To determine if distinct trajectories of coronary heart disease (CHD) self-care behaviours could be identified, linked to differences in quality of life (QoL), and predicted based on baseline characteristics. A secondary analysis of a prospective, longitudinal, observational study. Patients with CHD answered questionnaires at study enrolment and six months later: Self-Care of Coronary Heart Disease Inventory (three subscales: maintenance, management, and confidence, scored 0-100, higher score = better self-care), Hospital Anxiety and Depression Scale, 12-Item Short Form Survey, 16-Item European Health Literacy Survey Questionnaire, and CHD Education Questionnaire. Latent growth mixture modelling was used to identify distinct self-care trajectories over time. On average, patients (n = 430, mean age 64.3 ± 8.9, 79% male) reported inadequate self-care (maintenance 61.6 ± 15.4, management 53.5 ± 18.5) at enrolment. Two distinct trajectories of self-care behaviours were identified: first, an 'inadequate-and-worsening' (IN-WORSE) trajectory (57.2%), characterized by inadequate self-care, improvement in maintenance (4.0 ± 14.5-point improvement, P < 0.001), and worsening of management over time (6.3 ± 24.4-point worsening, P = 0.005). Second, an 'inadequate-but-maintaining' (IN-MAIN) trajectory (42.8%), characterized by inadequate self-care, improvement in maintenance (5.0 ± 16.2-point improvement, P < 0.001), and stability in management over time (0.8 ± 21.9-point worsening, P = 0.713). In comparison, patients in the IN-WORSE trajectory had less favourable characteristics (including lower health literacy, knowledge, confidence) and significantly lower QoL. Not attending rehabilitation (OR 2.175; CI 1.020-4.637, P = 0.044) and older age (OR 0.959; CI 0.924-0.994, P = 0.024) predicted (IN-WORSE) trajectory inclusion. Two self-care trajectories were identified, both suboptimal. Rehabilitation predicted membership in the more favourable trajectory and some positive characteristics were identified among patients in that group. Therefore, interventions supporting these factors may benefit patients' self-care and QoL.