Preventable Hospitalization Rates Research Articles

Impact of Family Doctor Contract Services on Preventable Hospitalizations Amongst Patients with Hypertension in Rural China: Mediating Role of Primary Healthcare Quality.

The objective of this study was to explore the connection between family doctor contract services and preventable hospitalizations. Additionally, we sought to examine the role of primary health care quality as a mediating factor in the link between family doctor contract services and preventable hospitalizations among patients with hypertension. This cross-sectional study was performed in Dangyang (Hubei Province, Central China) and Xishui (Guizhou Province, Western China) counties in July-August 2023. Participants comprised 625 patients selected via a multi-stage sampling method. Causal mediation analysis was conducted to explore the effect of family doctor contract services on preventable hospitalizations and the mediating effect of primary healthcare quality on this relationship. Utilization rate of family doctor contract services of hypertensive patients was 58.6%, score of primary health service quality was 70.75 and incidence of preventable hospitalizations was 28.2%. Amongst hypertensive patients, utilization of family doctor contract services decreased the occurrence of preventable hospitalizations, with a total effect of -0.22 (p < 0.001). Primary healthcare quality mediates the association, with a mediate effect of -0.05 (p < 0.001), explaining 22.73% of the total effect. Improving the utilization of family doctor contract services and primary healthcare quality may result in lower rates of preventable hospitalizations amongst hypertensive patients.

The variation in preventable hospitalization in patients with type 2 diabetes in Kentucky before and after the Medicaid expansion.

Hospitalizations are more resource intensive and expensive than outpatient care. Therefore, type 2 diabetes-related preventable hospitalization are a major topic of research efficiency in the healthcare system. Analyze county level variation in type 2 diabetes-related preventable hospitalization rates in Kentucky before the Medicaid expansion (2010-2013) and after the Medicaid expansion (2014-2017). Geographic mapping and cluster analysis. Data for a state of the United States of America. We used the KID data to generate geographic mapping for type 2 diabetes-related preventable hospitalizations to visualize rates. We included all Kentucky discharges of age 18 years and older with the ICD9/10 principal diagnosis code for type 2 diabetes. Then, we conducted cluster analysis techniques to compare county-level variation in type 2 diabetes-related preventable hospitalization rates across Kentucky counties pre- and post-Medicaid expansion. County type 2 diabetes-related preventable hospitalization pre- and post-Medicaid expansion. From 2010-2017, type 2 diabetes-related preventable hospitalization discharge rates reduced significantly in the period of the post-Medicaid expansion (P=.001). The spatial statistics analysis revealed a significant spatial clustering of counties with similar rates of type 2 diabetes-related preventable hospitalization in the south, east, and southeastern Kentucky pre- and post-Medicaid expansion (positive z-score and positive Moran's Index value (P>.05). Also, there was a significant clustering of counties with low type 2 diabetes-related preventable hospitalization rates in the north, west, and central regions of the state pre-Medicaid expansion and post-Medicaid expansion (positive z-score and positive Moran's Index value (P>.05). Kentucky counties in the southeast have experienced a significant clustering of highly avoidable hospitalization rates during both periods. Focusing on the vulnerable counties and the economic inequality in Kentucky could lead to efforts to lowering future type 2 diabetes-related preventable hospitalization rates. We used de-identified data which does not provide insights into the frequency of hospitalizations per patient. An individual patient may be hospitalized several times and counted as several individuals.

Impact of the COVID-19 Pandemic on Healthcare Utilization among Medically Underserved Patients with Ambulatory Care Sensitive Conditions.

