County-level Variables Research Articles

Abstract A104: EpiTracker Cancer: A new webtool to visualize disparities in cancer death rates across demographics and county-level characteristics

Abstract There are known disparities in cancer death rates in the U.S. by demographic, geographic and socioeconomic factors. While there are several widely available webtools to visualize U.S. cancer death rates, none allow for direct comparison of cancer death rates across county-level measures of social determinants of health. The National Cancer Institute has developed a new webtool, EpiTracker Cancer (https://episphere.github.io/epitracker), to allow for customizable visualization of cancer death rates across geography, demographics and county-level characteristics. Cancer death data and decedent race (i.e., American Indian and Alaska Native, Asian, Black, Native Hawaiian and Pacific Islander, White), ethnicity (Hispanic, non-Hispanic), sex, age and county of residence were ascertained from the National Center for Health Statistics (NCHS) death certificate data for 2018-2022. Counts and rates based on <10 deaths are suppressed per NCHS policy. Population data were abstracted from the U.S. Census. County-level characteristics were derived from multiple sources, and include metrics of socioeconomic status (e.g., median income, proportion with a college degree), health status (e.g., proportion who smoke, proportion who are obese), and healthcare access and utilization (e.g., proportion uninsured, proportion of women who have had a mammogram). EpiTracker Cancer is comprised of three main pages, each featuring a different type of visualization. The first page displays bar charts of age-standardized or crude death rates by cancer type, sex, race, ethnicity and age group. The second displays maps of cancer death rates at the county, state and national level. Users can generate national, state and county-level maps for each cause of cancer death that can be compared by race, ethnicity and sex, and can control how outlying rates are handled. The final visualization plots cancer death rates across quartiles, quintiles or deciles of county- level characteristics. The user can choose to display rate ratios comparing to either the highest or lowest quantile of the county-level metric. As an example, a user could visualize the association between county-level education, measured by the fraction of adults with at least some college education, and lung cancer mortality. Compared to counties in the highest quintile of educational attainment, those in the lowest quintile have lung cancer age-adjusted death rates that are 1.77- times (95%CI 1.72-1.82) higher among men and 1.56-times (95%CI 1.51-1.61) higher in women. Future iterations will include more county-level variables, death rates over time and additional visualizations, as well as user suggested improvements and innovations. Citation Format: Meredith Shiels, Lee Mason, Jonas Almeida, Sahar Behpour, Neal Freedman, Peter Kraft, Wayne Lawrence. EpiTracker Cancer: A new webtool to visualize disparities in cancer death rates across demographics and county-level characteristics [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr A104.

Abstract PR002: County-level food insecurity and hepatocellular carcinoma risk: A cross-sectional analysis

Abstract Background: Hepatocellular carcinoma (HCC), the most common type of primary liver cancer, is the sixth most common cancer and third leading cause of cancer deaths worldwide. Hepatitis B and/or C virus, alcohol use, and obesity are main drivers of HCC incidence and prognosis, with risk higher among Black and Hispanic individuals and those with lower socioeconomic status. HCC etiology recently shifted from viral to non-viral-related causes, with up to 40% of HCC cases unexplained by known risk factors. Food insecurity (FI) is associated with several known and emerging HCC risk factors, including poor diet quality, obesity, and diabetes. However, there is limited data on whether FI is in fact a social determinant of HCC risk. In this study, we conducted a county-level cross-sectional analysis to examine whether FI is associated with HCC risk in the United States. Methods: We used the Surveillance Epidemiology and End Results 22 database to obtain county-level counts of HCC cases from 2018-2021 stratified by age (20-64, ≥65 years), sex (female, male), and race and ethnicity (non-Hispanic White, non- Hispanic Black, non-Hispanic Asian or Pacific Islander, non-Hispanic American Indian or Alaska Native, and Hispanic). We obtained county-level FI rates for the years 2018-2021 from Feeding America’s Map the Meal Gap and calculated the average rate over these years. FI rates were calculated using imputation methods that consider state- and county-level sociodemographic variables associated with FI. We obtained additional county-level variables from the Behavior Risk Factor Surveillance System for the years 2018-2021. We used multi- level Poisson regression models with robust standard errors to estimate the association between FI and HCC risk. Models specified county population size as the offset term, and random effects for states and counties. We examined models adjusted for age, sex, race, and ethnicity (Model 1), and further adjusted for county-level binge drinking, smoking, obesity, and diabetes rates (Model 2). We tested multiplicative effect modification by individual-level factors. Results: We included 38,986 cases of HCC diagnosed in 1,085 counties. The mean county-level FI rate was 12.9% (standard deviation (SD)=3.6%). Overall, a 1-SD increase in FI was associated with a 9% increase in HCC risk in Model 1 (incidence rate ratio (IRR)=1.09, 95% CI=1.04-1.15) and a 5% increase in risk in Model 2 (IRR=1.05, 95% CI=1.02-1.09). When stratified by race and ethnicity, FI was only associated with HCC risk in non-Hispanic Whites in Model 2 (IRR=1.07, 95% CI=1.03-1.11; interaction p-value=.05). When stratified by age, FI was only associated with HCC risk in the <65 years group (IRR=1.13, 95% CI=1.08-1.18; interaction p-value <.001). The association did not differ by sex. Conclusion: This study provides some of the first data linking FI to increased HCC risk, in particular early onset HCC. If confirmed in other studies, these findings support the need for interventions that screen for, prevent, and address FI in populations at increased risk for HCC. Citation Format: Rebecca D. Kehm, Chrystelle L. Vilfranc, Jasmine A. McDonald, Hui Chen Wu. County-level food insecurity and hepatocellular carcinoma risk: A cross-sectional analysis [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr PR002.

