Clinical decision support-assisted pharmacotherapy optimisation for older hospitalised patients

Abstract

Worldwide, the population is ageing rapidly. Multimorbidity and polypharmacy become more prevalent with advancing age, increasing the risk of receiving potential inappropriate prescriptions (PIPs) and their associated negative health outcomes. Drug-related problems (DRPs) such as adverse drug reactions (ADRs) and drug-related hospital admissions occur more frequently in older patients with multimorbidity and polypharmacy. Many implicit and explicit tools are available for health care professionals to detect PIP in older patients. The Screening Tool of Older Persons' Prescriptions (STOPP) and Screening Tool to Alert to Right Treatment (START) criteria are the most frequently used and investigated explicit criteria in Europe. The research in this thesis promotes critical evaluation of the population-based STOPP/START recommendations by experienced clinicians when applying STOPP/START criteria to individual older patients to optimise pharmacotherapy, to prevent patient harm by inappropriate decisions. In addition, explicit criteria, such as STOPP/START criteria, can be converted into algorithms for software systems and integrated into CDSS to enhance application and help ensure greater efficiency for medication reviews. However, the clarity of presentation of the individual STOPP/START criteria can be improved on a language level. Explanations to justify the recommendations could help clinicians in deciding on the applicability of the recommendations to the individual older patient with polypharmacy. Furthermore, the involvement of an expert team to evaluate the applicability of CDSS- generated signals for individual patients is essential, as more than half of the signals for PIP were not deemed appropriate for patients in the hospital setting. Patients’ and physicians’ agreement with in-hospital pharmacotherapy optimisation recommendations in a clinical trial setting was >60%. Patients’ reluctance to change was the main reason for disagreement. Better patient and physician education regarding the benefit/risk balance of pharmacotherapy, in addition to more precise and up-to-date medical records to avoid irrelevant recommendations, will likely result in higher agreement with future pharmacotherapy optimisation recommendations. Based on a qualitative study, the most important resident-perceived barriers to in-hospital pharmacotherapy optimisation were: lack of time, patients’ lack of knowledge of their own treatment and reluctance to change, absence of a long-term patient relationship, the hospital setting in which patients are sick/unstable and the lack of knowledge/skills to perform medication reviews. Important facilitators include: the presence of certain triggers (e.g. suspected side effects) to evaluate the medication, a supervisor as a role model, more education and training about appropriate prescribing and consultation of medication optimisation teams. In conclusion, conducting clinical trials assessing the effect of optimising pharmacotherapy for multimorbid older patients with polypharmacy remains challenging, especially in the dynamic hospital setting. The studies presented in this thesis could help to deal with these challenges in future research and clinical practice. Good transitional care is important to follow-up on hospital-initiated medication changes after discharge. Close collaboration among all health care professionals involved in the care of older polypharmacy patients is indispensable to improve patient-related outcomes and to reduce pharmacotherapy related patient harm. In addition, active patient involvement in SDM is crucial to ensure adherence to implemented pharmacotherapy optimisation recommendations and to improve overall patient satisfaction.