Should Oral Nutritional Supplementation Be Given to Undernourished Older People upon Hospital Discharge? A Controlled Trial

OBJECTIVES: To determine whether the oral nutritional supplementation of undernourished older people upon discharge from hospital improves muscle function and reduces disability. DESIGN: Randomized controlled trial. SETTING: Community-based study in two centers in Scotland. PARTICIPANTS: Two hundred fifty-three people. INTERVENTION: Randomization to oral nutritional supplementation (600 kcal/d) or control supplement of 200 kcal/d. MEASUREMENTS: Primary outcome (20-point activity of daily living Barthel Index) and secondary outcomes (handgrip strength, Sit-to-Stand test, and Euroquol) were measured at baseline (after discharge from the hospital and before supplement was commenced) and 8 and 16 weeks and accelerometry-measured physical activity levels at baseline and 16 weeks. Falls were recorded prospectively. RESULTS: Mean age was 82. There was no significant difference in change in Barthel score between supplement and control groups (adjusted mean difference = 0.28, 95% confidence interval (CI) 0.28-0.84). Handgrip strength improved more in the supplemented group (adjusted mean difference = 1.52 kg, 95% CI = 0.50-2.55; P = .004). The supplemented group exhibited modestly greater vector movement (overall activity) than controls (P = .02). There were no significant between-group differences in Sit-to-Stand test, health-related quality of life, or falls. Adherence was 38.2% in the nutritional supplement group and 50.0% in the control supplement group. Weight did not increase in the nutritional supplement group as a whole, but oil-treatment analysis adjusting for adherence showed a mean weight gain of 1.17 kg (95% CI = 0.07-2.27; P = .04) more than in controls. CONCLUSION: Oral nutritional supplementation of undernourished older people upon hospital discharge did not reduce disability, despite improving handgrip strength and modestly increasing objectively measured physical activity levels. Lack of an effect of the nutritional supplement used in this Study may have been due to low adherence, suggesting that different approaches to nutritional supplementation need to be tested in this population. J Am Geriatr Soc 57:2239-2245, 2009.