Chronic obstructive pulmonary disease (COPD) is a major cause of morbidity and mortality in Australia and internationally. Pulmonary rehabilitation is an important component of COPD management due to the high quality evidence demonstrating that these programs can improve the burden of disease. However, there is considerable variation in the individual response of participants referred to pulmonary rehabilitation, with a significant number of individuals not achieving significant benefit in exercise capacity or quality of life post-program. Furthermore, despite access to pulmonary rehabilitation being limited, there is little evidence examining how pulmonary rehabilitation programs manage demand on their service. In addition, recent studies have reported non-completion rates for pulmonary rehabilitation programs between 20-30%.
The objectives of the first study (Chapter 3) were to determine the participant entry criteria used by Australian-based pulmonary rehabilitation programs and the factors that influence selection. This prospective cross-sectional observational study invited all programs registered on the Australian Lung Foundation’s pulmonary rehabilitation database to complete a specifically designed survey on participant selection. The response rate was 79/195 (41%). Forty respondents (51%) reported prioritising participant referrals commonly due to disease severity; requirement for a medical procedure; upon medical request; or if a participant was considered more likely to benefit. Fifty-eight (73%) respondents reported using used entry criteria to select participants for their program which was mainly for safety reasons and performance-based expectations including a participant’s likelihood to benefit. Increased demand on individual programs was related to prioritising referrals (p<0.001) and was reported by twelve programs as a reason for using participant entry criteria. The major findings were that program coordinators commonly prioritise referrals and use participant entry criteria to manage clinical demand with performance-based expectations an important consideration. However, programs do not report using the existing evidence identifying responders to pulmonary rehabilitation in selecting participants for program inclusion.
The main objective for the second study (Chapter 4) was to determine if existing baseline measures including the markers of disease severity, pre-program physical activity, hospitalisations and comorbidities can predict responders in six minute walk distance (6MWD) following pulmonary rehabilitation. All participants with COPD who attended a tertiary hospital’s pulmonary rehabilitation program between 2004 and 2009 were evaluated. A participant was classified a responder with improvement in 6MWD. Two hundred and three participants were assessed in this retrospective study. This study found that a younger age was an independent predictor of benefit in both 6MWD models, while participants with metabolic disease were identified as more likely to benefit when using a larger threshold to define responders.
The primary objective of the third study (Chapter 5) was to determine if the participant’s baseline measures can predict responders in 6MWD or quality of life using a prospective observational design. Participants with COPD who attended the pulmonary rehabilitation program between 2010 and 2012 were recruited. Baseline measures assessed included systemic inflammatory markers, physical activity, quadriceps strength, comorbidities and self-efficacy. A participant was classified a responder with improvement in 6MWD or improvement in the Chronic Respiratory Questionnaire (CRQ) quality of life questionnaire. Eighty-five participants were studied. Forty-nine (58%) and 19 (22%) participants were classified as a responder when using the 6MWD criteria of ≥25 m and ≥61.9 m respectively to define a responder. Forty-four (52%) participants were classified as a responder in the CRQ. In a multiple logistic regression model, quadriceps weakness (p=0.028) and higher self-efficacy scores (p=0.045) identified responders in 6MWD (≥25 m criteria); and quadriceps weakness (p=0.008) and participants with metabolic disease (p=0.019) identified responders in 6MWD (≥61.9 m criteria). No measure identified participants likely to respond in CRQ.
The objectives of the fourth study (Chapter 6) were to identify participant characteristics at baseline that independently predict pulmonary rehabilitation program non-completion; and to compare these characteristics against the participant’s reported reasons for non-completion. Participants with COPD who attended a standardised pulmonary rehabilitation program between 2010 and 2012 were recruited. Participants who attended <12/16 sessions were arbitrarily classified as a non-completer. Non-completers were asked to answer a survey about their pulmonary rehabilitation experience. Baseline measures used to assess for differences between the completer and non-completer cohorts included measures of disease severity, comorbidities, quadriceps strength, smoking history, social support and the season when each participant commenced rehabilitation. Twenty-six participants (23.4%) of the 111 eligible participants were classified as non-completers. The only independent predictors associated with program non-completion were participants living alone (p=0.042) and programs that commenced in winter (p=0.011) but neither measure was reported by participants as a reason for non-completion. The reasons given for non-completion by the twenty interviewed non-completers were grouped into: medical reasons (75%), other personal reasons (30%) and external barriers (45%).