The aim of this study was first, to assess the presence of medical conditions that might interfere with walking; second, to assess the differences in walking capacity, perceived exertion and physical complaints between lean, obese and morbidly obese women; and third, to identify anthropometric, physical fitness and physical activity variables that contribute to the variability in the distance achieved during a 6-minute walk test in lean and obese women. A total of 85 overweight and obese females (18-65 years, body mass index (BMI) > or = 27.5 kg m(-2)), 133 morbidly obese females (BMI > or = 35 kg m-2) and 82 age-matched sedentary lean female volunteers (BMI < or = 26 kg m(-2)) were recruited. Patients suffering from severe musculoskeletal and cardiopulmonary disease were excluded from the study. Prior to the test, conditions that might interfere with walking and hours of TV watching were asked for. Physical activity pattern was assessed using the Baecke questionnaire. Weight, height, body composition (bioelectrical impedance method), isokinetic concentric quadriceps strength (Cybex) and peak oxygen uptake (peakVO2_bicycle ergometer) were measured. A 6-minute walk test was performed and heart rate, walking distance, Borg rating scale of perceived exertion (RPE) and physical complaints at the end of the test were recorded. In obese and particularly in morbidly obese women suffering from skin friction, urinary stress incontinence, varicose veins, foot static problems and pain, wearing insoles, suffering from knee pain, low back pain or hip arthritis were significantly more prevalent than in lean women (P < 0.05). Morbidly obese women (BMI > 35 kg m(-2)N = 133) walked significantly slower (5.4 km h(-1)) than obese (5.9 km h(-1)) and lean women (7.2 km h(-1), P < 0.05), were more exerted (RPE 13.3, 12.8 and 12.4, respectively, P < 0.05) and complained more frequently of dyspnea (9.1%, 4.7% and 0% resp., P < 0.05) and musculoskeletal pain (34.9%, 17.7% and 11.4% resp., P < 0.05) at the end of the walk. In a multiple regression analysis, 75% of the variance in walking distance could be explained by BMI, peakVO2, quadriceps muscle strength age, and hours TV watching or sports participation. These data suggest that in contrast with lean women, walking ability of obese women is hampered not only by overweight, reduced aerobic capacity and a sedentary life style, but also by perceived discomfort and pain. Advice or programs aimed at increasing walking for exercise also need to address the conditions that interfere with walking, as well as perceived symptoms and walking difficulties in order to improve participation and compliance.