This study aims to examine (1) the impact of the pandemic phases on overall and preventable hospitalizations and emergency department (ED) visits, and (2) the effect of the pandemic on these outcomes within subgroup populations including gender, race, patients' residence in health professional shortage areas (HPSA), and residence in a federal poverty level. We used electronic medical record (EMR) data for the year 2019 and 2020 from a large health system predominantly serving medically underserved patients in the South. We used a difference-in-differences approach to examine changes in weekly rates of overall and preventable hospitalizations and ED visits in the pandemic phase 1 (Mid-March to June of 2020) and phase 2 (July-September of 2020) compared to the same period in 2019 after adjusting for weekly outcome rates in the baseline period (January to Mid-March of 2020) compared to the same period in 2019. The study sample included 1.4 million hospitalizations and ED encounters. In phase 1 of the pandemic, there were significant reductions in overall (-108) and preventable (-75.3) hospitalizations, and overall (-408) and preventable (-306) ED visits when compared to the same period in 2019. In phase 2 of the pandemic, there were significant reductions in overall (-60) and preventable (-43) hospitalizations and in overall (-360) and preventable (-258) ED visits as compared to 2019. We found greater reductions in ED visits, both overall and preventable, during the early pandemic phases among Black patients than among White patients. Similar patterns in the reduction of ED visits were found in Black versus White patients within subgroups of women, men, and those residing in a HPSA and low-income areas. Substantial reductions in utilization were observed in Black patients in comparison to white patients and these differences persisted among men, women, and those living in underserved and low-income areas.

On the Asymmetric Relationship Between Physician Mental Health Disorders on Quality of Healthcare Under the COVID-19 Pandemic in Taiwan: Quantile on Quantile Regression Analyses.

When examining the nexus of physician mental health disorders and healthcare quality from the empirical perspective, mental health disorders are frequently associated with cyclical patterns corresponding to cyclic seasonality, mood swings, emission of air pollution and business cycles, the potential asymmetric effects of physician mental health disorders on healthcare quality have not received adequate attention from researchers. Therefore, the purpose of this study is to explore the asymmetric relationship between physician mental health disorders and healthcare quality during the pandemic outbreak in Taiwan. Daily data for care quality indicators and physician mental health disorders were collected from the National Insurance Research Database in Taiwan, and the quantile-on-quantile regression model was applied to proceed with our analyses. Our results indicated that the overall aggregate effects of each quantile of physician mental health disorders on the cumulative quantiles of healthcare quality are negative (positive) for the 14-day readmission rate (preventable hospitalization rate and non-urgent ED-visit rate). Positively (negatively) cumulative effects of each quantile of physician mental health disorders were detected in the middle (low and high) quantiles of the preventable hospitalization rate. The cumulative effects of each quantile of physician mental health disorders on the high (low and middle) quantiles of the 14-day readmission rate are negative (positive), but the cumulative effects on various quantiles of the non-urgent ED-visit rate exhibit the opposite pattern. The observed variation in the relationship between physician mental health disorders and different quantiles of healthcare quality suggests the need for tailored strategic interventions based on distinct levels of healthcare quality when addressing the higher risk of physician mental health disorders during the pandemic outbreak conditions.

Association Between Community Social Vulnerability and Preventable Hospitalizations.

Preventable hospitalizations are common and costly events that burden patients and our health care system. While research suggests that these events are strongly linked to ambulatory care access, emerging evidence suggests they may also be sensitive to a patient's social, environmental, and economic conditions. This study examines the association between variations in social vulnerability and preventable hospitalization rates. We conducted a cross-sectional analysis of county-level preventable hospitalization rates for 33 states linked with data from the 2020 Social Vulnerability Index (SVI). Preventable hospitalizations were 40% higher in the most vulnerable counties compared with the least vulnerable. Adjusted regression results confirm the strong relationship between social vulnerability and preventable hospitalizations. Our results suggest wide variation in community-level preventable hospitalization rates, with robust evidence that variation is strongly related to a community's social vulnerability. The human toll, societal cost, and preventability of these hospitalizations make understanding and mitigating these inequities a national priority.

The Costs Of Disparities In Preventable Heart Failure Hospitalizations In The US South, 2015-17.

