Improving Patients’ Medication Adherence and Outcomes in Nonhospital Settings Through eHealth: Systematic Review of Randomized Controlled Trials

Abstract

BackgroundElectronic health (eHealth) refers to the use of information and communication technologies for health. It plays an increasingly important role in patients’ medication management. ObjectiveTo assess evidence on (1) whether eHealth for patients’ medication management can improve drug adherence and health outcomes in nonhospital settings and (2) which eHealth functions are commonly used and are effective in improving drug adherence. MethodsWe searched for randomized controlled trials (RCTs) on PubMed, MEDLINE, CINAHL, EMBASE, EmCare, ProQuest, Scopus, Web of Science, ScienceDirect, and IEEE Xplore, in addition to other published sources between 2000 and 2018. We evaluated the studies against the primary outcome measure of medication adherence and multiple secondary health care outcome measures relating to adverse events, quality of life, patient satisfaction, and health expenditure. The quality of the studies included was assessed using the Cochrane Collaboration’s Risk of Bias (RoB) tool. ResultsOur initial search yielded 9909 records, and 24 studies met the selection criteria. Of these, 13 indicated improvement in medication adherence at the significance level of P<.1 and 2 indicated an improved quality of life at the significance level of P<.01. The top 3 functions that were employed included mechanisms to communicate with caregivers, monitoring health features, and reminders and alerts. eHealth functions of providing information and education, and dispensing treatment and administration support tended to favor improved medication adherence outcomes (Fisher exact test, P=.02). There were differences in the characteristics of the study population, intervention design, functionality provided, reporting adherence, and outcome measures among the included studies. RoB assessment items, including blinding of outcome assessment and method for allocation concealment, were not explicitly reported in a large number of studies. ConclusionsAll the studies included were designed for patient home-based care application and provided a mechanism to communicate with caregivers. A targeted study population such as older patients should be considered to maximize the public health impact on the self-management of diseases. Trial RegistrationInternational Prospective Register of Systematic Reviews (PROSPERO) CRD42018096627; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=96627