The County Health Rankings: rationale and methods

Contribution of County Characteristics to Disparities in Rural Mortality After Cancer Diagnosis

INTRODUCTION

Rural disparities in cancer outcomes have been widely evaluated, but limited evidence is available to describe what characteristics of rural environments contribute to the increased risk of poor outcomes. Therefore, this manuscript sought to assess the mediating effects of county characteristics on the relationship between urban/rural status and mortality among patients with cancer, characterize county profiles, and determine at-risk county profiles alongside rural settings.

METHODS

Patients diagnosed with cancer between 2000 and 2016 were assessed using Surveillance, Epidemiology and End Results data linked to the 2010 Rural-Urban Commuting Codes and 2010 County Health Rankings. There were 757,655 patients representing 596 counties (of 3,143 in the U.S.) and 12 states. Mediation analyses, conducted in 2023, estimated the direct contribution of rurality to 5-year all-cause survival and the contribution of the rural effect indirectly through County Health Ranking domains. Latent class analysis and survival models identified county groupings and estimated the hazard of mortality associated with class membership.

RESULTS

Rankings for premature death, clinical care, and physical environment resulted in rural patients having 17.9%-20.2% less survival time than urban patients. Of this, 4.1%-12.6% of the total excess risk was mediated by these characteristics. Patients living in rural and high-risk county classes saw higher all-cause mortality than those in urban lower-risk counties (hazard ratio=1.04, 95% CI=1.01, 1.08 and 1.07, 95% CI=1.03, 1.11).

CONCLUSIONS

Counties with poorer health rankings had increased mortality risks regardless of rurality; however, the poor rankings, notably health behaviors and social and economic factors, elevated the risk for rural counties.

Beyond objective metrics: A comparative analysis of health care systems incorporating subjective dimensions to improve comparability of access and equity in healthcare assessments

Comparing health care systems is important for several reasons. E.g. lower-resource health care systems can learn from higher-resource ones, and country-specific progress can be made. Previous rankings of health care systems have been based on objective factors such as the number of available hospital beds or health care spending. An index is considered here that includes a subjective level that is intended to represent access to the health care system. Therefore, this study investigates the divergence between subjective and objective indices related to health care expenditure, with a focus on the influence of involuntary and voluntary payments. Utilizing the Rational Choice Theory as a framework, it explores how individual preferences and perceived benefits affect these indices. The analysis reveals that social insurance contributions, which are mandatory and beyond individual control, are evaluated differently in subjective indices compared to objective indices. This discrepancy is less pronounced for voluntary expenditures, where individuals have decision-making power. The findings highlight significant variations in the correlations between macroeconomic health care indicators and the indices, emphasizing the critical role of autonomy in financial decisions related to health care.

A two-stage bayesian model for assessing the geography of racialized economic segregation and premature mortality across US counties

Racialized economic segregation, a key metric that simultaneously accounts for spatial, social and income polarization in communities, has been linked to adverse health outcomes, including morbidity and mortality. Due to the spatial nature of this metric, the association between health outcomes and racialized economic segregation could also change with space. Most studies assessing the relationship between racialized economic segregation and health outcomes have always treated racialized economic segregation as a fixed effect and ignored the spatial nature of it. This paper proposes a two-stage Bayesian statistical framework that provides a broad, flexible approach to studying the spatially varying association between premature mortality and racialized economic segregation while accounting for neighborhood-level latent health factors across US counties. The two-stage framework reduces the dimensionality of spatially correlated data and highlights the importance of accounting for spatial autocorrelation in racialized economic segregation measures, in health equity focused settings.

Lifespan Socioeconomic Context Is Associated With Cytomegalovirus and Late-Differentiated CD8 + T and Natural Killer Cells: Initial Results in Older Adults

OBJECTIVE

Lower socioeconomic status (SES) can accelerate immune aging; however, it is unknown whether and how lifespan socioeconomic context (SEC)-the relative wealth and quality of the communities an individual lives in across their lifespan-impacts immune aging. We examined the effects of childhood and adulthood SEC on late-differentiated immune cells and investigated the mediating and moderating role of cytomegalovirus (CMV), a key driver of immune aging.

METHODS

Adults 60 years and older ( N = 109) reported their addresses from birth to age 60 years, which were coded for county-level employment, education, and income to construct a latent SEC variable, averaged across ages 0 to 18 years (childhood SEC) and 19 to 60 years (adulthood SEC). Blood was drawn semiannually for 5 years for CMV serostatus and flow cytometry estimates of late-differentiated CD8 + T and natural killer cells. Models were adjusted for chronological age, time, sex, and individual SES (current income and education).

RESULTS

Lower childhood SEC was associated with higher percentages of late-differentiated CD8 + T and natural killer cells via CMV seropositivity (indirect effects, p values = .015-.028). In addition, an interaction between CMV serostatus and SEC on CD8 + T-cell aging ( p = .049) demonstrated that adulthood SEC was negatively associated with immune aging among CMV- but not CMV+ adults.

CONCLUSIONS

Beyond current SES, SEC related to immune aging in distinct patterns by lifespan phase. Lower childhood SEC importantly may influence who acquires CMV, which in turn predicts higher levels of immune aging, whereas higher adulthood SEC was protective against immune aging among CMV- older adults. These initial results need to be explored in larger samples.

Spatiotemporal Patterns of Hospitalizations Among Older Adults With Co-Presence of Cancer and Dementia in US Counties: 2013-2018

We assessed the spatiotemporal patterns of hospitalization with comorbid cancer and dementia. Using the 2013-2018 inpatient claims data for Medicare fee-for-service (FFS) beneficiaries, we calculated hospitalization rates by dividing the total admissions from individuals with the co-presence of a major cancer (breast, prostate, lung, and colorectal) and dementia diagnoses with the total counts of FFS beneficiaries aged 65 or older. We identified 22 hotspots with high hospitalization rates that showed heterogeneous spatial and temporal utilization patterns. The odds of a county being a hotspot increased significantly with the county-level percentage of dual Medicare-Medicaid beneficiaries (aOR 1.05; 95% CI: 1.04-1.07) and the prevalence of cancer (aOR 1.73; 95% CI: 1.59-1.89), while decreased significantly with increasing degree of rurality (aOR .82; 95% CI: .79-.85) and decreased yearly over time (aOR .72; 95% CI: .68-.75). The identified hotspots and factors at the county-level may help understand healthcare utilization patterns and assess resource allocation for this unique patient group.

Multilevel Social Determinants of Patient-Reported Outcomes in Young Survivors of Childhood Cancer

In this study, the social determinants of patient-reported outcomes (PROs) in young survivors of childhood cancer aged <18 years are researched. This cross-sectional study investigated social determinants associated with poor PROs among young childhood cancer survivors. We included 293 dyads of survivors receiving treatment at St. Jude Children's Research Hospital who were <18 years of age during follow-up from 2017 to 2018 and their primary caregivers. Social determinants included family factors (caregiver-reported PROs, family dynamics) and county-level deprivation (socioeconomic status, physical environment via the County Health Rankings & Roadmaps). PROMIS measures assessed survivors' and caregivers' PROs. General linear regression tested associations of social determinants with survivors' PROs. We found that caregivers' higher anxiety was significantly associated with survivors' poorer depression, stress, fatigue, sleep issues, and reduced positive affect (p < 0.05); caregivers' sleep disturbances were significantly associated with lower mobility in survivors (p < 0.05). Family conflicts were associated with survivors' sleep problems (p < 0.05). Residing in socioeconomically deprived areas was significantly associated with survivors' poorer sleep quality (p < 0.05), while higher physical environment deprivation was associated with survivors' higher psychological stress and fatigue and lower positive affect and mobility (p < 0.05). Parental, family, and neighborhood factors are critical influences on young survivors' quality of life and well-being and represent new intervention targets.

Modeling health and well-being measures using ZIP code spatial neighborhood patterns

Individual-level assessment of health and well-being permits analysis of community well-being and health risk evaluations across several dimensions of health. It also enables comparison and rankings of reported health and well-being for large geographical areas such as states, metropolitan areas, and counties. However, there is large variation in reported well-being within such large spatial units underscoring the importance of analyzing well-being at more granular levels, such as ZIP codes. In this paper, we address this problem by modeling well-being data to generate ZIP code tabulation area (ZCTA)-level rankings through spatially informed statistical modeling. We build regression models for individual-level overall well-being index and scores from five subscales (Physical, Financial, Social, Community, Purpose) using individual-level demographic characteristics as predictors while including a ZCTA-level spatial effect. The ZCTA neighborhood information is incorporated by using a graph Laplacian matrix; this enables estimation of the effect of a ZCTA on well-being using individual-level data from that ZCTA as well as by borrowing information from neighboring ZCTAs. We deploy our model on well-being data for the U.S. states of Massachusetts and Georgia. We find that our model can capture the effects of demographic features while also offering spatial effect estimates for all ZCTAs, including ones with no observations, under certain conditions. These spatial effect estimates provide community health and well-being rankings of ZCTAs, and our method can be deployed more generally to model other outcomes that are spatially dependent as well as data from other states or groups of states.

