Exploring U.S. Stroke Epidemiology Across Demographic and Urban-Rural Divides

Thirty-day risk of readmission after ischemic stroke is significantly influenced by patients’ race and medical insurance but not by age, sex or household income, according to a large database analysis presented by Cleveland Clinic researchers at the 2020 International Stroke Conference this week.

A second nationwide database analysis reported at the conference by some of the same researchers finds that recent increases in the prevalence of risk factors for ischemic stroke are particularly pronounced among rural Americans compared with their urban counterparts.

“The observations from these analyses may help target interventions for stroke readmission reduction and stroke prevention efforts,” says Cleveland Clinic stroke neurologist Ken Uchino, MD, a co-investigator in both studies.

How social determinants of health impact stroke readmissions

The first study examined data on patients hospitalized for acute ischemic stroke in Florida and New York from February 2012 to November 2013. The analysis drew on the State Inpatient Databases for these two large states and linked these data with U.S. census data on social determinants of health at the ZIP code level. The objective was to identify how various patient factors were associated with risk of hospital readmission within 30 days of discharge following ischemic stroke.

Multivariate logistic regression analysis was performed to assess associations after adjusting for patient demographics, socioeconomic characteristics, comorbidities, proxies for stroke severity and complications during initial hospitalization.

Among the 127,290 patients included in the analysis, the overall readmission rate at 30 days was 23%.

Multivariable analysis revealed that risk for readmission was significantly associated with the following:

• Race, with black and Hispanic patients having higher risk than white patients
• Insurance status, with patients on Medicaid having higher risk than those on Medicare and with self-pay patients having lower risk

In contrast, age, sex and household income (at the ZIP code level) had no significant effect on readmission risk in both unadjusted and multivariable analyses.

“Readmission after hospitalization for stroke is an important topic, as it represents a huge burden for all involved — patient, family, hospital and payer,” says the study’s principal investigator, Shumei Man, MD, PhD, who serves as Medical Director of the Thrombectomy-Capable Stroke Center at Cleveland Clinic Fairview Hospital. “These findings can guide refinements of how we devote resources to prevent readmission.”

Stroke risk in urban vs. rural geographies

The second analysis by the Cleveland Clinic group used data from the National Inpatient Sample, a nationally representative database of hospitals covering more than 97% of the U.S. population.

The researchers harvested comorbid risk factors for stroke among all patients hospitalized for ischemic stroke in the database from 2000 to 2016. They calculated crude and age- and sex-standardized prevalence estimates for each risk factor over early (2000-2008) and late (2009-2016) segments of the study period. Prevalence rates were compared between the two time segments and by patients’ geographic location, which was classified as either urban (metropolitan areas with ≥ 1 million population) or rural (all other locations). Regression models were used to assess risk factor trends according to time period, location and interactions between the two.

Among the key findings:

• The prevalence of stroke risk factors rose significantly from 2000-2008 to 2009-2016, with most risk factors increasing in both urban and rural locations.
• On crude analysis, rates of increase in six risk factors — hypertension, hyperlipidemia, diabetes mellitus, coronary artery disease, smoking and alcohol use — were significantly higher among patients in rural locations versus those in urban locations.
• After standardization by age and sex, rates of increase in four risk factors — hypertension, hyperlipidemia, coronary artery disease and smoking — remained significantly higher among rural patients.

“These data show that there is a widening urban-rural divide in stroke risk factor prevalence,” notes the study’s senior author, Dr. Uchino, who serves as Head of Research and Education in Cleveland Clinic’s Cerebrovascular Center. “This geographic divide is a timely and relevant topic as we consider how to better address the social determinants of health in efforts to counter the overall rise in stroke risk factors across the U.S. population.”