Delivering a home-based medication review, process measures from the HOMER randomised controlled trial

Abstract
Objectives The HOme-based MEdication Review (HOMER) trial investigated whether home-based medication review by pharmacists could decrease hospital re-admission in older people. This trial demonstrated that the intervention increased admissions by 30% (P=0.009). This unexpected finding provoked significant interest. This paper describes the intervention in detail and the process measures recorded by review pharmacists, and investigates whether results differed according to pharmacist characteristics. Method 437 patients were randomised to the intervention, which involved two pharmacist home visits within two and eight weeks of discharge, and 435 were randomised to usual care. An analysis was undertaken of the process measures and to determine whether admission rates differed within the intervention group according to the type of pharmacist performing the review. Setting Norfolk or Suffolk patients aged over 80 years discharged to their own home after an emergency admission (any cause), and taking two or more medications daily. Key findings Twenty-two pharmacists participated. The majority (68%) were experienced community pharmacists (mean age = 42 years), 71% had a postgraduate qualification. Pharmacists identified adverse drug reactions in 33% of patients and made a mean of 1.6 recommendations/comments per visit undertaken. At least 35% of these were enacted. Pharmacists reduced inappropriate drug storage from 7% to 2% of visited patients by their second visit (P = 0.04), and reduced hoarding of unnecessary drugs from 40% of visited patients to 19% (P < 0.001). Finally, the rate of admission within the intervention group did not vary significantly according to experience or type of pharmacist delivering the intervention. Conclusion The HOMER intervention was conducted in a similar way to interventions in many other medication review studies. Given the HOMER trial's counter-intuitive findings it is clear that there is an urgent need to refine this intervention, identify the most suitable location for its delivery, and develop training that can ensure it is delivered to best effect.