Pryor 1994

Methods	RCT. Cross‐over design (no washout period). Location: single centre in UK. Duration: 2 days.
Participants	24 participants (14 males, 10 females) with positive sweat test for CF were randomised, but only 20 included in the study. 4 participants withdrew (3 males, 1 female); 2 had to have drug regimens changed; 2 withdrew due to technical problems with oximeter and sputum collection. Age range 16 ‐ 36 years; mean age 24.4 years. Stable as according to no clinical findings.
Interventions	ACBT versus flutter and ACBT. 2 supervised treatments per day then alternate treatment on following day. In addition 2 different postural drainage positions were used, but no statistical difference noted between treatments.
Outcomes	RFTs, wet sputum weight and participant satisfaction.
Notes	No statistical data presented on RFTs apart from there being no statistical significance in the results. Most found both regimens easy to use, with majority finding ACBT easier to clear secretions. 17 out of 20 felt they would continue with ACBT. On follow‐up the 3 participants who said they would continue with the flutter at home had discontinued it within the month and resumed ACBT.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Sequentially admitted into the study, randomised to treatment regimens, but method not stated.
Allocation concealment (selection bias)	Unclear risk	Not discussed.
Blinding (performance bias and detection bias) All outcomes	Low risk	Independent observer used to measure lung function, sputum weight and oxygen saturations.
Incomplete outcome data (attrition bias) All outcomes	Low risk	4 withdrawals after randomisation, (reasons given) analysis only on 20 remaining.
Selective reporting (reporting bias)	Unclear risk	Not discussed.
Other bias	Unclear risk	Not discussed.