How safe is mhealth? Concerns about patient-facing mobile health applications and their consequences

Research output: Contribution to journalMeeting abstractpeer-review


Objectives: In this digital age, mobile phones have become a resourceful tool for patients to manage their health. The numbers of mobile health applications, or apps and their use have soared in the past few years. These apps address populations in various disease groups along each level of the wellness continuum. While apps show much promise to help individuals manage their health, little is known about the risks they can pose to patient safety such as when they provide wrong advice or do not respond appropriately when users enter information indicating health dangers e.g. suicidal thoughts. The objective of this study was to review studies about health apps and to identify the types of safety concerns and their consequences.

Methods: We searched bibliographic databases including Scopus, PubMed, and Web of Science from 2013 to 2017 for articles about health apps. Randomized control trials, pilot studies, or reviews of patient-facing apps reporting safety concerns were reviewed. Strategies to engage patients were assessed using a published framework for evaluating health apps. The types of safety concerns were identified, and their consequences were assigned using a standard approach.
Results: Of the 48 studies identified, the majority were reviews of either a single app or a group of similar apps (n = 42, 88%). Most studies included apps that engaged consumers by tracking their information (n = 32, 67%) and displaying and summarizing the entered information (n = 31, 65%). Studies also included apps that provide educational information (n = 28, 58%) or guidance based on user entered information (n = 27, 56%).
The most frequently reported safety concern was the absence, or lack of sufficient scientific evidence to support the information content of the app (n = 18, 38%), followed by incorrect or incomplete information (n = 15, 31%). Variation in content quality, particularly in clinical information, was reported in 23% of studies, while 21% reported the lack of medical professional involvement in app development. Diagnostic errors such as misclassification of melanomas, and calculation errors, such as overestimation of Blood Alcohol Content (BAC) were reported in six and five studies respectively.
Twenty-two studies reported the consequences of safety concerns. Hazardous events, such as substituting a visit to a medical professional with the use of diagnostic apps that were found to be erratic, were most commonly reported. Potential or actual harms included increased alcohol consumption due to competitive drinking games offered by apps and increased anxiety due to excessive technical information. Other noticeable consequences were inability to accurately monitor step counts because of inconsistent functionalities, failure to incorporate important health information such as allergies in interventions recommended by apps, and insufficient response to consumers’ expression of suicidal thoughts.

Conclusion: The safety of health apps is an emerging public health issue. The available evidence shows that they can pose safety risks, and despite their growing use, most health apps available to US consumers are not regulated by the Food and Drug Administration or built to any common safety standard. There is a pressing need to address gaps in the current policy for development and safe use, particularly around increasing consumer awareness about safety risks.
Original languageEnglish
Article numberISQUA18-1534
Pages (from-to)49-50
Number of pages2
JournalInternational Journal for Quality in Health Care
Issue numberSupplement 2
Publication statusPublished - 18 Sep 2018
Event35th International Conference in Healthcare Quality and Safety (ISQUA) - Kuala Lumpur, Malaysia
Duration: 23 Sep 201826 Sep 2018


Dive into the research topics of 'How safe is mhealth? Concerns about patient-facing mobile health applications and their consequences'. Together they form a unique fingerprint.

Cite this