Eleven RCTs (n=11,710, range 61 to 3,305 participants) met the inclusion criteria. Seven studies assessed orlistat and two studies each assessed sibutramine and rimonabant. One study scored 5, three studies scored 4, five studies scored 3 and one study scored 2 on the Jadad scale (score for one study not reported). Eight studies were double blind. Nine studies were analysed by intention to treat (ITT). Baseline differences were reported to be significant. Withdrawal/dropout rate ranged from 10% to 66% for to four years of follow-up.
Percentage weight loss achieved at the end of two years with orlistat ranged from 5% to 12%. Mean weight loss was 8kg (seven studies), a difference of 3.7kg when compared to placebo.
Sibutramine achieved weight loss of 4% to 10% and mean weight loss of 7.4kg (two studies), a difference of 5.5kg when compared with placebo.
Rimonabant achieved 7% weight loss and mean weight loss of 7.3kg, a difference of 4.4 kg when compared with placebo.
Overall, 26% of participants who used orlistat reported adverse events (mostly gastrointestinal events), which led to many participants suspending treatment. The main adverse events with sibutramine and rimonabant were incidence of treatment-associated infections and other generalised symptoms. Specific adverse events and rate of occurrence were reported in the main paper.