Background Self-care is vital for patients with heart failure to maintain health and quality of life, and it is even more vital for those who are also affected by diabetes mellitus, since they are at higher risk of worse outcomes. The literature is unclear on the influence of diabetes on heart failure self-care as well as on the influence of socio-demographic and clinical factors on self-care. Objectives (1) To compare self-care maintenance, self-care management and self-care confidence of patients with heart failure and diabetes versus those heart failure patients without diabetes; (2) to estimate if the presence of diabetes influences self-care maintenance, self-care management and self-care confidence of heart failure patients; (3) to identify socio-demographic and clinical determinants of self-care maintenance, self-care management and self-care confidence in patients with heart failure and diabetes. Design Secondary analysis of data from a multicentre cross-sectional study. Setting Outpatient clinics from 29 Italian provinces. Participants 1192 adults with confirmed diagnosis of heart failure. Methods Socio-demographic and clinical data were abstracted from patients’ medical records. Self-care maintenance, self-care management and self-care confidence were measured with the Self-Care of Heart Failure Index Version 6.2; each scale has a standardized score from 0 to 100, where a score <70 indicates inadequate self-care. Multiple linear regression analyses were performed. Results Of 1192 heart failure patients, 379 (31.8%) had diabetes. In these 379, heart failure self-care behaviours were suboptimal (means range from 53.2 to 55.6). No statistically significant differences were found in any of the three self-care measures in heart failure patients with and without diabetes. The presence of diabetes did not influence self-care maintenance (p = 0.12), self-care management (p = 0.21) or self-care confidence (p = 0.51). Age (p = 0.04), number of medications (p = 0.01), presence of a caregiver (p = 0.04), family income (p = 0.009) and self-care confidence (p < 0.001) were determinants of self-care maintenance. Gender (p = 0.01), number of medications (p = 0.004) and self-care confidence (p < 0.001) were significant determinants of self-care management. Number of medications (p = 0.002) and cognitive function (p < 0.001) were determinants of self-care confidence. Conclusions Self-care was poor in heart failure patients with diabetes mellitus. This population needs more intensive interventions to improve self-care. Determinants of self-care in heart failure patients with diabetes mellitus should be systematically assessed by clinicians to identify patients at risk of inadequate self-care.