No abstract

Background: Polypharmacy and overprescribing pose an enormous challenge to safe healthcare and efficient use of resources. Patient record data could inform safer prescribing and deprescribing, but it is unclear how these complex data should be summarised and displayed to clinicians. The current study examined the perspectives of clinical pharmacists (CPs), a newly expanding workforce of primary care medication specialists, to explore ways that novel analytics could help improve health outcomes for frail and elderly (>65yrs) patients. Methods: The utility of novel analytics interventions were discussed in an exploratory scoping workshop. Health risk data for frail and elderly patients with polypharmacy (modelled from extensive national datasets) were presented to primary care clinical pharmacists (n=14). Verbal and textual comments were thematically analysed using the framework method (exploratory content analysis) combining inductive and deductive approaches. Results: Overarching themes of data use, data reservations and digital tools acceptance factors were identified. Respondents highlighted several uses for polypharmacy analytics interventions, including increased knowledge of clinical effects of drug-drug interactions, the ability to prioritise high-risk patients for reviews and medication to deprescribing (e.g., highlighting cumulative medication risk). Data reservations were linked to existing barriers (such as cognitive overload from existing systems and patient explainability) meaning that CPs’ acceptance of digital analytics tools is heavily contingent on facilitators such as ease of use, clear targeted purpose and the ability to support clinicians’ understanding and confidence in evaluation of analytics for patient care decisions. Conclusion: The workshop helped to identify promising analytics and features for polypharmacy intervention development. Patient record data could help address a concerning deficit in evidence of real-world medication interactions, and help clinicians prioritise medication reviews. Barriers to use of digital tools for novel analytics must be addressed and criteria for acceptable user-focused tools are suggested.

Background Polypharmacy and overprescribing is a growing concern due to increased risk of drug interaction and inefficient use of resources. Historical patient records offer a wealth of knowledge that could inform safer prescribing and deprescribing, but it is unclear how these complex data should be summarised and displayed to clinicians. The current study evaluated the perspectives of clinical pharmacists (CPs), a newly expanding workforce of primary care medication specialists, on ways that novel big data analytics could be used to improve health outcomes for frail and elderly (> 65yrs) patients. Methods Primary care clinical pharmacists (n = 14) took part in an exploratory participative design workshop to discuss examples and uses of novel analytics (including risk of negative outcomes for frail and elderly patients following treatment with a variety of medication combinations). Participant contributions (verbal and textual) were analysed using the framework method of exploratory content analysis. Results Overarching themes of data uses, reservations and digital tools acceptance factors were identified. Uses of polypharmacy analytics valued by CPs included the clinical effects of drug: drug interactions and patient risk scores to highlight patients at increased risk of harm from medication combinations (including cumulative risks). CPs were also positive about the creation of a dedicated online community of practice to enable wider peer support. CP’s current digital systems do not adequately facilitate patient medication reviews and contribute to cognitive overload. Experiencing these existing barriers mean that CPs’ acceptance of digital analytics tools is heavily contingent on ease of use, clear targeting and ability to support clinicians’ understanding and confidence in the strengths, limitations and validity of the analytics in the patient care context. Conclusions CPs would benefit from bespoke digital systems designed to meet the needs of their role. Analytics from patient records could help address a concerning deficit in wider knowledge of the real world clinical effects of polypharmacy medication interactions and help clinicians prioritise medication reviews. Based on the user acceptance themes identified, five criteria are suggested to inform user focused development of analytics tools for primary healthcare.

Background: Overprescribing of antibiotics is a major concern as it contributes to antimicrobial resistance. Research has found highly variable antibiotic prescribing in (UK) primary care, and to support more effective stewardship, the BRIT Project (Building Rapid Interventions to optimise prescribing) is implementing an eHealth Knowledge Support System. This will provide unique individualised analytics information to clinicians and patients at the point of care. The objective of the current study was to gauge the acceptability of the system to prescribing healthcare professionals and highlight factors to maximise intervention uptake.Methods: Two mixed-method co-design workshops were held online with primary care prescribing healthcare professionals (n=16). Usefulness ratings of example features were collected using online polls and online whiteboards. Verbal discussion and textual comments were analysed thematically using inductive (participant-centred) and deductive perspectives (using the Theoretical Framework of Acceptability). Results: Hierarchical thematic coding generated three overarching themes relevant to intervention use and development. Clinician concerns (focal issues) were safe prescribing, accessible information, autonomy, avoiding duplication, technical issues and time. Requirements were ease and efficiency of use, integration of systems, patient-centeredness, personalisation and training. Important features of the system included extraction of pertinent information from patient records (such as antibiotic prescribing history), recommended actions, personalised treatment, risk indicators and electronic patient communication leaflets. Anticipated acceptability and intention to use the knowledge support system was moderate to high. Time was identified as a focal cost/ burden, but this would be outweighed if the system improved patient outcomes and increased prescribing confidence.Conclusion: Clinicians anticipate that an eHealth knowledge support system will be a useful and acceptable way to optimise antibiotic prescribing at the point of care. The mixed method workshop highlighted issues to assist person-centred eHealth intervention development, such as the value of communicating patient outcomes. Important features were identified including the ability to efficiently extract and summarise pertinent information from the patient records, provide explainable and transparent risk information, and personalised information to support patient communication. The Theoretical Framework of Acceptability enabled structured, theoretically sound feedback and creation of a profile to benchmark future evaluations. This may encourage a consistent user-focused approach to guide future eHealth intervention development.

Background: Polypharmacy and overprescribing pose an enormous challenge to safe healthcare and efficient use of resources. Patient record data could inform safer prescribing and deprescribing, but it is unclear how these complex data should be summarised and displayed to clinicians. The current study examined the perspectives of clinical pharmacists (CPs), a newly expanding workforce of primary care medication specialists, to explore ways that novel analytics could help improve health outcomes for frail and elderly (>65yrs) patients.Methods: Primary care clinical pharmacists (n=14) discussed examples and uses of analytics in a participatory design workshop. Data presented comprised of risk scores of negative outcomes for frail and elderly with polypharmacy derived from patient record data. Verbal and textual contributions were analysed using the framework method of exploratory content analysis. Results: Overarching themes of data uses, reservations and digital tools acceptance factors were identified. Participants valued several uses of polypharmacy analytics including clinical effects of drug-drug interactions, and risk scores to highlight high-risk patients (including cumulative medication risks). They expressed concerns that their current digital systems contribute to cognitive overload, and whether analytics could be relied upon. Experience of existing barriers mean that CPs’ acceptance of digital analytics tools is heavily contingent on ease of use, clear targeting and ability to support clinicians’ understanding and confidence in the strengths, limitations, and validity of using the analytics in patient care decisions. Conclusions: CPs would benefit from digital systems designed to better meet the needs of their role. Analytics from patient records could help address a concerning deficit in evidence of the real world clinical effects of polypharmacy medication interactions and help clinicians prioritise medication reviews. Clinical digital tools must be carefully designed to help, not hinder clinicians. Criteria are suggested to inform user-focused onward development of acceptable analytics.