A Cochrane review [Abstract] 1 included 45 studies with a total of 3186 subjects. The duration of follow-up was from 18 days to 24 months. There was a significant change in weight found between groups when comparing dietary advice to no advice for interventions lasting greater than 12 months, mean difference 3.75 kg (95% confidence interval 0.97 to 6.53), and when all studies were combined, mean difference 1.47 kg (95% confidence interval 0.32 to 2.61) although there was significant heterogeneity in the combined analysis (I2 = 90%). Similar improvements in weight were found for the comparison of dietary advice with nutritional supplements if required versus no advice, mean difference 2.20 kg (95% confidence interval 1.16 to 3.25). Dietary advice compared with no advice was also associated with significantly improved mid-arm muscle circumference when all studies were combined, but with moderate heterogeneity, mean difference 0.81 mm (95% confidence interval 0.31 to 1.31). Dietary advice given with nutritional supplements compared with dietary advice alone resulted in improvements in: mid-arm muscle circumference, mean difference -0.89 mm (95% confidence interval -1.35 to -0.43); triceps skinfold thickness, mean difference -1.22 mm (95% confidence interval -2.34 to -0.09); and grip strength, mean difference -1.67 kg (95% confidence interval -2.96 to -0.37), although the effects on triceps skinfold thickness and grip strength were heterogeneous. Dietary advice with supplements if required resulted in a significant increase in triceps skinfold thickness compared with no advice, mean difference 0.40 mm (95% confidence interval 0.10 to 0.70), although these results are from a single trial with only 29 participants. No significant differences were found between groups for clinical outcomes. No comparison showed a significant difference in mortality.
Comment: The quality of evidence was downgraded by indirectness of evidence (no clinical outcomes reported).
Primary/Secondary Keywords