Monte Carlo–based Estimates
Published online by Cambridge University Press: 04 May 2001
Objectives: This article reports a cost-effectiveness analysis of targeted screening for abdominal aortic aneurysm (AAA). A major emphasis was on the estimation of distributions of costs and effectiveness.
Methods: We performed a Monte Carlo simulation using C programming language in a PC environment. Data on survival and costs, and a majority of screening probabilities, were from our own empirical studies. Natural history data were based on the literature.
Results: Each screened male gained 0.07 life-years at an incremental cost of FIM 3,300. The expected values differed from zero very significantly. For females, expected gains were 0.02 life-years at an incremental cost of FIM 1,100, which was not statistically significant. Cost-effectiveness ratios and their 95% confidence intervals were FIM 48,000 (27,000–121,000) and 54,000 (22,000–∞) for males and females, respectively. Sensitivity analysis revealed that the results for males were stable. Individual variation in life-year gains was high.
Conclusions: Males seemed to benefit from targeted AAA screening, and the results were stable. As far as the cost-effectiveness ratio is considered acceptable, screening for males seemed to be justified. However, our assumptions about growth and rupture behavior of AAAs might be improved with further clinical and epidemiological studies. As a point estimate, females benefited in a similar manner, but the results were not statistically significant. The evidence of this study did not justify screening of females.