Profits over care? An analysis of the relationship between corporate capitalism in the healthcare industry and cancer mortality in the United States

The characteristic features of 21st-century corporate capitalism - monopoly and financialization - are increasingly being recognized by public health scholars as undermining the foundations of human health. While the "vectors" through which this is occurring are well known - poverty, inequality, climate change among others - locating the root cause of this process in the nature and institutions of contemporary capitalism is relatively new. Researchers have been somewhat slow to study the relationship between contemporary capitalism and human health. In this paper, we focus on one of the leading causes of death in the United States; cancer, and empirically estimate the relationship between various measures of financialization and monopoly in the US healthcare system and cancer mortality. The measures we focus on are for the hospital industry, the health insurance industry, and the pharmaceutical industry. Using a fixed effects model with different specifications and control variables, our analysis is at the state level for the years 2012-2019. These variables include data on population demographic controls, social and economic factors, and health behavior and clinical care. We compare Medicaid expansion states with non-Medicaid expansion states to investigate variations in state-level funded health insurance coverage. The results show a statistically significant positive correlation between the HHI index in the individual healthcare market and cancer mortality and the opioid dispensing rate and cancer mortality.

Assessing health determinants worldwide: Econometric analysis of the Global Burden of Diseases Study 2000-18 - Highlighting impactful factors on DALY, YLL, and YLD indicators

Background

As the health status of a population is influenced by a variety of health determinants, we sought to assess their impact on health outcomes, both at the global and regional levels.

Methods

This ecological study encompassed all 194 member countries of the World Health Organization (WHO) from 2000 to 2018. We first identified all health determinants and then retrieved the related data from various global databases. We additionally considered three indicators - disability-adjusted life years (DALYs), years of life lost (YLL), and years lived with disability (YLD) - in evaluating health outcomes; we extracted their data from the Global Burden of Disease (GBD) 2019 study. We then applied econometric analyses using a multilevel mixed-effects linear regression model.

Results

The analysis using the DALY indicator showed that the variables of sexually transmitted infections, injuries prevalence, and urbanisation had the highest effect size or regression coefficients (β) for health outcomes. The variables of sexually transmitted infection (β = 0.75, P < 0.001) in the African region; drinking water (β = -0.60, P < 0.001), alcohol use (β = 0.20, P < 0.001), and drug use (β = 0.05, P = 0.036) in the Americas region; urbanisation (β = -0.34, P < 0.001) in the Eastern Mediterranean region; current health expenditure (β = -0.21, P < 0.001) in the Europe region; injuries (β = 0.65, P < 0.001), air pollution (β = 0.29, P < 0.001), and obesity (β = 0.92, P < 0.001) in the South-East Asia region; and gross domestic product (β = -0.25, P < 0.001), education (β = -0.90, P < 0.001), and smoking (β = 0.28, P < 0.001) in the Western Pacific region had the most significant role in explaining global health outcomes. Except for the drug use variable in regional findings, the role of other variables in explaining the YLL indicator was greater than that of the YLD indicator.

Conclusions

To address global health disparities and optimise resource allocation, global and interregional policymakers should focus on determinants that had the highest β with health outcomes in each region compared to other regions. These determinants likely have a higher marginal health product, and investing in them is likely to be more cost-effective.

Nationwide, County-Level Analysis of the Patterns, Trends, and System-Level Predictors of Opioid Prescribing in Surgery in the US: Social Determinants and Access to Mental Health Services Matter

BACKGROUND

The diversion of unused opioid prescription pills to the community at large contributes to the opioid epidemic in the US. In this county-level population-based study, we aimed to examine the US surgeons' opioid prescription patterns, trends, and system-level predictors in the peak years of the opioid epidemic.

STUDY DESIGN

Using the Medicare Part D database (2013 to 2017), the mean number of opioid prescriptions per beneficiary (OPBs) was determined for each US county. Opioid-prescribing patterns were compared across counties. Multivariable linear regression was performed to determine relationships between county-level social determinants of health (demographic, eg median age and education level; socioeconomic, eg median income; population health status, eg percentage of current smokers; healthcare quality, eg rate of preventable hospital stays; and healthcare access, eg healthcare costs) and OPBs.

RESULTS

Opioid prescription data were available for 1,969 of 3,006 (65.5%) US counties, and opioid-related deaths were recorded in 1,384 of 3,006 counties (46%). Nationwide, the mean OPBs decreased from 1.08 ± 0.61 in 2013 to 0.87 ± 0.55 in 2017; 81.6% of the counties showed the decreasing trend. County-level multivariable analyses showed that lower median population age, higher percentages of bachelor's degree holders, higher percentages of adults reporting insufficient sleep, higher healthcare costs, fewer mental health providers, and higher percentages of uninsured adults are associated with higher OPBs.

CONCLUSIONS

Opioid prescribing by surgeons decreased between 2013 and 2017. A county's suboptimal access to healthcare in general and mental health services in specific may be associated with more opioid prescribing after surgery.

Impact of alcohol driving-while-impaired license suspension duration on future alcohol-related license events and motor vehicle crash involvement in North Carolina, 2007 to 2016

BACKGROUND/PURPOSE

License suspensions are a strategy to address alcohol-impaired driving behavior and recidivism following an alcohol driving while impaired (alcohol-DWI) conviction. Little is known about the specific impacts of conviction-related suspensions on safety outcomes and given recent fluctuations in alcohol-impaired driving behavior, crashes, and suspension trends, updated and focused assessments of this intervention are necessary. This study aimed to 1) examine the association between type of recent alcohol-DWI suspension and having a secondary alcohol-related license outcome and/or future crash event in North Carolina (NC) between 2007 and 2016; and 2) assess potential modification of these associations by race/ethnicity.

METHODS

We used linked NC licensing data, NC crash data, and county-level contextual data from a variety of data sources. We compared individuals ages 21 to 64 who sustained initial (1-year) versus repeat (4-year) suspensions for alcohol-related license and crash involvement outcomes. We estimated unadjusted and adjusted hazard ratios (aHRs) using Cox proportional hazards models and produced Kaplan-Meier (KM) survival curves using a three-year follow-up period. After observing statistically significant modification by race/ethnicity, we calculated stratified aHRs for each outcome (Black and White subgroups only, as other subgroups had low numbers of outcomes).

RESULTS

122,002 individuals sustained at least one alcohol-DWI conviction suspension (117,244 initial, 4,758 repeat). Adjusted KM survival curves indicated that within three years of the index suspension, the predicted risks of having a license outcome and crash outcome were about 8 % and 15 %, respectively, among individuals with an initial suspension and 5 % and 10 %, respectively, among individuals with a repeat suspension. After adjusting for potential confounding, we found that compared to those with an initial suspension, those with repeat suspensions had a lower incidence of future license (aHR: 0.49; 95 % CI: 0.42, 0.57) and crash outcomes (aHR: 0.67; 95 % CI: 0.60, 0.75). Among Black individuals, license outcome incidence was 162 % lower among repeat versus initial index suspension groups (aHR: 0.38; 95 % CI: 0.26, 0.55), while for White individuals, the incidence was 87 % lower (aHR: 0.54; 95 % CI: 0.45, 0.64). Similarly, crash incidence for repeat versus initial suspensions among Black individuals was 56 % lower (aHR: 0.64; 95 % CI: 0.50, 0.83), while only 39 % lower among White individuals (aHR: 0.72; 95 % CI: 0.63, 0.81).

CONCLUSIONS

Decreased incidence of both license and crash outcomes were observed among repeat versus initial index suspensions. The magnitude of these differences varied by race/ethnicity, with larger decreases in incidence among Black compared to White individuals. Future research should examine the underlying mechanisms leading to alcohol-impaired driving behavior, convictions, recidivism, and crashes from a holistic social-ecological perspective so that interventions are designed to both improve road safety and maximize other critical public health outcomes, such as access to essential needs and services (e.g., healthcare and employment).

Social determinants of health and diabetes: using a nationally representative sample to determine which social determinant of health model best predicts diabetes risk

OBJECTIVES

Social determinants of health (SDOH) research demonstrates poverty, access to healthcare, discrimination, and environmental factors influence health outcomes. Several models are commonly used to assess SDOH, yet there is limited understanding of how these models differ regarding their ability to predict the influence of social determinants on diabetes risk. This study compares the utility of four SDOH models for predicting diabetes disparities.

STUDY DESIGN

We utilized The National Longitudinal Study of Adolescent to Adulthood (Add Health) to compare SDOH models and their ability to predict risk of diabetes and obesity.

METHODS

Previous literature has identified the World Health Organization (WHO), Healthy People, County Health Rankings, and Kaiser Family Foundation as the conventional SDOH models. We used these models to operationalize SDOH using the Add Health dataset. Add Health data were used to perform logistic regressions for HbA1c and linear regressions for body mass index (BMI).

RESULTS

The Kaiser model accounted for the largest proportion of variance (19%) in BMI. Race/ethnicity was a consistent factor predicting BMI across models. Regarding HbA1c, the Kaiser model also accounted for the largest proportion of variance (17%). Race/ethnicity and wealth was a consistent factor predicting HbA1c across models.

CONCLUSION

Policy and practice interventions should consider these factors when screening for and addressing the effects of SDOH on diabetes risk. Specific SDOH models can be constructed for diabetes based on which determinants have the largest predictive value.

