Background Progesterone is essential for a healthy pregnancy. Several small trials have suggested that progesterone therapy may rescue a pregnancy in women with early pregnancy bleeding, which is a symptom that is strongly associated with miscarriage. Objectives (1) To assess the effects of vaginal micronised progesterone in women with vaginal bleeding in the first 12 weeks of pregnancy. (2) To evaluate the cost-effectiveness of progesterone in women with early pregnancy bleeding. Design A multicentre, double-blind, placebo-controlled, randomised trial of progesterone in women with early pregnancy vaginal bleeding. Setting A total of 48 hospitals in the UK. Participants Women aged 16–39 years with early pregnancy bleeding. Interventions Women aged 16–39 years were randomly assigned to receive twice-daily vaginal suppositories containing either 400 mg of progesterone or a matched placebo from presentation to 16 weeks of gestation. Main outcome measures The primary outcome was live birth at ≥ 34 weeks. In addition, a within-trial cost-effectiveness analysis was conducted from an NHS and NHS/Personal Social Services perspective. Results A total of 4153 women from 48 hospitals in the UK received either progesterone (n = 2079) or placebo (n = 2074). The follow-up rate for the primary outcome was 97.2% (4038 out of 4153 participants). The live birth rate was 75% (1513 out of 2025 participants) in the progesterone group and 72% (1459 out of 2013 participants) in the placebo group (relative rate 1.03, 95% confidence interval 1.00 to 1.07; p = 0.08). A significant subgroup effect (interaction test p = 0.007) was identified for prespecified subgroups by the number of previous miscarriages: none (74% in the progesterone group vs. 75% in the placebo group; relative rate 0.99, 95% confidence interval 0.95 to 1.04; p = 0.72); one or two (76% in the progesterone group vs. 72% in the placebo group; relative rate 1.05, 95% confidence interval 1.00 to 1.12; p = 0.07); and three or more (72% in the progesterone group vs. 57% in the placebo group; relative rate 1.28, 95% confidence interval 1.08 to 1.51; p = 0.004). A significant post hoc subgroup effect (interaction test p = 0.01) was identified in the subgroup of participants with early pregnancy bleeding and any number of previous miscarriage(s) (75% in the progesterone group vs. 70% in the placebo group; relative rate 1.09, 95% confidence interval 1.03 to 1.15; p = 0.003). There were no significant differences in the rate of adverse events between the groups. The results of the health economics analysis show that progesterone was more costly than placebo (£7655 vs. £7572), with a mean cost difference of £83 (adjusted mean difference £76, 95% confidence interval –£559 to £711) between the two arms. Thus, the incremental cost-effectiveness ratio of progesterone compared with placebo was estimated as £3305 per additional live birth at ≥ 34 weeks of gestation. Conclusions Progesterone therapy in the first trimester of pregnancy did not result in a significantly higher rate of live births among women with threatened miscarriage overall, but an important subgroup effect was identified. A conclusion on the cost-effectiveness of the PRISM trial would depend on the amount that society is willing to pay to increase the chances of an additional live birth at ≥ 34 weeks. For future work, we plan to conduct an individual participant data meta-analysis using all existing data sets. Trial registration Current Controlled Trials ISRCTN14163439, EudraCT 2014-002348-42 and Integrated Research Application System (IRAS) 158326. Funding This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 24, No. 33. See the NIHR Journals Library website for further project information.