Black Americans in the US South have high rates of preventable heart failure hospitalizations, which reflects systemic inequities that also produce economic costs. We measured the direct medical costs of disparities in preventable heart failure admissions (that is, excess admissions) among Medicare beneficiaries living in six states in the US South (Kentucky, Arkansas, Florida, Georgia, Mississippi, and North Carolina). We used 2015-17 data from the Healthcare Cost and Utilization Project and constructed negative binomial models with state-level fixed effects to calculate adjusted admission rates with heart failure as the principal diagnosis. We calculated the number of these admissions that would have been avoided if Black, Hispanic, Asian/Pacific Islander, and American Indian/Alaska Native Medicare beneficiaries had the same admission rates as White beneficiaries. We found 28,213 excess admissions (48percent excess) with $60,845,855 annual costs among Black beneficiaries, 3,499 (14percent excess) with $8,179,381 annual costs among Hispanic beneficiaries, and 550 (51percent excess) with $1,093,472 in annual costs among American Indian/Alaska Native beneficiaries. Failure to address heart failure treatment inequities in the community has a high opportunity cost.

Increased female political representation associated with lower county-level uninsured and preventable hospitalizations rates in the United States, 2013–2018

ObjectiveAlthough women comprise 50% of the population, females remain underrepresented in government. Inequitable female political representation, a form of structural sexism, may impact population health. Previous studies focused primarily on individual health behaviors and low- or middle-income countries. To date, no study has examined the association between female political representation and healthcare access and utilization, immediately policy-amenable outcomes, in the United States. Study designThis was a repeated cross-sectional study. MethodsThis study uses 2013–2018 county-level data from the County Rankings. I performed multilevel analyses to determine the relationships between state-level female representation (% female state legislators) and two outcomes—the percentage of county-level population under age 65 years without health insurance (primary outcome) and the county-level preventable hospitalization rates (secondary outcome of interest). Potential confounders included county-level and state-level characteristics such as the unemployment rate. I also examined whether associations differed by political party control of the state legislature. ResultsIn the fully adjusted model, one standard deviation difference in female political representation was associated with a decrease of 0.22 percentage points in county-level uninsured (95% confidence interval = −0.32, −0.12). The association between female political representation and preventable hospitalization rate differed according to state political party in control, with a decrease found only among counties in democratic/split controlled states (−80.51, 95% confidence interval = −149.65, −11.38). ConclusionsThe results suggest that policy intervention addressing the underrepresentation of women in government may help increase the proportion of uninsured and, under certain circumstances, decrease county-level unnecessary hospitalizations. However, further research is needed to better understand the role of political party control in modifying noted associations.

The Appalachian Gap in Preventable Hospitalizations: Are We Seeing Any Progress?

Previous studies have documented geographic variation in preventable hospitalizations between rural and urban areas, but much less is known about preventable hospitalization patterns between heterogeneous rural areas. Unique challenges related to access of care and poverty may put the rural Appalachian Region at risk for higher rates of preventable hospitalizations. This study examines whether within-rural differences in Kentucky's preventable hospitalization rates exist and how these differences may be changing over time. Longitudinal and geographic trends in county-level preventable hospitalization rates were examined using Kentucky hospital discharge data from 2016 to 2019. Regression models were run to determine whether changes over time in preventable hospitalization rates led to an increasing or decreasing gap in outcomes between rural Appalachian counties and their urban and rural non- Appalachian counterparts. Rural Appalachian counties consistently had significantly higher preventable hospitalizations rates compared to their rural non-Appalachian and urban counterparts ( p < 0.01). A downward trend in overall preventable hospitalizations was observed for rural Appalachia over time, but trends were relatively stable for rural non-Appalachian and urban counties. Regression results indicate that there was no significant reduction in the "Appalachian gap" over time. The analyses confirm that rural areas within Kentucky experienced highly heterogeneous rates of preventable hospitalizations. Despite Medicaid expansion, there is little evidence of any narrowing of the "Appalachian gap." Focus on improving access to care alone may be insufficient to improve outcomes. Alternative strategies that leverage population health approaches may improve capacity to address complex health and social needs in rural Appalachia.

Racialized economic segregation and potentially preventable hospitalizations among Medicaid/CHIP-enrolled children.

