Background:
Medication adherence is a costly and damaging problem for both healthcare providers and patients alike. Patients adhere to only 50% of drugs prescribed for chronic diseases in developed nations. Digital health has paved the way for innovative smartphone solutions to tackle this challenge. However, despite the numerous applications (apps) available claiming to improve adherence, a thorough review of adherence applications has not been carried out to date.

Objective:
(i)To review medication adherence apps (otherwise known as mAdherence app) in the Apple App store and the Google Play repository in terms of their evidence base, medical professional involvement in development, and strategies used to facilitate behaviour change and improve adherence.
(ii)To provide a system of classification for these apps.

Methods:
In April 2015, relevant mAdherence apps were identified by systematically searching the Apple and Google Play app stores using a combination of relevant search terms.
Data extracted for each app included app store source, app price, documentation of healthcare professional (HCP) involvement during app development and evidence base for each respective app.
Free apps were downloaded to explore the strategies used to promote medication adherence. Testing involved a standardised medication regimen of three reminders over a four-hour period. Non-adherence features designed to enhance user experience were also documented.

Results:
The App repository search identified a total of 5889 applications. 806 fulfilled the inclusion criteria initially and were tested. 682 applications were further analysed for data extraction. Of these, 61.7% were free for testing, 8.5% were inaccessible and 29.8% required payment. Of the 421 free applications, 13.8% were developed with HCP involvement and an evidence base was identified in only 0.95%. Of the paid apps, 4.4% had HCP involvement, 0.5% had a documented evidence base and 0.5% had both. 31% of inaccessible apps were produced with HCP involvement while 3.4% had a documented evidence base.
The 421 free applications were further analysed to identify strategies used to improve medication adherence. This identified three broad categories of adherence strategies, ‘Reminder’ (92%), and ‘Behavioural’ (44.4%), Educational’ (0.1%). 59.4% apps utilised a single method, 35.4% used two methods, and only 5.2% apps utilised all three methods.

Conclusions:
This is the first study to systematically review all available medication adherence apps on the two largest app repositories. The results demonstrate a concerning lack of HCP involvement in app development and evidence base of effectiveness.
More collaboration is required between relevant stakeholders to ensure development of high quality and relevant adherence apps with well-powered and robust clinical trials investigating the effectiveness of these interventions. This will aid the adoption of such apps in to mainstream healthcare.