Structural Racism and HIV Pre-exposure Prophylaxis Use in the Nationwide US: A County-Level Analysis.

Structural racism contributes to geographical inequalities in pre-exposure prophylaxis (PrEP) coverage in the United States (US). This study aims to investigate county-level variability in PrEP utilization across diverse dimensions of structural racism. The 2013-2021 nationwide county-level PrEP rate and PrEP-to-need ratio (PNR) data were retrieved from AIDSVu. PrEP rate was defined as the number of PrEP users per 100,000 population, and PNR was defined as the ratio of PrEP users to new HIV diagnoses per calendar year. Linear mixed effect regression was employed to identify associations of county-level structural racism (e.g., structural racism in housing and socioeconomic status) with PrEP rate and PNR on a nationwide scale of the US. From 2013 to 2021, the mean PrEP rate and PNR increased from 3.62 to 71.10 and from 0.39 to 10.20, respectively. Counties with more structural racism in housing were more likely to have low PrEP rates (adjusted β = - 5.80, 95% CI [- 8.84, - 2.75]). Higher PNR was found in counties with lower structural racism in socioeconomic status (adjusted β = - 2.64, 95% CI [- 3.68, - 1.61]). Regionally, compared to the Midwest region, counties in the West region were more likely to have higher PrEP rate (adjusted β = 30.99, 95% CI [22.19, 39.80]), and counties in the South had lower PNR (adjusted β = - 1.87, 95% CI [- 2.57, - 1.17]). County-level structural racism plays a crucial role in understanding the challenges of scaling up PrEP coverage. The findings underscore the importance of tailored strategies across different regions and provide valuable insights for future interventions to optimize PrEP implementation.

The Association of Broadband Internet Use With Drug Overdose Mortality Rates in the United States: Cross-Sectional Analysis.

The availability and use of broadband internet play an increasingly important role in health care and public health. This study examined the associations between broadband internet availability and use with drug overdose deaths in the United States. We linked 2019 county-level drug overdose death data in restricted-access multiple causes of death files from the National Vital Statistics System at the US Centers for Disease Control and Prevention with the 2019 county-level broadband internet rollout data from the Federal Communications Commission and the 2019 county-level broadband usage data available from Microsoft's Airband Initiative. Cross-sectional analysis was performed with the fixed-effects regression method to assess the association of broadband internet availability and usage with opioid overdose deaths. Our model also controlled for county-level socioeconomic characteristics and county-level health policy variables. Overall, a 1% increase in broadband internet use was linked with a 1.2% increase in overall drug overdose deaths. No significant association was observed for broadband internet availability. Although similar positive associations were found for both male and female populations, the association varied across different age subgroups. The positive association on overall drug overdose deaths was the greatest among Hispanic and Non-Hispanic White populations. Broadband internet use was positively associated with increased drug overdose deaths among the overall US population and some subpopulations, even after controlling for broadband availability, sociodemographic characteristics, unemployment, and median household income.

Sampling coverage of the Arkansas all-payer claims database by County's persistent poverty designation.

To evaluate the quality of Arkansas All-Payer Claims Database (APCD) for disparity research in persistent poverty areas by determining (1) its representativeness of Arkansas population, (2) variation by county, and (3) differences in coverage between persistent poverty and other counties. Cross-sectional study using 2019 Arkansas APCD member enrollment data and county-level data from various agencies. An alias identifier linked persons across insurance plans. County FIPS codes were used to extract county-level variables. Cohort 1 included individuals with ≥1 day of medical coverage in 2019. Cohort 2 included individuals with medical coverage in June, 2019. Cohort 3 included individuals with continuous medical coverage in 2019. Sampling proportions of a county's population in the three cohorts were compared between persistent poverty and other counties. Inverse-variance weighted linear regression was used to identify county-level socioeconomic and demographic characteristics associated with inclusion in each cohort. In 2019, 73.6% of Arkansans had medical coverage for ≥1 day (Cohort 1), 66.3% had coverage in June (Cohort 2), and 58.8% had continuous coverage (Cohort 3) in APCD. Sampling proportions varied by county (median[range]: Cohort 1, 78% [58%-95%]; Cohort 2, 71% [51%-88%]; and Cohort 3, 64% [44%-80%]), and were higher among persistent poverty counties than others for all three cohorts (mean [SD], persistent poverty vs. other: Cohort 1: 80.9% [6.4%] vs. 77.1% [6.3%], p = 0.04; Cohort 2: 74.0% [6.4%] vs. 70.1% [6.2%], p = 0.03; Cohort 3: 66.4% [6.1%] vs. 62.7% [6.0%], p = 0.03). In the 2019 APCD, larger counties and those with higher proportions of females or persons 65+ years had higher coverage, whereas counties with higher per capita household income, median home value, or disproportionately more persons of other races (non-White and non-Black) had lower coverage (p < 0.05 for all three cohorts). The Arkansas APCD had good coverage of Arkansas population. Coverage was higher in persistent poverty counties than others.