To examine geographic variation in preventable hospitalizations among Medicaid/CHIP-enrolled children and to test the association between preventable hospitalizations and a novel measure of racialized economic segregation, which captures residential segregation within ZIP codes based on race and income simultaneously. We supplement claims and enrollment data from the Transformed Medicaid Statistical Information System (T-MSIS) representing over 12 million Medicaid/CHIP enrollees in 24 states with data from the Public Health Disparities Geocoding Project measuring racialized economic segregation. We measure preventable hospitalizations by ZIP code among children. We use logistic regression to estimate the association between ZIP code-level measures of racialized economic segregation and preventable hospitalizations, controlling for sex, age, rurality, eligibility group, managed care plan type, and state. We include children ages 0-17 continuously enrolled in Medicaid/CHIP throughout 2018. We use validated algorithms to identify preventable hospitalizations, which account for characteristics of the pediatric population and exclude children with certain underlying conditions. Preventable hospitalizations vary substantially across ZIP codes, and a quarter of ZIP codes have rates exceeding 150 hospitalizations per 100,000 Medicaid-enrolled children per year. Preventable hospitalization rates vary significantly by level of racialized economic segregation: children living in the ZIP codes that have the highest concentration of low-income, non-Hispanic Black residents have adjusted rates of 181 per 100,000 children, compared to 110 per 100,000 for children in ZIP codes that have the highest concentration of high-income, non-Hispanic white residents (p < 0.01). This pattern is driven by asthma-related preventable hospitalizations. Medicaid-enrolled children's risk of preventable hospitalizations depends on where they live, and children in economically and racially segregated neighborhoods-specifically those with higher concentrations of low-income, non-Hispanic Black residents-are at particularly high risk. It will be important to identify and implement Medicaid/CHIP and other policies that increase access to high-quality preventive care and that address structural drivers of children's health inequities.

Study: A Retrospective Claims Analysis of Preventable Hospitalizations Among Wisconsin Medicaid Beneficiaries with Dementia ‐ 2016 Through 2019

AbstractBackgroundOlder adults with Alzheimer’s Disease and Related Dementias (ADRD) are often more vulnerable to severe illness and injury than older adults without ADRD. A percentage of the causes of these hospitalizations are considered preventable with good care management. To quantify the impact of the resulting preventable hospitalizations, the Wisconsin Department of Health Services has performed an evaluation using claims data in a retrospective analysis.MethodUsing Wisconsin Medicaid claims records from the beginning of 2016 through the end of 2019, leading preventable causes of hospitalization among older Medicaid beneficiaries were identified, and rates of hospitalization for those with ADRD were compared against those without ADRD. Hospitalizations with certain diagnostic codes were considered “preventable” if timely outpatient care or simple precautions are known to resolve or avoid intensive hospital care. Direct costs to Medicaid for hospital stays were also tabulated, as well as differences by race/ethnicity and living arrangements.ResultsPreliminary aggregate results indicate higher rates of preventable hospitalization for older beneficiaries with ADRD than those without ADRD, especially for sepsis, urinary tract infections, and pneumonia. Rates of hospitalization were higher for racial and ethnic minority populations than for the white population, and an ADRD diagnosis was a larger risk factor for hospitalization in home and assisted living settings compared to a nursing home setting.ConclusionThe cost to the health of the person with dementia, the cost to the family, and health care costs associated with a few key preventable conditions is extraordinary and unnecessary. Use of this data can drive the development of policies and practices that address risk factors for these conditions that could improve the quality of life for the person living with dementia by preventing unnecessary illness and/or injury, but could also lower unnecessary costs for families and health care systems.

Preventable Hospitalization Trends Before and After the Affordable Care Act

The Patient Protection and Affordable Care Act aimed to increase the number of individuals with health insurance, which may lead to adequate primary care management and reduced rates of preventable hospitalizations. To investigate the rates of preventable hospitalization after the passing of the Affordable Care Act in 2010 and Medicaid expansion in 2014 across 26 states, a population-based study was conducted using the Healthcare Cost and Utilization Project National Inpatient Sample database from 2005-2017. A logistic regression and trend analysis was performed to assess the changes in preventable hospitalization rates over time and the impact of policy changes on the rate of preventable hospitalization. Individuals were included if they were aged between 18 and 64 years and had a preventable quality indicator International Classification of Diseases, Ninth or Tenth Revision code as determined by the Agency for Healthcare Research and Quality. More than 45 million preventable-hospitalization admissions were reported between 2005 and 2017. There was a significant decrease in preventable hospitalization rates after the passing of the Affordable Care Act from 12.0% to 10.8% (p<0.01) and from 11.5% to 10.6% (p<0.01) after Medicaid expansion. Bacterial pneumonia declined from 1.5% to 0.6% (p<0.01), along with chronic obstructive pulmonary disease and asthma in older adults from 1.9% to 1.7% (p=0.01) after the expansion. States that have not implemented Medicaid expansion should make it a priority because it may lead to a reduction in preventable hospitalization rates. Furthermore, preventable hospitalization rates may be considered a quality measure to examine the accessibility and effectiveness of primary care intervention.

