The beneficial effect of a 1-yr vegetarian diet in RA has recently been demonstrated in a clinical trial. We have analysed stool samples of the 53 RA patients by using direct stool sample gas-liquid chromatography of bacterial cellular fatty acids. Based on repeated clinical assessments disease improvement indices were constructed for the patients. At each time point during the intervention period the patients in the diet group were then assigned either to a group with a high improvement index (HI) or a group with a low improvement index (LI). Significant alteration in the intestinal flora was observed when the patients changed from omnivorous to vegan diet. There was also a significant difference between the periods with vegan and lactovegetarian diets. The faecal flora from patients with HI and LI differed significantly from each other at 1 and 13 months during the diet. This finding of an association between intestinal flora and disease activity may have implications for our understanding of how diet can affect RA.