The pilot study methodology is a before and after feasibility study that took place on four clinical teaching units in Quebec and Ontario with control for hospital level clustering. A second ongoing study is a stepped wedge cluster randomized controlled trial with three clusters and 11 hosptals across Quebec, Ontario, and Central/western Canada, that is powered for a reduction in adverse drug events at 30 days post hospital discharge (6000 patients; enrolment 30% complete).The results of the 1000 patient MedSafer trial will be presented either as an oral presentation or in workshop format. We will expore the path we took in developing the software and then deploying the intervention. We will discuss successes and barriers to implementing a large scale deprescribing intervention, that to date, has enrolled close to 3000 patients across the country in two official languages.Results
Pilot study: of 924 hospitalized eligible patients age>65 who took 5 or more medications, 779 (84%) had 1 or more PIMs identified at admission and were included in the primary outcome. Similar to the control population, the median age of patients was 81 (IQR 74–88) and 53% were female. Patients were highly co-morbid and more than 25% were moderate to severely frail. Baseline rates of deprescribing in the pilot study sites was high. Of 382 control patients, 179 (46.9%) had one or more PIMs stopped. In the intervention, 217 of 397 (54.7%) had one or more PIMs stopped. Odds ratio for discontinuation during the six-week intervention phase (after adjustment for hospital level clustering) was 1.24 (95% CI 1.02 to 1.50). This implied an absolute increase of 7.8% of patients having 1 or more PIMs stopped (Number needed to treat=13).Conclusions
MedSafer is an innovation that automates the time-consuming task of cross-referencing complex medical conditions with oftentimes upwards of 10–15 medications, provides tapering instructions when indicated, and pairs with patient directed educational brochures for deprescribing. MedSafer is extremely well suited to be highly scalable and spread across jurisdictions and facilities. The tool is fully bilingual making it usable across all provinces in Canada. We will present what is involved in the designing of an application programming interface (API) and how this technology can allow for the processing of thousands of patient records, in order to provide clinicians with concise, real-time, patient deprescribing opportunities. Future directions for the MedSafer API is the automation of deprescribing opportunities for all patients in long term care facilities in the province of New Brunswick, beginning in the Summer of 2018.