Abstract
We tested the efficacy of docosahexaenoic acid (DHA) supplementation on aggressive behavior among people with dementia. We used secondary data from a randomized, double-blind, placebo-controlled trial. Analyses included 402 individuals with dementia. We compared group differences in mean aggressive behavior change at baseline vs six (primary), 12, and 18 months (secondary outcome). We used independent samples t-tests with the aggressive behavior item from the Neuropsychiatric Inventory Questionnaire. We stratified group differences by ApoE-ε4 status, gender, baseline aggressive behavior, and dementia severity. All tests favored DHA but no differences reached statistical significance. For example, the group differences between baseline and month 6 aggressive behavior were 0.11 (p=0.22). Group differences nearly reached statistical significance with clinically important effect sizes when excluding participants without baseline aggressive behavior. Despite obtaining null results, these trends suggest avenues for future research that includes participants with higher baseline aggressive behavior levels and more sensitive outcome scales.