The association between primary care use and potentially-preventable hospitalization among dual eligibles age 65 and over

BackgroundIndividuals dually-enrolled in Medicare and Medicaid (dual eligibles) are disproportionately sicker, have higher health care costs, and are hospitalized more often for ambulatory care sensitive conditions (ACSCs) than other Medicare beneficiaries. Primary care may reduce ACSC hospitalizations, but this has not been well studied among dual eligibles. We examined the relationship between primary care and ACSC hospitalization among dual eligibles age 65 and older.MethodsIn this observational study, we used 100% Medicare claims data for dual eligibles ages 65 and over from 2012 to 2018 to estimate the likelihood of ACSC hospitalization as a function of primary care visits and other factors. We used linear probability models stratified by rurality, with subgroup analyses for dual eligibles with diabetes or congestive heart failure.ResultsEach additional primary care visit was associated with an 0.05 and 0.09 percentage point decrease in the probability of ACSC hospitalization among urban (95% CI: − 0.059, − 0.044) and rural (95% CI: − 0.10, − 0.08) dual eligibles, respectively. Among dual eligibles with CHF, the relationship was even stronger with decreases of 0.09 percentage points (95% CI: − 0.10, − 0.08) and 0.15 percentage points (95% CI: − 0.17, − 0.13) among urban and rural residents, respectively.ConclusionsIncreased primary care use is associated with lower rates of preventable hospitalizations for dual eligibles age 65 and older, especially for dual eligibles with diabetes and congestive heart failure. In turn, efforts to reduce preventable hospitalizations for this dual-eligible population should consider how to increase access to and use of primary care.

Segregation Predicts COVID-19 Fatalities in Less Densely Populated Counties.

AimIt is well known that social determinants of health (SDoH) have affected COVID-19 outcomes, but these determinants are broad and complex. Identifying essential determinants is a prerequisite to address widening health disparities during the evolving COVID-19 pandemic.MethodsCounty-specific COVID-19 fatality data from California, Illinois, and New York, three US states with the highest county-cevel COVID-19 fatalities as of June 15, 2020, were analyzed. Twenty-three county-level SDoH, collected from County Health Rankings & Roadmaps (CHRR), were considered. A median split on the population-adjusted COVID-19 fatality rate created an indicator for high or low fatality. The decision tree method, which employs machine learning techniques, analyzed and visualized associations between SDoH and high COVID-19 fatality rate at the county level.ResultsOf the 23 county-level SDoH considered, population density, residential segregation (between white and non-white populations), and preventable hospitalization rates were key predictors of COVID-19 fatalities. Segregation was an important predictor of COVID-19 fatalities in counties of low population density. The model area under the curve (AUC) was 0.79, with a sensitivity of 74% and specificity of 76%.ConclusionOur findings, using a novel analytical lens, suggest that COVID-19 fatality is high in areas of high population density. While population density correlates to COVID-19 fatality, our study also finds that segregation predicts COVID-19 fatality in less densely populated counties. These findings have implications for COVID-19 resource planning and require appropriate attention.

