Physiological sex steroid replacement in premature ovarian failure

amongst every day pragmatic practice such as ours. The difference between ITT and PP analyses is the number of subjects who were randomized but did not actually enter or complete the study. We clearly noted in our paper that only 1 of the 47 reasons for drop out following randomization was for a treatment-related reason (overstimulation in the FSH group; Fig 1). The large majority of the others did not enter the study following randomization or did not complete the study for purely personal reasons completely unrelated to the study itself. Thus, the PP analysis considered those who actually underwent one of the study treatments rather than including those who did not and so the PP analysis is closer to a comparison of the true efficacies of the treatments than ITT. We are, therefore, confident that our conclusion based on the PP analysis that low-dose FSH is a more successful first-line treatment than clomifene citrate for anovulatory PCOS is the correct one and that an even larger study, if ever performed, will confirm our findings.