Xiaoxiao Zhou, PhD
Assistant Professor
Department of Biostatistics
School of Public Health
The University of Alabama at Birmingham
Differences in cognitive decline across sociodemographic groups are well documented, but their estimated magnitude is inconsistent from study to study. Part of that inconsistency reflects who remains available for testing over time. Participants who die or drop out are a recognized source of bias, and statistical methods exist to account for them. Far less attention has been paid to test refusal, which occurs when participants stay enrolled but decline cognitive testing at a given visit, often refusing at one visit and completing testing at the next. This pilot project uses data from the REasons for Geographic And Racial Differences in Stroke (REGARDS) study to examine whether refusal signals underlying cognitive decline and mortality risk, and whether accounting for refusal alongside death changes what we conclude about cognitive decline by education, race, and gender.


