is unknown whether some of the observed relationship is explained by mediating behaviors related to alcohol use. OBJECTIVE: We assess the potential indirect role of smoking, physical inactivity, unhealthy diet, and poor medication adherence on the association between alcohol use and blood pressure among Black and White men and women with hypertension. DESIGN: Adjusted repeated-measures analyses using generalized estimating equations and mediation analyses using inverse odds ratio weighting. PARTICIPANTS: 1835 participants with hypertension based on ACC/AHA 2017 guidelines in three most recent follow-up exams of the longitudinal Coronary Artery Risk Development in Young Adults cohort study (2005-2016). MAIN MEASURES: Alcohol use was assessed using both self-reported average ethanol intake (drinks/day) and engagement in heavy episodic drinking (HED) in the past 30 days. Systolic and diastolic blood pressure (SBP, DBP) were measured by trained technicians (mmHg). Smoking, physical inactivity, and diet were self-reported and categorized according to American Heart Association criteria, and medication adherence was assessed using self-reported typical adherence to antihypertensive medications. KEY RESULTS: At baseline (2005-2006), 57.9% of participants were Black and 51.4% were women. Mean age (standard deviation) was 45.5 (3.6) years, mean SBP was 128.7 (15.5) mmHg, and mean DBP was 83.2 (10.1) mmHg. Each additional drink per day was significantly associated with higher SBP (ß = 0.713 mmHg, 95% confidence interval (CI): 0.398, 1.028) and DBP (ß = 0.398 mmHg, 95% CI: 0.160, 0.555), but there was no evidence of mediation by any of the behaviors. HED was not associated with blood pressure independent of average consumption. CONCLUSIONS: These findings support the direct nature of the association of alcohol use with blood pressure and the utility of advising patients with hypertension to limit consumption in addition to other behavioral and pharmacological interventions.

Cigarettes smoked among daily and non-daily smokers following CVS Health's tobacco-free pharmacy policy.

INTRODUCTION: In September 2014, CVS Health ceased tobacco sales in all of its 7700 pharmacies nationwide. We investigate the impact of the CVS policy on the number of cigarettes smoked per day among metropolitan daily and non-daily smokers, who may respond to the availability of smoking cues in different manners. METHODS: Data are from the US Census Bureau Tobacco Use Supplement to the Current Population Survey 2014-2015 and the Blue Cross and Blue Shield Institute Community Health Management Hub. Adjusted difference-in-difference (DID) regressions assess changes in the number of cigarettes smoked per day among daily smokers (n=10 759) and non-daily smokers (n=3055), modelling core-based statistical area (CBSA) level CVS pharmacy market share continuously. To assess whether the policy had non-linear effects across the distribution of CVS market share, we also examine market share using tertiles. RESULTS: CVS's tobacco-free pharmacy policy was associated with a significant reduction in the number of cigarettes smoked by non-daily smokers in the continuous DID (rate ratio=0.985, p=0.022), with a larger reduction observed among non-daily smokers in CBSAs in the highest third of CVS market share compared with those living in CBSAs with no CVS presence (rate ratio=0.706, p=0.027). The policy, however, was not significantly associated with differential changes in the number of cigarettes by daily smokers. CONCLUSION: The removal of tobacco products from CVS pharmacies was associated with a reduction in the number of cigarettes smoked per day among non-daily smokers in metropolitan CBSAs, particularly those in which CVS had a large pharmacy market share.

Washington's liquor license system and alcohol-related adverse health outcomes.

BACKGROUND AND AIMS: In June 2012, Washington state (USA) implemented Initiative 1183, privatizing liquor sales. As a result, off-premises outlets increased from 330 to over 1400 and trading hours lengthened. Increased availability of liquor may lead to increased consumption. This study examines the impact of Initiative 1183 on alcohol-related adverse health outcomes, measured by inpatient hospitalizations for alcohol-related disorders and accidental injuries. It further assesses heterogeneity by urbanicity, because outlets increased most in metropolitan-urban areas. DESIGN: County-by-quarter difference-in-difference linear regression models, estimated statewide and within metropolitan/rural strata. SETTING AND PARTICIPANTS: Data are from AHRQ Healthcare Cost and Utilization State Inpatient Database 2010-2014 and HHS Area Health Resource File 2010-2014. Changes in the rates of hospitalizations in the 2.5 years following Initiative 1183 in Washington (n = 39 counties) are compared with changes in Oregon (n = 36 counties). MEASUREMENTS: County rates of hospitalizations per 1000 residents, including all records with any-listed ICD-9 Clinical Classification Software code denoting an alcohol-related disorder, and all records with any-listed external cause of injury code denoting an accidental injury. FINDINGS: The increase in the rate of accidental injury hospitalizations in Washington's metropolitan-urban counties was on average 0.289 hospitalizations per 1000 county residents per quarter greater than the simultaneous increase observed in Oregon (P = 0.017). This result was robust to alternative specifications using a propensity score matched sample and synthetic control methods with data from other comparison states. The evidence did not suggest that Initiative 1183 was associated with differential changes in the rate of hospitalizations for alcohol-related disorders in metropolitan-urban (P = 0.941), non-metropolitan-urban (P = 0.162), or rural counties (P = 0.876). CONCLUSIONS: Implementing Washington's Initiative 1183 (privatizing liquor sales) appears to have been associated with a significant increase in the rate of accidental injury hospitalizations in urban counties in that state but does not appear to be significantly associated with changes in the rate of hospitalizations specifically for alcohol-related disorders within 2.5 years.

