BackgroundObesity in pregnancy is associated with insulin resistance, which underpins many common complications including gestational diabetes mellitus (GDM) and fetal macrosomia.ObjectivesTo assess the effect of a complex behavioural intervention based on diet and physical activity (PA) on the risk of GDM and delivery of a large-for-gestational age (LGA) infant.DesignThree phases: (1) the development phase, (2) the pilot study and (3) a multicentre randomised controlled trial (RCT) comparing a behavioural intervention to improve glycaemic control with standard antenatal care in obese pregnant women. A cost–utility analysis was undertaken to estimate the cost-effectiveness of the health training (intervention) over and above standard care (control).SettingPilot study: antenatal clinics in four inner-city UK hospitals. RCT: eight antenatal clinics in eight UK inner-city hospitals.ParticipantsWomen were eligible for inclusion if they had a body mass index of ≥ 30 kg/m2, were pregnant with a single fetus and at 15+0to 18+6weeks’ gestation, were able to give written informed consent and were without predefined disorders.InterventionThe intervention comprised an initial session with a health trainer, followed by eight weekly sessions. Dietary advice recommended foods with a low dietary glycaemic index, avoidance of sugar-sweetened beverages and reduced saturated fats. Women were encouraged to increase daily PA.Main outcome measuresDevelopment phase: intervention development, acceptability and optimal approach for delivery. Pilot study: change in dietary and PA behaviours at 28 weeks’ gestation. RCT: the primary outcome of the RCT was, for the mother, GDM [as measured by the International Association of the Diabetes and Pregnancy Study Groups (IADPSG)’s diagnostic criteria] and, for the infant, LGA delivery (i.e. customised birthweight ≥ 90th centile for gestational age).ResultsDevelopment phase: following a literature meta-analysis, a study of dietary intention questionnaires and semistructured interviews, an intervention based on behavioural science was developed that incorporated optimal and acceptable methods for delivery. Pilot study: the pilot study demonstrated improvement in dietary behaviours in the intervention compared with the standard care arm but no increase in objectively measured PA. Process evaluation demonstrated feasibility and general acceptability. RCT: the RCT showed no effect of the intervention on GDM in obese pregnant women or the number of deliveries of LGA infants. There was a reduction in dietary glycaemic load (GL) and reduced saturated fat intake, an increase in PA and a modest reduction in gestational weight gain, all secondary outcomes. Lower than expected was the number of LGA infant deliveries in all women, which suggested that universal screening for GDM with IADPSG’s diagnostic criteria, and subsequent treatment, may reduce the number of deliveries of LGA infants. According to the cost–utility analysis, the estimated probability that the UK Pregnancies Better Eating and Activity Trial (UPBEAT) behavioural intervention is cost-effective at the £30,000/quality-adjusted life-year willingness-to-pay threshold was 1%.LimitationsIncluded the high refusal rate for participation and self-reported assessment of diet and PA.ConclusionsThe UPBEAT intervention, an intense theoretically based intervention in obese pregnant women, did not reduce the risk of GDM in women or the number of LGA infant deliveries, despite successfully reducing the dietary GL. Based on total cost to the NHS provider and health gains, the UPBEAT intervention provided no supporting evidence to suggest that the intervention represents value for money based on the National Institute for Health and Care Excellence benchmarks for cost-effectiveness.Future workAlternative strategies for reducing the risk of GDM in obese pregnant women and the number of LGA infant deliveries should be considered, including development of clinically effective interventions to prevent obesity in women of reproductive age, of clinically effective interventions to reduce weight retention following pregnancy and of risk stratification tools in early pregnancy.Trial registrationCurrent Controlled Trials ISRCTN89971375 and UK Clinical Research Network Portfolio 5035.FundingThis project was funded by the NIHR Programme Grant for Applied Research programme and will be published in full inProgramme Grants for Applied Research, Vol. 5, No. 10. See the NIHR journals library website for further project information. Contributions to funding were also provided by the Chief Scientist Office CZB/4/680, Scottish Government Health Directorates, Edinburgh; Guys and St Thomas’ Charity, Tommy’s Charity (Lucilla Poston, Annette L Briley, Paul T Seed) and the NIHR Biomedical Research Centre at Guy’s and St Thomas’ NHS Foundation Trust and King’s College London, UK and the Academy of Finland, Finland. Keith M Godfrey was supported by the National Institute for Health Research through the NIHR Southampton Biomedical Research Centre. Lucilla Poston and Keith M Godfrey were supported by the European Union’s Seventh Framework Programme (FP7/2007-2013), project EarlyNutrition under grant agreement number 289346.