Background: To examine whether the nutritional status of aged undernourished residents in care could be improved through dietary modification to increase energy intake but not portion size. Methods: A 12-week cluster randomised controlled trial was carried out in 21 residential care homes. Participants comprised undernourished residents with a body mass index (BMI) <18.5 kg m–2. All menus were analysed to evaluate nutrient provision. Energy and macronutrient intakes of undernourished residents were estimated using 3-day weighed food intake diaries. Those resident in homes randomised to intervention had their usual meals enriched with energy-dense foods to a maximum of +1673 kJ day−1. Results: Of 445 residents screened, 41 (9%) had a BMI <18.5 kg m–2 and entered the study. Despite adequate food provision, energy and macronutrient intakes were below UK dietary reference values. Mean (SEM) energy intake increased [+556 (372) kJ, P = 0.154] in residents allocated to intervention but fell in those residents in ‘control homes’ receiving usual care [−151 (351) kJ, P = 0.676]. Weight change [+1.3 (0.53) kg, P = 0.03] was seen in intervention residents but not in controls [−0.2 (1.5) kg, P = 0.536]. Between-group differences for changes in weight and energy intake were not significant (P = 0.08 and 0.20, respectively). Six residents allocated to the intervention increased their BMI >18.5 kg m–2 (P = 0.018). Conclusions: Achieving weight gain in frail older people is difficult. These results suggest that enriching food could help address undernutrition and slow chronic weight loss. Interventions of a longer duration are needed to confirm or exclude the value of food enrichment.