Forty-six RCTs met the inclusion criteria. These provided 63 intervention groups (approximately 6,386 participants received dietary counselling and 5,467 usual care).
Of the 46 studies, 9% were considered good quality, 63% were considered fair quality and 28% poor quality. Flaws in the poor-quality studies included high rates of withdrawal, incomplete reporting and unclear analyses.
Change in BMI at single time points.
In the active phase, data for the greatest number of participants were available at 6 and 12 months, when dietary counselling resulted in significant reductions in BMI: mean net changes of –1.39 (5% confidence interval, CI: –1.79, -1.00; 3,888 participants) and –1.88 (95% CI: –2.29, -1.46; 1,363 participants), respectively. A significant reduction in weight associated with dietary counselling was also found at 3, 4, 5, 7, 8, 13, 36 and 42 months, but not at 18, 24 or 30 months (most of these analyses were based on a small number of studies).
In the maintenance phase, significant reductions in BMI were associated with counselling between 6 and 42 months; the response became steadily smaller over that period of time despite most of the intermediate time points having substantially greater numbers of participants (up to 2,274): mean net changes were –2.87 at 6 months (95% CI: –4.50, -1.25; 228 participants) and –0.83 (95% CI –1.29, -0.37; 250 participants) at 42 months. After 42 months, the response was mixed with low numbers of participants.
Change in BMI across multiple time points.
During the first year of the active phase, 60% of slopes were positive (weight loss) and 40% were flat or negative (weight gain). After 12 months, 8 out of 9 slopes were positive.
During the maintenance phase, 85% of slopes were positive. However, most studies showed a steady regain in weight over time.