Feasibility of pediatric obesity and prediabetes treatment support through Tess, the AI behavioral coaching chatbot

Autor: Lloyd N. Werk, Michiel Rauws, Taylor N. Stephens, Angela Joerin
Rok vydání: 2019
Předmět:
Zdroj: Translational Behavioral Medicine. 9:440-447
ISSN: 1613-9860
1869-6716
DOI: 10.1093/tbm/ibz043
Popis: Behavioral intervention technologies (BITs) are unique ways to incorporate the benefits of technology and psychology to address differing health needs through various media, including Internet interventions, mobile apps, and video games. BITs present several possible benefits, including increased dissemination and accessibility, cost-effectiveness, increased engagement, and decreased stigma, especially among youth. A behavioral coaching chatbot, Tess, addresses different facets of behavioral health, such as depression and anxiety. Available 24/7, Tess delivers customized integrative support, psychoeducation, and interventions through brief conversations via existing communication channels (i.e., SMS text messaging and Facebook Messenger). This study assessed the feasibility of integrating Tess in behavioral counseling of adolescent patients (n = 23; Mage = 15.20 years; Rangeage = 9.78-18.54 years; 57% female) coping with weight management and prediabetes symptoms. Tess engaged patients via a preferred method of communication (SMS text messaging) in individualized conversations to promote treatment adherence, behavior change, and overall wellness. Adolescent patients reported experiencing positive progress toward their goals 81% of the time. The 4,123 messages exchanged and patients' reported usefulness ratings (96% of the time) illustrate that adolescents engaged with and viewed this chatbot as helpful. These results highlight the feasibility and benefit of support through artificial intelligence, specifically in a pediatric setting, which could be scaled to serve larger groups of patients. As a partner to clinicians, Tess can continue the therapeutic interaction outside office hours while maintaining patient satisfaction. Due to Tess's capacity for continuous learning, future iterations may have additional features to increase the user experience.
Databáze: OpenAIRE