Community Culture Survey – Revised: Measuring neighborhood culture and exploring geographic, socioeconomic, and cultural determinants of health in samples across the United States and in Thailand

ObjectivesResearch on links between social, geographic, and cultural determinants of health has been thwarted by inadequate measures of culture. The purpose of this study was to improve the measurement of community culture, defined as shared patterns of attitudes and behaviors among people within a neighborhood that distinguish it from others, and to examine dimensions of culture, independent of socioeconomic and demographic factors, and their relationships with health. Study designA survey research design with correlational analyses was used. MethodsA survey packet including the Community Culture Survey – Revised (CCS-R), demographic, health, and other individual-level measures was administered through convenience sampling across the United States (US) and to a sample in Thailand from 2016 to 2018. US county-level variables were obtained from zip codes. Results1930 participants from 49 US states (n = 1592) and Thailand (n = 338) completed all CCS-R items, from which 12 subscales were derived: Social Support & Connectedness, Responsibility for Self & Others, Family Ties & Duties, Social Distress, Urban Diversity, Discontinuity, Church-Engaged, External Resource-Seeking, Locally Owned Business-Active, Power Deference, Next Generation Focus, and Self-Reliance. Neighborhood culture subscale scores varied more by geography than by participant's demographics. All subscales predicted one or more health indicator, and some of these relationships were significant after adjusting for participant age and county-level socioeconomic variables. Most of the significant differences on subscales by race/ethnicity were no longer significant after adjusting for participant's age and county-level socioeconomic variables. Most rural/urban and regional differences in culture within the US persisted after these adjustments. Based on correlational analyses, Social Support & Connectedness and Responsibility for Self & Others were the best predictors of participants' overall health and quality of life, and Responsibility for Self & Others was the best predictor (inversely) of the CDC's measures of social vulnerability. ConclusionsNeighborhood culture is measurable, multi-dimensional, distinct from race/ethnicity, and related to health even after controlling for age and socioeconomic factors. The CCS-R is useful for advancing research and practice addressing the complex interactions between individuals, their neighborhood communities, and health outcomes.

US County-Level Variation in Availability and Prevalence of Black Physicians in 1906

Black physicians are substantially underrepresented in the US health care workforce, with detrimental effects on the health and health care experiences of Black individuals. These contemporary gaps can be traced to the early days of the medical profession using the first edition of the American Medical Directory (AMD). To identify state- and county-level patterns related to the training and availability of Black physicians relative to their White counterparts in the 1906 AMD. For this cross-sectional study, data for 41 828 physician entries in 18 US states in or adjacent to the South as well as the District of Columbia were extracted from the 1906 AMD and aggregated to 1570 counties. Data analysis was performed between September 2023 and January 2024. County-level exposure variables included population density, racial composition, and illiteracy rate among US-born White residents as well as an index of terrain ruggedness and the number of lynchings in the previous decade. Median values of physicians' distance from place of practice to place of medical training (by race of physician) were also used as an exposure variable. There were 4 county-level outcomes: (1) presence of any Black physician, (2) proportion of Black physicians per Black population, (3) proportion of White physicians per White population, and (4) community representativeness (reported as the community representativeness ratio). The cross-sectional analysis used generalized additive mixed models with state-level random effects. Across 1570 counties, Black physicians comprised 746 (1.8%) of the 41 828 physicians in the dataset. Black physicians tended to train further from their place of practice than their White counterparts. The proportion of Black physicians per 1000 Black residents was 0.08 compared with 1.62 for White physicians; these proportions varied substantially by state. At the county level, the presence of any Black physician was associated with percentage Black population (odds ratio [OR], 28.94 [95% CI, 9.77 to 85.76]; P ≤ .001), population density (OR, 2.63 [95% CI, 2.03 to 3.40]; P ≤ .001), and distance to the nearest Black medical school (OR, 0.62 [95% CI, 0.42 to 0.92]; P = .02). A variety of structural disadvantages are illustrated in this cross-sectional study of county-level sociodemographic and geographic characteristics associated with the prevalence of Black physicians in the earliest days of the profession. To demonstrate its broader utility for health disparities research, the dataset has been made publicly available with a visualization platform.

