Background: Preterm birth is common in twins and accounts for significant mortality and morbidity. There are no effective preventative treatments. Some studies have suggested that, in twin pregnancy complicated by a short cervix, the Arabin pessary, which fits around the cervix and can be inserted as an outpatient procedure, reduces preterm birth and prevents neonatal morbidity. Objective: STOPPIT 2 aimed to evaluate the clinical utility of the Arabin cervical pessary in preventing preterm birth in women with a twin pregnancy and a short cervix. Design: STOPPIT 2 was a pragmatic, open label, multicentre randomised controlled trial with two treatment group – the Arabin pessary plus standard care (intervention) and standard care alone (control). Participants were initially recruited into the screening phase of the study, when cervical length was measured. Women with a measured cervical length of ≤ 35 mm were then recruited into the treatment phase of the study. An economic evaluation considered cost-effectiveness and a qualitative substudy explored the experiences of participants and clinicians. Setting: Antenatal clinics in the UK and elsewhere in Europe. Participants: Women with twin pregnancy at < 21 weeks’ gestation with known chorionicity and gestation established by scan at ≤ 16 weeks’ gestation. Interventions: Ultrasound scan to establish cervical length. Women with a cervical length of ≤ 35 mm at 18+ 0–20+ 6 weeks’ gestation were randomised to standard care or Arabin pessary plus standard care. Randomisation was performed by computer and accessed through a web-based browser. Main outcome measures: Obstetric – all births before 34+ 0 weeks’ gestation following the spontaneous onset of labour; and neonatal – composite of adverse outcomes, including stillbirth or neonatal death, periventricular leukomalacia, early respiratory morbidity, intraventricular haemorrhage, necrotising enterocolitis or proven sepsis, all measured up to 28 days after the expected date of delivery. Results: A total of 2228 participants were recruited to the screening phase, of whom 2170 received a scan and 503 were randomised: 250 to Arabin pessary and 253 to standard care alone. The rate of the primary obstetric outcome was 18.4% (46/250) in the intervention group and 20.6% (52/253) in the control group (adjusted odds ratio 0.87, 95% confidence interval 0.55 to 1.38; p = 0.54). The rate of the primary neonatal outcome was 13.4% (67/500) and 15.0% (76/506) in the intervention group and control group, respectively (adjusted odds ratio 0.86, 95% confidence interval 0.54 to 1.36; p = 0.52). The pessary was largely well tolerated and clinicians found insertion and removal ‘easy’ or ‘fairly easy’ in the majority of instances. The simple costs analysis showed that pessary treatment is no more costly than standard care. Limitations: There was the possibility of a type II error around smaller than anticipated benefit. Conclusions: In this study, the Arabin pessary did not reduce preterm birth or adverse neonatal outcomes in women with a twin pregnancy and a short cervix. The pessary either is ineffective at reducing preterm birth or has an effect size of < 0.4. Future work: Women with twin pregnancy remain at risk of preterm birth; work is required to find treatments for this. Trial registration: Current Controlled Trials ISRCTN98835694 and ClinicalTrials.gov NCT02235181. 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. 25, No. 44. See the NIHR Journals Library website for further project information.