Balestri 2004.

Study characteristics
Methods	RCT Cross‐over design. No mention of a washout period Single centre: Centro Fibrosi Cistica, Cesena Duration: 2 weeks
Participants	Inclusion criteria: clinically stable Exclusion criteria: not specified 13 participants Age, range: 10 to 41 years Sex: 10 males, 3 females Disease status: all stable at time of study Lung function, FEV1 % predicted (range): 54% to 95%
Interventions	Intervention: exercise (cycle ergometer constant effort 1/2 W/kg) 1 x 30‐minute session per week where participant undertakes 5 minutes exercise followed by a 2.5 minute break, then 3 bouts of cough, repeated 4 times Comparator: chest physiotherapy (underwater PEP), 1 x 30‐minute session per week where, for each of the 4 positions, participant performs 15 expirations against 10 cm water resistance, then 3 bouts of cough, repeated 4 times
Outcomes	Sputum expectorated during treatment (wet weight (g)), acceptability of treatment (VAS), adverse events (SPO2 % desaturation), peak heart rate
Identification	Sponsorship source: not stated Country: Italy Authors name: Elena Balestri Institution: Centro Fibrosi Cistica, Cesena Email: elenabalestri66@gmail.com Address: Centro Fibrosi Cistica, Cesena, FC, Italy
Notes	Abstract only published. Lead author was contacted on 16 April 2019 for further information and results. They confirmed that no further results are available, and no manuscript will be published.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Process of randomisation not described: "13 CF patients underwent the 4 sequences randomly assigned for the 4 treatments."
Allocation concealment (selection bias)	Unclear risk	Allocation concealment not discussed.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding was not possible due to the type of intervention used.
Blinding of participants and personnel (performance bias ‐ subjective outcomes) Subjective outcomes	High risk	Blinding was not possible due to the type of intervention used. Participant awareness of the intervention may have led to some impact on reporting of subjective measures, such as QoL, due to pre‐existing perceptions or preferences on the benefits of exercise versus traditional airway clearance techniques.
Blinding of participants and personnel (performance bias ‐ objective outcomes) Objective outcomes	Unclear risk	Blinding was not possible due to the type of intervention used; however, this was unlikely to confer any impact on objective outcome measures such as lung function and exercise testing.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	It is unknown whether or not the outcome assessors were aware of the intervention assignments. It is unclear who the personnel responsible for overseeing the intervention and those responsible for collection of data were.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	All outcomes reported in narrative form only or partially reported with final mean values and standard deviations. No access to raw data. There seems to be zero attrition in this study, with all 13 participants completing the trial; however, there was no reference to participant withdrawal or intention‐to‐treat analysis.
Selective reporting (reporting bias)	Unclear risk	No protocol for comparison. Only methodology available in abstract. This may have been limited by the abstract word count.
Other bias	Unclear risk	Abstract only available. There is no mention of how participants were recruited into this study. The washout period used during the study was not reported, as methods stated only that 4 treatment days were completed during a 2‐week period, with no explicit statement about whether the days were conducted with fixed or variable gaps.  There was no reported power calculation. There is an unclear risk associated with the potentially insufficient sample size.