Reporting multiple cycles in trials on medically assisted reproduction

Trials assessing effectiveness in medically assisted reproduction (MAR) should aim to study the desired effect over multiple cycles, as this reflects clinical practice and captures the relevant perspective for the couple. The aim of this study was to assess the extent to which multiple cycles are reported in MAR trials. A sample of randomized controlled trials (RCT) was collected on MAR, published in four time periods, in 11 pre-specified peer-reviewed journals; 253 trials were included: 196 on IVF, 37 on intrauterine insemination and 20 on ovulation induction. Forty-eight (19%) reported on multiple cycles, which was significantly more common in trials on intrauterine insemination and ovulation induction compared with trials on IVF (P < 0.01). Both trials on IVF were multi-centre trials, and those using live birth as primary outcome, reported significantly more often on multiple cycles (OR 3.7 CI 1.1 to 12.5) and (OR 8.7 CI 1.8 to 40.3), respectively. Trials designed to compare protocol variations reported multiple cycles less often (OR 0.07 CI 0.01 to 0.74). Most RCT on MAR, especially those on IVF, do not report cumulative pregnancy rates. As not all women become pregnant in their first cycle, the clinical significance of these trials is limited.