The Centers for Medicaid and Medicare Services State Innovation Models Initiative and Social Risk Factors: Improved Diagnosis Among Hospitalized Adults With Diabetes.

INTRODUCTION: Unaddressed social risks among hospitalized patients with chronic conditions contribute to costly complications and preventable hospitalizations. This study examines whether the Centers for Medicaid and Medicare Services State Innovation Models initiative, through payment and delivery system reforms, accelerates the diagnosis of social risk factors among hospitalized adults with diabetes. METHODS: Encounter-level data were from the Agency for Healthcare Research and Quality Healthcare Cost and Utilization Project State Inpatient Databases (2010-2015, N=5,040,456). Difference-in-difference logistic regression estimated the extent to which hospitalized adults with diabetes in 4 State Innovation Models states (Arkansas, Massachusetts, Oregon, and Vermont) had increased odds of having a social risk factor diagnosed with an ICD-9 V code compared with hospitalized adults with diabetes in 4 comparison states (Arizona, Georgia, New Jersey, and New Mexico) 2 years after implementation. Data were analyzed between June and December 2019. RESULTS: Adults with diabetes who were hospitalized in State Innovation Models states had a 30% greater increase in the odds of having a V code documented after implementation than adults with diabetes who were hospitalized in comparison states (AOR=1.29, 95% CI=1.07, 1.56). However, V code use remained infrequent, with only 2.05% of encounters, on average, having any V codes on record in State Innovation Models states after implementation. CONCLUSIONS: The State Innovation Models initiative slightly but significantly improved the diagnosis of social risks among hospitalized adults with diabetes. State-led delivery system and payment reform may help support movement of hospitals toward better recognition and management of social determinants of health.

The Early Impact of the Centers for Medicare & Medicaid Services State Innovation Models Initiative on 30-Day Hospital Readmissions Among Adults With Diabetes.

BACKGROUND: The Centers for Medicare & Medicaid Services (CMS) State Innovation Models (SIM) Initiative funds states to accelerate delivery system and payment reforms. All SIM states focus on improving diabetes care, but SIM's effect on 30-day readmissions among adults with diabetes remains unclear. METHODS: A quasi-experimental research design estimated the impact of SIM on 30-day hospital readmissions among adults with diabetes in 3 round 1 SIM states (N=671,996) and 3 comparison states (N=2,719,603) from 2010 to 2015. Difference-in-differences multivariable logistic regression models that incorporated 4-group propensity score weighting were estimated. Heterogeneity of SIM effects by grantee state and for CMS populations were assessed. RESULTS: In adjusted difference-in-difference analyses, SIM was associated with an increase in odds of 30-day hospital readmission among patients in SIM states in the post-SIM versus pre-SIM period relative to the ratio in odds of readmission among patients in the comparison states post-SIM versus pre-SIM (ratio of adjusted odds ratio=1.057, P=0.01). Restricting the analyses to CMS populations (Medicare and Medicaid beneficiaries), resulted in consistent findings (ratio of adjusted odds ratio=1.057, P=0.034). SIM did not have different effects on 30-day readmissions by state. CONCLUSIONS: We found no evidence that SIM reduced 30-day readmission rates among adults with diabetes during the first 2 years of round 1 implementation, even among CMS beneficiaries. It may be difficult to reduce readmissions statewide without greater investment in health information exchange and more intensive use of payment models that promote interorganizational coordination.

U.S. county "food swamp" severity and hospitalization rates among adults with diabetes: A nonlinear relationship.