Evaluating county-level lung cancer incidence from environmental radiation exposure, PM2.5, and other exposures with regression and machine learning models

Characterizing the interplay between exposures shaping the human exposome is vital for uncovering the etiology of complex diseases. For example, cancer risk is modified by a range of multifactorial external environmental exposures. Environmental, socioeconomic, and lifestyle factors all shape lung cancer risk. However, epidemiological studies of radon aimed at identifying populations at high risk for lung cancer often fail to consider multiple exposures simultaneously. For example, moderating factors, such as PM2.5, may affect the transport of radon progeny to lung tissue. This ecological analysis leveraged a population-level dataset from the National Cancer Institute's Surveillance, Epidemiology, and End-Results data (2013-17) to simultaneously investigate the effect of multiple sources of low-dose radiation (gross [Formula: see text] activity and indoor radon) and PM2.5 on lung cancer incidence rates in the USA. County-level factors (environmental, sociodemographic, lifestyle) were controlled for, and Poisson regression and random forest models were used to assess the association between radon exposure and lung and bronchus cancer incidence rates. Tree-based machine learning (ML) method perform better than traditional regression: Poisson regression: 6.29/7.13 (mean absolute percentage error, MAPE), 12.70/12.77 (root mean square error, RMSE); Poisson random forest regression: 1.22/1.16 (MAPE), 8.01/8.15 (RMSE). The effect of PM2.5 increased with the concentration of environmental radon, thereby confirming findings from previous studies that investigated the possible synergistic effect of radon and PM2.5 on health outcomes. In summary, the results demonstrated (1) a need to consider multiple environmental exposures when assessing radon exposure's association with lung cancer risk, thereby highlighting (1) the importance of an exposomics framework and (2) that employing ML models may capture the complex interplay between environmental exposures and health, as in the case of indoor radon exposure and lung cancer incidence.

Revealing spatiotemporal inequalities, hotspots, and determinants in healthcare resource distribution: insights from hospital beds panel data in 2308 Chinese counties

BACKGROUND

Ensuring universal health coverage and equitable access to health services requires a comprehensive understanding of spatiotemporal heterogeneity in healthcare resources, especially in small areas. The absence of a structured spatiotemporal evaluation framework in existing studies inspired us to propose a conceptual framework encompassing three perspectives: spatiotemporal inequalities, hotspots, and determinants.

METHODS

To demonstrate our three-perspective conceptual framework, we employed three state-of-the-art methods and analyzed 10 years' worth of Chinese county-level hospital bed data. First, we depicted spatial inequalities of hospital beds within provinces and their temporal inequalities through the spatial Gini coefficient. Next, we identified different types of spatiotemporal hotspots and coldspots at the county level using the emerging hot spot analysis (Getis-Ord Gi* statistics). Finally, we explored the spatiotemporally heterogeneous impacts of socioeconomic and environmental factors on hospital beds using the Bayesian spatiotemporally varying coefficients (STVC) model and quantified factors' spatiotemporal explainable percentages with the spatiotemporal variance partitioning index (STVPI).

RESULTS

Spatial inequalities map revealed significant disparities in hospital beds, with gradual improvements observed in 21 provinces over time. Seven types of hot and cold spots among 24.78% counties highlighted the persistent presence of the regional Matthew effect in both high- and low-level hospital bed counties. Socioeconomic factors contributed 36.85% (95% credible intervals [CIs]: 31.84-42.50%) of county-level hospital beds, while environmental factors accounted for 59.12% (53.80-63.83%). Factors' space-scale variation explained 75.71% (68.94-81.55%), whereas time-scale variation contributed 20.25% (14.14-27.36%). Additionally, six factors (GDP, first industrial output, local general budget revenue, road, river, and slope) were identified as the spatiotemporal determinants, collectively explaining over 84% of the variations.

CONCLUSIONS

Three-perspective framework enables global policymakers and stakeholders to identify health services disparities at the micro-level, pinpoint regions needing targeted interventions, and create differentiated strategies aligned with their unique spatiotemporal determinants, significantly aiding in achieving sustainable healthcare development.

Stigma as a local process: Stigma associated with opioid dependency in a rural-mixed Indiana county

BACKGROUND

Because the nature and magnitude of stigmatizing views associated with opioid dependency vary by social, cultural, and structural factors, strategies to reduce public stigma towards opioid dependency should vary by context. We leverage a unique dataset with evidence of multiple stigmatizing views to understand how to target interventions to reduce stigma in a state disproportionately impacted by the opioid epidemic, with a specific focus on a rural-mixed county.

METHODS

Data come from the representative Person-to-Person Health Study (2018-2020) of 2,050 Indiana residents, 224 from the rural-mixed Fayette County. Bivariate statistics and multivariate regression analyses were used to evaluate the association between Fayette County and measures of stigma (e.g., desire for social distance, prejudice, causal attributions) relative to the rest of Indiana.

RESULTS

Fayette County statistically differed from the rest of Indiana on most demographic characteristics and measures of stigmatizing views. Multivariate regressions revealed that compared to the rest of Indiana, residence in Fayette County was associated with a higher desire for social distance, perceptions of unpredictability, and attributing opioid dependency to genetics and the way the person was raised.

CONCLUSION

Our results contribute to growing evidence supporting the need for local approaches to address stigma. Stigma in Fayette County primarily reflects concerns about how people manage their opioid dependency. Strategies focusing on treatment and recovery potential, accompanied by extending the influence of supportive stakeholders and policies, will become important to address this stigma.

Language use on Twitter reflects social structure and social disparities

Large-scale mental health assessments increasingly rely upon user-contributed social media data. It is widely known that mental health and well-being are affected by minority group membership and social disparity. But do these factors manifest in the language use of social media users? We elucidate this question using spatial lag regressions. We examined the county-level (N = 1069) associations of lexical indicators linked to well-being and mental health, notably depression (e.g., first-person singular pronouns, negative emotions) with markers of social disparity (e.g., the Area Deprivation Index-3) and ethnicity, using a sample of approximately 30 million content-coded tweets (U.S. county-level aggregation). Results confirmed most expected associations: County-level lexical indicators of depression are positively linked with county-level area disparity (e.g., economic hardship and inequity) and percentage of ethnic minority groups. Predictive validity checks show that lexical indicators are related to future health and mental health outcomes. Lexical indicators of depression and adjustment coded from tweets aggregated at the county level could play a crucial role in prioritizing public health campaigns, particularly in socially deprived counties.

Regional and racial/ethnic inequalities in public drinking water fluoride concentrations across the US

BACKGROUND

Although the US Centers for Disease Control and Prevention (CDC) considers fluoridation of community water systems (CWSs) to be a major public health achievement responsible for reducing dental disease, recent epidemiologic evidence suggests that chronic exposure to population-relevant levels of fluoride may also be associated with adverse child neurodevelopmental outcomes. To our knowledge, a nationally representative database of CWS fluoride concentration estimates that can be readily linked to US epidemiologic cohorts for further study is not publicly available. Our objectives were to evaluate broad regional and sociodemographic inequalities in CWS fluoride concentrations across the US, and to determine if county-level racial/ethnic composition was associated with county-level CWS fluoride.

METHODS

We generated CWS-level (N = 32,495) and population weighted county-level (N = 2152) fluoride concentration estimates using over 250,000 routine compliance monitoring records collected from the US Environmental Protection Agency's (EPA) Third Six Year Review (2006-2011). We compared CWS-level fluoride distributions across subgroups including region, population size served, and county sociodemographic characteristics. In county-level spatial error models, we also evaluated geometric mean ratios (GMRs) of CWS fluoride per 10% higher proportion of residents belonging to a given racial/ethnic subgroup.

RESULTS

4.5% of CWSs (serving >2.9 million residents) reported mean 2006-2011 fluoride concentrations ≥1500 µg/L (the World Health Organization's guideline for drinking water quality). Arithmetic mean, 90th, and 95th percentile contaminant concentrations were greatest in CWSs reliant on groundwater, located in the Southwest and Eastern Midwest, and serving Semi-Urban, Hispanic communities. In fully adjusted spatial error models, the GMR (95% CI) of CWS fluoride per a 10% higher proportion of county residents that were Hispanic/Latino was 1.16 (1.10, 1.23).

IMPACT STATEMENT

We find that over 2.9 million US residents are served by public water systems with average fluoride concentrations exceeding the World Health Organization's guidance limit. We also find significant inequalities in community water system fluoride concentration estimates (2006-2011) across the US, especially for Hispanic/Latino communities who also experience elevated arsenic and uranium in regulated public drinking water systems. Our fluoride estimates can be leveraged in future epidemiologic studies to assess the potential association between chronic fluoride exposure and related adverse outcomes.

TRansit ACessibility Tool (TRACT): Developing a novel scoring system for public transportation system accessibility

Introduction

Although federal laws require equal access to public transportation for people with disabilities, access barriers persist. Lack of sharing accessibility information on public transportation websites restricts people with disabilities from making transportation plans and effectively using public transportation systems. This project aims to document information provided about public transportation systems accessibility and share this information using an open data platform.

Methods

We reviewed the top twenty-six public transportation systems in the United States based on federal funding in fiscal year 2020. Information about accessibility was abstracted from each public transportation system website by two independent reviewers from February-March 2022. Informed by universal design principles, public transportation systems were scored across six dimensions: facility accessibility (0-22 points), vehicle accessibility (0-11 points), inclusive policies (0-12 points), rider accommodations (0-9 points), paratransit services (0-6 points), and website accessibility (0-2 points). Total scores were calculated as the sum of each dimension (0-62 points). Data and findings were publicly disseminated (https://disabilityhealth.jhu.edu/transitdashboard/).

Results

The average overall accessibility information score was 31.9 (SD=6.2) out of 62 possible points. Mean scores were 8.4 (SD=2.9) for facility accessibility, 4.5 (SD=2.1) for vehicle accessibility, 7.8 (SD=1.6) for inclusive policies, 4.9 (SD=1.6) for rider accommodations, 4.5 (SD=2.0) for paratransit services, and 1.8 (SD=0.4) for website accessibility. Eleven public transportation systems (42%) received the maximum score for paratransit services and 20 (77%) received the maximum score for website accessibility. No public transportation system received the maximum score for any of the other dimensions.

