Executive function and mortality in homebound elderly adults

Objectives: To determine the association between executive function and mortality in homebound elderly adults.

Design: Longitudinal study.

Setting: Four home care agencies in the Boston area.

Participants: Homebound adults aged 60 and older with 8-year follow-up for mortality (N=1,172).

Measurements: Cognitive domains including executive, memory, and language functions were evaluated at baseline. Executive function was measured using the Trail-Making Test Part B (TMT B), and subjects were divided into four subgroups from lowest to highest TMT B score. The second cross-group analyses were used to compare those who were alive and those who had died. Multivariate logistic regression analysis was used to determine whether there was an association between TMT B scores and mortality.

Results: At baseline, 436 (37.2%) homebound elderly adults had the highest TMT B scores (≥300), which indicated the slowest performance. At 8-year follow-up, 381 (32.5%) participants had died. Participants with the highest TMT B scores were more than twice as likely to have died as those with the lowest scores (0–99) (odds ratio=2.39, 95% confidence interval=1.27–4.52, P=.003) after adjusting for confounders including medical comorbidities related to death. The other cognitive domains, including memory and language, were not associated with mortality in the same model.

Conclusion: Many homebound elderly adults have multiple medical conditions, and executive function may be critical in their ability to manage their medical conditions and may affect the outcome of death.