Repeated implantation failure at the crossroad between statistics, clinics and over-diagnosis.

The most common definition of repeated implantation failure (RIF) is the failure to obtain a clinical pregnancy after three completed IVF cycles. This definition, however, may lead to misuse of the diagnosis. To disentangle this, we set up a mathematical model based on the following main assumptions: rate of success of IVF constant and set at 30%; and RIF postulated to be a dichotomous condition (yes or no) with a prevalence of 10%. On this basis, the expected cumulative chance of pregnancy after three and six cycles was 59% and 79%, respectively. Consequently, the false-positive rate of a diagnosis of RIF is 75% and 51%, respectively. Increasing the rate of success of IVF or the prevalence of RIF lowers but does not make unremarkable the rate of false-positive diagnoses. Overall, this model shows that the commonly used definition of RIF based on three failed attempts in a standard population with good prognosis leads to over-diagnosis and, potentially, to over-treatments.