Premium
A competing risk approach to the analysis of trials of alternative intra‐uterine devices (IUDs) for fertility regulation
Author(s) -
Tai BeeChoo,
Peregoudov Alexander,
Machin David
Publication year - 2001
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.1079
Subject(s) - discontinuation , medicine , context (archaeology) , fertility , family planning , randomized controlled trial , gynecology , intensive care medicine , population , research methodology , surgery , paleontology , environmental health , biology
The methods of survival analysis have had a profound influence on the way that studies concerned with the safety and efficacy of intra‐uterine devices (IUDs) for fertility control have been designed, conducted, analysed and reported. For example, the Kaplan–Meier technique has been used to summarize the results of controlled clinical trials of alternative devices and the logrank test used to make any comparisons. A particular feature of IUD studies is the large number of possible causes of failure (reasons for discontinuation of the device). These lead to considerations of competing risks, where in this framework, the discontinuations are considered as competing causes of contraceptive failure. In this context, we compare the net and crude probability estimates of discontinuation rates using continuous time and argue for the routine use of the latter. We use data from a randomized multi‐centre trialon the long‐term safety and efficacy of two IUDs, TCu220C and TCu380A, for illustration. Copyright © 2001 John Wiley & Sons, Ltd.