County-Level Atrazine Use and Gastroschisis

The incidence of gastroschisis, a birth defect involving the herniation of the small bowel through the abdominal wall, has increased in the US since the 1960s. The pesticide atrazine is a hypothesized cause of gastroschisis; however, examination of the association between atrazine and gastroschisis has been limited. To evaluate national trends in gastroschisis incidence, maternal and infant characteristics associated with gastroschisis, and whether county-level atrazine use is associated with gastroschisis. This retrospective, repeated cross-sectional study examined birth certificate data of all live births in the US and data on atrazine use from the US Geological Survey from January 1, 2009, through December 31, 2019. The data analysis was performed between August 5, 2021, and May 26, 2023. County-level atrazine use. The primary outcome was gastroschisis incidence. Covariates included maternal age, race and ethnicity, body mass index (measured by weight in kilograms divided by height in meters squared), parity, insurance type, Chlamydia infection during pregnancy, smoking, and rurality. Mixed-effects logistic regression models (year fixed effects and county random effects) were constructed using different county-level atrazine exposure variables (1-, 5-, and 10-year means). Between 2009 and 2019, 39 282 566 live births were identified, with 10 527 infant diagnoses of gastroschisis. Infants with gastroschisis were more likely to have mothers who identified as non-Hispanic White (61% vs 54%; P < .001), had a lower body mass index (median [IQR], 23.4 [20.8-27.2] vs 25.4 [22.0-30.8]; P < .001), were more likely to be nulliparous (median [IQR], 0 [0-1] vs 1 [0-2]; P < .001), and were more commonly covered by Medicaid (63% vs 43%; P < .001). During the study period, the rate (per 1000 live births) of gastroschisis decreased from 0.31 (95% CI, 0.29-0.33) to 0.22 (95% CI, 0.21-0.24). The median (IQR) county-level atrazine use estimates were higher among infants with gastroschisis (1 year, 1389 [IQR, 198-10 162] vs 1023 [IQR, 167-6960] kg; 5 years, 1425 [IQR, 273-9895] vs 1057 [IQR, 199-6926] kg; 10 years, 1508 [IQR, 286-10 271] vs 1113 [IQR, 200-6650] kg; P < .001). In adjusted models, higher county levels of atrazine (each 100 000-kg increase) were associated with a higher incidence of gastroschisis (1 year: adjusted odds ratio [AOR], 1.12 [95% CI, 1.01-1.24]; 5 years: AOR, 1.15 [95% CI, 1.02-1.30]; 10 years: AOR, 1.21 [95% CI, 1.07-1.38]). In this cross-sectional study, higher county levels of atrazine were associated with infant diagnoses of gastroschisis. While atrazine is the second-most used herbicide in the US, numerous countries around the world have banned it out of concern for adverse effects on human health. These findings suggest that exploring alternatives to atrazine in the US may be warranted.

The associations of food environment with gastrointestinal cancer outcomes in the United States.

Social conditions and dietary behaviors have been implicated in the rising burden of gastrointestinal cancers (GIC). The "food environment" reflects influences on a community level relative to food availability, nutritional assistance, and social determinants of health. Using the US Department of Agriculture-Food Environment Atlas (FEA), we sought to characterize the association of food environment on GIC presenting stage and long-term survival. Patients diagnosed with GIC between 2013 and 2017 were identified using the SEER database. FEA-scores were based on 282 county-level food security variables, store-restaurant availability, SNAP/WIC enrollment, pricing/taxes, and producer vicinity adjusted-for factors of socioeconomic status, race-ethnicity, transportation access, and comorbidities. Relative FEA rankings across US counties were averaged into a composite score and assigned to patients by county-of-residence. The association of FEA, cancer stage, and survival were analyzed using multiple logistic regression and cox-proportional hazard models relative to White/non-White race/ethnicity. Among 287,148 patients, the most common GIC-sites were colon (n = 97,942, 34%), pancreas (n = 49,785, 17.3%), liver (n = 31,098, 11.0%) and esophagus (n = 16,271, 5.7%). A worse food environment was independently associated with increased odds of late-stage diagnosis (esophageal odds ratio [OR]: 1.03, 95% confidence interval [CI]: 1.01-1.05; hepatic OR: 1.06, 95% CI: 1.03-1.08; pancreatic OR: 1.04, 95% CI: 1.01-1.06) among all patients; in contrast, food environment was associated with colorectal cancer stage among non-White patients only (OR: 1.04, 95% CI: 1.03-1.06). Worse food environment was associated with worse 3-year survival (colon OR: 1.03, 95% CI: 1.01-1.04; hepatic OR: 1.12, 95% CI: 1.08-1.17; gastric OR: 1.07, 95% CI: 1.01-1.13). Similar associations were noted relative to overall survival among the entire cohort (biliary tract hazard ratio [HR]: 1.03, 95% CI: 1.01-1.05; esophageal HR: 1.02, 95% CI: 1.01-1.04; hepatic HR: 1.07, 95% CI: 1.06-1.09; pancreatic HR: 1.04, 95% CI: 1.02-1.05; rectum HR: 1.03, 95% CI: 1.01-1.04; gastric HR: 1.05, 95% CI: 1.03-1.07), as well as among non-White patients (biliary HR: 1.04, 95% CI: 1.01-1.07; colon HR: 1.03, 95% CI: 1.01-1.05; esophageal HR: 1.05, 95% CI: 1.02-1.08; hepatic HR: 1.08, 95% CI: 1.06-1.10) (all p < 0.003). Food environment was independently associated with late-stage tumor presentation and worse 3-year and overall survival among GIC patients. Interventions to address inequities across communities relative to food environments are needed to alleviate disparities in cancer care.

