Using a mobile application to reduce potentially inappropriate prescribing for older Brazilian adults in primary care: a triple-blind randomised clinical trial.

ABSTRACT

BACKGOUND Potentially inappropriate prescribing (PIP) has been evaluated in several countries, and several strategies have been devised for deprescribing drugs in older adults. The aim of this study was to evaluate the efficacy of a mobile application in reducing PIP for older adults in primary care facilities in Brazil. METHODS: This randomised, triple-blind, parallel-group trial was conducted in 22 public primary care facilities in Brazil. During the intervention phase, the general practitioners (GPs) were randomly allocated to the intervention (MPI Brasil app provides information about PIP, therapeutic alternatives and deprescribing) or control (MedSUS app provides general information about medications) group. All GPs were trained on the Clinical Decision-Making Process and how to access an Evidence-Based Health website. The GPs received an Android tablet with an installed mobile application depending on their allocated group, which they used when caring for older patients over at least 3 months. At the end of this period, a sample of older patients aged ≥ 60 years who had been awaiting medical consultation by the participating GPs were interviewed and their prescriptions analysed. The primary outcome was the frequency of PIP in and between the groups. RESULTS: Among 53 GPs who were administered the baseline survey, 14 were included in the clinical trial. At baseline, 146 prescriptions were analysed the PIP overall was 37.7% (55/146), in the intervention group was 40.6% (28/69), and in the control group was 35.1% (27/77). After the intervention, 284 prescriptions were analysed the PIP overall was 31.7% (90/284), in the intervention group was 32.2% (46/143), and in the control group was 31.2% (44/141) (RR 1.16; 95% CI, 0.76-1.76). In the within-group analysis, the PIP reduced from before to after the intervention in both groups-more significantly in the intervention than in the control group (p < 0.001). In the stratified analysis of PIP frequency by GPs, there was a relative risk reduction in 86% (6/7) of GPs in the intervention group compared to 71% (5/7) in the control group. CONCLUSION: We found that the MPI Brasil app effectively reduced PIP, suggesting that it may be useful to incorporate this tool into clinical practice. TRIAL REGISTRATION The study was registered at ClinicalTrials.gov (NCT02918643). First registration on 22/09/2016.