Effectiveness of nutritional interventions in older adults at risk of malnutrition across different health care settings: Pooled analyses of individual participant data from nine randomized controlled trials

Background & aims: Protein-energy malnutrition is a health concern among older adults. Improving nutritional status by increasing energy and protein intake likely benefits health. We therefore aimed to investigate effects of nutritional interventions in older adults (at risk of malnutrition) on change in energy intake and body weight, and explore if the intervention effect was modified by study or participants’ characteristics, analysing pooled individual participant data. Methods: We searched for RCTs investigating the effect of dietary counseling, oral nutritional supplements (ONS) or both on energy intake and weight. Principle investigators of eligible studies provided individual participant data. We investigated the effect of nutritional intervention on meaningful increase in energy intake (>250 kcal/day) and meaningful weight gain (>1.0 kg). Logistic generalized estimating equations were performed and ORs with 95% CIs presented. Results: We included data of nine studies with a total of 990 participants, aged 79.2 ± 8.2 years, 64.5% women and mean baseline BMI 23.9 ± 4.7 kg/m2. An non-significant intervention effect was observed for increase in energy intake (OR:1.59; 95% CI 0.95, 2.66) and a significant intervention effect for weight gain (OR:1.58; 95% CI 1.16, 2.17). Stratifying by type of intervention, an intervention effect on increase in energy intake was only observed for dietary counseling in combination with ONS (OR:2.28; 95% CI 1.90, 2.73). The intervention effect on increase in energy intake was greater for women, older participants, and those with lower BMI. Regarding weight gain, an intervention effect was observed for dietary counseling (OR:1.40; 95% CI 1.14, 1.73) and dietary counseling in combination with ONS (OR:2.48; 95% CI 1.92, 3.31). The intervention effect on weight gain was not influenced by participants’ characteristics. Conclusions: Based on pooled data of older adults (at risk of malnutrition), nutritional interve