There are 20 questions which are grouped into one of four domains: dyspnoea (5 individualised dyspnoea questions), fatigue (4 questions), emotional function (7 questions), and mastery (4 questions), as well as total score. Each question was scored from one to seven, with higher scores indicating less impairment Raf inhibitor in health status. A change of 0.5 in the mean score per domain (calculated by dividing the overall score

by the number of questions) has been shown to be associated with a minimal important difference in health status (Jaeschke et al 1989). This means that a minimal important difference would be 2.5 for dyspnoea, 2 for fatigue, 3.5 for emotional function, 2 for mastery, and 10 for the total Chronic Respiratory Disease Questionnaire score. The minimal important difference of the endurance shuttle walk test has not yet been published. However, based on previous studies using other endurance tests, an improvement of 105 seconds has been suggested as meaningful (Casaburi

2004). We sought to detect a minimum difference of 120 seconds in the endurance shuttle walk test between groups. Assuming a SD of 108 seconds (Sewell et al 2006), 36 participants (18 per group) would provide 85% power to detect as significant, at the two-sided 5% level, a 120-second difference in endurance shuttle walk test time between the walk and cycle groups, allowing for a 15% loss to follow-up. Repeated-measures analysis of variance was used to compare the changes between groups from pre- to post-training. The standardised response mean (SRM) was Gefitinib used to assess responsiveness of the endurance shuttle walk test using data from all participants. The SRM is the ratio of change in average scores over time to the SD of change (mean endurance shuttle walk test score at the end

of training minus mean endurance shuttle walk test score at baseline/SD of the change). An SRM of approximately 0.2 is small, 0.5 is moderate, and greater than 0.8 is highly responsive (Garratt et al 1994). The flow of participants is presented in Figure 1. Thirtysix participants were recruited PAK6 and 32 (89%) completed the study with 17 in the walk group and 15 in the cycle group. Baseline characteristics of participants are presented in Table 1. Participants were trained by the same physiotherapist in a rehabilitation gymnasium at Concord Repatriation General Hospital, Sydney. The training therapist was a qualified physiotherapist with extensive experience in exercise training in people with COPD. The mean attendance of participants for both groups was 23 sessions (SD 1) and no adverse events were reported. All participants were able to achieve the prescribed increments in duration at the appropriate time points before training intensity was progressed. The progression of training intensity is presented in Figure 2.