The cost-effectiveness of progesterone in preventing miscarriages in women with early pregnancy bleeding: an economic evaluation based on the PRISM trial

Objectives: To assess the cost-effectiveness of progesterone compared with placebo in preventing pregnancy loss in women with early pregnancy vaginal bleeding. Design: Economic evaluation alongside a large multi-centre randomised placebo-controlled trial. Setting: Forty-eight UK NHS early pregnancy units. Population: Four thousand one hundred and fifty-three women aged 16–39 years with bleeding in early pregnancy and ultrasound evidence of an intrauterine sac. Methods: An incremental cost-effectiveness analysis was performed from National Health Service (NHS) and NHS and Personal Social Services perspectives. Subgroup analyses were carried out on women with one or more and three or more previous miscarriages. Main outcome measures: Cost per additional live birth at ≥34 weeks of gestation. Results: Progesterone intervention led to an effect difference of 0.022 (95% CI −0.004 to 0.050) in the trial. The mean cost per woman in the progesterone group was £76 (95% CI −£559 to £711) more than the mean cost in the placebo group. The incremental cost-effectiveness ratio for progesterone compared with placebo was £3305 per additional live birth. For women with at least one previous miscarriage, progesterone was more effective than placebo with an effect difference of 0.055 (95% CI 0.014–0.096) and this was associated with a cost saving of £322 (95% CI −£1318 to £673). Conclusions: The results suggest that progesterone is associated with a small positive impact and a small additional cost. Both subgroup analyses were more favourable, especially for women who had one or more previous miscarriages. Given available evidence, progesterone is likely to be a cost-effective intervention, particularly for women with previous miscarriage(s). Tweetable abstract: Progesterone treatment is likely to be cost-effective in women with early pregnancy bleeding and a history of miscarriage.