Association of number of primary care physicians with preventable hospitalizations and premature deaths

Background Primary Care Physicians (PCPs) may be the first providers for the patients in a healthcare interaction, putting them in a unique position which may determine health trajectory of a patient. Assessing whether PCPs improve the overall health of a community through reducing preventable hospital stays and premature deaths may provide necessary information towards improving the health outcomes at grass roots. Methods County-level data on the number of primary care physicians, preventable hospital stays and ‘years of potential life lost’ (YPLL) were obtained from the Physician Master File data of the American Medical Association, Centers for Medicare & Medicaid Services Office of Minority Health’s Mapping Medicare Disparities data, and Center for Disease Control and Prevention’s WONDER database, respectively. We employed linear regression model to assess the association of PCP rate with preventable hospital stays and YPLL. Results Preventable hospitalization rate in the United States was 6303.4 (95% CI, 6212.5-6394.3) hospitalizations per 100,00 population per year while the average YPLL across the counties in the United States was 7792.9 (95% CI, 7697.6-7888.3) years per 100,000 population. For an increase of 1 PCP in a county, around 16 hospitalizations were prevented per 100,000 population (P=0.001) each year. Furthermore, around 14 years of life were saved per 100,000 population for every additional PCP in a county across the United States (P <0.001). Conclusion Higher number of PCPs in a county was associated with lower hospitalizations for preventable causes and lower premature deaths. Increasing PCPs may be important metric to improve overall health in a community.

Hearing Loss and Preventable Hospitalizations

Nearly half of all adults over the age of 60 years have hearing loss. Recent research suggests adults with hearing loss experience increased health care expenditures and hospitalization. However, little is known about whether these are preventable hospitalizations which may indicate poorer healthcare system engagement. In this cross-sectional analysis, we examined data from combined 2016-2018 Medicare Current Beneficiary Survey (MCBS) datasets. Participants are asked to describe their self-perceived trouble hearing. Preventable hospitalizations were defined and generated from administrative claims files based on the Agency for Healthcare Research and Quality identified conditions that should be manageable in ambulatory care settings. Multivariate regression models adjusted for demographic/socioeconomic characteristics and general health determinants were used to explore the association between trouble hearing and outcomes. The combined 2016-2018 MCBS administrative claims files included 18,814 participant-years, 49.8% reported no trouble hearing, 43.4% reported a little trouble and 6.8% a lot of trouble hearing, respectively. A higher proportion of those with a lot of trouble hearing (6.8%) experienced at least one preventable hospitalization compared to those with a little trouble hearing (3.4%) and no trouble hearing (2.5%). In a fully adjusted logistic regression model, hearing loss was associated with 1.35 times the odds of experiencing at least one preventable hospitalization per year (OR=1.35; 95% CI=1.03-1.77). Medicare beneficiaries with hearing loss experience higher rates of preventable hospitalizations. This may be due to avoidance of care due to communication barriers. Further work is needed to understand underlying reasons and whether addressing hearing loss modifies the observed association.

Abstract P178: Social Determinants Of Health And Preventable Hospitalization For Hypertension

Introduction: Social determinants of health (SDH) can have a significant impact on the risk of hypertension related hospitalization. Understanding the drivers of potentially preventable hospitalizations for hypertension improves health outcomes and reduce associated costs. Identifying place and associated SDH has implications for tailored interventions to address disparities for high burden population for preventable hypertension hospitalization as hypertension and associated cardiovascular disease. Methods: County level preventable hospitalization rates for hypertension for Virginia was obtained using prevention quality indicators published by the Agency for Healthcare Research and Quality. The Health Opportunity Index (HOI), developed by the Virginia Department of Health to identify vulnerable populations is a multivariate tool that uses complex SDH indicators of a community and comprises of 13 indices - affordability, income inequality, Townsend/Material deprivation, job participation, employment access, education, air quality, segregation, food accessibility, population density, population churning, walkability, and access to care. Principal component analysis was used to develop the composite HOI and further aggregated into simple quintiles at the county level. Step-wise multiple regression analysis was performed to explore SDH and preventable hospitalization for hypertension. Results: Material deprivation index ( r = -.44), affordability index ( r = -.42) and air quality index ( r = -.23) were found to be significantly associated with preventable hospitalization rate for hypertension. Together the model accounted for 23% of the variability in the preventable hypertension hospitalization rate ( p &gt; 0.05). Conclusions: Findings indicate that neighborhoods that spend most on housing and transportation and that are materially deprived of goods, services, amenities and resources and physical environment have higher rates of preventable hospitalization for hypertension. Areas with higher air pollution may result in hypertension. Addressing these disparities by targeted approach is one possible approach to reducing the burden of preventable hospitalization for hypertension in Virginia.

