Probability of pregnancy after sterilization: a comparison of hysteroscopic versus laparoscopic sterilization

An illuminating paper and good that it promotes “perfect” and “typical” use failure rates for sterilizations [ ]. The authors claim that they started the Markov analysis (see their Fig. 1) with “Desires sterilization”. Incorrectly so: that would be a far more complicated intention-to-be-sterilized model. Instead, their simulation starts when all sorts of barriers have already been conquered: inertia, deferment, fear, urban legends, finances, patriarchalism, culture, transport, waiting lists, civil war, regulations, religion of provider or institution [ ], negotiations with employer, insurer and babysitter. These obstacles have a much larger impact on the overall number of unintended pregnancies than disparities in tubal occlusion (TO) methods, however important those are for the individual women/families. For Medicaid beneficiaries, just one of those barriers — government regulations — results in 10,000 abortions and 19,000 originally unintended births annually in the US [ ]. If Essure provision would remove a few of those barriers much better than access to laparoscopic TO does, then the factor 4 difference over 10 years in typical failure rates will be more than offset. It sometimes works like that in the Netherlands. More often however, the mere availability of the Essure method causes unintended effects. In abortion clinics, relevant women are enthusiastically informed about Essure as one of their options. At that time, some embrace the idea, but later, the above barriers kick in. This might result in more unintended pregnancies than combining the suction curettage with an intrauterine device insertion, for which, at that time, the hurdle is quite low. Similarly, Dutch cesareans are rarely combined with effective, convenient, very low failure, cost-free TO partly because TO deferment is often advised with the “there is this smart new technique” argument [ ]. However, women are not informed about the much, much higher failure rates when compared on an intention-to-be-sterilized basis [ ].