To examine the relationship of fatigue in people with rheumatoid arthritis (RA) with self-efficacy, positive and problematic aspects of social support, and demographic and disease-related variables. Out-patients with at least 5 yr RA were studied. Fatigue was measured with a visual analogue scale. Other variables included were: positive social support [Social Support List-Interactions (SSL12-I)] and problematic social support; self-efficacy towards coping with RA and towards mobilizing support; health status (Dutch-AIMS2); and laboratory tests: erythrocyte sedimentation rate (ESR), haemoglobin (Hb) and rheumatoid factor (RF); and disease duration. A total of 229 out-patients were included. Fatigue correlated with all scales of the Dutch-AIMS2: with pain, physical function and affect (P < 0.001). There was no significant correlation with social support, but there was a highly significant correlation of fatigue with problematic social support (P < 0.001). Both forms of self-efficacy correlated strongly with fatigue: patients with high self-efficacy expectations towards coping with RA, and towards mobilizing the social network (P < 0.001), had less fatigue. In the regression analysis to explain the variation in fatigue, only pain, self-efficacy expectations towards coping with RA, and towards asking for help and problematic social support remained significant. Fatigue can to a large extent (37%) be explained by pain, self-efficacy towards coping with RA, and towards asking for help and problematic social support. It is known that self-efficacy can be enhanced by self-management courses and it may thus be possible to improve fatigue.