Milos 2013.

Methods	Randomised trial
Participants	374 elderly patients (intervention 185; control 189) ≥ 75 years, and living in nursing homes or the community 4 pharmacists with at least 4 years' experience of performing medication reviews Skåne County, Sweden Year of study: September 2011 to February 2012.
Interventions	Pharmacists conducted a medication review for patients based on electronic medical records without interaction. Recommendations were sent to the patient's physician by team rounds, written contact, personal contact or phone.
Outcomes	Drug‐related problems Number of patients with potentially inappropriate medications Number of patients with unplanned admissions All‐cause mortality
Notes	Funding source: The study was conducted with government funding for projects involving improvement of drug therapy in the elderly. Conflict of interest: None stated
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The randomisation was performed using a random‐number generator and stratified only for geographic area.
Allocation concealment (selection bias)	Low risk	After inclusion, the pharmacist used closed, non‐transparent envelopes to randomise the patient to 1 of 2 groups: control or intervention.
Blinding of participants and personnel (performance bias) All Outcomes/Outcome 1	Unclear risk	Unclear from information provided
Blinding of outcome assessment (detection bias) All Outcomes/Outcome 1	Low risk	Drug‐related outcomes (number of drugs, drug‐related problems, etc.) unlikely to be biased.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Between group attrition < 10%. Overall completion rate > 80%.
Selective reporting (reporting bias)	Low risk	All outcomes reported.
Other bias	Low risk	None identified