What types of recommendations do pharmacists make, do GPs action them and do they reduce complexity? Findings from the 3D Study

Talk Code: 
Polly Duncan
Line Guenette, Debbie McCahon, Daisy Gaunt, Jack Collins, Shuip Kobeci, Margaret Watson, Chris Salisbury.
Author institutions: 
University of Bristol (PD, DM, DG, CS), University of Bath (MW), University of Reading (SK), University of Sydney (JC), University of Laval (LG)


In UK General Practice, there is a trend for practices to employ a pharmacist as part of their primary care team. Research into how best to utilise their skills is lacking. This was a sub-study of the 3D Study, a randomised controlled trial for patients with multimorbidity (three or more long term conditions, mean age 71 years). As part of the intervention, one of nine pharmacists reviewed patients’ GP computer records, conducted a remote medication review and made up to four recommendations for the GP to discuss with the patient. We explored the reasons why the 3D intervention had no effect on the number of drugs prescribed and examined the impact of the intervention on medication regimen complexity. The study aimed to:(i) categorize the types of recommendations made by the pharmacists and the reasons for the recommendations;(ii) examine what proportion of recommendations were actioned by GPs; and(iii) assess whether medication regimen complexity reduced in those receiving the intervention in comparison with usual care


Phase ITwo coders independently coded the free text of each pharmacist recommendation using an adapted version of the validated PharmDISC framework. The type of intervention (e.g. stop/switch) and the reason for the intervention (e.g. cost, safety) were coded. Discrepancies between the two coders were identified and discussed with the wider team where consensus could not be agreed. The framework and coding rules were updated iteratively.Phase IITo assess whether the recommendations were actioned by the GPs, we focused on recommendations which advised a change in prescribing. We examined the prescribing data between the date of pharmacist review and 90 days after GP review.Phase IIIMedication Regimen Complexity Index (MRCI) scores were calculated for patients in the intervention and usual care arms of the study at baseline, 6 and 15 months follow-up.


609/797 (76%) of intervention patients had a pharmacist review for whom 115/609 (19%) no pharmacist recommendation was made. Of 1100 recommendations made, 218 (20%) were either vague, indirect or a question. The most common interventions were to stop/reduce a medication (26%), switch a medication within the same class (18%) or ‘review’ a medication (16%). Of the recommendations advising changes to prescriptions, over half were not actioned by the GPs. Analysis of MRCI scores will be complete by July 2019.


A high proportion of pharmacist recommendations were vague or indirect and over half of the recommendations that advised a change in prescribing were not actioned by the GPs. This explains why the 3D intervention had no effect on the number of medications prescribed. It is possible that the 3D Study intervention improved the medication complexity (e.g. the number of doses per day). Findings from the analysis of MRCI scores will be presented.

Submitted by: 
Polly Duncan
Funding acknowledgement: 
The 3D Study was funded by the National Institute for Health Research (NIHR) Health Services and Delivery Research Programme (project number 12/130/15). Dr Polly Duncan is a NIHR In-Practice Fellow (IPF-2016-10-05).