Ischemic Stroke: Epidemiologic Research of Primary and Secondary Prevention

Abstract

With over 7 million stroke survivors in the United States, characterizing and understanding risk factors and treatment gaps in stroke primary and secondary prevention is important. Using data from the Atherosclerosis Risk in Communities (ARIC) study we examined risk factors for ischemic stroke (IS) and atherosclerotic cardiovascular disease (ASCVD). Using the MarketScan health claims data, we explored trends and factors in post-stroke statin treatment. Manuscript 1 assessed the association between endogenous sex hormones and IS and silent cerebrovascular disease markers. Endogenous hormones were associated with higher odds of silent brain infarcts and microhemorrhages and reduced white matter hyperintensities. However, there was no association between endogenous hormone levels and IS. Manuscript 2 evaluated and compared four different statistical methods for atherosclerotic (ASCVD) and global (gCVD) cardiovascular disease risk prediction among the elderly. Improvement in elderly ASCVD risk prediction, which includes IS, may provide insights into the prevention of disease in late life. We found that the machine learning methods for CVD risk prediction did not improve the model performance compared to the risk prediction using Cox Proportional Hazards model. All methods were suboptimal (area under the receiving operating curve <0.7) for CVD risk prediction among the elderly. The third manuscript examined trends and factors associated with post-stroke statin treatment. We found that statin fills among stroke survivors within 30-days of hospital discharge increased from 2013 to 2019 overall and by geographical and sex subgroups. Women and individuals with co-morbidities of CVD were less likely to fill a statin prescription within 30-days of an IS hospital discharge. Overall, this dissertation extends our understanding of risk factors and prediction for IS in the middle-aged and elderly. We identify subgroups who may have sub-optimal rates of filling a statin prescription after stroke.