The purpose of this study was to test the relationships between perceived self-efficacy, functional status and depression in a group of chronic heart failure patients. The conceptual model for this study was primarily derived from Bandura's social learning theory. The study is based on an exploratory, correlational design. A sample of 100 chronic heart failure patients who met the inclusion criteria were recruited from four cardiovascular clinics of major hospitals in northern Taiwan. The instruments used in this study included Jenkins' Self-Efficacy Expectation Scales, Seattle Angina Inventory, and Geriatric Depression Scale. Results demonstrate a significant inverse relationship between perceived self-efficacy and depressive symptoms (r = -.61, p < .001), a significant positive relationship between perceived self-efficacy and functional status (r = .55, p < .001), and a significant negative relationship between functional status and depressive symptoms (r = -.33, p < .001). The findings of the path analysis support the hypothesis that after controlling for disease severity, perceived self-efficacy has a direct and negative effect on depressive symptoms. An indirect effect of perceived self-efficacy on depression through functional status in heart failure patients was also revealed in the finding. All paths in the model were significant (p < .05). Results of the study can help to identify heart failure patients who are at risk for low perceived self-efficacy, low functional status, and a depressed mood. Nursing intervention to address these factors can then be developed and tested.