In a controlled, single blind clinical trial we have demonstrated recently a beneficial effect of fasting and vegetarian diet in RA. In the present study we compared 53 patients who participated in this clinical trial with 71 other RA patients with regard to some psychological parameters. The patients who participated in the clinical trial differed significantly from other RA patients. Firstly, they had a higher internal score and a lower chance score on the Multi-dimensional Health Locus of Control Scale (MHLCS). Secondly, their belief in the effect of ordinary medical treatment, evaluated by a 10-cm visual analogue scale, was lower, and their belief in the effect of 'alternative', unconventional forms of treatment was higher. Of the patients who were randomized to a vegetarian diet, there was no significant difference between diet responders and diet non-responders with regard to the MHLCS scores. But, diet responders had a significantly lower belief in the effect of ordinary medical treatment compared with diet non-responders. The psychological distress imposed on the patients by changing from an omnivorous diet to a vegetarian diet was monitored during the clinical trial by means of the General Health Questionnaire. Throughout the clinical trial, this variable favoured the vegetarians compared with the omnivorous and the diet responders vs the diet non-responders. We conclude, firstly, that patients with certain psychological characteristics were selected to the clinical trial; secondly, that the MHLCS scores could not explain the clinical improvement, but it may have been influenced by the patients' beliefs in ordinary and 'alternative' forms of treatment; and thirdly, that dietary treatment decreased psychological distress.