Abstract TP184: Urban-rural Influence On Readmission After Acute Ischemic Stroke In Younger Adults

Abstract

Introduction: Complications and mortality after ischemic stroke are higher in rural areas. Identifying predictors of readmission is important, especially for younger patients who represent a growing share of new strokes, but studies differ in how they define rurality. Hypothesis: There is an association between urban-rural influence and readmission after acute ischemic stroke among younger adults, and this association depends on how urban-rural influence is defined. Methods: Data from the AHRQ State Inpatient Databases for ischemic stroke survivors aged 19-64 years during 2012-2014 were analyzed for six states: Arkansas, Florida, Georgia, Maryland, New Mexico, and Washington. In-state hospitalization within 30 days of discharge was deemed a readmission, excluding admissions for rehabilitation, psychiatry, or cancer treatment. A four-level urban influence code was used as the patient location variable. To further explore the impact of how rurality was defined, this variable was dichotomized three ways: large metropolitan vs. all others, metropolitan vs. non-metropolitan, and rural vs. all others. Chi-square tests were used to determine if planned and unplanned readmission were associated with each urban-rural variable to determine which variables to advance to multivariable analysis. Logistic regression models were then fitted for both planned and unplanned readmission, using the large metropolitan variable and adjusting for length of stay of the first hospitalization, age, race, sex, payer type, and number of chronic conditions. Results: Patients who lived in a large metropolitan area compared to those who did not had 0.885 (95% CI: 0.697, 1.122) times the odds of planned readmission when controlling for covariates; however, this difference was not statistically significant. The odds of unplanned readmission, however, were 1.071 times higher (95% CI: 1.004, 1.142) if the patient lived in a large metropolitan area compared to those who did not, controlling for covariates. Conclusions: Urban residence can be described as an independent predictor of unplanned readmission for younger adult ischemic stroke survivors. However, researchers should be attentive to the impact of how urban-rural influence is operationalized.