ObjectiveTo evaluate the effect, on birth weight and birth weight centile, of use of the PrenaBelt, a maternal positional therapy device, during sleep in the home setting throughout the third trimester of pregnancy.DesignA double-blind, sham-controlled, randomised clinical trial.SettingConducted from September 2015 to May 2016, at a single, tertiary-level centre in Accra, Ghana.ParticipantsTwo-hundred participants entered the study. One-hundred-eighty-one participants completed the study. Participants were women, 18 to 35 years of age, with low-risk, singleton, pregnancies in their third-trimester, with body mass index <35 kg/m2at the first antenatal appointment for the index pregnancy and without known foetal abnormalities, pregnancy complications or medical conditions complicating sleep.InterventionsParticipants were randomised by computer-generated, one-to-one, simple randomisation to receive either the PrenaBelt or sham-PrenaBelt. Participants were instructed to wear their assigned device to sleep every night for the remainder of their pregnancy (approximately 12 weeks in total) and were provided a sleep diary to track their use. Allocation concealment was by unmarked, security-tinted, sealed envelopes. Participants and the outcomes assessor were blinded to allocation.Primary and secondary outcome measuresThe primary outcomes were birth weight and birth weight centile. Secondary outcomes included adherence to using the assigned device nightly, sleeping position, pregnancy outcomes and feedback from participants and maternity personnel.ResultsOne-hundred-sixty-seven participants were included in the primary analysis. The adherence to using the assigned device nightly was 56%. The mean ±SD birth weight in the PrenaBelt group (n=83) was 3191g±483 and in the sham-PrenaBelt group (n=84) was 3081g±484 (difference 110 g, 95% CI −38 to 258, p=0.14). The median (IQR) customised birth weight centile in the PrenaBelt group was 43% (18 to 67) and in the sham-PrenaBelt group was 31% (14 to 58) (difference 7%, 95% CI −2 to 17, p=0.11).ConclusionsThe PrenaBelt did not have a statistically significant effect on birth weight or birth weight centile in comparison to the sham-PrenaBelt.Trial registration numberNCT02379728.