Population and community predictors of preventable hospitalizations in Texas
Abstract
Background: Preventable hospitalizations (PH) can be averted with proper access to preventive and outpatient primary care. The rates of PH are affected by population and community elements. Realizing the importance of these elements relative to PH rates is vital for enhancement of the safety-net care delivery organization. Research Objectives: To determine the variations in county-level PH rates in Texas and the relationship between PH and population factors and community features. Methods: The study was conducted using cross-sectional analysis. .County-level data from the Texas Health Information Collection (THCIC), Area Resource File (ARF), and Texas Indigent Health Program. Hot-spot and outlier/cluster analyses were conducted to determine the spatial variations of all PH in Texas. To determine the relationship between population factors and PH rates and community features, spatial lag regression was applied to address spatial autocorrelation in the dependent variable. Two-way interaction terms were developed to examine the relative effects of healthcare coverage and safety-net factors on rates of PH. Study Population: The study population was non-elderly adult residents from 254 counties in Texas who were hospitalized at least one time in 2011. Results: This study found that bacterial pneumonia, chronic obstructive pulmonary disease (COPD), urinary tract infection (UTI), perforated appendix, lower-extremity amputation associated with diabetes, and diabetes with long-term complications were major contributors of overall PH in Texas. The study results indicated that the county-level proportional increase in males, people in the 50-64 year age group, and Medicaid beneficiaries significantly (p-value<0. 05) increased the PH rates by 0.073%, 0.145%, and 0.095%, respectively. Contrarily, the PH rates significantly (p-value<0. 05) decreased by 0.058%, with a proportional increase in privately insured individuals by 0.460% with an increased FQHC per 10,000, and by 0.137% with an ordinal increase in the indigent health program generosity. There was a significant negative association between the rates of PH and the effect of interactions between indigent health program generosity and private health insurance, and a positive association between the rates of PH and the effect of interactions between indigent health program generosity and the uninsured. The unit increase in indigent health generosity decreased the effect of a proportional increase of private health insurance on rate of PH by 0.019% (p-value<0.05). The effect of a percentage increase in uninsured on rate of PH increased by 0.041 % (p-value<0.05) with a unit increase in indigent health generosity. Conclusion: An increase in indigent health generosity, private health insurance, and FQHC were inversely associated to PH in Texas with other variables held constant. As indigent health programs become more generous, the effect of private health insurance on the rate of PH decreased, while the effect of uninsured on rate of PH increased. These interaction findings indicate the role of more generous indigent health in granting access to uncompensated inpatient and outpatient care for the uninsured. Development of more generous hospital district indigent health programs in Borden, Red River, Franklin, and other counties with high PH rates and strengthening of the outpatient care capacity of hospital districts in counties such as Collingsworth, Marion, Houston, and San Augustine can be expected to reduce PH in Texas.
Subject Area
Public health|Public policy|Health care management
Recommended Citation
Nkem, Malachy, "Population and community predictors of preventable hospitalizations in Texas" (2014). Texas Medical Center Dissertations (via ProQuest). AAI3639362.
https://digitalcommons.library.tmc.edu/dissertations/AAI3639362