The relationship between food environments and diabetes morbidity is vastly understudied, despite the well-recognized linkage between dietary quality and diabetes complications. Further, literature demonstrates that attributes of places can have nonlinear relationships with health outcomes. This study examines the extent to which "food swamps" are associated with greater rates of hospitalizations for complications among adults with diabetes over time as well as the linearity of this relationship. We conduct a longitudinal county-level analysis of 832 counties across 16 U.S. states in 2010, 2012, and 2014 using data from the USDA Food Environment Atlas and the AHRQ Health Care Cost and Utilization Project State Inpatient Databases. Food swamp severity is measured as the percentage of food outlets in a county that sell primarily unhealthy foods. Hierarchical linear mixed models with county random intercepts are estimated, controlling for area-level covariates and state and year fixed effects. Curvilinear relationships are explored by additively incorporating quadratic terms. We find that, over the study period, mean food swamp severity remained relatively stable. Mean hospitalization rates decreased from 296.72 to 262.82 hospitalizations per 1000 diabetic adults (p < 0.001). In adjusted models, greater food swamp severity was associated with higher hospitalization rates in a curvilinear manner (severity: ß = 2.181, p = 0.02; severity2: ß = -0.017, p = 0.04), plateauing at approximately 64% unhealthy outlets, a saturation point observed in 17% of observations. Policies that limit saturation of the environment with unhealthy outlets may help in the prevention of diabetic complications, but more saturated counties will likely require more extensive intervention.

Implementation Variation in Natural Experiments of State Health Policy Initiatives.

OBJECTIVES: An increasing number of federal initiatives allow states flexibility in selecting the strategies used to achieve initiative-specific goals. Variation in the foci and intensity of implementation may explain why federal policy initiatives succeed in some states and fail in others. The CMS State Innovation Models (SIM) initiative is a complex policy intervention implemented with substantial variation across states and may have variable impacts. This paper presents a method to characterize and account for that variation in states' implementation foci and intensity in natural policy experiments. STUDY DESIGN: A combination of quantitative and qualitative measures of SIM implementation was used to characterize the foci of payment and delivery system reforms across states. METHODS: A modified Delphi expert panel process was used to prioritize the features of SIM implementation that would differentiate grantee states with respect to improved health outcomes. Three researchers then reviewed summaries of published evaluations and reports to characterize and score states on each implementation feature. Expert panelists guided the researchers on developing the criteria and weights applied to the focus areas when calculating SIM implementation intensity scores for states. RESULTS: Over 3 years of an expert panel process, 4 dimensions of SIM implementation that would most affect health outcomes were prioritized: 1) extent and breadth of stakeholder engagement, (2) extent that SIM implementation was focused on improving behavioral health, (3) amount of SIM funding per capita, and (4) breadth and depth of value-based payment reforms. Scoring states based on the prioritized factors resulted in composite scores that differentiated states into 3 categories: high, moderate, and low implementation intensity. CONCLUSIONS: We developed a stakeholder-driven method to measure and account for variation in implementation foci and intensity in a federal policy initiative that was implemented heterogeneously across grantee states. Our method for characterizing state implementation variation may be useful for natural policy experiments examining the variable impact of policy initiatives.

Adults with diabetes residing in "food swamps" have higher hospitalization rates.

OBJECTIVE: To examine the relationship between food swamps and hospitalization rates among adults with diabetes. DATA SOURCES: Blue Cross Blue Shield Association Community Health Management Hub® 2014, AHRQ Health Care Cost and Utilization Project state inpatient databases 2014, and HHS Area Health Resources File 2010-2014. STUDY DESIGN: Cross-sectional analysis of 784 counties across 15 states. Food swamps were measured using a ratio of fast food outlets to grocers. Multivariate linear regression estimated the association of food swamp severity and hospitalization rates. Population-weighted models were controlled for comorbidities; Medicaid; emergency room utilization; percentage of population that is female, Black, Hispanic, and over age 65; and state fixed effects. Analyses were stratified by rural-urban category. PRINCIPAL FINDINGS: Adults with diabetes residing in more severe food swamps had higher hospitalization rates. In adjusted analyses, a one unit higher food swamp score was significantly associated with 49.79 (95 percent confidence interval (CI) = 19.28, 80.29) additional all-cause hospitalizations and 19.12 (95 percent CI = 11.09, 27.15) additional ambulatory care-sensitive hospitalizations per 1000 adults with diabetes. The food swamp/all-cause hospitalization rate relationship was stronger in rural counties than urban counties. CONCLUSIONS: Food swamps are significantly associated with higher hospitalization rates among adults with diabetes. Improving the local food environment may help reduce this disparity.