PCV98 - Statin Cost-Effectiveness in Patients with Previous Coronary Heart Disease: A Systematic Review of the Cost-Effectiveness Analysis Derived from Single Randomised Clinical Trials

Abstract

Large randomized clinical trials (RCT) evidenced the benefits of statins in reducing major cardiovascular events in patients with established coronary heart disease (CHD). These multinational trials are likely to provide internally valid evidence. Cost-effectiveness analysis based on single trials costs and effects are common and represent the potential net benefit of an intervention in a well-controlled environment. The aim of this study is to systematically review the cost-effectiveness studies based on statins single trials in patients with previous CHD. We searched to identify all literature relating to the cost-effectiveness of statins in the secondary prevention in patients with established CHD. Only studies with the effectiveness data extracted from a single RCT and clinical outcomes such as quality assessment, mortality or cardiovascular events rate were included. The cost per QALY was classified according to the WHO, following three categories of cost-effectiveness, Highly cost-effective, Cost-effective and Not cost-effective, adjusted with GDP per capita based on purchasing power parity (constant 2005 international USD). Twenty-one studies were included in the final analysis, covering a period range from 1996 to 2009. 7 large RCTs represented the origin of efficacy data. Most of studies assumed a full compliance, the Markov models were used in 11 out 21 studies. Time horizon ranged from 5 years to time life, with 10 years being the predominant choice. 9 studies performed a cost-utility analysis and showed the average cost per QALY, 8 of them classified as highly cost-effective and 1 cost-effective. Cost per QALY was sensitive for drug price, time horizon and event rates, 6 of this models models worked with composite endpoints. Statins are highly cost-effective in patients with CHD when effect size came from single well designed RCTs. Models heterogeneity and composite endpoints can decrease the robustness of the results.