Human Reproduction · 1995 · 144 citations · 0 references
MedicineBiostatisticsGynecologyEmbryo TransferEmbryologyInfertilityAssisted Reproductive Technology OutcomesFemale InfertilityPregnancyAssisted Reproduction Outcome PredictionEmbryo Transfer PrognosticationFertility PredictionLogistic RegressionLogistic Regression AnalysisMultivariable Prediction ModelsPositive Predictive ValuePredictive Modeling In Pregnancy
Because the process of conception is affected by many variables, a multiple logistic regression analysis was performed to assess (i) the impact and relative weight of both patient and embryo variables and (ii) their possible effects on the probability of a vital pregnancy after embryo transfer. A statistical model was constructed predicting the probability of pregnancy after embryo transfer. The variables that contributed significantly to the predictive value of the model were the age of the patient, the cause of infertility, the number of embryos transferred and the average morphology score of the transferred embryos. Embryo variables appeared to have a significant but modest value in predicting the probability of pregnancy after embryo transfer. Other variables, such as the thickness of the endometrium, were found to have no prognostic value. Moreover, we found that their effect could be explained by the variables already included in the model.