Researchers working in the field, the places where research-relevant activity happens, are essential to recruitment and data collection in randomised controlled trials (RCTs). This study aimed to understand the nature of this often invisible work. Data were generated through an RCT of a pharmacist-led medication management service for older people in care homes. The study was conducted over three years and employed seven Research Associates (RA) working in Scotland, Northern Ireland, and England. Weekly research team meetings and Programme Management Group meetings naturally generated 129 sets of minutes. This documentary data was supplemented with two end-of-study RA debriefing meetings. Data were coded to sort the work being done in the field, then deductively explored through the lens of Normalization Process Theory to enable a greater understanding of the depth, breadth and complexity of work carried out by these trial delivery RAs. Results indicate RAs helped stakeholders and participants make sense of the research, they built relationships with participants to support retention, operationalised complex data collection procedures and reflected on their own work contexts to reach agreement on changes to trial procedures. The debrief discussions enabled RAs to explore and reflect on experiences from the field which had affected their day-to-day work. The learning from the challenges faced in facilitating care home research may be useful to inform future research team preparation for complex interventions. Scrutinising these data sources through the lens of NPT enabled us to identify RAs as linchpins in the successful conduct of a complex RCT study.
OBJECTIVE: To estimate the effectiveness, cost effectiveness (to be reported elsewhere), and safety of pharmacy independent prescribers in care homes. DESIGN: Cluster randomised controlled trial, with clusters based on triads of a pharmacist independent prescriber, a general practice, and one to three associated care homes. SETTING: Care homes across England, Scotland, and Northern Ireland, their associated general practices, and pharmacy independent prescribers, formed into triads. PARTICIPANTS: 49 triads and 882 residents were randomised. Participants were care home residents, aged ≥65 years, taking at least one prescribed drug, recruited to 20 residents/triad. INTERVENTION: Each pharmacy independent prescriber provided pharmaceutical care to approximately 20 residents across one to three care homes, with weekly visits over six months. Pharmacy independent prescribers developed a pharmaceutical care plan for each resident, did medicines reviews/reconciliation, trained staff, and supported with medicines related procedures, deprescribing, and authorisation of prescriptions. Participants in the control group received usual care. MAIN OUTCOMES MEASURES: The primary outcome was fall rate/person at six months analysed by intention to treat, adjusted for prognostic variables. Secondary outcomes included quality of life (EQ-5D by proxy), Barthel score, Drug Burden Index, hospital admissions, and mortality. Assuming a 21% reduction in falls, 880 residents were needed, allowing for 20% attrition. RESULTS: The average age of participants at study entry was 85 years; 70% were female. 697 falls (1.55 per resident) were recorded in the intervention group and 538 falls (1.26 per resident) in the control group at six months. The fall rate risk ratio for the intervention group compared with the control group was not significant (0.91, 95% confidence interval 0.66 to 1.26) after adjustment for all model covariates. Secondary outcomes were not significantly different between groups, with exception of the Drug Burden Index, which significantly favoured the intervention. A third (185/566; 32.7%) of pharmacy independent prescriber interventions involved medicines associated with falls. No adverse events or safety concerns were identified. CONCLUSIONS: Change in the primary outcome of falls was not significant. Limiting follow-up to six months combined with a small proportion of interventions predicted to affect falls may explain this. A significant reduction in the Drug Burden Index was realised and would be predicted to yield future clinical benefits for patients. This large trial of an intensive weekly pharmacist intervention with care home residents was also found to be safe and well received. TRIAL REGISTRATION: ISRCTN 17847169.
Pharmaceutical Services , Pharmacists , Humans , Female , Aged, 80 and over , Male , Quality of Life , Northern Ireland , Scotland
BACKGROUND: Residents in care homes are often very frail, have complex medicine regimens and are at high risk of adverse drug events. It has been recommended that one healthcare professional should assume responsibility for their medicines management. We propose that this could be a pharmacist independent prescriber (PIP). This feasibility study aimed to test and refine the service specification and proposed study processes to inform the design and outcome measures of a definitive randomised controlled trial to examine the clinical and cost effectiveness of PIPs working in care homes compared to usual care. Specific objectives included testing processes for participant identification, recruitment and consent and assessing retention rates; determining suitability of outcome measures and data collection processes from care homes and GP practices to inform selection of a primary outcome measure; assessing service and research acceptability; and testing and refining the service specification. METHODS: Mixed methods (routine data, questionnaires and focus groups/interviews) were used in this non-randomised open feasibility study of a 3-month PIP intervention in care homes for older people. Data were collected at baseline and 3 months. One PIP, trained in service delivery, one GP practice and up to three care homes were recruited at each of four UK locations. For ten eligible residents (≥ 65 years, on at least one regular medication) in each home, the PIP undertook management of medicines, repeat prescription authorisation, referral to other healthcare professionals and staff training. Outcomes (falls, medications, resident's quality of life and activities of daily living, mental state and adverse events) were described at baseline and follow-up and assessed for inclusion in the main study. Participants' views post-intervention were captured in audio-recorded focus groups and semi-structured interviews. Transcripts were thematically analysed. RESULTS: Across the four locations, 44 GP practices and 16 PIPs expressed interest in taking part; all care homes invited agreed to take part. Two thirds of residents approached consented to participate (53/86). Forty residents were recruited (mean age 84 years; 61% (24) were female), and 38 participants remained at 3 months (two died). All GP practices, PIPs and care homes were retained. The number of falls per participating resident was selected as the primary outcome, following assessment of the different outcome measures against predetermined criteria. The chosen secondary outcomes/outcome measures include total falls, drug burden index (DBI), hospitalisations, mortality, activities of daily living (Barthel (proxy)) and quality of life (ED-5Q-5 L (face-to-face and proxy)) and selected items from the STOPP/START guidance that could be assessed without need for clinical judgement. No adverse drug events were reported. The PIP service was generally well received by the majority of stakeholders (care home staff, GPS, residents, relatives and other health care professionals). PIPs reported feeling more confident implementing change following the training but reported challenges accommodating the new service within their existing workload. CONCLUSION: Implementing a PIP service in care homes is feasible and acceptable to care home residents, staff and clinicians. Findings have informed refinements to the service specification, PIP training, recruitment to the future RCT and the choice of outcomes and outcome measures. The full RCT with internal pilot started in February 2016 and results are expected to be available in mid late 2020.
