OBJECTIVE: Malnutrition affects 5-10% of elderly people living in the community. A few studies suggest that nutritional intervention may reduce health care costs. The present study included malnourished elderly patients living at home. It aimed to compare health care costs between patients that were prescribed ONS by their general practitioner and those who were not, and to assess the effect of ONS prescription on the risk of hospitalisation. METHODS: This prospective multicentre observational study included malnourished patients ≥70 years old who lived at home. Patients were defined as malnourished if they presented with one or more of the following criteria: weight loss ≥5% in 1 month, weight loss ≥10% in 6 months, BMI <21 kg/m2 RESULTS: We analysed 191 patients. At baseline, the 133 patients (70%) who were prescribed ONS were more disabled (p < 0.001) and had poorer perception of their health (p = 0.02), lower QoL (p = 0.04) and lower appetite (p < 0.001) than the 58 patients (30%) who were not prescribed ONS. At 6 months, appetite had improved more in the ONS prescription group (p = 0.001). Weight change was not different between groups. Patients prescribed ONS were more frequently hospitalised (OR 2.518, 95% CI: [1.088; 5.829] hosp; p = 0.03). Analyses of adjusted populations revealed no differences in health care costs between groups. In the ONS prescription group, we identified that health care costs were lower (p = 0.042) in patients with an energy intake from ONS ≥ 500 kcal/d (1389 ± 264 €) vs. < 500 kcal/d (3502 ± 839 €). The risk of hospitalisation was reduced 3 and 5 times when the intake from ONS was ≥30 g of protein/day or ≥500 kcal/d, respectively. CONCLUSIONS: ONS prescription in malnourished elderly patients generated no extra heath care cost. High energy and protein intake from ONS was associated with a reduced risk of hospitalisation and health care costs.