Various mental health apps have been developed to support clinical professionals and patients. What is already possible today and what does it have to do with augmented intelligence?

In Switzerland, 25.9% of the Swiss population reported mild, 5.9% moderate and 2.8% severe depression symptoms in 2017, an increase compared to the previous survey in 2012 [1]. It is estimated that 7.8% of the Swiss population has an unmet need for psychiatric-psychotherapeutic treatment, which corresponds to almost 0.7 million people [2]. In Germany, the average waiting time for psychotherapy is 6 months [3], a long time during which patients remain untreated, although capacities have been expanded in recent years. This time can be bridged through the use of digital technologies. One example of this is the so-called mental health apps. In the USA, the chatbots Woebot and Noni have been particularly successful [4], as they address the problems of lack of time, lack of therapy offers and the question of costs (e.g. in the absence of insurance coverage). The Woebot [5] was programmed by psychologists at the University of Stanford with the aim of supporting students and doctoral students with regard to anxiety disorders or depression. The chatbot checks in regularly, asks how you are feeling and gives you tips to improve your mood. For example, by suggesting three things that went well; or identifying three strengths.

Scientific studies

A study [6] has shown that such applications can indeed have a measurable benefit. Participants were divided into two groups – one group used the Woebot and the other group received an e-book with instructions on how to help themselves. Participants were tested at the beginning and after 2-3 weeks using clinical psychology questionnaires for depression and anxiety disorders. While a reduction in anxiety disorders was observed in both groups, a decrease in symptoms could only be measured in the Woebot group, especially for depression. A similar study was conducted in Germany [7], for the product moodgym [8]. This computer-based, interactive and easy-to-understand training programme is designed to prevent and reduce depressive symptoms.

Originally developed in Australia, moodgym has since been translated into various languages, including German. This study also found that the use of internet-based self-management programmes can be a promising measure against depressive symptoms as an addition to inpatient care. There are now numerous such applications on the market: somnio [9] supports people with insomnia, and has been tested and approved in Germany by the Federal Institute for Drugs and Medical Devices. This is possible due to the Digital Care Act, which came into force in December 2019 and is intended to advance digitalisation in the healthcare sector [10]. Another example is deprexis for the support of depression, burnout and anxiety disorders, for which studies have also shown that it can improve the success of therapy [11] [12].

Limitations and risks

As with many other products, such apps come in different quality levels. While some apps are based on clinical studies and experiences from therapy, there are other apps available on the market. The question of data protection also arises. In particular, the apps Pocketcoach and Woebot, for example, can be used without registration, while other apps use Facebook Messenger, for example, as a platform [13], which raises the question of how and for what purpose these chat histories are then stored.

Augmented Intelligence – no substitute for therapy

While such technologies can be very successful in supporting patients while waiting for a therapy place or parallel to therapy, it must be noted that they cannot replace therapy or the making of a diagnosis. This is also pointed out by the Woebot at the beginning of the conversation. Experts also largely agree that such tools should not further isolate sufferers socially, but rather inspire them to get in touch with other people, as the chatbot cannot offer an emotional relationship [4]. As in other areas, we are therefore talking about Augmented Intelligence instead of Artificial Intelligence. The innovative technologies and tools support humans instead of replacing them, and expand their capabilities. While they help the patient to better reflect on himself, they help the therapist to provide efficient counselling and support to her patients.

References

1. Schuler et al (2020). Mental health in Switzerland. Swiss Health Observatory, OBSAN Report 15/2020.
2. Künzi, K., Stocker, D. & Schläpfer, D. (2020). Economic benefits of the order model and thus of lower-threshold access to outpatient care for mental illness. Bern: Federation of Swiss Psychologists.
3. Ute Eppinger, Stress and depressive mood: What mental health apps do for care – and which questions are open, Medscape, 2021, https://deutsch.medscape.com/artikelansicht/4910308 (last accessed 5.11.21)
4. Ute Eppinger, With Artificial Intelligence against Depression and Anxiety: What can “Dr Bot” do?, Medscape, 2018 https://deutsch.medscape.com/artikelansicht/4906868 (last accessed 5.11.21)
5. Christiane Löll, Woebot – Can a grief chatbot help with depression?, Geo Magazine, 2017 https://www.geo.de/wissen/gesundheit/17146-rtkl-woebot-kann-ein-kummer-chatbot-bei-depressionen-helfen (last accessed 5.11.21)
6. Fitzpatrick KK, Darcy A, Vierhile M, Delivering Cognitive Behavior Therapy to Young Adults With Symptoms of Depression and Anxiety Using a Fully Automated Conversational Agent (Woebot): A Randomized Controlled Trial, JMIR Ment Health 2017;4(2):e19
7. Dorow, M., Stein, J., Foerster, F., Loebner, M., Franz, M., Guenther, R., … & Riedel-Heller, S. G. (2017). Implementation of the Internet-Based Self-Management Program” moodgym” in Patients with Depressive Disorders in Inpatient Clinical Settings-Patient and Expert Perspectives. Psychiatric Practice, 45(5), 256-262.
8. http://www.moodgym-deutschland.de/was-ist-moodgym/
9. https://somn.io/fachpersonen/
10. https://www.stiftung-gesundheitswissen.de/gesundes-leben/e-health-trends/digitale-gesundheit-was-aendert-das-digitale-versorgung-gesetz
11. Berger, T., Krieger, T., Sude, K., Meyer, B., & Maercker, A. (2018). Evaluating an e-mental health program (“deprexis”) as adjunctive treatment tool in psychotherapy for depression: results of a pragmatic randomized controlled trial. Journal of affective disorders, 227, 455-462.
12. Meyer, B., Bierbrodt, J., Schröder, J., Berger, T., Beevers, C. G., Weiss, M., … & Klein, J. P. (2015). Effects of an Internet intervention (Deprexis) on severe depression symptoms: Randomized controlled trial. Internet Interventions, 2(1), 48-59.
13. Andreas Grote, Apps gegen Depression und Ängste – Was Dr. Chatbot kann – und was nicht, Beobachter, 2021 https://www.beobachter.ch/gesundheit/medizin-krankheit/apps-gegen-depressionen-und-angste-was-dr-chatbot-kann-und-was-nicht (last accessed 5.11.21)