Meta-analysis of studies of a specific delivery mode for a modified-carbohydrate diet

Abstract

Obesity is highly prevalent throughout the world. Although modified-carbohydrate diets (MCDs) comprise one popular approach, questions remain about their utility for weight loss. The objective of the present study was to conduct a meta-analysis of randomised controlled trials (RCTs) of a specific MCD compared with various control diets on weight loss. Data from four RCTs (three obtained from the sponsor and one indentified through literature searches) were included. Intent-to-treat analyses were conducted using multiple imputation to handle missing data, where possible. Because inter-study heterogeneity was demonstrated with fixed-effects meta-analysis, a random-effects meta-analysis also was conducted. When considered separately, all four studies showed greater reduction in body weight with the MCD compared to control diets at 12-week follow-up; the results at 24 weeks (available for three of the studies) were not as consistent. Results for body mass index (BMI) were similar. Greater reductions in waist circumference with the MCD were seen at either time point in only one study. When fixed-effects meta-analysis was applied, significantly greater reductions in weight, BMI and waist circumference with the MCD at both 12 weeks (1.66 kg, 0.53 kg m(-2) and 1.02 cm, respectively) and 24 weeks (1.20 kg, 0.43 kg m(-2) and 0.69 cm, respectively) were evident. Random-effects meta-analysis revealed similar results; however, the 24-week difference for a reduction in waist circumference was no longer statistically significant. Meta-analysis of individual RCT results demonstrated consistent benefits of this MCD compared to control diets on weight loss up to 24 weeks and waist circumference up to 12 weeks.