Background Adults with severe obesity [body mass index (BMI) of ≥ 35 kg/m2] have an increased risk of comorbidities and psychological, social and economic consequences. Objectives Systematically review bariatric surgery, weight-management programmes (WMPs) and orlistat pharmacotherapy for adults with severe obesity, and evaluate the feasibility, acceptability, clinical effectiveness and cost-effectiveness of treatment. Data sources Electronic databases including MEDLINE, EMBASE, PsycINFO, the Cochrane Central Register of Controlled Trials and the NHS Economic Evaluation Database were searched (last searched in May 2017). Review methods Four systematic reviews evaluated clinical effectiveness, cost-effectiveness and qualitative evidence for adults with a BMI of ≥ 35 kg/m2. Data from meta-analyses populated a microsimulation model predicting costs, outcomes and cost-effectiveness of Roux-en-Y gastric bypass (RYGB) surgery and the most effective lifestyle WMPs over a 30-year time horizon from a NHS perspective, compared with current UK population obesity trends. Interventions were cost-effective if the additional cost of achieving a quality-adjusted life-year is < £20,000–30,000. Results A total of 131 randomised controlled trials (RCTs), 26 UK studies, 33 qualitative studies and 46 cost-effectiveness studies were included. From RCTs, RYGB produced the greatest long-term weight change [–20.23 kg, 95% confidence interval (CI) –23.75 to –16.71 kg, at 60 months]. WMPs with very low-calorie diets (VLCDs) produced the greatest weight loss at 12 months compared with no WMPs. Adding a VLCD to a WMP gave an additional mean weight change of –4.41 kg (95% CI –5.93 to –2.88 kg) at 12 months. The intensive Look AHEAD WMP produced mean long-term weight loss of 6% in people with type 2 diabetes mellitus (at a median of 9.6 years). The microsimulation model found that WMPs were generally cost-effective compared with population obesity trends. Long-term WMP weight regain was very uncertain, apart from Look AHEAD. The addition of a VLCD to a WMP was not cost-effective compared with a WMP alone. RYGB was cost-effective compared with no surgery and WMPs, but the model did not replicate long-term cost savings found in previous studies. Qualitative data suggested that participants could be attracted to take part in WMPs through endorsement by their health-care provider or through perceiving innovative activities, with WMPs being delivered to groups. Features improving long-term weight loss included having group support, additional behavioural support, a physical activity programme to attend, a prescribed calorie diet or a calorie deficit. Limitations Reviewed studies often lacked generalisability to UK settings in terms of participants and resources for implementation, and usually lacked long-term follow-up (particularly for complications for surgery), leading to unrealistic weight regain assumptions. The views of potential and actual users of services were rarely reported to contribute to service design. This study may have failed to identify unpublished UK evaluations. Dual, blinded numerical data extraction was not undertaken. Conclusions Roux-en-Y gastric bypass was costly to deliver, but it was the most cost-effective intervention. Adding a VLCD to a WMP was not cost-effective compared with a WMP alone. Most WMPs were cost-effective compared with current population obesity trends. Future work Improved reporting of WMPs is needed to allow replication, translation and further research. Qualitative research is needed with adults who are potential users of, or who fail to engage with or drop out from, WMPs. RCTs and economic evaluations in UK settings (e.g. Tier 3, commercial programmes or primary care) should evaluate VLCDs with long-term follow-up (≥ 5 years). Decision models should incorporate relevant costs, disease states and evidence-based weight regain assumptions. Study registration This study is registered as PROSPERO CRD42016040190. Funding The National Institute for Health Research Health Technology Assessment programme. The Health Services Research Unit and Health Economics Research Unit are core funded by the Chief Scientist Office of the Scottish Government Health and Social Care Directorate.

Background Weight loss interventions for obesity, such as bariatric surgery, are associated with reductions in bone mineral density and may increase the risk of fractures. We undertook a systematic review and meta-analysis of bariatric surgery and lifestyle weight management programs (WMPs) with fracture outcomes. Methods We searched MEDLINE, Embase, the Cochrane Central Register of Controlled Trials from 1966 to 2018, and our trial registry of WMP randomized controlled trials (RCTs). We included RCTs, non-randomized trials, and observational studies of bariatric surgery, and RCTs of WMPs. Studies had follow-up ≥ 12 months, mean group body mass index ≥ 30 kg/m 2 . The primary outcome measure was incidence of any type of fracture in participants, and the secondary outcome was weight change. We used random effects meta-analysis for trial data. Results Fifteen studies were included. Three small trials provided short-term evidence of the association between bariatric surgery and participants with any fracture (365 participants; RR 0.82; 95% CI 0.29 to 2.35). Four out of six observational studies of bariatric surgery demonstrated significantly increased fracture risk. Six RCTs of WMPs with 6214 participants, the longest follow-up 11.3 years, showed no clear effect on any type of fracture (RR 1.04; 95% CI 0.91 to 1.18), although authors of the largest RCT reported an increased risk of frailty fracture by their definition (RR 1.40; 95% CI 1.04 to 1.90).Conclusion Bariatric surgery appears to increase the risk of any fracture; however, longer-term trial data are needed. The effect of lifestyle WMPs on the risk of any fracture is currently unclear.

Background: Weight loss could improve fertility, perhaps by reducing insulin resistance.Objectives: To assess the effect of weight loss interventions on fertility in women with obesity not recruited because of known infertility.Search Strategy: Three databases during 1966-2020, trial registry.Selection Criteria: Randomized controlled trials (RCTs) with a follow-up of 1 year or more, with a mean cohort BMI of 30 kg/m 2 or above. Data Collection and Analysis:A systematic review and meta-analysis was conducted.The primary outcome was pregnancy. The secondary outcome was weight change.Main Results: A total of 27 RCTs (5938 women) were included. Weight loss interventions showed no statistically significant increase in pregnancies compared to control interventions (24 trials, 97 women with pregnancy; risk ratio [RR] 1.43, 95% confidence interval [CI] 0.91-2.23); weight change (mean difference [MD] −2.36 kg, 21 trials, 95% CI −3.17 to −1.55). Compared with low-fat diets, very-low-carbohydrate diets showed no statistically significant effect on women with pregnancy (three trials, 14 women with pregnancy; RR 1.37, 95% CI 0.49-3.84) or weight change (MD −0.32 kg, 95% CI −3.84 to 3.21). Conclusions: Diet-based weight loss interventions for women with obesity not recruited because of infertility were effective at producing long-term weight loss. The effects on fertility were not statistically significant, but few trials provided data. Weight loss trials should routinely collect fertility outcomes. Systematic review registration: PROSPERO CRD42017078819.

Correction to Declared competing interests of authors.