Apps and websites such as WebMD, Ada, Babylon Health, and FamilyDoctor aim to provide diagnoses and triage advice to patients, given self-reported symptom(s), demographics, and other information. Such tools have the potential to enable health screening and let patients take charge of their own health. Patients would have more information on when to administer self-care and avoid unnecessary health bills, while providers could devote more time to patients who genuinely need care. However, obstacles such as mediocre accuracy, low usability, and low awareness among potential users might hamper the adoption of these tools in the short term.
If apps for symptom analysis and advice are less accurate than recommendations from general practitioners (GPs), patients could become misinformed about their own health or fail to pursue necessary care. One can measure accuracy in several ways: accuracy of diagnosis, accuracy of triage advice, and the breadth of symptom coverage. In a study of eight symptom analysis apps, one app had accuracy on par with GPs across all three metrics, while two other apps had on-par accuracy with respect to two metrics . In another study, 23 symptom analysis apps were found to recommend emergency or primary care in cases where self-care would be appropriate . Although these apps might be generally inaccurate today, their developers will continue to make improvements, and the apps will gain accuracy over time.
Research has shown that patients also care about factors beyond accuracy. A study of young adult users of symptom analysis apps emphasized the importance of personalization, security, and privacy. Participants perceived symptom analysis apps to be more useful for triage advice than diagnosis . In a separate study, participants tended to perceive Google as a more familiar and flexible tool than symptom analysis apps. After using several symptom analysis tools, they realized that Google was not as usable as they thought . These findings suggest that patients would respond well to apps with strong outreach campaigns and transparency about their underlying processes.
To summarize the aforementioned findings, symptom analysis apps for physical health are well–studied, with some being almost as effective as a general practitioner. The question remains whether symptom analysis apps for mental health are equally effective. A review of sixteen such apps found that clinical research backed fourteen of those apps, but none of the apps covered more than one or two mental disorders. Furthermore, roughly half of the apps were targeted to a specific population, like veterans or university students. This lack of breadth is a far cry from the breadth of coverage displayed in most apps for physical symptom analysis. To overcome this challenge, developers will need to release mental health apps for wider populations and types of disorders. At the same time, the proliferation of mental health apps in recent years has led researchers to urge caution. One study claims that “the majority” of mental health apps have no evidence of efficacy, calling for greater clinician involvement before the disparity between apps and practitioners widens too much . As with physical health, inaccurate mental health advice might mislead or harm patients.
By bringing medical expertise to mobile devices, apps for symptom analysis are already widening access to medical advice. With the proper technical and legal protections, trustworthy stakeholders could leverage search data from symptom analysis apps to monitor public health. From March to April 2020, such data from nearly 100,000 participants in Germany and the United Kingdom proved effective in flagging the COVID-19 pandemic as it unfolded . As these apps become more popular, data-driven predictions in public health may become more accurate and more mainstream.
 Gilbert S., et al. How Accurate Are Digital Symptom Assessment Apps for Suggesting Conditions and Urgency Advice? A Clinical Vignettes Comparison to GPs. The BMJ 2020. DOI:10.1136/bmjopen-2020-040269.
 Semigran H. L., et al. Evaluation of Symptom Checkers for Self Diagnosis and Triage: Audit Study. The BMJ 2015. DOI:10.1136/bmj.h3480.
 Aboueid S., et al. Young Adults’ Perspectives on the Use of Symptom Checkers for Self-Triage and Self-Diagnosis: Qualitative Study. JMIR Public Health and Surveillance 2021; 7: 1. DOI:10.2196/22637.
 Li H., and Salah H. Comparing Four Online Symptom Checking Tools: Preliminary Results. Proceedings of the 2017 Industrial and Systems Engineering Conference.
 Wang K., et al. A Systematic Review of the Effectiveness of Mobile Apps for Monitoring and Management of Mental Health Symptoms or Disorders. Journal of Psychiatric Research 2018; 107. DOI:j.jpsychires.2018.10.006.
 Marshall J. M., et al. Clinical or Gimmickal: The Use and Effectiveness of Mobile Mental Health Apps for Treating Anxiety and Depression. Australian & New Zealand Journal of Psychiatry 2020; 54: 1. DOI:10.1177/0004867419876700.
 Mehl A., et al. Syndromic Surveillance Insights from a Symptom Assessment App Before and During COVID-19 Measures in Germany and the United Kingdom: Results From Repeated Cross-Sectional Analyses. JMIR mHealth and uHealth 2020; 8: 10. DOI:2020/10/e21364.