Leveraging real-world data for population health outcomes research: evaluating all-cause mortality in stroke hospitalization using a statewide health information exchange

Inpatient stroke mortality remains high at 2–5% in the U.S. This retrospective cohort study evaluated all-cause mortality in ischemic stroke hospitalizations using real-world data from a mature health information exchange (HIE) over a 17-year period (2004–2020). We analyzed 88,397 unique inpatient encounters corresponding to 62,271 adult patients aged 45 and over. Ischemic stroke hospitalizations were identified using diagnostic codes. We modeled all-cause inpatient mortality using multivariable logistic regression, adjusting for socio-demographics, Charlson Comorbidity Index, and other relevant factors. Time was included as year fixed effects. Overall, between 2004 and 2020 all-cause inpatient ischemic stroke mortality was 2.36% in Indiana. Compared to 2004, the odds of inpatient mortality among stroke patients were significantly lower for most years. Rural residence was associated with higher adjusted odds (OR 1.31, 95% CI: 1.17, 1.47) for all-cause inpatient ischemic stroke mortality. Compared to commercial insurance, Medicare insurance (OR 1.80, 95% CI: 1.49, 2.17) was associated with higher adjusted odds of all-cause inpatient ischemic stroke mortality. Our findings suggest that inpatient ischemic stroke mortality declined through 2014 relative to 2004 in Indiana. Real-world evidence from our study may inform care delivery especially for rural residents and patients on Medicare in Indiana.