How consistent are cost-effectiveness estimates of a cardiovascular polypill strategy for the secondary prevention of cardiovascular disease across different cardiovascular risk equations?

Abstract

Abstract Funding Acknowledgements Type of funding sources: Private company. Main funding source(s): Ferre Internacional Background Risk equations (RE) are crucial to individualise estimates and properly adjust preventive treatments in patients with previous cardiovascular (CV) disease. RE are also routinely incorporated into health economic assessments but it is unknown if the cost-effectiveness results vary according to the RE applied. Purpose To determine the cost-effectiveness of a CV polypill (ASA 100mg, atorvastatin 20/40mg and ramipril 2.5/5/10mg) strategy compared to usual practice of combining monocomponents in the prevention of recurrent events in patients with previous coronary heart disease (CHD) or stroke applying two different CV RE: SMART and FRAMINGHAM, respectively. Methods A Markov cost-effectiveness model (1-year cycles; 4 health states: stable disease, subsequent CHD, subsequent stroke; death; payer perspective; direct medical costs; lifetime horizon; 4% discount rate) was developed for Portugal. Transition probability between health states was based on the SMART RE and an adaptation for secondary CV prevention of the FRAMINGHAM RE, respectively. Cost-effectiveness was calculated for a mixed cohort of secondary prevention patients (weighed post-CHD: 57.9%; post-stroke: 42.1%). Systematic literature reviews, Portuguese registries, mortality tables and official reports ware used to identify effectiveness, epidemiological, costs and utility data. Outcomes were costs (€, 2020) per life year (LY) and Quality Adjusted Life Year (QALY) gained. One-way (OWA) and probabilistic (PSA) sensitivity analyses tested the consistency of results. Assumptions were validated by experts. Results Applying the SMART RE, the incremental cost-effectiveness ratio (ICER) is 1,555€/LY gained and the incremental cost-utility ratio (ICUR) is 1,785€/QALY gained for the polypill strategy. The incremental costs of adopting the polypill strategy are 171,378€. Recurrent CV events (550.68 vs 642.13) and CV deaths (106.05 vs 122.81) are also less frequent with the polypill strategy compared with monocomponents. Assuming a willingness-to-pay (WTP) threshold of 30.000 €/QALY gained, there is a 77.80% probability for the polypill strategy to be cost-effective and 43.00% chances to be costs saving when used in a mixed cardiovascular and cerebrovascular disease population. Applying the FRAMINGHAM RE, the ICER is 998€/LY gained and the ICUR is 1,242€/QALY. The incremental costs amount 175,122€. Recurrent CV events (452.66 vs 563.48) and CV deaths (104.77 vs 127.32) are less frequent with the polypill strategy. The PSA shows a 99.5% probability for the polypill strategy to be cost-effective and 46.8% chances to be costs saving. Conclusion Both risk equations result in comparable results on the cost-effectiveness of interventions for the secondary prevention of CV disease. The polypill strategy remains cost-effective compared to the common practice of using individual monocomponents concomitantly, reducing recurrent CV events at a moderately higher cost.