Skip to main content

Table 2 The main predictive data from the logistic regression model

From: Predicting live birth chances for women with multiple consecutive failing IVF cycles: a simple and accurate prediction for routine medical practice

Effects Logistic Model Linear Simplified Model
  OR 95% CI P Incr 95% CI P
Intercept 0.044 0.006; 0.392 0.004 0.045 0; 0.350 0.047
Women age 0.868 0.782; 0.964 0.008 −0.023 −0.039; −0.006 0.009
FSH < 10 IU/L 4.222 1.267; 14.077 0.019 0.200 0.039; 0.361 0.016
Male Infertility Etiology (MI) 3.862 1.097; 13.595 0.035 0.172 0.011; 0.334 0.038
occurrence of at least one GS (OGS) 3.099 1.395; 6.884 0.005 0.174 0.047; 0.302 0.008
Mean number of GQE (MQE) 2.422 1.322; 4.437 0.004 0.168 0.064; 0.263 0.003
  1. Proportion of variability due to covariates R2=0.2955; ROC curve AUC (C) = 0.76 (95% CI [0.71–0.81]) and the simplified model (from an ordinary least squares procedure suited to binary data). OR: odds ratio; CI: confidence interval. Logistic model coefficients can be assimilated to OR, and linear model coefficients to incremental values (Incr). For interpreting OR and Incr, age, number of cycles with at least one sac, and mean number are continuous values, while FSH<10 IU/L and Male infertility are binary values.