The aim of this study was to compare the effects of interval training (3-min intervals) with continuous training on peak exercise capacity (W peak), physiological response, functional capacity, dyspnoea, mental health and health-related quality of life (HRQoL) in patients with moderate or severe COPD. Sixty patients exercised twice weekly for 16 weeks after randomisation to interval- or continuous training. Target intensity was 80% of baseline W peak in the interval group (I-group) and 65% in the continuous group (C-group). Patients were tested by spirometry, ergometer cycle test, cardiopulmonary test and a 12 min walk test. Dyspnoea was measured by the dyspnoea scale from Chronic Obstructive Disease Questionnaire (CRDQ), mental health by Hospital Anxiety and Depression scale (HAD) and HRQoL by the Medical Outcomes Survey Short Form 36 (SF-36). After training, W peak, peak oxygen uptake (VO(2) peak) and exhaled carbon dioxide (VCO(2) peak) increased significantly in both groups, no significant differences between the groups. Minute ventilation (V(E) peak) increased only in the C-group. At identical work rates (isotime) VO(2), VCO(2) and V(E) were significantly more decreased in the I-group than in the C-group (p<0.05). Functional capacity, dyspnoea, mental health, and HRQoL improved significantly in both groups, no difference between the groups. Interval training and continuous training were equally potent in improving peak exercise capacity, functional exercise capacity, dyspnoea, mental health and HRQoL in patients with moderate or severe COPD. At isotime, the physiological response to training differed between the groups, in favour of the interval training.