400 CHAPTER 5 and control group was found in the occurrence of ADRs (24.5% vs 24.8%; OR 0.98; 95% CI 0.77–1.24; p = 0.88) [47]. The results of this SENATOR trial were attributed to the poor implementation of CDSS-generated STOPP/START signals, approximately 15%, while a retrospective examination of signals found that 39% of all presented signals were deemed clinically relevant [48]. The results of this large, multicentre trial contrasted with the results of a smaller single-center cluster randomised controlled trial, in which a significant effect on ADR incidence and medication costs was found after applying STOPP/START criteria in admitted older adults [11]. However, the implementation rates of STOPP and START recommendations in single-centre trials were much higher (STOPP: 81%; START: 87%) [10]. In contrast to the SENATOR trial, CDSS-generated signals for OPERAM intervention patients were first reviewed for clinical applicability by a pharmacotherapy team before discussing with the attending physician and patient. Nevertheless, pharmacotherapy team’s recommended drug changes were moderate, with an overall implementation rate of recommendations of 42% at two months after the in-hospital medication review. Again, there was high variability in implementing recommendations for different drug classes, ranging from 13% to 65% for the top ten most prevalent recommended drug changes (Chapter 3.2) [16]. 2.2. P erspectives on how to improve the implementation of medication optimisation recommendations The reasons for the non-implementation of proposed recommendations were investigated in Chapter 4.2 [44]. Barriers to changing medication differed between prescribers and patients and varied per drug class. An important factor significantly impacting patient’s and prescriber’s barriers to implementation is the clinical setting in which these recommendations are presented. The OPERAM trial was performed in a hospital setting, while decisions to accept or reject STOPP/START signals might have been different in other healthcare settings, as well as the willingness of patients and attending physicians to change long-termmedication use. For instance, results from a study investigating the impact of STOPP/START criteria (v1) in frail geriatric chronic care residents found that 82.4% of STOPP and 92.6% of START recommendations by a research pharmacist were implemented by the attending physician [49]. Another important condition for implementing medication changes is trusting relationships, on the patient-physician level and among healthcare professionals [50]. For instance, previous research found that attending physicians’ implementations of STOPP/START recommendations were significantly higher if the recommendations were discussed by a physician rather than a pharmacist [51]. In addition, a multicentre mixed-methods interview study among OPERAMpatients (n = 48) found that trusting
RkJQdWJsaXNoZXIy MTk4NDMw