OBJECTIVES: Non-adherence to cardiovascular disease (CVD) treatments leads to suboptimal health outcomes and increased healthcare and societal costs. We assessed the long-term effects of adherence to CVD and diabetes medications using population data and microsimulation modelling. METHODS: We developed a CVD microsimulation model using individual participant data from the SIDIAP database (2012-2021) for 152,117 adults who received new prescriptions for antihypertensive, lipid-lowering, oral glucose-lowering or antiplatelet treatments in Catalonia between January 2012 and December 2013. Model inputs included demographic and clinical characteristics, medication adherence and cardiovascular events. Costs (€, 2025) and treatment effects were sourced from the literature, and utilities were estimated using national population-based surveys. Model validity was assessed by comparing simulated and observed cumulative incidences over 8 years. The model simulated life-years (LYs), quality-adjusted life-years (QALYs) and healthcare and societal costs under three scenarios: non-adherent, observed adherence and full adherence. We estimated the maximum per-patient cost at which adherence-enhancing interventions would remain cost-effective. RESULTS: Simulated cumulative incidences of cardiovascular events and all-cause death closely matched observed data. Improved adherence increased survival by 0.19-0.58 years and QALYs by 0.25-0.70, while increasing lifetime healthcare costs by €2,431-€8,093 per patient. The additional cost per QALY ranged from €8,946 to €12,614 per QALY, indicating that improving adherence is likely to be a cost-effective if achieved at additional cost of up to €4,041 to €10,098 per patient. CONCLUSIONS: Long-term extrapolation of real-world data using microsimulation modelling shows that optimising adherence to CVD and diabetes medications can enhance health outcomes cost-effectively.