Cost effectiveness of ovarian reserve testing in in vitro fertilization: a Markov decision-analytic model.

OBJECTIVE To compare the cost effectiveness of ovarian reserve testing in in vitro fertilization (IVF). DESIGN A Markov decision model based on data from the literature and original patient data. SETTING Decision analytic framework. PATIENT(S) Computer-simulated cohort of subfertile women aged 20 to 45 years who are eligible for IVF. INTERVENTION(S) [1] No treatment, [2] up to three cycles of IVF limited to women under 41 years and no ovarian reserve testing, [3] up to three cycles of IVF with dose individualization of gonadotropins according to ovarian reserve, and [4] up to three cycles of IVF with ovarian reserve testing and exclusion of expected poor responders after the first cycle, with no treatment scenario as the reference scenario. MAIN OUTCOME MEASURE(S) Cumulative live birth over 1 year, total costs, and incremental cost-effectiveness ratios. RESULT(S) The cumulative live birth was 9.0% in the no treatment scenario, 54.8% for scenario 2, 70.6% for scenario 3 and 51.9% for scenario 4. Absolute costs per woman for these scenarios were €0, €6,917, €6,678, and €5,892 for scenarios 1, 2, 3, and 4, respectively. Incremental cost-effectiveness ratios (ICER) for scenarios 2, 3, and 4 were €15,166, €10,837, and €13,743 per additional live birth. Sensitivity analysis showed the model to be robust over a wide range of values. CONCLUSION(S) Individualization of the follicle-stimulating hormone dose according to ovarian reserve is likely to be cost effective in women who are eligible for IVF, but this effectiveness needs to be confirmed in randomized clinical trials.