Bidirectional Associations Between End-of-Life Communication and Depressive Symptoms in Older Adults

Context: Advance care planning (ACP) may prepare older adults for future health problems including cognitive and physical decline. Depressive symptoms are observed among those engaging in ACP. However, the directionality of the association between ACP engagement and depressive symptoms remains uncertain. Objectives: To investigate bidirectional associations between engaging in ACP and depressive symptoms in older adults at a within-person level after accounting for cognitive and physical decline. Methods: A longitudinal study design using random intercept cross-lagged panel models. Community-dwelling older adults from the Longitudinal Aging Study Amsterdam (LASA) with three assessments at T0 (2015–2016), T1 (2018–2019), and T2 (2021–2022). Discussion of end-of-life wishes with a physician was used as a proxy measure for ACP engagement. Results: Participants included 1669 older adults (mean [SD] age, 69.7 [7.7] years at T0; 923 [55.3%] were women). Poor cognition at T0 was associated with having discussed end-of-life wishes at T1 (β = −0.12 [95% CI, −0.23 to −0.01]), that was prospectively associated with increased depressive symptoms (0.12 [0.01 to 0.24]), decreased cognition (−0.13 [−0.22 to −0.04]), and decreased physical performance (−0.16 [−0.27 to −0.04]) at T2. Conclusion: Older adults appear at risk for having depressive symptoms following discussing end-of-life wishes with their physicians. End-of-life communication may reflect an ongoing trajectory of cognitive decline which could result in further functional decline and increased depressive symptoms. A better understanding of how and which elements of end-of-life discussions are associated with the mood of older adults will provide implications for the implementation of ACP.