Behavioral intervention technology
Artificial intelligence in CBT
According to the World Health Organization, there is a global shortage of mental health-trained health workers. Many mental health initiatives do not reach those in need, with nearly 70% having no access to these services. In 2017, 42.6% of adults with mental illnesses in the United States received mental health services. In particular, 75% of patients with depression in primary care settings have one or more structural or psychological barriers that prevent them from receiving behavioral treatments. To address these issues, Kazdin and Rabbitt proposed new models of psychosocial intervention delivery. According to Mohr et al., behavioral intervention technologies (BITs) may offer a potential solution to resolve barriers to access and expand mental health care.
Behavioral intervention technology
Behavioral Intervention Technology (BIT) includes using technology features that address behavioral, cognitive, and affective components that support physical, behavioral, and mental health to implement behavioral and psychological intervention strategies. Internet interventions for anxiety and depression have factual support with outcomes equivalent to therapist-delivered cognitive behavioral therapy (CBT). Several BITs use the same content as face-to-face CBT programs, allowing them to reach a larger population at a lower cost.
Chatbots are BITs that are used in addressing mental health conditions. Chatbots are software programs that perform text-based or voice-activated conversations with users and respond to them using preprogrammed responses or artificial intelligence (AI). Woebot, Shim, KokoBot, Wysa, Vivibot, Pocket Skills, and Tess are some of the most reported mental health chatbots in the literature.
Artificial intelligence in CBT
Scott et al. (2015) conducted a systematic review of studies of pure Technology-based CBT (TB-CBT) intervention for dementia carers dating back to 1995. The databases Cochrane Reviews, PsycINFO, MedLine, and Scopus, were searched using keywords related to CBT, carers, and dementia. They identified 440 articles, and inclusion/exclusion criteria were applied; only studies with quantitative data available and no active therapist contact were retained. Two randomized trials and two waitlist control trials were chosen. Quality of methodology and reportings were evaluated. For the outcome measures of caregiver depression, meta-analyses were performed. The meta-analysis found that pure TB-CBT interventions for depression had small but significant post-intervention effects, similar to face-to-face interventions. There is, however, no evidence of the long-term efficacy of pure TB-CBT for dementia carers. The systematic review also identified critical methodological and reporting flaws in these studies.
Another study was conducted by Fitzpatrick et al. in 2017. The study was designed to find the feasibility, acceptability, and preliminary efficacy of an entirely automated conversational agent in delivering a self-help program to college students who self-identify as experiencing symptoms of anxiety and depression. In this unblinded trial, 70 participants between the ages of 18 and 28, with an average age of 22.2 years old (SD 2.33), 67% female (47/70), Caucasian (79%, 46/58), and non-Hispanic (93%, 54/58) were recruited online from a university community social media site and were randomly allocated to either two weeks (up to 20 sessions) of self-help content derived from CBT principles in a text-based conversational agent (Woebot) (n=34) or the ebook "Depression in College Students" by the National Institute of Mental Health (NIMH) as an information-only control group (n=36).
At baseline and 2-3 weeks later, all participants have completed Web-based versions of the 7-item Generalized Anxiety Disorder scale (GAD-7), the 9-item Patient Health Questionnaire (PHQ-9), and the Positive and Negative Affect Scale (T2). Interaction of participants in the Woebot group with the conversational agent was observed at an average of 12.14 (SD 2.23) times throughout the study. There was no notable difference between the groups at baseline, and 83% (58/70) of participants provided data at 17% attrition (T2).
Aims to treat univariate covariance analysis showed a significant group difference in depression, with those in the Woebot group greatly reducing their depressive symptoms as measured by the PHQ-9 over the study period (F=6.47; P=.01). However, the participants in the information control group did not. As measured by the GAD-7 (F1,54= 9.24; P=.004), observation showed that both groups significantly reduced anxiety. According to the comments of those who completed the study successfully, process factors were more influential on their acceptance of the program than content factors similar to traditional therapy.
Åho et al. proposed a paper on Virtual Reality (VR) based treatment for Social Anxiety Disorder and speech anxiety using CBT. CBT is the most widely used psychological treatment for Social Anxiety Disorder (SAD) and speech anxiety. The core principle of graduated exposure to a feared phenomenon is effectively reducing stress and anxiety symptoms. A fundamental principle of exposure-based therapy is that the exposure should trigger an emotional and physiological response. It is built on the concept that an effective VR-enhanced CBT (VR-CBT) treatment should cause physiological responses like elevated heart rate and skin conductance. Several studies indicate that VR-based therapy can produce such responses.
To summarize, VR technology appears to recreate phobic situations with sufficient accuracy to evoke fear responses similar to real-world situations. A program that promotes inhibitory learning by disproving catastrophic beliefs. It should be conducted by exposing patients to distressing speech situations in virtual reality environments. The concept also aims at self-attention and biased recall by mental imagery-enhanced audio feedback. The paper also suggests treatment prediction based on the brain data collected, using AI for new treatment strategies. According to research, VR-CBT appears to be effective in improving the quality of life by reducing depression and anxiety.
AI-based programs like chatbots appear to be feasible, engaging, and effective methods of delivering CBT. These programs effectively reduced depression and anxiety. However, to understand better how AI-based information could strengthen CBT, researchers should focus on how much trust or confidence individuals undergoing treatment can have in information derived from AI applications. If AI-derived information is trusted to the same extent as traditional information from psychologists, potential doors may unlock for CBT design.
- Åhs, F., Mozelius, P., & Dobslaw, F. (2020). Artificial Intelligence Supported Cognitive Behavioral Therapy for Treatment of Speech Anxiety in Virtual Reality Environments. In ECIAIR 2020 (Vol. 2). Academic Conferences and Publishing International Limited.
- Dosovitsky, G., Pineda, B. S., Jacobson, N. C., Chang, C., & Bunge, E. L. (2020). Artificial intelligence chatbot for depression: Descriptive study of usage. JMIR Formative Research, 4(11), e17065.
- Fitzpatrick, K. K., Darcy, A., & Vierhile, M. (2017). 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 mental health, 4(2), e7785.
- Kazdin, A. E., & Rabbitt, S. M. (2013). Novel models for delivering mental health services and reducing the burdens of mental illness. Clinical Psychological Science, 1(2), 170-191.
- Mohr, D. C., Burns, M. N., Schueller, S. M., Clarke, G., & Klinkman, M. (2013). Behavioral intervention technologies: evidence review and recommendations for future research in mental health. General hospital psychiatry, 35(4), 332-338.
- Scott, J. L., Dawkins, S., Quinn, M. G., Sanderson, K., Elliott, K. E. J., Stirling, C., … & Robinson, A. (2016). Caring for the carer: a systematic review of pure technology-based cognitive behavioral therapy (TB-CBT) interventions for dementia carers. Aging & mental health, 20(8), 793-803.
- All Artificial Intelligence Content
- Artificial Intelligence in Cardiology
- Artificial Intelligence and Deep Learning in Medicine
- Machine Learning in Drug Development
- Deep-Learning in Radiology
Last Updated: Jul 13, 2022
Danielle graduated with a 2:1 in Biological Sciences with Professional Training Year from Cardiff University. During her Professional Training Year, Danielle worked with registered charity the Frozen Ark Project, creating and promoting various forms of content within their brand guidelines.Danielle has a great appreciation and passion for science communication and enjoys reading non-fiction and fiction in her spare time. Her other interests include doing yoga, collecting vinyl, and visiting museums.
Source: Read Full Article