Elemental diets provide food in its simplest formulation and have been used in the treatment of rheumatoid arthritis (RA) and other chronic inflammatory diseases. Such a diet is supposed to be less antigenic to the human immune system than normal food. The aim of this study was to evaluate the clinical effect of an artificial peptide diet as a temporary supplement to conventional treatment. Patients with active RA were single-blindly randomized either to a liquid elemental peptide-diet for four weeks or to continuation of the usual food (control group). In the diet group all normal foods were renounced. Thirty patients were included and followed for six months. The outcome measurements were pain intensity, morning stiffness, HAQ-score, number of swollen joints, joint tenderness, erythrocyte sedimentation rate, and patient's global assessment of health. Two of the fifteen patients assigned to the diet dropped out. The diet resulted in a transient but statistically significant improvement in the average level of pain (P = 0.02), in HAQ-score (P=0.03), and a significant reduction in Body Mass Index (P=0.001). Only one patient in the diet group had a clear remission. Side-effects were frequent but compliance good. The study showed that the peptide diet can improve some subjective and objective disease parameters. Due to the low remission ratio the peptide diet is not a treatment of choice in unselected RA-patients. but the peptide diet might be beneficial to a subset of RA-patients, e.g. patients where foods aggravate disease activity.