Spatial and temporal analysis of HIV clinical outcomes in Florida reveals counties with persistent racial and ethnic disparities during 2012-2019

BackgroundRacial/ethnic disparities in the HIV care continuum have been well documented in the US, with especially striking inequalities in viral suppression rates between White and Black persons with HIV (PWH). The South is considered an epicenter of the HIV epidemic in the US, with the largest population of PWH living in Florida. It is unclear whether any disparities in viral suppression or immune reconstitution—a clinical outcome highly correlated with overall prognosis—have changed over time or are homogenous geographically. In this analysis, we 1) investigate longitudinal trends in viral suppression and immune reconstitution among PWH in Florida, 2) examine the impact of socio-ecological factors on the association between race/ethnicity and clinical outcomes, 3) explore spatial and temporal variations in disparities in clinical outcomes.MethodsData were obtained from the Florida Department of Health for 42,369 PWH enrolled in the Ryan White program during 2008-2020. We linked the data to county-level socio-ecological variables available from County Health Rankings. GEE models were fit to assess the effect of race/ethnicity on immune reconstitution and viral suppression longitudinally. Poisson Bayesian hierarchical models were fit to analyze geographic variations in racial/ethnic disparities while adjusting for socio-ecological factors.ResultsProportions of PWH who experienced viral suppression and immune reconstitution rose by 60% and 45%, respectively, from 2008-2020. Odds of immune reconstitution and viral suppression were significantly higher among White [odds ratio =2.34, 95% credible interval=2.14-2.56; 1.95 (1.85-2.05)], and Hispanic [1.70 (1.54-1.87); 2.18(2.07-2.31)] PWH, compared with Black PWH. These findings remained unchanged after accounting for socio-ecological factors. Rural and urban counties in north-central Florida saw the largest racial/ethnic disparities.ConclusionsThere is persistent, spatially heterogeneous, racial/ethnic disparity in HIV clinical outcomes in Florida. This disparity could not be explained by socio-ecological factors, suggesting that further research on modifiable factors that can improve HIV outcomes among Black and Hispanic PWH in Florida is needed.

Abstract TMP30: The Impact of Social Influences on Rehabilitation Services Utilization After Stroke

Introduction: Evidence supports that early initiation of rehabilitation is important for patient outcomes after stroke. Thus, determination of patients needing rehabilitation during their acute hospitalization is critical. We assessed the impact of demographic, clinical, and county-level social influences on rate of transfer to a rehabilitation facility after acute stroke hospitalizations. Methods: Using Get with the Guidelines-Stroke registry data and Institute for Health Metrics and Evaluation data, we fitted univariate and multivariate generalized linear mixed models (GLMM) with demographic, clinical, and social variables to determine how social influences of health impacted variances in transfers to rehabilitation facilities. Specific county-level variables assessed were related to education, income, employment, home ownership, and being foreign born. Results: Among 912,488 patients who were assessed or received rehabilitation services due to acute ischemic or intracerebral hemorrhagic stroke, 33.4% of them were transferred to rehabilitation facilities. Multivariate analyses of patient-level demographic and clinical variables achieved an Akaike Information Criterion (AIC) of 925780 (21.0% reduction from the base model) largely driven by clinical factors (mRS at discharge, initial NIHSS, and initial exam findings). When county-level variables related to social influences of health were incorporated, AIC decreased by 23.4% to 871806, indicating a better model fit largely driven by education level. Conclusions: Social influences of health impact transfer to rehabilitation facilities after acute stroke hospitalization. Future studies should focus on individual (versus county-level) metrics of social influences of health and their impact on access to rehabilitation. This information should be used to develop interventions to improve the equity of rehabilitation utilization after stroke.

The impact of county-level factors on meaningful use of electronic health records (EHRs) among primary care providers.

