Abstract P214: Twenty-five Years Trends in Stroke Incidence in the Atherosclerosis Risk in Communities (ARIC) Study
Introduction: Prior studies have shown a decrease in stroke mortality rates, but data on validated stroke incidence and trends by race are more limited. We aimed to study trends in stroke incidence by age and race in the Atherosclerosis Risk in Communities (ARIC) population from 1987 to 2010.
Hypothesis: We assessed the hypothesis that total stroke incidence rates are decreasing over calendar time in all age groups, and in whites and African-Americans.
Methods: ARIC participants were recruited in 1987-1989 and followed until 2010. A computer algorithm and physician reviewers identified and adjudicated definite and probable stroke as ischemic, intracerebral hemorrhage and subarachnoid hemorrhage (total stroke). A person-years (PY) approach using 5-year periods for age and calendar time was used for studying rates of total stroke incidence and trends. Poisson regression modeled calendar time in 5-year periods and linearly per decade, adjusted for 5-year age group.
Results: From 1987 to 2010, 994 strokes occurred over 274,837 PY in 14,145 participants free of stroke at baseline. Total stroke incidence rates by age and calendar time increased with age and generally decreased with time (Table). Poisson regression adjusted for age group showed incidence rate ratios of 2.28 (95% CI 2.01-2.59) for African-Americans, compared to whites. Using the 1992-1996 as reference period, incidence rate ratios were 0.90 (0.70-1.16) for 1987-1991, 0.84 (0.69-1.02) for 1997-2001, 0.69 (0.56-0.85) for 2002-2006, and 0.59 (0.46-0.75) for 2007-2010. Analysis stratified by race shows an incidence rate ratio decrease per decade of -27% (-13% to - 39%) in whites and -22% (-5% to -34%) in African-Americans.
Conclusions: Validated total stroke incidence rates decreased over calendar time from 1992 to 2010 in the ARIC cohort in both whites and African-Americans, after taking the aging of the cohort into account. Cohorts like ARIC only represent a limited age range but provide an ability to use validated events in updating stroke incidence trends.
Author Disclosures: S. Koton: None. A.L.C. Schneider: None. W.D. Rosamond: None. R.F. Gottesman: None. J. Coresh: None.
- © 2014 by American Heart Association, Inc.