OBJECTIVE In patients with intermittent claudication (IC) a structured walking exercise program improves exercise performance. However, few studies have evaluated the effects of exercise training on functional status during daily activities. We hypothesized that a supervised exercise training program would improve functional status in patients with IC, with 24 weeks of training more beneficial than 12 weeks. A secondary aim was to evaluate the effects of strength training and combinations of strength and treadmill training on functional status. METHODS Twenty-nine men with disabling IC were randomized to 12 weeks of either supervised treadmill training (3 hr/wk at a work intensity sufficient to produce claudication), strength training (3 hr/wk of resistive training of six muscle groups of each leg), or to a nonexercising control group. Functional status was assessed by questionnaires characterizing walking ability (Walking Impairment Questionnaire, WIQ), habitual physical activity level (Physical Activity Recall, PAR), and physical, social, and role functioning, well-being, and overall health (Medical Outcomes Study SF-20, MOS). Patients alos had their activity levels monitored with an activity monitor (Vitalog). RESULTS After 12 weeks of treadmill training PAR scores increased by 48 metabolic equivalent hr/wk, the MOS physical functioning score by 24 percentage points, and the number of bouts of walking activity measured by the Vitalog by 4.5 bouts/hr (all p < 0.05). No changes were seen in WIQ scores. After 12 additional weeks of treadmill training improvements initially observed in the PAR, MOS, and Vitalog scores were maintained, and in addition the ability to walk distances (WIQ) improved by 31 percentage points, and the IC severity score had improved by 29 percentage points (both p < 0.05). After 12 weeks of strength training patients improved their WIQ walking speed, stair climbing scores, and MOS well-being scores with no other changes in functional status. Subjects in the control group did not improve functional status by any measure. Twelve weeks of treadmill training after the strength training program maintained WIQ walking speed scores, and activity level defined by Vitalog improved. Twelve weeks of combined treadmill and strength training after the control period had no effect on functional status. CONCLUSIONS A supervised treadmill training program improved functional status during daily activities, with 24 weeks more effective than 12. In addition, treadmill training alone was more effective in improving functional status in patients with IC than strength training or combinations of the training modalities.