The effects of detraining subsequent to strength training on neuromuscular function were examined in 39 recent-onset rheumatoid arthritis (RA) patients. Eighteen age- and sex-matched healthy people (H) served as controls. Patients were randomly allocated either to the experimental group (PE), who carried out progressive strength training for 6 months, or to the control group (PC), who maintained only their habitual physical activities. After 6 months, PE returned to their earlier physical activities and strength training was terminated. At baseline, the maximal strength of the trunk extensors (not significant), grip strength and maximal dynamic strength and the shape of the force-time curve of the knee extensors were lower in PE and PC (P < 0.05-0.001) than in H. Strength training in PE led to remarkable increases (P < 0.05-0.001) in the maximal strength of all muscle groups without changes in the shape of the force-time curve. The increases in muscle strength in PE obtained by strength training were lost to a great degree during the detraining period for the isometric trunk extension (P < 0.01) and flexion (P < 0.01) strength and for the dynamic knee extension strength (P < 0.05), but not for the grip strength. In PC, trunk extension and flexion strength decreased significantly throughout the study period. At the post-test, all the strength values in both patient groups were much lower than in H. RA is a chronic disease which seems to need continuous physical exercise with sufficient intensity to minimize/prevent the loss of muscle strength and functional capacity.