Smart About Meds (SAM) RCT

Introduction:Almost half of patients discharged from hospital are re‐admitted or return to theemergency department (ED) within 90 days. Patient nonadherence to medication changesduring hospitalization and the use of potentially inappropriate medications (PIMs) bothcontribute to the risk of adverse events post‐discharge. Smart About Meds (SAM) is apatient‐centered mobile application designed to target medication nonadherence and PIMsuse. This protocol describes a randomized controlled trial (RCT) to evaluate theeffectiveness of SAM.Methods & Analysis:A pragmatic, stratified RCT will be conducted among 3,250 patients discharged frominternal medicine, cardiac care, cardiac surgery, vascular surgery, and respiratory unitsof the Royal Victoria Hospital and the Montreal General Hospital. At discharge, patientswill be randomized 1:1 to usual care or the SAM intervention. SAM integrates noveluser‐centered features (e.g. continuously updated medication list with pill images,side‐effect checker, interaction checker) with pharmacist monitoring to tacklepost‐discharge nonadherence to new medication regimens. SAM also notifies patients ofPIMS in their regimen, with advice to discuss with their physician.Following discharge, patients will be followed for 90 days, during which the occurrenceof the composite outcome of ED visits, hospital readmissions, or death will be measured.Secondary outcomes will include the individual components of the composite outcome,nonadherence to medication changes, defined as failure to fill a new prescription,filling a modified prescription at the incorrect dose, or filling discontinuedmedications, secondary medication adherence, patient empowerment, and health‐relatedquality of life.An intention‐to‐treat analysis will evaluate the effectiveness of SAM. Multivariablelogistic regression will estimate differences between treatment groups in the proportionof patients nonadherent to at least one medication change. With a sample size of 3,250,there will be 80% power to detect a 5% absolute reduction in this outcome. Two‐wayinteraction terms will test hypothesized modifiers of SAM's effectiveness, includinghospital, unit, age, sex, gender and comorbidity burden. Binary and continuous secondaryoutcomes will be assessed using multivariable logistic and linear regression,respectively.