Reply of the authors

We thank Drs. Verit and Verit for their interest in our paper (1) and their insightful comments. We agree that many factors would have contributed to the observed increase in IVF effectiveness between the two time periods in our study; especially those they highlighted (refer also to the fifth paragraph in our discussion). However, the main focus of our paper was not on a comparison between the two time periods, and we did not base the conclusions of our study on such a comparison. Rather, our main analysis focussed on data from the later time period (1993–2002).
We used these data to estimate IVF effectiveness according to the woman’s age at the start of IVF treatment, as shown in Figure 1A (1). We then calculated the cumulative probability of IVF success according to the number of cycles (Fig. 2) (1) and, based on the results of these two analyses combined with the observation that couples undertake, on average, three cycles, we concluded that an increase in the average number of cycles undertaken would likely lead to a further overall increase in IVF effectiveness beyond the improvements already observed.
Drs. Verit and Verit make the point that IVF effectiveness could also be estimated on the basis of many factors other than age at start of treatment, including cause and duration of infertility; we agree with them. It would, however, require a much larger study population to estimate IVF effectiveness within finely stratified subgroups with a reasonable degree of precision.