Objectives To examine the relationship between psychological well‐being and depression in older adults and the relative contribution these psychological factors have on risk of functional disability, frailty, and mortality. Methods This is a secondary analysis of 1668 community‐dwelling older adults without dementia who participated in the second wave of the Canadian Study of Health and Aging. Baseline assessments of psychological well‐being (Ryff scale) and depression (Geriatric Depression Scale; GDS) were collected. At 5‐year follow‐up, mortality data were collected; frailty and disability in activities of daily living were evaluated using the frailty index (FI) and the Lawton‐Brody scale, respectively. Results Area under the receiver‐operating characteristic curve indicated that GDS and Ryff scores were able to independently discriminate whether individuals were considered frail (C = 0.66; C = 0.59, respectively), had limitations in basic (C = 0.64; C = 0.57, respectively) or instrumental (C = 0.70; C = 0.57, respectively) activities of daily living, or had died (C = 0.63; C = 0.57) at follow‐up (all P < 0.01). Regression models in which the Ryff and GDS were included in the same model demonstrated that the GDS significantly predicted frailty, disability, and mortality, whereas the Ryff effect was not significant. Mediation analysis determined that the effect of psychological well‐being on adverse outcomes was fully mediated by depression. Conclusions Our results suggest that although both depression and psychological well‐being appear to modulate risk for adverse physical health outcomes, depression mediates this relationship. Detecting and treating depressive symptoms should be of high priority in older patients to mitigate risk of future physical health adversities including mortality. Copyright © 2016 John Wiley & Sons, Ltd.