To investigate the effect of reduced frequency of training and detraining on lumbar extension strength, 50 subjects (34 men, aged 34 +/- 11 yrs; and 16 women, aged 33 +/- 11 yrs) were recruited from ongoing strength training programs. Initial training consisted of 10 or 12 weeks of variable resistance lumbar extension strength exercise to volitional fatigue 1, 2, or 3 times a week. After the initial training, subjects reduced the frequency of training to once every 2 weeks (n = 18) or once every 4 weeks (n = 22) for 12 weeks. Only the frequency of training was changed; the mode, volume, and intensity of exercise remained constant for both reduced frequency of training groups. An additional ten subjects terminated training and acted as controls (detraining group). Isometric lumbar extension strength was evaluated at seven angles through a 72 degree range-of-motion before training, after training, and after reduced frequency of training or detraining. Analysis of variance with repeated measures indicated that lumbar extension strength improved (P < or = 0.05) for all groups after the initial 10 or 12 weeks of training. After 12 weeks of reduced training, the once every 2 weeks and once every 4 weeks groups showed no significant reduction in lumbar extension strength at any angle tested, whereas the detraining group demonstrated an average 55% reduction in strength. These findings indicate that isometric lumbar extension strength can be maintained for up to 12 weeks with a reduced frequency of training as low as once every 4 weeks when the intensity and the volume of exercise are maintained.