Can AI-chatbots promote health-lifestyle changes?

By Suchandrima BhowmikJul 11 2022Reviewed by Benedette Cuffari, M.Sc.

Artificial intelligence (AI) chatbots are capable of mimicking human interactions with the help of oral, written, or verbal communication with the user. AI chatbots can provide important health-related information and services, ultimately leading to promising technology-facilitated interventions.

Study: Artificial Intelligence (AI)-based Chatbots in Promoting Health Behavioral Changes: A Systematic Review. Image Credit: TippaPatt / Shutterstock.com

This news article was a review of a preliminary scientific report that had not undergone peer-review at the time of publication. Since its initial publication, the scientific report has now been peer reviewed and accepted for publication in a Scientific Journal. Links to the preliminary and peer-reviewed reports are available in the Sources section at the bottom of this article. View Sources

AI chatbots in healthcare

Current digital telehealth and therapeutic interventions are associated with several challenges including unsustainability, low adherence, and inflexibility. AI chatbots are capable of overcoming these challenges and providing personalized on-demand support, higher interactivity, and higher sustainability.

AI chatbots utilize data input from various sources, which is followed by data analysis that is completed through natural language processing (NLP) and machine learning (ML). Data output then helps users achieve their health behavior goals.

Thus, AI chatbots are capable of promoting diverse health behaviors by effectively delivering interventions. Moreover, this technology can provide additional benefits to health behavior changes by integrating into embodied functions.

Most previous studies conducted on AI chatbots aimed to improve mental health outcomes. Comparatively, recent studies are increasingly focused on the use of AI chatbots for inciting health behavior changes.

However, one systematic review on the impact of AI chatbots on modifying lifestyles was associated with several limitations. These include the inability of the authors to differentiate AI chatbots from other chatbots. Furthermore, this study targeted only a limited set of behaviors and did not discuss all potential platforms that could utilize AI chatbots.

A new systematic review published on the preprint server medRxiv* discusses the results of previous studies on AI chatbot intervention characteristics, functionality, and components, as well as their impact on a wide range of health behaviors.

About the study

The current study was carried out in June 2022 and followed the PRISMA guidelines. Herein, three authors searched seven bibliographic databases including IEEE Xplore, PubMed, JMIR publications, EMBASE, ACM Digital Library, Web of Science, and PsychINFO.

The search involved a combination of keywords that belonged to three categories. The first category included keywords that were related to AI-based chatbots, the second included keywords related to health behaviors, and the third focused on interventions.

The inclusion criteria for the search were studies that involved intervention research focused on health behaviors, those that were developed on existing AI platforms or AI algorithms, empirical studies that used chatbots, English articles that were published between 1980 and 2022, as well as studies that reported quantitative or qualitative intervention results. All data were extracted from these studies and underwent quality assessment according to the National Institutes of Health (NIH) Quality Assessment Tool.

Study findings

A total of 15 studies matched the inclusion criteria, most of which were distributed across developed countries. The median sample size was 116 participants, while the mean was 7,200 participants.

Most of the studies included adult participants, while only two involved participants less than 18 years of age. All study participants had pre-existing conditions and included individuals with lower physical exercise, obese, smokers, substance abusers, breast cancer patients, and Medicare recipients.

The target health behaviors included cessation of smoking, promotion of a healthy lifestyle, reduction of substance abuse, and adherence to medication or treatment. Moreover, only four studies were reported to use randomized control trials (RCTs), while others used a quasi-experimental design.

The risk of reporting outcomes and bias in the randomization process was low, the risk of bias from intended interventions was low to moderate, the risk of bias in outcome measurement was moderate, and the risk of outcomes from unintended sources was high. All factors for the description of AI components were sufficient, except the handling of unavailable input data and input data characteristics.

Out of 15 studies, six reported feasibility in terms of the mean number of messages that were exchanged with the chatbot per month and safety. Moreover, 11 studies reported usability in terms of usability of the content, ease of using the chatbot, user-initiated conversation, non-judgmental safe space, and outside-office support. Acceptability and engagement were reported in 12 studies in terms of satisfaction, rate of retention, technical issues, and duration of the engagement.

An increase in physical activity was reported in six studies, along with an improvement in the diet in three studies through chatbot-based interventions. Smoking cessation was reported in the four assessed studies, whereas one study reported a reduction in substance use and two studies reported an increase in adherence to treatment or medication through the use of chatbots.

Several behavioral change theories were integrated into the chatbots including the transtheoretical model (TTM), cognitive behavioral therapy (CBT), social cognitive theory (SCT), habit formation model, motivational interviewing, Mohr’s Model of Supportive Accountability, and emotionally focused therapy to provide motivational support and track the behavior of participants. Most of the studies targeted behavioral goal setting, used behavioral monitoring, and offered behavior-related information, while four studies also provided emotional support.

Most studies used different AI techniques such as ML, NLP, Hybrid Health Recommender Systems (HHRS), hybrid techniques (ML and NLP), and face-tracking technology to provide personalized interventions. The chatbots primarily used text-based communication and were either integrated into pre-existing platforms or delivered as independent platforms. Additionally, most chatbots required data on the users’ background information, their goals, and feedback on behavioral performance to ensure the delivery of personalized services.

Conclusions

Taken together, AI chatbots can efficiently promote a healthy lifestyle, the cessation of smoking, and adherence to treatment or medications. Additionally, the current study found that AI chatbots demonstrated significant usability, feasibility, and acceptability.

Taken together, AI chatbots are capable of providing personalized interventions and could be scalable to diverse and large populations. However, further studies are needed to acquire an accurate description of AI-related processes, since AI chatbot interventions are still at a nascent stage.

Limitations

The current study did not include a meta-analysis and focused only on three behavioral outcomes. Additionally, articles from unselected databases, articles in other languages, grey literature, and unpublished articles were not included in the study.

An additional limitation was that the interventions could not provide a clear description of excluded AI chatbots. The study also lacked generalizability and patient safety information was limited.

This news article was a review of a preliminary scientific report that had not undergone peer-review at the time of publication. Since its initial publication, the scientific report has now been peer reviewed and accepted for publication in a Scientific Journal. Links to the preliminary and peer-reviewed reports are available in the Sources section at the bottom of this article. View Sources