Conclusions

Using a novel scoring system, we found significant variation in the accessibility information presented on public transportation system websites. Websites are a primary mode where users obtain objective information about public transportation systems and are therefore important platforms for communication. Absence of accessibility information creates barriers for the disability community and restricts equal access to public transportation.

Amputation Rates and Associated Social Determinants of Health in the Most Populous US Counties

Importance

Social Determinants of Health (SDOH) have been found to be associated with health outcome disparities in patients with peripheral artery disease (PAD). However, the association of specific components of SDOH and amputation has not been well described.

Objective

To evaluate whether individual components of SDOH and race are associated with amputation rates in the most populous counties of the US.

Design, Setting, and Participants

In this population-based cross-sectional study of the 100 most populous US counties, hospital discharge rates for lower extremity amputation in 2017 were assessed using the Healthcare Cost and Utilization Project State Inpatient Database. Those data were matched with publicly available demographic, hospital, and SDOH data. Data were analyzed July 3, 2022, to March 5, 2023.

Main outcome and Measures

Amputation rates were assessed across all counties. Counties were divided into quartiles based on amputation rates, and baseline characteristics were described. Unadjusted linear regression and multivariable regression analyses were performed to assess associations between county-level amputation and SDOH and demographic factors.

Results

Amputation discharge data were available for 76 of the 100 most populous counties in the United States. Within these counties, 15.3% were African American, 8.6% were Asian, 24.0% were Hispanic, and 49.6% were non-Hispanic White; 13.4% of patients were 65 years or older. Amputation rates varied widely, from 5.5 per 100 000 in quartile 1 to 14.5 per 100 000 in quartile 4. Residents of quartile 4 (vs 1) counties were more likely to be African American (27.0% vs 7.9%, P < .001), have diabetes (10.6% vs 7.9%, P < .001), smoke (16.5% vs 12.5%, P < .001), be unemployed (5.8% vs 4.6%, P = .01), be in poverty (15.8% vs 10.0%, P < .001), be in a single-parent household (41.9% vs 28.6%, P < .001), experience food insecurity (16.6% vs 12.9%, P = .04), or be physically inactive (23.1% vs 17.1%, P < .001). In unadjusted linear regression, higher amputation rates were associated with the prevalence of several health problems, including mental distress (β, 5.25 [95% CI, 3.66-6.85]; P < .001), diabetes (β, 1.73 [95% CI, 1.33-2.15], P < .001), and physical distress (β, 1.23 [95% CI, 0.86-1.61]; P < .001) and SDOHs, including unemployment (β, 1.16 [95% CI, 0.59-1.73]; P = .03), physical inactivity (β, 0.74 [95% CI, 0.57-0.90]; P < .001), smoking, (β, 0.69 [95% CI, 0.46-0.92]; P = .002), higher homicide rate (β, 0.61 [95% CI, 0.45-0.77]; P < .001), food insecurity (β, 0.51 [95% CI, 0.30-0.72]; P = .04), and poverty (β, 0.46 [95% CI, 0.32-0.60]; P < .001). Multivariable regression analysis found that county-level rates of physical distress (β, 0.84 [95% CI, 0.16-1.53]; P = .03), Black and White racial segregation (β, 0.12 [95% CI, 0.06-0.17]; P < .001), and population percentage of African American race (β, 0.06 [95% CI, 0.00-0.12]; P = .03) were associated with amputation rate.

Conclusions and Relevance

Social determinants of health provide a framework by which the associations of environmental factors with amputation rates can be quantified and potentially used to guide interventions at the local level.

Prioritizing determinants of cognitive function in healthy middle-aged and older adults: insights from a machine learning regression approach in the Canadian longitudinal study on aging

Introduction

The preservation of healthy cognitive function is a crucial step toward reducing the growing burden of cognitive decline and impairment. Our study aims to identify the characteristics of an individual that play the greatest roles in determining healthy cognitive function in mid to late life.

Methods

Data on the characteristics of an individual that influence their health, also known as determinants of health, were extracted from the baseline cohort of the Canadian Longitudinal Study of Aging (2015). Cognitive function was a normalized latent construct score summarizing eight cognitive tests administered as a neuropsychological battery by CLSA staff. A higher cognitive function score indicated better functioning. A penalized regression model was used to select and order determinants based on their strength of association with cognitive function. Forty determinants (40) were entered into the model including demographic and socioeconomic factors, lifestyle and health behaviors, clinical measures, chronic diseases, mental health status, social support and the living environment.

Results

The study sample consisted mainly of White, married, men and women aged 45-64 years residing in urban Canada. Mean overall cognitive function score for the study sample was 99.5, with scores ranging from 36.6 to 169.2 (lowest to highest cognitive function). Thirty-five (35) determinants were retained in the final model as significantly associated with healthy cognitive functioning. The determinants demonstrating the strongest associations with healthy cognitive function, were race, immigrant status, nutritional risk, community belongingness, and satisfaction with life. The determinants demonstrating the weakest associations with healthy cognitive function, were physical activity, greenness and neighborhood deprivation.

Conclusion

Greater prioritization and integration of demographic and socioeconomic factors and lifestyle and health behaviors, such greater access to healthy foods and enhancing aid programs for low-income and immigrant families, into future health interventions and policies can produce the greatest gains in preserving healthy cognitive function in mid to late life.

Perceived Quality of Life Is Related to a Healthy Lifestyle and Related Outcomes in Spanish Children and Adolescents: The Physical Activity, Sedentarism, and Obesity in Spanish Study

BACKGROUND

Maintaining a healthy lifestyle is crucial for safeguarding the well-being and quality of life perception, appropriate growth, and development of children and adolescents, while also mitigating the risk of future adult-onset diseases.

OBJECTIVE

To assess associations between perceived quality of life and healthy lifestyle and related outcomes in Spanish children and adolescents.

METHODS

Cross-sectional analysis of 8-16-year-old children and adolescents (n = 3534) were included in the nationwide study of Physical Activity, Sedentarism, and Obesity in Spanish Youth (PASOS). Data were collected through (1) questionnaires on health-related quality of life (HRQoL), healthy lifestyle outcomes (dietary intake, physical fitness, sleep, and screen time), and (2) anthropometric measurements for weight status assessment. Data were analysed by logistic regression, using the health-related quality of life (HRQoL) as the grouping variable.

RESULTS

Participants with a lower HRQoL were those with a lower adherence to the MedDiet and lower achievement of the recommended daily intake of fruit and vegetables. They were also less likely to follow the recommendations for screen time and sleep (with the exception of the weekend) compared to participants with a higher HRQoL. Participants with a lower HRQoL showed a lower healthy weight status and poorer physical fitness than those with a higher HRQoL.

CONCLUSIONS

Healthy eating habits, healthy weight status (normal weight), appropriate sleep time, physical fitness, and limited screen time play a crucial role in the perceived quality of life in children and adolescents.

Understanding the association between county-level unemployment and health stratified by education and income in the southwestern United States

Past research on the relationship between unemployment rates and population health has produced mixed findings. The relationship can be influenced by the kinds of health outcomes observed, time frame, level of geographic aggregation, and other factors. Given these mixed findings, there is a need to add to our knowledge about how unemployment rates and population health are related. There is limited research that examines the association of unemployment rates with both physical and mental health, while simultaneously stratifying populations by income and education levels. Using survey-based self-reported data, this first population-based study examined the association between unemployment rates and physically and mentally unhealthy days in the southwestern United States, by county-level stratification of income (high and low) as well as education (high and low), from 2015 to 2019. After controlling for covariates, associations were modelled using negative binomial regression, with autocorrelative residuals, and were reported as rate ratios (RR). Overall, we found that a 1% rise in unemployment rates was significantly associated with an increase in physically unhealthy days [adjusted RR 1.007; 95% CI, 1.004-1.011, P < 0.001] and mentally unhealthy days [RR 1.006; 95% CI, 1.003-1.009, P  < 0.001]. Upon stratification, a significant risk was found among the high education and high income category [RR 1.035; 95% CI, 1.021-1.049, P  < 0.001], as well as for the high education and low income category [RR 1.026; 95% CI, 1.013-1.040, P  < 0.001]. A better understanding of how unemployment is associated with the health of communities with different education and income levels could help reduce the burden on society through tailored interventions and social policies not only in the United States, but also in other developed nations.

Maria Theresia und ihre Kinder

Maria Theresia, Regentin von Österreich, Böhmen und Ungarn und ab 1745 auch Kaiserin des Heiligen Römischen Reichs, hatte kein leichtes Amt. Als sie 1780 im Alter von 63 Jahren starb, hatte sie ihren Herrschaftsraum – aus einer feudalen Staatsordnung kommend – im Sinne eines „aufgeklärten Absolutismus“ umgestaltet. Ihr Mann, Kaiser Franz, war bereits 1765 im Alter von 57 Jahren gestorben und ihr gemeinsamer Sohn Joseph II. war als Kaiser und Mitregent an dessen Stelle getreten – er starb 1790 mit 48 Jahren. Von ihren 16 Kindern erreichten nur zwei mit jeweils 65 Jahren ein höheres Lebensalter als sie selbst. Ein Kind war im ersten Lebensjahr gestorben, fünf weitere Kinder vor ihrem 18. Geburtstag 1. Für Aufsehen sorgte auch der frühe Tod ihrer 38jährigen Tochter Marie Antoinette 1793 durch eine Guillotine der Französischen Revolution: Diese hatte dem dortigen, unaufgeklärt gebliebenen Absolutismus ein Ende gesetzt. „Media vita in morte sumus – Mitten im Leben sind wir im Tod“, singt ein mittelalterlicher gregorianischer Choral.

