We have previously reported that a significant improvement can be obtained in rheumatoid arthritis patients by fasting followed by an individually adjusted vegetarian diet for one year. The patients who changed their diet could be divided into diet responders and diet nonresponders. After the clinical trial the patients were free to change diet or medication and after approximately one year they were asked to attend a new clinical examination. We compared the change from baseline (i.e. at the time of study entry) to the time of the follow-up examination for diet responders, diet nonresponders and controls who ate an omnivorous diet. The following variables favoured diet responders: pain score, duration of morning stiffness, Stanford Health Assessment Questionnaire index, number of tender joints, Ritchie's articular index, number of swollen joints, ESR and platelet count [corrected]. The difference between the three groups were significant for all the clinical variables, except for grip strength. There was no significant difference between the groups with regard to laboratory or anthropometric variables. At the time of the follow-up examination all diet responders but only half of the diet nonresponders still followed a diet. Our findings indicate that a group of patients with rheumatoid arthritis benefit from dietary manipulations and that the improvement can be sustained through a two-year period.