Fig 1.
A flow diagram is shown which depicts the process of study selection.
Table 1.
Overview of study characteristics included in NMAs (n = 75).
Fig 2.
Network diagram for change in body weight (kg): 18 interventions and standard care, 11,267 patients.
The evidence network of the available studies and interventions for change in body weight is shown. Joining lines denote intervention comparisons where one or more trials were available. Nodes are proportionally sized to reflect the numbers of patients studied with each intervention. Edge width reflects the number of RCTs for each comparison. Nodes coloured green represent interventions considered to belong to the exercise group, while red nodes reflect the dietary group and the blue node denotes the diet/exercise combination group.
Fig 3.
Network diagram for change in BMI: 16 interventions and standard care, 6,857 patients.
The evidence network of the available studies and interventions for change in BMI is shown. Joining lines denote intervention comparisons where one or more trials were available. Nodes are proportionally sized to reflect the numbers of patients studied with each intervention. Edge width reflects the number of RCTs for each comparison. Nodes coloured green represent interventions considered to belong to the exercise group, while red nodes reflect the dietary group and the blue node denotes the diet/exercise combination group.
Fig 4.
Network diagram for change in waist circumference: 15 interventions and standard care, 1,835 patients.
The evidence network of the available studies and interventions for change in waist circumference is shown. Joining lines denote intervention comparisons where one or more trials were available. Nodes are proportionally sized to reflect the numbers of patients studied with each intervention. Edge width reflects the number of RCTs for each comparison. Nodes coloured green represent interventions considered to belong to the exercise group, while red nodes reflect the dietary group and the blue node denotes the diet/exercise combination group.
Fig 5.
Estimated differences in weight change (kg) compared to standard care from NMA (18 interventions and standard care, 65 studies, 11,267 patients).
The estimated differences (2.5% and 97.5% quantiles) of interventions versus standard care from random effects consistency model are displayed. Colored summary estimates represent estimated treatment effects of the groups of interventions versus standard care.
Fig 6.
Estimated differences in BMI change compared to standard care from NMA (16 interventions and standard care, 47 studies, 6,824 patients).
The estimated differences (2.5% and 97.5% quantiles) of interventions versus standard care from random effects consistency model are displayed. Colored summary estimates represent estimated treatment effects of the groups of interventions versus standard care.
Fig 7.
Estimated differences in waist circumference change (cm) compared to standard care from NMA (15 interventions and standard care, 32 studies, 1,875 patients).
The estimated differences (2.5% and 97.5% quantiles) of interventions versus standard care from random effects consistency model are displayed. Coloured summary estimates represent estimated treatment effects of the groups of interventions versus standard care.
Fig 8.
League tables, comparisons between group.
League tables of estimated posterior median pairwise differences in weight change (kg) with credible intervals / 2.5% and 97.5% quantiles (lower triangle), and the pairwise probabilities that a treatment is better than another (upper triangle) are presented. A complete summary of estimates for efficacy from the RE consistency model assuming vague priors is displayed. Statistically significant differences between intervention categories are shown in bold, underlined font. The lower/right-most comparison for each comparison is the reference treatment.