Longitudinal Psychosocial and Sleep Assessments for Long-Term Cardiovascular Risk Stratification After First Myocardial Infarction in Rural India

OBJECTIVES: The risk of myocardial infarction (MI) evolves beyond the acute phase, and in-hospital tools, such as the Global Registry of Acute Coronary Events (GRACE) score, may not fully capture long-term vulnerability. We evaluated whether simple longitudinal clinical and patient-reported measures improved the prediction of major adverse cardiovascular events (MACE). SUBJECTS AND METHODS: In this prospective cohort study, 351 patients with first MI were enrolled at a rural tertiary center in South India, of whom 319 formed the analytic cohort. MACE was defined as cardiovascular death, non-fatal MI, stroke, heart failure hospitalization, or urgent coronary revascularization, as ascertained through hospital records and structured follow-up. Baseline clinical, biochemical, echocardiographic, and interviewer-administered psychosocial (PHQ-9 and GAD-7) and sleep variables were collected during hospitalization and follow-up (1, 6, and 12 months). Cox models were developed in stages, with the GRACE score as a baseline anchor and follow-up variables incorporated using the last available observations. Model performance was assessed using Harrell's concordance index (C-index), time-dependent area under the curve (AUC), and bootstrap validation methods. RESULTS: Over a median follow-up of 5.8 years, 94 patients (29.5%) experienced MACE. The baseline model modestly improved discrimination over GRACE alone (C-index, 0.622 vs. 0.556). The extended model showed higher discrimination (apparent C-index 0.684; optimism-corrected 0.660). Follow-up sleep quality, psychosocial symptoms, glycemic measures, and global longitudinal strain contributed to the model, although the estimates were imprecise. CONCLUSION: Longitudinal follow-up measures provided modest incremental prognostic information beyond a GRACE-based benchmark. Findings are exploratory and require external validation before clinical application.