Empirical evidence supporting the inclusion of multi-axes segregation in assessing US county health

To facilitate community action toward health equity, the County Health Rankings & Roadmaps program (CHR&R) assigns health rankings to US counties. The CHR&R conceptual model considers White-Black and White-non-White dissimilarity values to represent residential segregation as part of the family and social support subcomponent. As the US is greying and becoming more multi-racial-ethnic, the two-group White-centered segregation measures are inadequate to capture segregation among population subgroups in the US. Thus, we evaluate the relevancy of segregation measures that consider multiple racial, ethnic, and age groups in assessing US county health. Besides using the two-group dissimilarity index to measure White-centered racial segregation as conceptualized by CHR&R, the study also uses the multi-group generalized dissimilarity index to measure racial-ethnic-age segregation by counties, employing both aspatial and spatial versions of these measures. These indices are computed for counties using the 2015-2019 American Community Survey data at the census tract level. Descriptive statistics and regressions controlling for sociodemographic factors and healthcare access are used to assess the contributions of individual segregation measures to mortality (life expectancy, years of potential life lost and premature mortality) and morbidity (frequent mental distress, frequent physical distress, and low birth weight) indicators representing county health. Overall, correlations between these indicators and most segregation measures are significant but weak. Regression results show that many segregation measures are not significantly related to mortality indicators, but most are significantly associated with morbidity indicators, with the magnitudes of these associations higher for the multi-group racial-ethnic-age segregation index and its spatial version. Results provide evidence that racial-ethnic-age segregation is associated with county-level morbidity and that spatial measures capturing segregation of multiple population axes should be considered for ranking county health.

Features associated with travel distance for radical cystectomy in Florida: Implications for access to care

INTRODUCTION

Characteristics associated with travel distance for radical cystectomy (RC) remain incompletely defined but are needed to inform efforts to bridge gaps in care. Therefore, we assessed features associated with travel distance for RC in a statewide dataset.

METHODS

We identified RC patients in the Florida Inpatient Discharge dataset from 2013 to 2019. Travel distance was estimated using zip code centroids. The primary outcome was travel >50 miles for RC. Secondary outcomes included inpatient mortality, nonhome discharge, and inpatient complications. U.S. County Health Rankings were included as model covariates. Mixed effects logistic regression models accounting for clustering within hospitals were utilized.

RESULTS

We identified 4,209 patients, of whom 2,284 (54%) traveled <25 miles, 654 (16%) traveled 25 to 50 miles, and 1271 (30%) traveled >50 miles. Patients who traveled >50 miles primarily lived in central and southwest Florida. Following multivariable adjustment, patients traveling >50 miles were less likely to be Hispanic/Latino (odds ratio [OR] 0.35, 95% CI: 0.23-0.51), and more likely to reside in a county with the lowest health behavior (OR 6.48, 95% CI: 3.81-11.2) and lowest socioeconomic (OR 7.63, 95% CI: 5.30-11.1) rankings compared to those traveling <25 miles (all P < 0.01). Travel distance >50 miles was associated with treatment at a high-volume center and significantly lower risks of inpatient mortality, nonhome discharge, and postoperative complications (all P < 0.02).

CONCLUSION

These data identify characteristics of patients and communities in the state of Florida with potentially impaired access to RC care and can be used to guide outreach efforts designed to improve access to care.

Relationships between COVID-19 healthcare outcomes and county characteristics in the U.S. for Delta (B.1.617.2) and Omicron (B.1.1.529 and BA.1.1) variants

Background

COVID-19 is constantly evolving, and highly populated communities consist of many different characteristics that may contribute to COVID-19 health outcomes. Therefore, we aimed to (1) quantify the relationships between county characteristics and severe and non-severe county-level health outcomes related to COVID-19. We also aimed to (2) compare these relationships across time periods where the Delta (B.1.617.2) and Omicron (B.1.1.529 and BA.1.1) variants were dominant in the U.S.

Methods

We used multiple regression to measure the strength of relationships between healthcare outcomes and county characteristics in the 50 most populous U.S. counties.

Results

We found many different significant predictors including the proportion of a population vaccinated, median household income, population density, and the proportion of residents aged 65+, but mainly found that socioeconomic factors and the proportion of a population vaccinated play a large role in the dynamics of the spread and severity of COVID-19 in communities with high populations.

Discussion

The present study shines light on the associations between public health outcomes and county characteristics and how these relationships change throughout Delta and Omicron's dominance. It is important to understand factors underlying COVID-19 health outcomes to prepare for future health crises.

Comparison of SIREN social needs screening tools and Simplified Omaha System Terms: informing an informatics approach to social determinants of health assessments

OBJECTIVE

Numerous studies indicate that the social determinants of health (SDOH), conditions in which people work, play, and learn, account for 30%-55% of health outcomes. Many healthcare and social service organizations seek ways to collect, integrate, and address the SDOH. Informatics solutions such as standardized nursing terminologies may facilitate such goals. In this study, we compared one standardized nursing terminology, the Omaha System, in its consumer-facing form, Simplified Omaha System Terms (SOST), to social needs screening tools identified by the Social Interventions Research and Evaluation Network (SIREN).

MATERIALS AND METHODS

Using standard mapping techniques, we mapped 286 items from 15 SDOH screening tools to 335 SOST challenges. The SOST assessment includes 42 concepts across 4 domains. We analyzed the mapping using descriptive statistics and data visualization techniques.

RESULTS

Of the 286 social needs screening tools items, 282 (98.7%) mapped 429 times to 102 (30.7%) of the 335 SOST challenges from 26 concepts in all domains, most frequently from Income, Home, and Abuse. No single SIREN tool assessed all SDOH items. The 4 items not mapped were related to financial abuse and perceived quality of life.

DISCUSSION

SOST taxonomically and comprehensively collects SDOH data compared to SIREN tools. This demonstrates the importance of implementing standardized terminologies to reduce ambiguity and ensure the shared meaning of data.

CONCLUSIONS

SOST could be used in clinical informatics solutions for interoperability and health information exchange, including SDOH. Further research is needed to examine consumer perspectives regarding SOST assessment compared to other social needs screening tools.

Filling in the white space: Spatial interpolation with Gaussian processes and social media data

Full national coverage below the state level is difficult to attain through survey-based data collection. Even the largest survey-based data collections, such as the CDC's Behavioral Risk Factor Surveillance System or the Gallup-Healthways Well-being Index (both with more than 300,000 responses p.a.) only allow for the estimation of annual averages for about 260 out of roughly U.S. 3,000 counties when a threshold of 300 responses per county is used. Using a relatively high threshold of 300 responses gives substantially higher convergent validity-higher correlations with health variables-than lower thresholds but covers a reduced and biased sample of the population. We present principled methods to interpolate spatial estimates and show that including large-scale geotagged social media data can increase interpolation accuracy. In this work, we focus on Gallup-reported life satisfaction, a widely-used measure of subjective well-being. We use Gaussian Processes (GP), a formal Bayesian model, to interpolate life satisfaction, which we optimally combine with estimates from low-count data. We interpolate over several spaces (geographic and socioeconomic) and extend these evaluations to the space created by variables encoding language frequencies of approximately 6 million geotagged Twitter users. We find that Twitter language use can serve as a rough aggregate measure of socioeconomic and cultural similarity, and improves upon estimates derived from a wide variety of socioeconomic, demographic, and geographic similarity measures. We show that applying Gaussian Processes to the limited Gallup data allows us to generate estimates for a much larger number of counties while maintaining the same level of convergent validity with external criteria (i.e., N = 1,133 vs. 2,954 counties). This work suggests that spatial coverage of psychological variables can be reliably extended through Bayesian techniques while maintaining out-of-sample prediction accuracy and that Twitter language adds important information about cultural similarity over and above traditional socio-demographic and geographic similarity measures. Finally, to facilitate the adoption of these methods, we have also open-sourced an online tool that researchers can freely use to interpolate their data across geographies.

Unmet Health Care Needs of the Older Population in European Countries Based on Indicators Available in the Eurostat Database

Access to healthcare may affect the health of the population, especially older people. The aim of this study is to analyze the reasons and factors influencing the unmet healthcare needs (UHCN) of the older population in the context of differences between age groups for 28 European countries. A self-reported UHCN indicator obtained from Eurostat database was used. The share of people with healthcare needs reporting distance/transportation issues was significantly different in the younger and older groups, as well as in age groups within the older population. The differences in other reasons were not so considerable. Problems with UHCN were observed more often in the older population with lower rather than with higher income and with more severe activity limitations rather than with none/moderate limitations (differences statistically significant, except for income for 75+). In most countries, the UHCN dependence on income/activity limitation is higher in the age group of 15-64 than for the older population. To plan/introduce/monitor appropriate, tailored actions for improving healthcare access for the older population, a detailed analysis of the UHCN prevalence, reasons, and determinants in this age group is needed; it is insufficient to analyze only the population as a whole. Additionally, the group of older people is not homogeneous in terms of UHCN.

An interpretable machine learning model of cross-sectional U.S. county-level obesity prevalence using explainable artificial intelligence

BACKGROUND

There is considerable geographic heterogeneity in obesity prevalence across counties in the United States. Machine learning algorithms accurately predict geographic variation in obesity prevalence, but the models are often uninterpretable and viewed as a black-box.

