Anxious, Depressed, and Planning for the Future: Advance Care Planning in Diverse Older Adults

To determine whether depression and anxiety are associated with advance care planning (ACP) engagement or values concerning future medical care.

Cross-sectional.

English- and Spanish-speaking patients, aged 55 years and older, from a San Francisco, CA, county hospital.

Depression was measured by the Patient Health Questionnaire 8-item scale, and anxiety was measured by the Generalized Anxiety Disorder 7-item scale, using standardized cutoffs of 10 or more for moderate-to-severe symptoms. ACP engagement was measured using validated surveys of ACP behavior change (e.g., self-efficacy and readiness; mean five-point Likert score) and ACP actions (e.g., ask, discuss, and document wishes; 0- to 25-point scale), with higher scores representing higher engagement. In addition, we asked a question about valuing life extension ("some health situations would make life not worth living"). We used adjusted linear and logistic regression.

Mean age of 986 participants was 63 years, 81% were non-White, 39% had limited health literacy, 45% were Spanish speaking, 13% had depression, and 10% had anxiety. After adjustment for demographic and health status variables, participants who were depressed versus not depressed had higher ACP behavior change scores (0.2 points; 95% confidence interval (CI) = 0.06-0.38; P = .007), higher ACP action scores (1.5 points; 95% CI = 0.51-2.57; P = .003), and higher odds of not valuing life extension (odds ratio (OR) = 2.5; 95% CI = 1.5-4.3; P < .001). Results were similar in participants with versus without anxiety (ACP behavior change: 0.2 points; 95% CI = 0.05-0.40; P = .01; ACP action scores: 1.2 points; 95% CI = 0.14-2.32; P = .028; odds of not valuing life extension: OR = 2.3; 95% CI = 1.3-3.9; P = .004).

Depression and anxiety were associated with greater ACP engagement and not valuing life extension. Although the direction of association between ACP engagement and values with anxiety and depression cannot be determined in this cross-sectional study, these conditions may influence ACP preferences. Future studies should assess whether changes in anxiety or depression affect ACP preferences over time.