Can AI-chatbots promote health-lifestyle changes?

Artificial intelligence (AI) chatbots have the ability to mimic human interaction with the user through verbal, written or oral communication. AI chatbots can provide important health information and services, ultimately leading to promising technology-facilitated interventions.

AI chatbots in healthcare
Current digital telehealth and therapeutic interventions are associated with several challenges, including instability, low compliance, and flexibility. Artificial intelligence chatbots are able to overcome these challenges and provide customized support, higher interactivity and higher durability.

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

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

Most of the previous research on AI chatbots aimed to improve mental health outcomes. In comparison, recent research has increasingly focused on the use of AI chatbots to promote health behavior changes.

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

A new systematic review published on the preprint server medRxiv* Discusses findings from previous research on AI chatbot intervention features, functionality, and components, as well as their implications for broader health behaviors.

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

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

Inclusion criteria for the search were studies involving intervention studies targeting health behaviors, those developed on existing AI platforms or AI algorithms, empirical studies using chatbots, articles in English published between 1980 and 2022, as well as studies. reported quantitative or qualitative intervention results. All data were extracted from these studies and quality assessed according to the National Institutes of Health (NIH) Quality Assessment Tool.

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

Most of the studies involved adult participants, with only two involving participants under 18 years of age. All study participants had pre-existing conditions and included those who were less physically active, obese, smokers, substance abusers, breast cancer patients, and Medicare recipients.

Target health behaviors include smoking cessation, promoting a healthy lifestyle, reducing substance abuse, and adherence to medication or treatment. Moreover, only four studies were reported to have used randomized control trials (RCTs), while the others used a quasi-experimental design.

The randomization process had low risk of reporting bias, low to moderate risk of bias from intended interventions, moderate risk of bias in outcome measurement, and high risk of bias from unexpected sources. All factors were sufficient for the description of AI components, non-accessible input data management and input data characteristics.

Six of the 15 studies reported the average number of messages exchanged with a chatbot per month and its feasibility in terms of security. Moreover, 11 studies reported usability of the content, ease of use of the chatbot, user-initiated conversation, non-judgmental safe space, and out-of-office support. Acceptability and engagement in terms of satisfaction, retention rate, technical issues, and task duration were reported in 12 studies.

Increases in physical activity were reported in six studies, along with improved diet through chatbot-based interventions in three studies. Four of the evaluated studies reported smoking cessation, while one study reported a reduction in substance use and two studies reported an increase in treatment or medication adherence with the use of chatbots.

Several behavior change theories have been integrated into chatbots, including the transtheoretical model (TTM), cognitive behavioral therapy (CBT), social cognitive theory (SCT), the habit formation model, motivational interviewing, Mohr’s supportive accountability model, and emotionally focused therapy. motivational support and monitoring of participants’ behavior. Most studies aimed to set behavioral goals, used behavioral monitoring, and offered behavioral information, while four studies provided emotional support.

Most studies have used various artificial intelligence techniques such as ML, NLP, Hybrid Health Recommender Systems (HHRS), hybrid methods (ML and NLP), and face tracking technology to provide personalized interventions. Chatbots primarily used text-based communication and were either integrated into pre-existing platforms or delivered as stand-alone platforms. In addition, most chatbots required information about users’ information, goals, and behavioral performance feedback to ensure the delivery of personalized services.

Results
Taken together, AI chatbots can effectively promote healthy lifestyles, smoking cessation, and adherence to treatment or medication. In addition, the current study found that AI chatbots demonstrate a significant degree of usability, feasibility, and acceptability.

Taken together, AI chatbots are capable of delivering personalized interventions and are scalable to diverse and large populations. However, further research is needed to obtain an accurate picture of AI-related processes, as AI chatbot interventions are still in their infancy.

Restrictions
The current study did not include a meta-analysis and focused on only three behavioral outcomes. In addition, articles from unselected databases, articles in other languages, gray literature, and unpublished articles were excluded from the study.

An additional limitation was that the interventions may not provide an accurate representation of excluded AI chatbots. The study also lacked generalizability and data on patient safety were limited. Getabout Columbi