OBJECTIVE

The goal of this study is to extract knowledge from machine learning models for county-level variation in obesity prevalence.

METHODS

This study shows the application of explainable artificial intelligence methods to machine learning models of cross-sectional obesity prevalence data collected from 3,142 counties in the United States. County-level features from 7 broad categories: health outcomes, health behaviors, clinical care, social and economic factors, physical environment, demographics, and severe housing conditions. Explainable methods applied to random forest prediction models include feature importance, accumulated local effects, global surrogate decision tree, and local interpretable model-agnostic explanations.

RESULTS

The results show that machine learning models explained 79% of the variance in obesity prevalence, with physical inactivity, diabetes, and smoking prevalence being the most important factors in predicting obesity prevalence.

CONCLUSIONS

Interpretable machine learning models of health behaviors and outcomes provide substantial insight into obesity prevalence variation across counties in the United States.

Estimating the US Baseline Distribution of Health Inequalities Across Race, Ethnicity, and Geography for Equity-Informative Cost-Effectiveness Analysis

OBJECTIVES

Information on how life expectancy, disability-free life expectancy, and quality-adjusted life expectancy varies across equity-relevant subgroups is required to conduct distributional cost-effectiveness analysis. These summary measures are not comprehensively available in the United States, given limitations in nationally representative data across racial and ethnic groups.

METHODS

Through linkage of US national survey data sets and use of Bayesian models to address missing and suppressed mortality data, we estimate health outcomes across 5 racial and ethnic subgroups (non-Hispanic American Indian or Alaska Native, non-Hispanic Asian and Pacific Islander, non-Hispanic black, non-Hispanic white, and Hispanic). Mortality, disability, and social determinant of health data were combined to estimate sex- and age-based outcomes for equity-relevant subgroups based on race and ethnicity, as well as county-level social vulnerability.

RESULTS

Life expectancy, disability-free life expectancy, and quality-adjusted life expectancy at birth declined from 79.5, 69.4, and 64.3 years, respectively, among the 20% least socially vulnerable (best-off) counties to 76.8, 63.6, and 61.1 years, respectively, among the 20% most socially vulnerable (worst-off) counties. Considering differences across racial and ethnic subgroups, as well as geography, gaps between the best-off (Asian and Pacific Islander; 20% least socially vulnerable counties) and worst-off (American Indian/Alaska Native; 20% most socially vulnerable counties) subgroups were large (17.6 life-years, 20.9 disability-free life-years, and 18.0 quality-adjusted life-years) and increased with age.

CONCLUSIONS

Existing disparities in health across geographies and racial and ethnic subgroups may lead to distributional differences in the impact of health interventions. Data from this study support routine estimation of equity effects in healthcare decision making, including distributional cost-effectiveness analysis.

Composite Indices of Social Determinants of Health: Overview, Measurement Gaps, and Research Priorities for Health Equity

The goal of health equity is for all people to have opportunities and resources for optimal health outcomes regardless of their social identities, residence in marginalized communities, and/or experience with oppressive systems. Social determinants of health (SDOH)-the conditions in which we are born, grow, live, work, and age-are inextricably tied to health equity. Advancing health equity thus requires reliable measures of SDOH. In the United States, comprehensive individual-level data on SDOH are difficult to collect, may be inaccurate, and do not capture all dimensions of inequitable outcomes. Individual area-based indicators are widely available, but difficult to use in practice. Numerous area-level composite indices are available to describe SDOH, but there is no consensus on which indices are most appropriate to use. This article presents an analytic taxonomy of currently available SDOH composite indices and compares their components and predictive ability, providing insights into gaps and areas for further research.

Current evidence for designing self-management support for underserved populations: an integrative review using the example of diabetes

AIMS

With numerous and continuing attempts at adapting diabetes self-management support programmes to better account for underserved populations, its important that the lessons being learned are understood and shared. The work we present here reviews the latest evidence and best practice in designing and embedding culturally and socially sensitive, self-management support programmes.

METHODS

We explored the literature with regard to four key design considerations of diabetes self-management support programmes: Composition - the design and content of written materials and digital tools and interfaces; Structure - the combination of individual and group sessions, their frequency, and the overall duration of programmes; Facilitators - the combination of individuals used to deliver the programme; and Context - the influence and mitigation of a range of individual, socio-cultural, and environmental factors.

RESULTS

We found useful and recent examples of design innovation within a variety of countries and models of health care delivery including Brazil, Mexico, Netherlands, Spain, United Kingdom, and United States of America. Within Composition we confirmed the importance of retaining best practice in creating readily understood written information and intuitive digital interfaces; Structure the need to offer group, individual, and remote learning options in programmes of flexible duration and frequency; Facilitators where the benefits of using culturally concordant peers and community-based providers were described; and finally in Context the need to integrate self-management support programmes within existing health systems, and tailor their various constituent elements according to the language, resources, and beliefs of individuals and their communities.

CONCLUSIONS

A number of design principles across the four design considerations were identified that together offer a promising means of creating the next generation of self-management support programme more readily accessible for underserved communities. Ultimately, we recommend that the precise configuration should be co-produced by all relevant service and patient stakeholders and its delivery embedded in local health systems.

Twenty-five years of the Medicare Rural Hospital Flexibility Program: The past as prologue

PURPOSE

The Medicare Rural Hospital Flexibility (Flex) Program and the Critical Access Hospital (CAH) provider type are now 25 years old. Since the inception of the program, the needs of CAHs have evolved greatly. This article describes the history of the limited-service hospital model that led to the creation of CAHs, the evolution and impact of the Flex Program on CAHs, and the trends likely to impact CAHs and rural healthcare in the future. It concludes with recommendations to address these future needs.

METHODS

This review of the 25-year history of the Flex Program and CAHs is based on a detailed analysis of the literature on the limited-service hospital model and CAHs, the evaluation reports of the Flex Tracking and Flex Monitoring Teams, and the author's 25-year history with the program.

FINDINGS

The Flex Program has made important contributions to the viability of rural hospitals through the conversion of 1,360 CAHs. The program has encouraged attention on CAH quality of care and the role of CAHs in addressing the population health needs of their communities. It has further encouraged the development of a robust rural health policy and advocacy infrastructure that has heightened attention on the needs of rural providers and communities.

CONCLUSIONS

The needs of CAHs and rural delivery systems have evolved greatly since the implementation of the Flex Program. The 25th anniversary of the program is an ideal time to re-evaluate and update the program to support CAHs in adapting to the fast-changing healthcare environment.

Food Is Medicine Movement-Key Actions Inside and Outside the Government

This JAMA Forum discusses the key food is medicine (FIM) actions being taken by the federal government and individual state governments and key nongovernmental actions that are advancing FIM.

The impact of screening for social risks on OBGYN patients and providers: A systematic review of current evidence and key gaps

BACKGROUND

Increasingly, policymakers and professional organizations support screening for social assets and risks during clinical care. Scant evidence exists on how screening impacts patients, providers, or health systems.

OBJECTIVE

To systematically review published literature for evidence of the clinical utility of screening for social determinants of health in clinical obstetric and gynecologic (OBGYN) care.

SEARCH STRATEGY

We systematically searched Pubmed (March 2022, 5,302 identified) and identified additional articles using hand sorting (searching articles citing key articles (273 identified) and through bibliography review (20 identified)).

SELECTION CRITERIA

We included all articles that measured a quantitative outcome of systematic social determinants of health (SDOH) screening in an OBGYN clinical setting. Each identified citation was reviewed by two independent reviewers at both the title/abstract and full text stages.

DATA COLLECTION AND ANALYSIS

We identified 19 articles for inclusion and present the results using narrative synthesis.

MAIN RESULTS

The majority of articles reported on SDOH screening during prenatal care (16/19) and the most common SDOH was intimate partner violence (13/19 studies). Overall, patients had favorable attitudes towards SDOH screening (in 8/9 articles measuring attitudes), and referrals were common following positive screening (range 5.3%-63.6%). Only two articles presented data on the effects of SDOH screening on clinicians and none on health systems. Three articles present data on resolution of social needs, with inconsistent results.

CONCLUSIONS

Limited evidence exists on the benefits of SDOH screening in OBGYN clinical settings. Innovative studies leveraging existing data collection are needed to expand and improve SDOH screening.

Association of Place With Adolescent Obesity

Importance

Despite strong evidence linking place and obesity risk, the extent to which this link is causal or reflects sorting into places is unclear.

Objective

To examine the association of place with adolescents' obesity and explore potential causal pathways, such as shared environments and social contagion.

Design, Setting, and Participants

This natural experiment study used the periodic reassignment of US military servicemembers to installations as a source of exogenous variation in exposure to difference places to estimate the association between place and obesity risk. The study analyzed data from the Military Teenagers Environments, Exercise, and Nutrition Study, a cohort of adolescents in military families recruited from 2013 through 2014 from 12 large military installations in the US and followed up until 2018. Individual fixed-effects models were estimated that examined whether adolescents' exposure to increasingly obesogenic places over time was associated with increases in body mass index (BMI) and probability of overweight or obesity. These data were analyzed from October 15, 2021, through March 10, 2023.

Exposure

Adult obesity rate in military parent's assigned installation county was used as a summary measure of all place-specific obesogenic influences.

Main Outcomes and Measures

Outcomes were BMI, overweight or obesity (BMI in the 85th percentile or higher), and obesity (BMI in the 95th percentile or higher). Time at installation residence and off installation residence were moderators capturing the degree of exposure to the county. County-level measures of food access, physical activity opportunities, and socioeconomic characteristics captured shared environments.

