RESUMO
OBJECTIVE: To conduct an economic evaluation estimating the cost per live birth after controlled ovarian stimulation (COS) using Follitropin Alpha (Gonal-F), in the Greek National Health System setting. A secondary objective was to predict the live birth rateof the In Vitro Fertilization (IVF) procedure. METHODS: A single arm, multi-center, prospective, non-interventional study was conducted on which economic, efficacy and safety data were collected by six of the largest IVF centers. The participants were 350 female patients. Three statistical methods were employed for the analysis of the study outcomes, namely (a) Generalized Linear Modeling for the estimation of the costs of IVF treatment, (b) multivariable logistic regression and (c) an Artificial Neural Network (ANN) model for live birth prediction. RESULTS: The mean total cost of IVF therapy per patient was estimated at 3728 (95% CI: 3679-3780), while the total cost per live birth was 14,872 (95% CI: 12,441-17,951). The live birth rate after 3 complete IVF cycles was estimated at 22.9%, while the percentage of those suffering from OHSS was limited at 0.57%. In logistic regression, the Ovarian Sensitivity Index (OSI) was a factor found to be positively associated with live birth (OR 7.39, 95% CI: 1.84-29.71). For the ANN, important predictors included number of gestational sacs and the duration of infertility. CONCLUSION: The present study constitutes the largest single-arm study based on real data in Greece to date. The cost of IVF treatment and the cost per live birth are not insignificant in this NHS setting. The live birth rate, cost per oocyte, and the cost per live birth are in line with literature. OSI was a main contributing factor to the accurate prediction of the live birth rate, while age and BMI were found to be negatively correlated.
