John: Complementing a number of other Cochrane Reviews of ways to improve the prescription of medicines to people as they get older, the review focused on prescribing for people in care homes was updated in February 2016. David Alldred from the School of Heathcare at the University of Leeds in England summarises the latest evidence.

David: As a clinical pharmacist, I have seen first hand the consequences of inappropriate prescribing for older people living in care homes; these include adverse effects, hospital admissions and reduced quality of life. Older people in care homes are prescribed an average of eight medicines for a number of illnesses and because they are frail, this means that each of these medicines needs to be carefully considered. We need to look at the choice of medicine and its dose, the ways for monitoring for benefits and side effects, and decide with the resident whether a medicine should be continued or stopped. Given the substantial evidence that prescribing is poor for these people, we wanted to gather together the research evidence to see the effect of interventions to optimise prescribing for people living in care homes. We did this first with the original review in 2013 and have now updated it, using a comprehensive search strategy.
We have included 4 new studies, bringing us to a total of 12 randomised trials involving over 10,000 residents living in 355 care homes in 10 countries. 9 of the trials were of a cluster design, in which randomisation was by care home, and the other 3 studies randomised individual residents to different strategies. Ideally, randomisation by care homes is preferred, because it reduces the risk that there will contamination between the strategies being compared, if both are used in the same home.
The interventions tested in the trials were varied and often had several components. For example, medication review was a part of 10 studies, 5 used education, 4 used case-conferencing, and one used clinical decision support technology.
We chose adverse drug events, hospital admissions and mortality as the primary outcomes for the review. We also had secondary outcomes of quality-of-life, medicine-related problems, medication appropriateness and medicine costs but we weren’t able to look at all of these in detail because of a lack of information in the included trials. Differences in how the outcomes were measured also made things difficult and we were disappointed not to be able to pool results quantitatively in a meta-analysis.
Looking at the picture as a whole, there was no evidence of benefit in relation to adverse drug events or death, and the great majority of studies failed to show a reduction in hospital admissions. However, the interventions did lead to the identifying and resolving of medication related problems, and there was evidence from 5 trials that medication appropriateness was improved.
As in the first version of the review, we continue to be disappointed by how rarely quality-of-life was measured in the trials. Quality-of-life was an outcome measure for only 2 studies. And although we recognise that it may be challenging to measure quality of life in a population with a high level of cognitive impairment, this is an important resident-related outcome and should be investigated in future studies.
Overall, our updated review has shown that the quality of the current evidence is still relatively low and more high-quality randomised trials with a cluster design are needed to test interventions to optimise prescribing in care homes. Trials should focus on outcomes that are important to residents and test the effectiveness of technology, such as clinical decision support, along with multidisciplinary interventions.

John: Details of the different strategies that have been assessed and other ideas for future studies are available in the full review at Cochrane Library dot com, which you can find with a simple search for 'prescribing for older people in care homes'.