Basheti 2008.

Study characteristics
Methods	Design: single‐blind parallel‐group cluster‐randomised controlled trial (i.e. each randomised pharmacist is a 'cluster') Duration: 26 weeks Setting: community pharmacies, Australia Trial registration: not reported
Participants	Population: 31 pharmacists randomised; 97 patients with asthma subsequently enrolled. 16 pharmacists (53 patients) randomised to pharmacist–participant educational intervention and inhaler technique labels; 15 pharmacists (44 participants) randomised to peak flow measurement training (control) Age: mean participant age (SD) in the intervention group: Diskus 51.4 (8.3) years and Turbuhaler 45.48 (19.7) years. Mean participant age (SD) in the control group: Diskus 41.1 (20.0) years and Turbuhaler 38.85 (18.4) years Baseline asthma severity: intervention group: mean peak flow variability (Min%Max) (SD) Diskus 71.5 (9.7) and Tubuhaler 74.8 (9.2); control group: mean peak flow variability (Min%Max) (SD) Diskus 76.0 (7.2) and Tubuhaler 71.2 (8.7) Inclusion criteria: aged 14 years, doctor‐diagnosed asthma, use of inhaled corticosteroid by Turbuhaler or Diskus with or without long‐acting beta2‐agonist, no change in asthma medication or dose for 1 month Exclusion criteria: did not self‐administer medication, did not speak or understand English, were not able to return for all visits, were involved in another study Percentage withdrawn: 1 pharmacist withdrew from the intervention group, and 3 from the control group. 16% of participants withdrew across both groups Other allowed medication: not reported
Interventions	Intervention summary: Both groups of pharmacists educated participants on measurement of peak expiratory flow (PEF) for assessment of peak flow variability. Active pharmacists also assessed participants' inhaler technique, without giving them feedback. At visit 2, active pharmacists again assessed participants' inhaler technique, then educated participants using a specialised "Show and Tell" Inhaler Technique counselling service, going through each step on the checklist to describe and demonstrate correct use, repeated up to 3 times if necessary, until the participant had correct technique on all steps. The pharmacist then used a highlighter pen to identify incorrect steps from that day’s initial assessment on an "Inhaler Technique Label", which was attached to the highlighted label on the participant's own controller medication inhaler. At each subsequent visit, active pharmacists repeated inhaler technique assessment/education and placed a new label on the participant's replacement inhaler (or on the old one if still in use). If no steps were incorrect on the initial assessment at any visit, the label was attached to the participant's inhaler with no highlighting. Five visits over 6 months Control summary: Both groups of pharmacists educated participants on measurement of peak expiratory flow (PEF) for assessment of peak flow variability. During the run‐in period, 1 researcher independently assessed inhaler technique for all control participants (to establish their baseline inhaler technique without providing information about correct inhaler technique to control pharmacists). No education was provided to participants during these assessments. For control participants, the researcher re‐assessed inhaler technique at the end of the study, then provided inhaler technique counselling
Outcomes	Outcomes measured: peak flow variability, calculated as Min%Max (lowest morning PEF over 2 weeks, as percentage of highest PEF over the same period), Inhaler Technique Score, categorisation of asthma severity based on Australian Asthma Management Handbook, asthma‐related quality of life (AQOL), perceived control (PC) Technique assessment method used: 9‐step inhaler technique checklist
Notes	Type of publication: 1 conference abstract, 1 peer‐reviewed journal article, 1 letter to the editor Funding: funded by the Faculty of Pharmacy, University of Sydney. HK Reddel was funded by the Asthma Foundation of NSW
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"After arrival, pharmacists were allocated randomly by computer‐generated list to Active or Control groups"
Allocation concealment (selection bias)	Unclear risk	No details
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participating pharmacists could not be blinded to group allocation. It is unclear whether participants were aware of their group allocation, although the trial is reported as 'single‐blind', and control participants did not have their inhaler technique assessed at 3 months "in order to avoid drawing their attention to inhaler technique"
Blinding of outcome assessment (detection bias) All outcomes	High risk	No description of procedures intended to blind outcome assessors to group assignment, although the study is reported as 'single‐blind'. Patient‐reported outcomes may have been affected by knowledge of group assignment. It is not clear whether the unblinded participating pharmacist, or a blinded researcher, assessed objective outcomes such as inhaler technique
Incomplete outcome data (attrition bias) All outcomes	High risk	Although only 5 pharmacists dropped out, report states that 116 participants were enrolled, but only 97 returned from the second visit, at which the first intervention was delivered. It is not clear how many participants were initially enrolled into each arm. It is also unclear how many participants were assessed at 6 months for each outcome. Although it is stated that 97 'completed the study', the total n for those with correct technique at the end equals 84. It is not clear on what number the continuous outcomes were based
Selective reporting (reporting bias)	Unclear risk	No prospective trial registration identified. All outcomes listed in Methods appear to be reported in a peer‐reviewed publication and a letter to the editor
Other bias	Low risk	None noted