This study examines the impact of county-level factors on "meaningful use" (MU) of electronic health records (EHRs) for 8415 primary care providers (PCPs) that enrolled in the Florida Medicaid EHR Incentive Program through adopting, improving, or upgrading (AIU) a certified EHR technology. PCPs received incentive payments at enrollment and if they used their EHRs in meaningful ways; ways that benefit patients and providers alike they received additional payments. We conducted a retrospective cohort study of these providers over the 2011-2018 period while linking their records to other state data. We used the core constructs of the resource dependence theory (RDT), a well-established organization theory in business management, to operationalize the county-level variables. These variables were rurality, poverty, educational attainment, managed care penetration, changes in population, and number of PCPs per capita. The unit of analysis was provider-years. For practical and computational purposes, all the county variables were dichotomized. We used analysis of variance (ANOVA) to test for differences in MU attestation rates across each county variable. Odds ratios and corresponding 95% confidence intervals were derived from pooled logistic regressions using generalized estimated equations (GEE) with the binomial family and logit link functions. Clustered standard errors were used. Approximately 42% of these providers attested to MU after receiving first-year incentives. Rurality and poverty were significantly associated with MU. To some degree, managed care penetration, change in population size, and number of PCPs per capita were also associated with MU. Policy makers and healthcare managers should not ignore the contribution of county-level factors in the diffusion of EHRs among physician practices. These county-level findings provide important insights about EHR diffusion in places where traditionally underserved populations live. This county-perspective is particularly important because of the potential for health IT to enable public health monitoring and population health management that might benefit individuals beyond the patients treated by the Medicaid providers.

Parental Influence and Religious Context: A Cross-Level Perspective on Volunteering during Adolescence and Early Adulthood

Different religions may have different cultural values toward volunteering in the wider community. Although prior research has examined how individual religious characteristics may influence volunteering, comparatively less is known about how different religious contexts in which one is embedded may bring about variations in volunteering. To fill in this gap, this study applied multilevel analyses to merged data from the National Study of Youth and Religion (NSYR), the U.S. Census, and the Religious Congregations and Membership Study (RCMS). Results suggest that a county’s Catholic population share was positively associated with the frequency of youth volunteering even after controlling for a wide range of individual and county-level variables. Surprisingly, an increasing county-level Catholic population share may even strengthen the positive effect of parental volunteering influence on youth volunteering. In contrast, the county-level evangelical Protestant population share bear a negative relationship with youth volunteering frequency, which was mediated by county-level socioeconomic controls.

Physician selection for hospital integration: Theoretical considerations and empirical findings.

The U.S. health care system has seen an increase in hospital-physician integration, with hospitals acquiring increasing numbers of physician practices. This shift has been linked to higher costs without significant improvements in quality. This study sought to identify the characteristics of physicians who transitioned from independent practice to hospital integration. We used physician variables, including quality scores, medical school rankings, years of experience, experience treating socially or medically complex patients, practice style, and location, as well as health care market and county-level variables to understand these determinants using a fixed-effects logistic regression model. A total of 101,746 physicians representing 66 clinical specialties satisfied our inclusion criteria, of which 3,656 became hospital-integrated between 2018 and 2020. The integrating physicians were generally less experienced, had lower quality scores, and generated less revenue per Medicare patient. Their patients, on average, had higher comorbidity scores, were more likely to be dually eligible, and resided in counties with higher poverty rates. Our findings indicate that the physicians most likely to become hospital integrated are those facing reimbursement pressures due to a complex case mix and the associated challenges of performing well on the quality metrics. We also found some support for the anticompetitive aspects of hospital-physician integration. Our results suggest that hospitals are integrating with a relatively less experienced physician workforce but one that is perhaps more capable of treating clinically and socioeconomically complex patients. Hospitals interested in using physician integration strategically to improve care quality should put more emphasis on physician quality. Such an approach has the potential to increase efficiency without sacrificing quality of care.

Spatial, temporal, and demographic nonstationary dynamics of COVID-19 exposure among older adults in the U.S.

This study examines demographic disparities in COVID-19 exposures across older adults age 60-79 and older adults age 80 and over, and explores the factors driving these dynamics in the United States (U.S.) from January 2020 to July 2022. Spatial clusters were identified, and 14 main health determinants were synthesized from 62 pre-existing county-level variables. The study also assessed the correlation between these health determinants and COVID-19 incidence rates for both age groups during the pandemic years. Further examination of incidence rates in relation to health determinants was carried out through statistical and spatial regression models. Results show that individuals aged 80 and over had much higher hospitalization rates, death rates, and case-fatality rates in 2020-2022. Spatial results indicate that the geographical cluster of high incidence rates for both groups shifted from the Midwest at the beginning of the pandemic to the Southwest in 2022. The study revealed marked spatial, temporal, and demographic nonstationary dynamics in COVID-19 exposures, indicating that the health effects of contextual factors vary across age groups. COVID-19 incidence rates in older adults were strongly influenced by race, healthcare access, social capital, environment, household composition, and mobility. Future public health policies and mitigations should further their efforts by considering temporal and demographic nonstationarity as well as local conditions.