Results

A cohort of 970 adolescents had a baseline mean age of 13.7 years and 512 were male (52.8%). A 5 percentage point-increase over time in the county obesity rate was associated with a 0.19 increase in adolescents' BMI (95% CI, 0.02-0.37) and a 0.02-unit increase in their probability of obesity (95% CI, 0-0.04). Shared environments did not explain these associations. These associations were stronger for adolescents with time at installation of 2 years or longer vs less than 2 years for BMI (0.359 vs. 0.046; P value for difference in association = .02) and for probability of overweight or obesity (0.058 vs. 0.007; P value for difference association = .02), and for adolescents who lived off installation vs on installation for BMI (0.414 vs. -0.025; P value for association = .01) and for probability of obesity (0.033 vs. -0.007; P value for association = .02).

Conclusion and Relevance

In this study, the link between place and adolescents' obesity risk is not explained by selection or shared environments. The study findings suggest social contagion as a potential causal pathway.

Extracting social determinants of health from clinical note text with classification and sequence-to-sequence approaches

OBJECTIVE

Social determinants of health (SDOH) are nonmedical factors that can influence health outcomes. This paper seeks to extract SDOH from clinical texts in the context of the National NLP Clinical Challenges (n2c2) 2022 Track 2 Task.

MATERIALS AND METHODS

Annotated and unannotated data from the Medical Information Mart for Intensive Care III (MIMIC-III) corpus, the Social History Annotation Corpus, and an in-house corpus were used to develop 2 deep learning models that used classification and sequence-to-sequence (seq2seq) approaches.

RESULTS

The seq2seq approach had the highest overall F1 scores in the challenge's 3 subtasks: 0.901 on the extraction subtask, 0.774 on the generalizability subtask, and 0.889 on the learning transfer subtask.

DISCUSSION

Both approaches rely on SDOH event representations that were designed to be compatible with transformer-based pretrained models, with the seq2seq representation supporting an arbitrary number of overlapping and sentence-spanning events. Models with adequate performance could be produced quickly, and the remaining mismatch between representation and task requirements was then addressed in postprocessing. The classification approach used rules to generate entity relationships from its sequence of token labels, while the seq2seq approach used constrained decoding and a constraint solver to recover entity text spans from its sequence of potentially ambiguous tokens.

CONCLUSION

We proposed 2 different approaches to extract SDOH from clinical texts with high accuracy. However, accuracy suffers on text from new healthcare institutions not present in the training data, and thus generalization remains an important topic for future study.

Racial and ethnic disparities in psychosocial evaluation and liver transplant waitlisting

Health disparities have been well-described in all stages of the liver transplantation (LT) process. Using data from psychosocial evaluations and the Stanford Integrated Psychosocial Assessment, our objective was to investigate potential racial and ethnic inequities in overall LT waitlisting and not waitlisting for medical or psychosocial reasons. In a cohort of 2271 candidates evaluated for LT from 2014 to 2021 and with 1-8 years of follow-up, no significant associations were noted between race/ethnicity and overall waitlisting and not waitlisting for medical reasons. However, compared with White race, Black race (odds ratio [OR], 1.65; 95% confidence interval [CI], 1.07-2.56) and Hispanic/Latinx ethnicity (OR, 2.10; 95% CI, 1.16-3.78) were associated with not waitlisting for psychosocial reasons. After adjusting for sociodemographic variables, the relationship persisted in both populations: Black (OR, 1.95; 95% CI, 1.12-3.38) and Hispanic/Latinx (OR, 2.29; 95% CI, 1.08-4.86) (reference group, White). High-risk Stanford Integrated Psychosocial Assessment scores were more prevalent in Black and Hispanic/Latinx patients, likely reflecting upstream factors and structural racism. Health systems and LT centers should design programs to combat these disparities and improve equity in access to LT.

A systems-level model of direct and indirect links between environmental health, socioeconomic factors, and human mortality

Major efforts are being made to better understand how human health and ecosystem health are influenced by climate and other environmental factors. However, studies that simultaneously address human and ecosystem health within a systems-level framework that accounts for both direct and indirect effects are rare. Using path analysis and a large database of environmental and socioeconomic variables, we create a systems-level model of direct and indirect effects on human and ecosystem health in counties throughout the conterminous United States. As indicators of human and ecosystem health, we use age-adjusted mortality rate and an index of biological integrity in streams and rivers, respectively. We show that: (i) geology and climate set boundary conditions for all other variables in the model; (ii) hydrology and land cover have predictable but distinct effects on human and ecosystem health; and (iii) forest cover is a key link between the environment and the socioeconomic variables that directly influence human health.

Longitudinal trends (2011-2020) of premature mortality and years of potential life loss (YPLL) and associated covariates of the 62 New York State counties

BACKGROUND

New York State (NYS) is the 27^th largest state and the 4^th most populous state in the U.S., with close to 20 million people in 62 counties. Territories with diverse populations present the best opportunity to study health outcomes and associated covariates, and how these differ across different populations and groups. The County Health Ranking and Roadmaps (CHR&R) ranks counties by linking the population's characteristics and health outcomes and contextual factors in a synchronic approach.

METHODS

The goal of this study is to analyze the longitudinal trends in NYS counties of age-adjusted premature mortality rate and years of potential life loss rate (YPLL) from 2011-2020 using (CHR&R) data to identify similarities and trends among the counties of the state. This study used a weighted mixed regression model to analyze the longitudinal trend in health outcomes as a function of the time-varying covariates and clustered the 62 counties according to the trend over time in the covariates.

RESULTS

Four clusters of counties were identified. Cluster 1, which represents 33 of the 62 counties in NYS, contains the most rural counties and the least racially and ethnically diverse counties. Clusters 2 and 3 mirror each other in most covariates and Cluster 4 is comprised of 3 counties (Bronx, Kings/Brooklyn, Queens) representing the most urban and racial and ethnic diverse counties in the state.

CONCLUSION

The analysis clustered counties according to the longitudinal trends of the covariates, and by doing so identified clusters of counties that shared similar trends among the covariates, to later examine trends in the health outcomes through a regression model. The strength of this approach lies in the predictive feature of what is to come for the counties by understanding the covariates and setting prevention goals.

Predictive Modeling to Identify Children With Complex Health Needs At Risk for Hospitalization

BACKGROUND

Identifying children at high risk with complex health needs (CCHN) who have intersecting medical and social needs is challenging. This study's objectives were to (1) develop and evaluate an electronic health record (EHR)-based clinical predictive model ("model") for identifying high-risk CCHN and (2) compare the model's performance as a clinical decision support (CDS) to other CDS tools available for identifying high-risk CCHN.

METHODS

This retrospective cohort study included children aged 0 to 20 years with established care within a single health system. The model development/validation cohort included 33 months (January 1, 2016-September 30, 2018) and the testing cohort included 18 months (October 1, 2018-March 31, 2020) of EHR data. Machine learning methods generated a model that predicted probability (0%-100%) for hospitalization within 6 months. Model performance measures included sensitivity, positive predictive value, area under receiver-operator curve, and area under precision-recall curve. Three CDS rules for identifying high-risk CCHN were compared: (1) hospitalization probability ≥10% (model-predicted); (2) complex chronic disease classification (using Pediatric Medical Complexity Algorithm [PMCA]); and (3) previous high hospital utilization.

RESULTS

Model development and testing cohorts included 116 799 and 27 087 patients, respectively. The model demonstrated area under receiver-operator curve = 0.79 and area under precision-recall curve = 0.13. PMCA had the highest sensitivity (52.4%) and classified the most children as high risk (17.3%). Positive predictive value of the model-based CDS rule (19%) was higher than CDS based on the PMCA (1.9%) and previous hospital utilization (15%).

CONCLUSIONS

A novel EHR-based predictive model was developed and validated as a population-level CDS tool for identifying CCHN at high risk for future hospitalization.

Using the Community Readiness Model and Stakeholder Engagement to Assess a Health System's Readiness to Provide LGBTQ+ Healthcare: A Pilot Study

Background

Despite broad social and policy changes over the past several decades, many LGBTQ+ people face barriers to healthcare and report mistreatment and disrespect in healthcare settings. Few health systems level interventions have been shown to improve sexuality- and gender-related health disparities. Using the Community Readiness Model, we developed and implemented a rigorous assessment and priority-setting intervention at one mid-sized health system in the midwestern US. We evaluated the system's readiness to provide LGBTQ+ healthcare and developed immediate action steps that are responsive to local context. We engaged diverse stakeholder groups throughout the process.

Methods

Led by the Community Readiness Model, we identified key groups within the health system and conductedstructured interviews with 4-6 key informants from each group. Two trained scorers external to the study team individually scored each interview on a numerical scale ranging from 1 (no awareness of the problem) to 9 (community ownership of the problem) and discussed and reconciled scores. Group scores were averaged for each dimension of readiness and overall readiness, and then triangulated with stakeholders to ensure they reflected lived experiences. Finally, specific recommendations were generated to match the needs of the system and move them towards higher levels of readiness.

Results

We convened an advisory committee of LGBTQ+ patients of the health system and a panel of local experts on LGBTQ+ wellness. Both groups contributed significantly to research processes. 28 interviews across 6 staff subcommunities indicated readiness levels ranging from "3: Vague Awareness" of the issue, and the "4: Preplanning" stage. Discrepancies across staff groups and dimensions of readiness suggested areas of focus for the health system. The evaluation process led to immediately actionable recommendations for the health system.

