Weight loss is associated with improvements in glycaemic control and cardiovascular disease risk factors. However, in the diabetic population, weight management is more challenging, in part because of the weight-promoting effects of the majority of glucose-lowering therapies. This review summarizes evidence from 23 placebo-controlled randomized trials, of at least 1 year duration, on the effects of drugs promoting weight loss (orlistat, sibutramine and rimonabant) on glycaemic variables, diabetes incidence and diabetes control. Fifteen studies of non-diabetic subjects were found, eight of which included a longer treatment period. Eight studies in diabetic patients were reviewed. In non-diabetic subjects, weight loss agents led to a significant improvement in fasting glucose, fasting insulin and insulin resistance. In the diabetic population, glycated haemoglobin decreased by 0.28–1.1% with orlistat and 0.6% with sibutramine and rimonabant. Orlistat reduces progression to diabetes in patients with glucose intolerance treated for 4 years (risk reduction of 45%). In summary, despite leading to only modest weight loss after 12 months, agents promoting weight loss have beneficial effects on glycaemic parameters, glycaemic control and progression to diabetes. These additional benefits of weight loss agents need to be highlighted in order to increase their judicious use in clinical practice, although this may be limited by their well-known adverse side effects. The longer-term safety of these agents beyond a few years is yet to be established.