Forty studies were included in the review: 39 RCTs and one controlled before-and-after study (sample size ranged from 200 to approximately 49,000 participants). Follow-up ranged from 24 to 97 months. Twelve studies had good allocation concealment, two studies had blinded outcome assessors and 10 studies used ITT.
Interventions with a definite intention to lose weight (compared to control):
A 600kcal/day deficit diet was associated with a statistically significant reduction in weight at 36 months (WMD -3.49kg, 95% CI -4.63 to -2.35; one study). A low calorie diet was associated with a statistically significant reduction in weight at 24 months (WMD -7.00kg, 95% CI -10.99 to -3.01; one study) and at 36 months (WMD -6.10kg, 95% CI -10.71 to -1.49; one study in women treated for breast cancer). Weight Watchers was associated with a statistically significant reduction in weight at 24 months (WMD -2.70kg, 95% CI -3.95 to -1.45; one study). A 600kcal/day deficit diet or low fat diet that used meal replacements was associated with a statistically significant reduction in weight at 24 months (WMD -11.49kg, 95% CI -12.98 to -10.00; one study).
Diet and exercise was associated with a statistically significant reduction in weight at 24 months when compared with self-help (WMD -2.56 kg, 95% CI -3.34 to -1.77; three studies). Diet and exercise and behaviour therapy was associated with a statistically significant reduction in weight at 24 months (WMD -2.47kg, -3.18 to -1.77; two studies), at 30 to 36 months (WMD -2.04kg, 95% CI -2.70 to -1.39; two studies) and at 54 months (WMD -2.50kg, 95% CI -3.59 to -1.41; one study).
Interventions with no target for weight loss (compared to control):
A low fat non-reducing diet was associated with a statistically significant reduction in weight at 24 months (WMD -1.42kg, 95% CI -2.10 to -0.74; two studies). Diet and behaviour therapy was associated with a statistically significant reduction in weight at 24 months (WMD -1.01kg, 95% CI -1.34 to -0.68; two studies), at 36 months (WMD -1.77kg, 95% CI -1.94 to -1.59; three studies), at 48 months (WMD -0.52kg, 95% CI -0.85 to -0.19; two studies) and at 90 months (WMD -0.70kg, 95% CI -0.90 to -0.50; one study).
A number of other comparisons were reported. Most showed no statistically significant difference in weight.
There was no statistically significant difference between lifestyle interventions and control groups for death, stroke, heart disease and cancer outcomes.
Three studies (diet versus control, diet and exercise versus control, and diet and exercise and behaviour therapy versus control) showed a statistically significant reduced risk of type 2 diabetes up to six years (RR 0.68, 95% CI 0.57 to 0.82).Two studies (intensive low fat diet, exercise and psychological support, and Mediterranean diet and behaviour therapy) showed a statistically significant higher resolution of metabolic syndrome for up to three years (RR 2.52, 95% CI 2.08 to 3.05).