Measuring the impact of the Affordable Care Act Medicaid expansion on access to primary care using an interrupted time series approach

BackgroundThe Patient Protection and Affordable Care Act of 2010, commonly referred to as the Affordable Care Act (ACA), was created to increase access to primary care, improve quality of care, and decrease healthcare costs. A key provision in the law that mandated expansion of state Medicaid programme changed when states were given the option to voluntarily expand Medicaid. Our study sought to measure the impact of ACA Medicaid expansion on preventable hospitalization (PH) rates, a measure of access to primary care.MethodsWe performed an interrupted time series analysis of quarterly hospitalization rates across eight states from 2012 to 2015. Segmented regression analysis was utilized to determine the impact of policy reform on PH rates.ResultsThe Affordable Care Act’s Medicaid expansion led to decreased rates of PH (improved access to care); however, the finding was not significant (coefficient estimate: −0.0059, CI −0.0225, 0.0107, p = 0.4856). Healthcare system characteristics, such as Medicaid spending per enrollee and Medicaid income eligibility, were associated with a significant decrease in rates of PH (improved access to care). However, the Medicaid-to-Medicare fee index (physician reimbursement) and states with a Democratic state legislature had a significant increase in rates of PH (poor access to care).ConclusionHealth policy reform and healthcare delivery characteristics impact access to care. Researchers should continue evaluating such policy changes across more states over longer periods of time. Researchers should translate these findings into cost analysis for state policy-makers to make better-informed decisions for their constituents.Contribution to knowledgeAmbulatory care-sensitive conditions are a feasible method for evaluating policy and measuring access to primary care. Policy alone cannot improve access to care. Other factors (trust, communication, policy-makers’ motivations and objectives, etc.) must be addressed to improve access.

Comparing Preventable Acute Care Use of Rural Versus Urban Americans: an Observational Study of National Rates During 2008-2017.

Rural Americans have less access to care than urban Americans. Preventable acute care use is a marker of unmet ambulatory healthcare needs, but little is known about how such utilization has differed between rural and urban areas over time. Compare preventable emergency department (ED) visit and hospitalization rates among rural versus urban residents over the past decade. Observational study using a validated algorithm to compute age-sex-adjusted rates per 100,000 individuals of preventable ED visits and hospitalizations. Differences in overall, annual, and condition-specific rates for rural versus urban residents were assessed and linear regression was used to assess 10-year trends. Nationwide Emergency Department Sample, National Inpatient Sample, and US Census, 2008-2017. US adults, an annual average of 241.3 million individuals. Preventable ED visits and hospitalizations. Compared to urban residents, rural residents had 45% higher rates of preventable ED visits in 2008 (3003 vs. 2070 per 100,000, adjusted difference [AD]: 933; 95% CI: 928-938) and 44% higher rates of preventable ED visits in 2017 (3911 vs. 2708 per 100,000, AD: 1202; 95% CI: 1196-1208). Rural residents had 26% higher rates of preventable hospitalizations in 2008 (2104 vs. 1666 per 100,000, AD: 439; 95% CI: 434-443) and 13% higher rates in 2017 (1634 vs. 1440 per 100,000, AD: 194; 95% CI: 190-199). Preventable ED visits increased more in absolute terms in rural versus urban residents, but the percentage increase was similar (30% vs. 31%) because rural residents started at a higher baseline. Preventable hospitalizations decreased at a faster rate (22% vs. 14%) among rural versus urban residents. Observational study; unable to infer causality. Rural disparities in acute care use are narrowing for preventable hospitalizations but have persisted for all preventable acute care use, suggesting unmet demand for high-quality ambulatory care in rural areas.