Abstract B076: Influence of residential segregation and women’s health provider density on advanced stage endometrial cancer diagnoses

Abstract Background. Endometrial cancer (EC) is the most common gynecologic malignancy in the US, with an increasing incidence. Non-Hispanic Black (NHB) women are disproportionately impacted by EC but the complex mechanisms by which this is occurring are not yet fully understood. Taking a social ecological approach to this issue, our objective was to determine if there are associations between county-level OB-GYN provider density and residential segregation as measured by the Index of Concentration at the Extremes (ICE) on late-stage EC diagnoses in Florida. Methods. All malignant EC cases were identified from 2001 to 2016 in the Florida Cancer Database System (FCDS). Using 5-year estimates from the 2013-2017 American Community Survey, five county-level ICE variables were calculated: economic (high vs low), race and/or ethnicity (non-Hispanic white (NHW) vs. NHB and NHW vs. Hispanic), and racialized economic segregation (low-income NHB vs. high-income NHW and low-income Hispanic vs. high-income NHW). County-level provider density was calculated as the number of OB-GYN providers divided by the female population of each county multiplied by 100,000. Early stage was defined as local and late stage was defined as regional/distant. Multivariable-adjusted logistic regression models were specified to estimate the association between each ICE variable and provider density separately on late-stage diagnosis of EC. Results. There were a total of 50,363 EC cases in Florida from 2001 to 2016 with 44,678 (88.7%) having stage information. Of those with stage information, the mean age at diagnosis was 64.1 years (SD: 11.9), 71.8% were NHW, 11.5% were NHB, 14.0% were Hispanic, and 2.8% were other race. The majority of individuals had government insurance (50.6%). More NHB women (27.1%) were diagnosed with aggressive EC histologies relative to NHW (16.4%) and Hispanic women (15.5%) (p&amp;lt;0.001). 14,366 (32.2%) were diagnosed with late-stage EC. A larger proportion of NHB women were diagnosed with later-stage EC compared to NHW women and Hispanic women (43.7% vs. 30.3% and 32.4%, respectively, p&amp;lt;0.001). NHB and Hispanic women had significantly greater odds of being diagnosed with later-stage EC compared to NHW women, regardless of residential segregation (OR: 1.46, 95% CI: 1.36, 1.56 and OR: 1.09, 95% CI: 1.01, 1.17, respectively). Women living in more economically disadvantaged Hispanic segregated counties had a greater odds of being diagnosed with later-stage EC compared to those living in more NHW segregated areas (OR: 1.16, 95% CI: 1.00, 1.65). Provider density was not found to be associated with later-stage diagnosis. Discussion. Advanced stage EC at diagnosis among Black and Hispanic women in Florida seems to be largely independent of provider density and residential segregation, though NHB more commonly present with metastatic disease. Biologic drivers of oncogenesis and barriers to timely care in this group require further exploration. Given the diversity and representation from Afro-Caribbeans in Florida, community-level investigations are required. Citation Format: Ashly Westrick, Zinzi Bailey, Matthew Schlumbrecht. Influence of residential segregation and women’s health provider density on advanced stage endometrial cancer diagnoses [abstract]. In: Proceedings of the 16th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2023 Sep 29-Oct 2;Orlando, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2023;32(12 Suppl):Abstract nr B076.

Association of area-level structural racism with racial disparities in cancer mortality.

172 Background: Structural racism (SR) refers to interactive systems of discrimination that perpetuate advantage for one racial group over others. Single-variable SR measures, such as residential segregation and historical red-lining, are associated with cancer disparities but do not represent all aspects of SR. We aimed to develop a composite measure of SR for cancer research and examine the association of county-level SR with disparities in cancer mortality. Methods: Adapting methods from obesity and public safety research, we developed a SR composite score (SRCS) by calculating the ratio of Black to White performance on eight county-level variables across five domains (housing, education, employment, income and healthcare) using race-stratified census data from 2006-2020. Using North Carolina as a test set, we rank-ordered counties on a min-max scale 1-100 for each domain and averaged scores across domains to create the SRCS, where higher SRCS suggests greater SR. Outcomes were county age-standardized mortality rates (per 100k) for all cancers combined, breast, prostate, colorectal and lung cancers, extracted from CDC Wonder for non-Hispanic Blacks and Whites (2000-2020). We used generalized estimating equations to regress mortality rates on SRCS, race, and their interaction. Results: County-level SRCS was positively associated with county-level racial disparities in all-cancer mortality; for each 10-unit increase in SRCS, Black-White difference in rates increased by 15.3/100k (p&lt;0.001). For individual cancers, a 10 unit rise in county SRCS was associated with a 1.93/100k increase in the colorectal cancer mortality gap (p=0.003) and a 6.18/100k increase in the lung cancer mortality gap (p&lt;0.001). SRCS was not associated with mortality disparities for breast and prostate cancers, although these disparities were large. For all-cancer mortality and all tumor types except breast cancer, mortality rates of Whites significantly decreased as SRCS increased, while mortality rates for Blacks remained roughly stable with rising SRCS. Adjustment for county-level measures of social deprivation, unemployment, primary care provider ratio, and severe housing problems did not attenuate the relationship of SRCS with mortality. Conclusions: A composite measure of SR is associated with racial disparities in overall cancer mortality and mortality of specific solid tumors. Continued research is needed to further understand the role of area-level social determinants, ecologic fallacy, and differences in the relationship of SRCS to mortality across cancers. Nevertheless, increased SR is associated with lower county-level cancer mortality among Whites, suggesting that SR may perpetuate cancer mortality disparities via relative advantage to White residents rather than direct disadvantage to Black residents.

