Malnutrition has been often suggested as contributing to both the high incidence of hip fracture in elderly people and its complications. In a recent prospective controlled randomized study, the clinical outcome of elderly patients with osteoporotic fracture of the proximal femur (hip fracture) improved by giving a simple oral dietary supplement. This study, however, did not prove that protein was responsible for the clinical improvement since the supplement also contained vitamins and minerals. We addressed this question by comparing the clinical outcome and bone mineral density (BMD) changes in elderly patients with hip fracture, receiving two different dietary supplements with different protein contents. Sixty-two patients (mean age 82) admitted into the orthopedic ward for fracture of the proximal femur were randomized into two groups. One group (n = 33) received 250 ml/day of an oral nutritional supplement containing protein (20.4 g), mineral salts (Ca: 0.525 g) and vitamins A = 750 IU; D3 = 25 IU) for a mean of 38 days. A control group (n = 29) received the same supplement dose, but with no protein, for the same period of time. The clinical course was significantly better in the group receiving protein, with 79% having a favorable course as compared to 36% (p less than 0.02) in the control group during the stay in the recovery hospital. The rate of complications and deaths was also significantly lower in the protein-supplemented vs the control group (52 vs 80%, p less than 0.05) 7 months after hip fracture. The median hospital stay was significantly lower in the protein-supplemented group (69 vs 102 days, p less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)