Trends in Stroke Incidence by Neighborhood Socioeconomic Status in the Greater Cincinnati/Northern Kentucky Region
Christopher Becker1, Heidi Sucharew2, David Robinson2, Brian Stamm1, Regina Royan1, Lisa Nobel2, Brett Kissela2, Robert Stanton2, Kathleen Alwell2, Daniel Woo2, Felipe De Los Rios La Rosa3, Jason Mackey4, Simona Ferioli2, Eva Mistry2, Stacie Demel2, Mary Haverbusch2, Elisheva Coleman5, Adam Jasne6, Sabreena Slavin7, Kyle B Walsh2, Michael Star8, Sharyl Martini9, Matthew Flaherty2, Opeolu Adeoye10, Pooja Khatri2, Joseph Broderick2, Dawn Kleindorfer1
1University of Michigan, 2University of Cincinnati, 3Miami Neuroscience Institute, 4Indiana University, 5University of Chicago Medical Center, 6Yale, 7University of Kansas Hospital, 8Soroka Medical Center, 9VHA Neurology, 10Washington University School of Medicine
Objective:

To evaluate whether disparities in stroke incidence by neighborhood socioeconomic status (nSES) have increased, decreased, or remained unchanged from 2010 to 2015.

Background:

Poorer neighborhood socioeconomic status (nSES) is associated with greater stroke incidence. It is unknown whether this has changed over time.

Design/Methods:

The Greater Cincinnati/Northern Kentucky region includes a population of approximately 1.3 million, similar to the US population in sociodemographics and percent Black race. All hospitalized first-ever ischemic and hemorrhagic stroke cases in the region were ascertained in calendar years 2010 and 2015, and confirmed by physician review. Patient home addresses were geocoded using DeGAUSS. Neighborhood SES was estimated by the percentage below poverty in each census tract. Census tract percentage below poverty, population estimates, and demographic information were obtained from American Community Survey 5-year data. Neighborhood stroke incidence rates were adjusted for age, sex, and race using Poisson regression models. An nSES-by-year interaction term was included to evaluate whether the impact of nSES on stroke incidence changed between the study periods.

Results:

There were a total of 1,968 patients with incident stroke in 2010, and a total of 2,125 in 2015. There was no change in overall stroke incidence between the study periods (p-value=0.43). In each study period, poorer nSES was associated with greater stroke incidence after adjusting for covariates (p<0.01 for each year). The difference in incidence by nSES remained unchanged across study periods (nSES-by-year interaction p-value=0.37).

Conclusions:

Poorer nSES is associated with greater stroke incidence, and the gap between the richest and poorest neighborhoods has remained unchanged from 2010 to 2015. Further efforts are needed to address SES-based disparities in stroke incidence.

10.1212/WNL.0000000000205752