Racial Disparities Associated with Increased Burden of Sexually Transmitted Infections in North Carolina, Southeastern United States

Sexually transmitted infections (STIs) constitute a major public health issue in the United States. North Carolina (NC) in the southeastern U.S. ranks among the highest in STI incidence. We aimed to describe the incidence rates and identify potential risk factors of STIs in NC. The STI data reported by the NC Department of Health and Human Services for 2018 were compiled for chlamydia, gonorrhea, syphilis, and HIV infections for all 100 NC counties. Linear regression modeling was used to assess the association of STIs with predefined county-level variables. The mean STI incidence rates per 100,000 persons were highest for chlamydia (592.43 ± 30.02), followed by gonorrhea (212.06 ± 13.75), HIV (12.66 ± 0.947), and syphilis (3.33 ± 0.439). For chlamydia, higher risk was significantly associated with income (β = −0.008, SE = 0.003; p = 0.006), education (some college; β = 10.02, SE = 3.15, p = 0.002), race (Black; β = 12.17, SE = 1.57, p &lt; 0.0001), and number of truck stops (β = 20.20, SE = 6.75, p = 0.004). The same variables, except for education, were associated with higher gonorrhea risk. Only race (being Black) was significantly associated with higher syphilis risk. Racial disparities in STI burden were significant, with race (being Black) constituting a risk factor for all four STIs. Interventions targeted to identified risk factors may help to reduce the STI burden in NC.

Disparities in Access to Robotic Knee Arthroplasty: A Geospatial Analysis

BackgroundThe utilization of robotic knee arthroplasty (RKA) continues to increase across the United States. The aim of this geospatial analysis was to elucidate if RKA is distributed uniformly across the United States or if disparities exist in patient access. MethodsPublicly available provider-finding functions for 5 major manufacturers of RKA systems were used to obtain the practice locations of surgeons performing RKA along with their associated RKA system manufacturer. The average travel distance for each county to the nearest RKA surgeon was calculated and Moran’s index clustering analysis was used to find hotspots and coldspots of RKA access. A logistic regression model was used to identify the predictive odds ratios between robotic hotspots and coldspots with county-level sociodemographic variables. Of the 34,216 currently practicing orthopedic surgeons in 2022, 2,571 have access to robotic assistance for knee arthroplasty. ResultsHotspots of increased travel time were predominantly in West South Central and West North Central census regions. Hotspots were significantly more rural and consisted of predominantly White populations, with lower median income and health insurance coverage. ConclusionsThe results of the current study align with existing literature, demonstrating absolute geographic access disparities for rural and economically disadvantaged populations. Additionally, relative access disparities persist for minority populations and individuals with high comorbidity burdens residing in urban areas.

Different volatile organic compounds in local point source air pollution pose distinctive elevated risks for respiratory disease-associated emergency room visits

Air pollution increases risk of respiratory disease but prior research has focused on particulate matter and criteria air pollutants, and there are few studies on respiratory effects of volatile organic compounds (VOC). We examined zip code level relationships between emergency room (ER) visits for respiratory illness and VOC pollution in New York State from 2010 to 2018. Detailed information on VOC pollution was derived from the National Emissions Inventory, which provides point source information on VOC emissions at the zip code level. We considered four respiratory diseases: asthma, acute upper respiratory infections, chronic obstructive pulmonary disease (COPD), and lower respiratory disease, using mixed effects regression with a random intercept to account for county level variability in single pollutant models, and Random Forest Regression (RFR) to assess relative importance of VOC exposures when considered together in multipollutant models.Single pollutant models show associations between respiratory-related ER visits with all pollutants of interest across all study years, even after adjusting for poverty and smoking by zip code. The largest relative single pollutant effect sizes considered included benzene, ethylbenzene, and total (summed) VOCs. Results from RFR including all VOC exposures indicate that ethylbenzene has the greatest variable importance for asthma, acute upper respiratory infections, and COPD, with toluene and benzene most important for lower respiratory ailments. RFR results also demonstrate presence of pairwise interactive effects between VOC pollutants. Our findings show that local VOC pollution may offer a significant contribution to the risk of respiratory disease-related ER visits, and that effects vary by illness and by VOC compound. ER visit rates for respiratory illness were elevated in high poverty zip codes, although this may be attributable to the fact that the poor lack basic access to health care and use ERs more frequently for routine care.