Comparison of Data Science and Qualitative Approaches for Variable Selection of County‐Level Social Determinants of Health

Research ObjectiveHealth services researchers' use of social determinants of health (SDOH) variables in quantitative models is increasing, and many publicly available data sources contain scores of high‐quality, complete SDOH variables. However, determining which SDOH variables are most important to include among those available creates challenges for variable selection. One approach is relying on a conceptual framework, prior research, and intuition. But, often conceptual framework domains broadly describe “external context” or “community factors” that provide little help with identifying specific variables to use. Data science methods, particularly random forest regression, are a potential data‐driven approach for SDOH variable selection. This study compared a qualitative approach and a data‐driven approach to SDOH variable selection to identify key SDOH predictors of county‐level health outcomes.Study DesignWe constructed an initial dataset of county‐level SDOH variables compiled from the following data sources: Area Health Resources File, County Health Rankings, American Community Survey, Picture of Subsidized Households, Penn State University’s Social Capital Index, and the Food Environment Atlas. We then employed a qualitative variable selection approach using the Healthy People 2020 organizing framework for SDOH. We purposively selected 6 variables that touched on all 5 domains of the framework, had sufficient variation across counties, were relatively normally distributed, and had established associations with health outcomes in the literature. Next, we employed a data‐driven variable selection approach using random forest regression. We used 3 random forest regression models, each with a different county‐level health outcome specified, and determined the top 6 SDOH predictors driving each outcome. We used the following outcomes: premature death (days of life lost), proportion of the population reporting fair or poor health, and preventable hospitalization rate (ambulatory care sensitive conditions). We identified overlap among the 6 SDOH predictors determined from each random forest model to determine the final set of variables using the data‐driven approach. We then compared the SDOH variables determined using the data‐driven approach to those selected using the qualitative approach.Population StudiedWe included all 3142 U.S. counties in the analysis, and our dataset contained 81 SDOH variables.Principal FindingsWe selected the following SDOH variables using the qualitative approach: median household income, poverty rate, primary care physician‐to‐population ratio, social deprivation index, food environment index, and proportion of the population that reports severe housing problems. The following SDOH variables were selected using the data‐driven approach: median household income (3 models), poverty rate (2 models), proportion of the population with some college (2 models), proportion of the population who report excessive drinking (2 models), proportion of the population who identifies as American Indian or Alaskan Native (2 models), and social capital index (2 models). Two of the 6 variables selected using the qualitative approach (median household income and poverty rate) were validated by the data‐driven approach.ConclusionsRandom forest models can assist with SDOH variable selection for quantitative analysis. However, variables selected using these techniques may not align well with those selected using qualitative approaches.Implications for Policy or PracticeResearchers should consider using data science approaches to validate and compliment—rather than supplement—qualitative approaches to variable selection.

A Nationwide Study Assessing Preventable Hospitalization Rate on Mortality After Major Cardiovascular Surgery

Despite the use of various factors to measure hospital quality, most measures have not resulted in long-term improvements in patient outcomes. This study's purpose is to determine the effect of a previously unassessed measure of quality of care-a hospital's preventable hospitalization rate-on 30-day mortality at both the hospital and individual levels after three major cardiovascular surgery procedures. This is a population-based study using Taiwan's National Health Insurance database. We retrieved data from 2001 to 2014 for patients who had undergone abdominal aortic aneurysm (AAA) repair, coronary artery bypass graft, or aortic valve replacement (AVR). Preventable hospitalizations are hospitalizations for 11 chronic conditions that are considered preventable with effective primary care. The outcome was 30-day surgical mortality. Our dataset contained 65,863 patients who had undergone surgery for one of the three cardiovascular procedures. Preventable hospitalization rate was significantly associated with higher hospital mortality rates for all procedures. At the patient level, the adjusted odds of mortality after AAA repair were increased 55% (P < 0.01) for every 2% increase in the preventable hospitalization rate. For coronary artery bypass graft, preventable hospitalization was not a significant predictor of mortality, but rather patient factors and surgeon factors were significant. For AVR, the adjusted odds of mortality were increased 7% (P < 0.01) for every 1% increase in preventable hospitalization rate. High preventable hospitalization rate may serve as a hospital quality measure that could signal increased odds of mortality for selected cardiovascular procedures, especially for higher risk-lower volume procedures such as AAA repair and AVR.