Background
The economic burden of asthma, which relates to the degree of control, is €5 billion annually in Italy. Pharmacists could help improve asthma control, reducing this burden. 
Objective To evaluate the effectiveness and cost-effectiveness of Medicines Use Reviews provided by community pharmacists in asthma. 
Methods
Design, setting and participants Cluster randomised, multi-centre, controlled trial in adult patients with asthma, conducted in 15 of the 20 regions of Italy between September 2014 and July 2015. After stratification by region, community pharmacists were randomly allocated to group A (trained in and delivered the intervention at baseline) or B (training and delivery three months later), using computerised random number generation in blocks of 10. Each recruited up to five patients, with both groups followed for nine months.
Intervention Systematic, structured face-to-face consultation with a pharmacist, covering asthma symptoms, medicines used, attitude towards medicines and adherence, recording pharmacist-identified pharmaceutical care issues (PCIs).
Primary outcome Asthma control, assessed using the Asthma-Control-Test (ACT) score (ACT?20 represents good control). Secondary outcomes  Number of active ingredients, adherence, number of PCIs, cost-effectiveness compared with usual care. 
Blinding was not possible for either pharmacists or patients. Assessment of outcomes was conducted by researchers blind to group allocation.

Results Numbers of pharmacists and patients enrolled were 283 (A=136; B=147) and 1263 (A=600; B=663), numbers completing were 201 (A=97; B=104) and 816 (A=400; B=416), respectively. Patients were similar in age and gender and 56.13% (458/816) had poor/partial asthma control. Median ACT score at baseline differed between groups (A=19, B=18; p<0.01). Odds ratio for improved asthma control was 1.76 (95% CI 1.33 to 2.33) and number needed to treat 10 (95% CI 6 to 28). Number of active ingredients reduced by 7.9% post-intervention (p<0.01). Adherence improved by 35.4% three months post-intervention and 40.0% at six months (p<0.01). Pharmacists identified 1256 PCIs (mean 1.54/patient), mostly need for education, monitoring and potentially ineffective therapy. The probability of the intervention being more cost-effective than usual care was 100% at nine months. 
Conclusions This community pharmacist-based intervention demonstrated both effectiveness and cost-effectiveness. It has since been implemented as the first community pharmacy cognitive service in Italy.
TRN: ISRCTN72438848 (registered 5th January 2015)