Evaluation of the effectiveness and acceptability of the Bristol Medication Review toolkit

INTERVENTION: This is a large, highly pragmatic, cluster randomised controlled trial with an embedded mixed‐methods process evaluation. The trial will be conducted in around 1,000 general practices in England contributing data to the Royal College of General Practitioners (RCGP) Research and Surveillance Centre (RSC), based in Oxford. The trial will compare the BRISMED toolkit to usual clinical care. Practices will be recruited before randomisation. Randomisation will be generated using a computer algorithm. Randomistion will be by Primary Care Network (PCN) rather than individual practice to reduce inter‐arm contamination risk as practice pharmacists generally work within PCNs. Practices will not be informed of whether they will be in the intervention arm before agreeing to participate. It is not possible to blind practices to randomisation status after randomisation, and although patients could discover which arm their practice has been randomised to, we will not draw attention to this. Analysis will be conducted in a blinded manner. The BRISMED toolkit will be provided electronically to 500 randomly selected practices, with practices able to implement any or all of the resources as they see fit. Medication reviews are recommended for any patient receiving long‐term medication, so there are no constraints on which patients need to be reviewed. The BRISMED toolkit will be used at intervention practices over a 12‐month period. A mixed‐method process evaluation will be conducted in a sub‐set of 56 practices (e.g. 28 intervention and 28 control sites) to understand users’ experience of the reviews that are being carried out, to explain the success or otherwise of the intervention, and to explore its acceptability to users. Implementation and adoption of the Bristol toolkit will be CONDITION: Medication optimisation ; Not Applicable PRIMARY OUTCOME: Potentially inappropriate prescribing (PIP) at 12 months will be assessed using routine electronic health records. SECONDARY OUTCOME: 1. Number of long‐term medicines currently used at 12 months, captured via routine electronic health records; 2. Medication Treatment Burden at 12 months, measured using Multimorbidity treatment burden questionnaire (MTBQ); 3. Medication Regimen Complexity Inde Xat 12 months, captured via routine electronic health records; 4. Medication adherence at 12 months, captured via routine electronic health records; 5. Quality of life at 3 months post‐review, measured using the EQ5D‐5L; 6. Rates of medication review at 12 months, assessed via routine electronic health records; 7. Health service utilisation including GP consultations and unplanned hospitalisation over the past 12 months, captured via routine electronic health records and Hospital Episode Statistics (national administrative records); 8. All cause mortality at 12 months captured via routine electronic health records and Hospital Episode Statistics (national administrative records); 9. NHS costs (i.e. sum of GP prescribing, consultations, and hospital use) assessed using routine electronic health records and Hospital Episode Statistics (national administrative records) at 12 months; INCLUSION CRITERIA: 1. General practices using EMIS or SystmOne clinical systems, and contributing to the secure data processing environment of the nationally representative Oxford‐Royal College of GPs Clinical Informatics Digital Hub (ORCHID), will be eligible to participate in the trial. All patients aged =18 years who have not registered with their practice an opt‐out of sharing data outside of the practice for purposes of research or planning, will be included in the main trial analysis. 2. Within the 56 practices participating in the process evaluation, all adult patients (aged =18 years) receiving a structured medication review during four 2‐week blocks within the 12‐month intervention period will be sent a patient‐experience questionnaire, and all adult patients receiving a review within the 12‐month intervention period will be sent a quality of life questionnaire. Additionally, a smaller number of patients will be approached and invited to provide consent to participate in