Eighteen RCTs (4,405 participants) were included in the review. There were 1,275 ethnic minority participants and 3,130 white participants. Most trials adequately defined the method of randomisation, used allocation concealment, had similar baseline data and used intention-to-treat analysis; four trials failed on more than five criteria.
Orlistat (12 trials, 3,100 participants): Compared with placebo, orlistat had a statistically significantly greater weight loss in both ethnic minorities (mean difference -2.3kg, 95% CI -2.6 to -2.0, I2=0%; 12 RCTs) and white Caucasians (mean difference -2.8kg, 95% CI -5.1 to -0.5, I2=0%; nine RCTs). There was no difference between the two groups in terms of weight loss.
Sibutramine (six trials, 1,305 participants): Compared with placebo, sibutramine had a statistically significantly greater weight loss in both ethnic minorities (mean difference -2.7kg, 95% CI -3.1 to -2.3, I2=0%; six RCTs) and white Caucasians (mean difference -4.5kg, 95% CI -5.1 to -3.8, I2=54%; six RCTs). There was a significant difference between the two groups in terms of weight loss.
Secondary outcomes were presented in the review. There was no evidence of publication bias.