Conclusions

This pilot study demonstrates the potential impact of the Community Readiness Model on improving health systems' readiness to provide LGBTQ+ healthcare. This model combines strengths from community-based research and implementation science approaches to form an intervention that can be widely disseminated and maintain the flexibility and agility to meet local needs. Future research will evaluate changes in readiness at the same health system and test the process in additional health systems.

The Association Between Social Determinants of Health and Population Health Outcomes: Ecological Analysis

BACKGROUND

With the increased availability of data, a growing number of studies have been conducted to address the impact of social determinants of health (SDOH) factors on population health outcomes. However, such an impact is either examined at the county level or the state level in the United States. The results of analysis at lower administrative levels would be useful for local policy makers to make informed health policy decisions.

OBJECTIVE

This study aimed to investigate the ecological association between SDOH factors and population health outcomes at the census tract level and the city level. The findings of this study can be applied to support local policy makers in efforts to improve population health, enhance the quality of care, and reduce health inequity.

METHODS

This ecological analysis was conducted based on 29,126 census tracts in 499 cities across all 50 states in the United States. These cities were grouped into 5 categories based on their population density and political affiliation. Feature selection was applied to reduce the number of SDOH variables from 148 to 9. A linear mixed-effects model was then applied to account for the fixed effect and random effects of SDOH variables at both the census tract level and the city level.

RESULTS

The finding reveals that all 9 selected SDOH variables had a statistically significant impact on population health outcomes for ≥2 city groups classified by population density and political affiliation; however, the magnitude of the impact varied among the different groups. The results also show that 4 SDOH risk factors, namely, asthma, kidney disease, smoking, and food stamps, significantly affect population health outcomes in all groups (P<.01 or P<.001). The group differences in health outcomes for the 4 factors were further assessed using a predictive margin analysis.

CONCLUSIONS

The analysis reveals that population density and political affiliation are effective delineations for separating how the SDOH affects health outcomes. In addition, different SDOH risk factors have varied effects on health outcomes among different city groups but similar effects within city groups. Our study has 2 policy implications. First, cities in different groups should prioritize different resources for SDOH risk mitigation to maximize health outcomes. Second, cities in the same group can share knowledge and enable more effective SDOH-enabled policy transfers for population health.

Your neighborhood matters: an ecological social determinant study of the relationship between residential racial segregation and the risk of firearm fatalities

BACKGROUND

Firearm fatalities are a major public health concern, claiming the lives of 40,000 Americans each year. While firearm fatalities have pervasive effects, it is unclear how social determinants of health (SDOH) such as residential racial segregation, income inequality, and community resilience impact firearm fatalities. This study investigates the relationships between these SDOH and the likelihood of firearm fatalities.

METHODS

County-level SDOH data from the Agency for Health Care Research and Quality for 2019 were analyzed, covering 72 Wisconsin counties. The dependent variable was the number of firearm fatalities in each county, used as a continuous variable. The independent variable was residential racial segregation (Dissimilarity Index), defined as the degree to which non-White and White residents were distributed across counties, ranging from 0 (complete integration) to 100 (complete segregation), and higher values indicate greater residential segregation (categorized as low, moderate, and high). Covariates were income inequality ranging from zero (perfect equality) to one (perfect inequality) categorized as low, moderate, and high, community resilience risk factors (low, moderate, and high risks), and rural-urban classifications. Descriptive/summary statistics, unadjusted and adjusted negative binomial regression adjusting for population weight, were performed using STATA/MPv.17.0; P-values ≤ 0.05 were considered statistically significant. ArcMap was used for Geographic Information System analysis.

RESULTS

In 2019, there were 802 firearm fatalities. The adjusted model demonstrates that the risk of firearm fatalities was higher in areas with high residential racial segregation compared to low-segregated areas (IRR.:1.26, 95% CI:1.04-1.52) and higher in areas with high-income inequality compared to areas with low-income inequality (IRR.:1.18, 95% CI:1.00-1.40). Compared to areas with low-risk community resilience, the risk of firearm fatalities was higher in areas with moderate (IRR.:0.61, 95% CI:0.48-0.78), and in areas with high risk (IRR.:0.53, 95% CI:0.41-0.68). GIS analysis demonstrated that areas with high racial segregation also have high rates of firearm fatalities.

CONCLUSION

Areas with high residential racial segregation have a high rate of firearm fatalities. With high income inequality and low community resilience, the likelihood of firearm fatalities increases.

Measuring disadvantage: A systematic comparison of United States small-area disadvantage indices

Extensive evidence demonstrates the effects of area-based disadvantage on a variety of life outcomes, such as increased mortality and low economic mobility. Despite these well-established patterns, disadvantage, often measured using composite indices, is inconsistently operationalized across studies. To address this issue, we systematically compared 5 U.S. disadvantage indices at the county-level on their relationships to 24 diverse life outcomes related to mortality, physical health, mental health, subjective well-being, and social capital from heterogeneous data sources. We further examined which domains of disadvantage are most important when creating these indices. Of the five indices examined, the Area Deprivation Index (ADI) and Child Opportunity Index 2.0 (COI) were most related to a diverse set of life outcomes, particularly physical health. Within each index, variables from the domains of education and employment were most important in relationships with life outcomes. Disadvantage indices are being used in real-world policy and resource allocation decisions; an index's generalizability across diverse life outcomes, and the domains of disadvantage which constitute the index, should be considered when guiding such decisions.

Non-pharmaceutical Interventions and COVID-19: Do County- and State-Level Policies Predict the Spread of COVID-19?

This study examines the impact of county- and state-level policies on the spread and severity of COVID-19 in communities in the USA during the first wave of the COVID-19 pandemic. We use county-level COVID-19 death and case data to examine the impact of county- and state-level mandates and non-pharmaceutical interventions (NPIs) on the spread and severity of COVID-19. Following previous work by Amuendo-Dorantes et al. (2020), we utilize a strategy that incorporates the duration of NPI implementation within a county. Specifically, we examine aggregated measures of mask mandates, daycare closures, stay-at-home orders, and restaurant and bar closures. In addition to the implementation and duration of NPI policy, we examine the role of pre-existing factors that contribute to social determinants of health in a locality. We incorporate information on the incidence of prior health conditions, socio-economic factors, and demographics including racial and ethnic composition, share of immigrant population of counties, and state governance in our estimations. To alleviate the possible endogeneity of COVID-19 outcomes and NPIs, we use instrumental variable estimation and our results show that collectively NPIs decreased the intensity of the pandemic by decreasing the total deaths and cases. Furthermore, we find the magnitude of the impact of NPIs increases the longer they are implemented. We also estimate a specification that allows for heterogeneity of NPI impact based on the racial and ethnic composition of counties. Our results suggest that NPIs have a non-uniform impact in counties with different racial and ethnic compositions.

Data resource profile: the Korean Community Health Status Indicators (K-CHSI) database

Korean Community Health Status Indicators (K-CHSI) is a model-based database containing annual data on health outcomes and determinants at the municipal level (si/gun/gu-level regions, including mid-sized cities, counties, and districts). K-CHSI's health outcomes include overall mortality, disease incidence, prevalence rates, and self-reported health. Health determinants were measured in 5 domains: socio-demographic factors, health behaviors, social environment, physical environment, and the healthcare system. The data sources are 71 public databases, including Causes of Death Statistics, Cancer Registration Statistics, Community Health Survey, Population Census, and Census on Establishments and Statistics of Urban Plans. This dataset covers Korea's 17 metropolitan cities and provinces, with data from approximately 250 municipal regions (si/gun/gu). The current version of the database (DB version 1.3) was built using 12 years of data from 2008 to 2019. All data included in K-CHSI may be downloaded via the Korea Community Health Survey site, with no login requirement (https://chs.kdca.go.kr/chs/recsRoom/dataBaseMain.do). K-CHSI covers extensive health outcomes and health determinants at the municipal level over a period of more than 10 years, which enables ecological and time-series analyses of the relationships among various health outcomes and related factors.

The Impact of Funding Inpatient Treatments for COVID-19 on Health Equity in the United States: A Distributional Cost-Effectiveness Analysis

OBJECTIVES

We conducted a distributional cost-effectiveness analysis (DCEA) to evaluate how Medicare funding of inpatient COVID-19 treatments affected health equity in the United States.

METHODS

A DCEA, based on an existing cost-effectiveness analysis model, was conducted from the perspective of a single US payer, Medicare. The US population was divided based on race and ethnicity (Hispanic, non-Hispanic black, and non-Hispanic white) and county-level social vulnerability index (5 quintile groups) into 15 equity-relevant subgroups. The baseline distribution of quality-adjusted life expectancy was estimated across the equity subgroups. Opportunity costs were estimated by converting total spend on COVID-19 inpatient treatments into health losses, expressed as quality-adjusted life-years (QALYs), using base-case assumptions of an opportunity cost threshold of $150 000 per QALY gained and an equal distribution of opportunity costs across equity-relevant subgroups.

RESULTS

More socially vulnerable populations received larger per capita health benefits due to higher COVID-19 incidence and baseline in-hospital mortality. The total direct medical cost of inpatient COVID-19 interventions in the United States in 2020 was estimated at $25.83 billion with an estimated net benefit of 735 569 QALYs after adjusting for opportunity costs. Funding inpatient COVID-19 treatment reduced the population-level burden of health inequality by 0.234%. Conclusions remained robust across scenario and sensitivity analyses.

CONCLUSIONS

To the best of our knowledge, this is the first DCEA to quantify the equity implications of funding COVID-19 treatments in the United States. Medicare funding of COVID-19 treatments in the United States could improve overall health while reducing existing health inequalities.