Allostatic load measurement and cognitive function among midlife and older Black and White adults in the health and retirement study

Allostatic load (AL), a composite of biological measures, is frequently used to capture the physiological burden of chronic stress and is associated with cognitive decline. Black Americans are more likely to develop Alzheimer's disease and related dementias than White Americans, but racial disparities in the AL-cognition relationship have been under explored. Additionally, AL is not consistently measured in literature. The aim of this research was to examine racial variation of the AL-cognition relationship using various AL scoring methods. Data were from the 2006-2016 waves of the Health and Retirement Study, and the sample included 29,809 person-waves of White adults and 7612 person-waves of Black adults. Cognitive function was measured using a modified Telephone Interview for Cognitive Status (mTICS). AL calculation methods included quartile, z-score, and clinical cut-offs. Each measure defined biomarker levels as either high risk or low risk. For 'one-tailed' measures, higher values indicated higher risk for all biomarkers except HDL, for which lower values indicated higher risk. For quartile and z-scores, additional two-tailed measures were calculated which define risk at both unusually low and high biomarker levels. Mixed models were run examining both within-person and between-person change in allostatic load measures in relation to cognitive function. Two-tailed measures more strongly predicted cognitive function both within- and between-persons than traditional quartile or clinical cut-offs, particularly in the Black sample. Findings suggest that nonlinear relationships should be given greater consideration when examining AL, particularly in diverse populations of older adults.