LAUSR

Cystic fibrosis (CF) is the most common lifelimiting genetic disease. The treatment landscape of CF lung disease is changing dramatically with widespread availability of CF transmembrane conductance regulator (CFTR) modulator treatments targeting the primary protein defect. The latest generation of CFTR modulators (elexacaftor–tezacaftor–ivacaftor) has shown remarkable improvements in lung function, sweat chloride concentration, exacerbation rates and healthrelated quality of life of people with CF (pwCF), 2 which in turn impacts individuals’ daily airway clearance therapy (ACT) and physical activity routine. In light of the improvements in the overall health status of pwCF, investigating the role of physical activity and exercise is timely and desired by the CF community. Two papers in this issue of Thorax focus on physical activity and exercise in pwCF, respectively, either as a substitute for traditional ACT or as an intervention to promote physical activity. 5 Saynor et al addressed 1 of the top 10 research questions in CF ‘Can exercise replace chest physiotherapy’ to support removal and prevent accumulation of secretions. The research group conducted a UKbased eDelphi survey among physicians, physiotherapists, caregivers of pwCF and pwCF aiming to establish consensus on the role of exercise as ACT as well as on the type, intensity and duration of exercise to be tested against traditional ACTs in a prospective feasibility trial. The work by Saynor et al is of great interest to the CF community as a substantial proportion of pwCF (44%–48%) already substitute traditional ACTs with physical activity and exercise, 7 although the current evidence on the longterm effects supporting this practice is limited. The study team developed statements that were scored by the Delphi participants on a 0–9 scale (0—total disagreement, 9—total agreement). Consensus was defined a priori as >70%. In a threeround eDelphi, experts concluded that exercise could be used as a form of ACT in CF (75% consensus). To be considered as an alternative to traditional ACTs, the authors suggest (weightbearing) aerobictype activities lasting 20 min and longer, performed at an intensity that causes deep breathing, and integrates huffs and coughs during or after exercise facilitating removal of secretions. The study by Saynor et al is an important step providing the methodological framework for future trial designs aiming to study the efficacy and safety of exercise as a substitute for traditional ACTs. Designing such a trial is a challenge and requires careful conceptualisation and definition of trial eligibility criteria and type of ACT to be studied. In addition, individuals’ experiences and preferences for exercise need to be considered. Saynor et al recruited a sufficiently sized, multidisciplinary expert panel. The authors used various recruitment channels to recruit their expert panel, although, potential selection bias and lack of representativeness of panellists representing the four expert groups is difficult to ascertain. Finally, the study by Saynor et al provides a basis for testing exercise as a substitute for traditional ACT in a prospective fashion; its efficacy, safety and applicability to different target groups including those with advanced CF lung disease and severe deconditioning remains to be investigated. In the second paper, Cox et al reported on the effects of a 12week webbased randomised controlled trial (RCT) using elements of behaviour change theory in pwCF with a 6 months followup. Study participants were recruited during hospital admission for a pulmonary exacerbation across eight CF centres in Australia. The intervention group participants received an activity monitor and access to a webbased application to set individualised physical activity goals, and to record and monitor their physical activity. The control group received usual care. The primary endpoint was change in objectively measured moderatetovigorous physical activity from baseline to the end of the 12week intervention. The intentiontotreat analysis (n=63, n=107 randomised) did not reveal any betweengroup differences in the change in moderatetovigorous physical activity either at 12 weeks or at 6 months posthospital discharge. Secondary endpoints (eg, change in lung function, exercise capacity, healthrelated quality of life) were not different between groups. The study by Cox et al is an outcome assessor blinded RCT targeting pwCF admitted for inhospital treatment of a pulmonary exacerbation. This subgroup of patients has received little attention in physical activity trials so far, although it is a particularly interesting subgroup. A substantial proportion of pwCF do not fully recover short term with respect to lung function, 10 showing remarkably reduced physical activity levels post hospitalisation for pulmonary exacerbation, with potential consequences for their longterm physical activity behaviour. The trial investigators faced several challenges; study limitations were comprehensively addressed. Recruitment of study participants turned out to be demanding, a wellknown problem in physical activity intervention trials in this population. 13 Both study groups appeared to be physically active and their measured baseline physical activity levels were—on average—above guideline recommended levels. Possible selfselection of physically active individuals limits the applicability of the study findings to less active individuals. Moreover, despite positive feedback on the web platform from pwCF in a pilot trial, access rates in this study were low with only 40% of intervention group participants using the platform (ie, at least one login), and with large variation in the number of individual entries and goal sets. This is different to a large international, multicentre, partially supervised physical activity intervention using a webbased activity diary and lowthreshold supervision by professionals. Access rates were >95%, and 58% of the participants were able to achieve the intervention targets, an increase of at least 2 hours of reported vigorous physical activity per week relative to baseline levels over a period of 6 months. Intervention delivery and access rates to the web platform could have possibly been improved by personal contact with the study team providing individualised motivational support. Finally, the analysis was based on a reduced study sample (59% of randomised participants) Correspondence to Dr Thomas Radtke, Epidemiology, Biostatistics and Prevention Institute, University of Zurich, Zurich, Switzerland; thomas. radtke@ uzh. ch Editorial