Voice-Based Conversational Agents for the Prevention and Management of Chronic and Mental Health Conditions: Systematic Literature Review

Background Chronic and mental health conditions are increasingly prevalent worldwide. As devices in our everyday lives offer more and more voice-based self-service, voice-based conversational agents (VCAs) have the potential to support the prevention and management of these conditions in a scalable manner. However, evidence on VCAs dedicated to the prevention and management of chronic and mental health conditions is unclear. Objective This study provides a better understanding of the current methods used in the evaluation of health interventions for the prevention and management of chronic and mental health conditions delivered through VCAs. Methods We conducted a systematic literature review using PubMed MEDLINE, Embase, PsycINFO, Scopus, and Web of Science databases. We included primary research involving the prevention or management of chronic or mental health conditions through a VCA and reporting an empirical evaluation of the system either in terms of system accuracy, technology acceptance, or both. A total of 2 independent reviewers conducted the screening and data extraction, and agreement between them was measured using Cohen kappa. A narrative approach was used to synthesize the selected records. Results Of 7170 prescreened papers, 12 met the inclusion criteria. All studies were nonexperimental. The VCAs provided behavioral support (n=5), health monitoring services (n=3), or both (n=4). The interventions were delivered via smartphones (n=5), tablets (n=2), or smart speakers (n=3). In 2 cases, no device was specified. A total of 3 VCAs targeted cancer, whereas 2 VCAs targeted diabetes and heart failure. The other VCAs targeted hearing impairment, asthma, Parkinson disease, dementia, autism, intellectual disability, and depression. The majority of the studies (n=7) assessed technology acceptance, but only few studies (n=3) used validated instruments. Half of the studies (n=6) reported either performance measures on speech recognition or on the ability of VCAs to respond to health-related queries. Only a minority of the studies (n=2) reported behavioral measures or a measure of attitudes toward intervention-targeted health behavior. Moreover, only a minority of studies (n=4) reported controlling for participants’ previous experience with technology. Finally, risk bias varied markedly. Conclusions The heterogeneity in the methods, the limited number of studies identified, and the high risk of bias show that research on VCAs for chronic and mental health conditions is still in its infancy. Although the results of system accuracy and technology acceptance are encouraging, there is still a need to establish more conclusive evidence on the efficacy of VCAs for the prevention and management of chronic and mental health conditions, both in absolute terms and in comparison with standard health care.

Propose the possibility of automated discourse that could mimic patient-provider dialogue and provide consistent responses Support Targeted health information based on health status Automated counseling on HPV vaccination compare voice-activated internet searches by smartphone (two digital assistants) with laptop ones for information and advice related to smoking cessation Support Look-up of health information Providing information or advice on smoking cessation.
Overcome lack of effectiveness of self-management mobile apps in elderly with a Google Home assistant application

Monitoring and Support
Look-up of health information Active data capture/documentation Targeted alerts and reminders healthy coping and monitoring surveys, medication reminders, information on activity and nutrition Feasibility evaluation of whether the system could identify with high accuracy the information the patient provides in the interview (compared to nurse practitioners) Monitoring Active data capture/documentation conducting self-care checkup via health monitoring interviews with chronic heart failure patients Describes the concept and implementation of an interactive storytelling application for conversational-voice interfaces and reports the preliminary results of a pilot test Support Targeted health information based on health status Provide people with intellectual disabilities with a playful, accessible, and cost-effective training platform to improve their social interaction abilities.
Detailing the realization of the concept of a conversational agent for the remote monitoring of audio and conversation dialogues in the form of a smartphone application Monitoring Self monitoring of health or diagnostic data Remote monitoring of audio and conversation dialogues Preliminary evaluation of a knowledge-enables personalized conversational agent to assist pediatric asthmatic patients

Monitoring and Support
Look-up of health information Active data capture/documentation Self monitoring of health or diagnostic data Targeted alerts and reminders Monitors asthma symptoms and checks both co-occurrence of factors potentially triggering asthma symptoms to then deliver warnings whenever the factors enter the unhealthy range To design, develop and assess the usability of the concept of CARMIE as a virtual medication advisor Monitoring and Support Look-up of health information Active data capture/documentation Targeted health information based on health status 1) Deliver information and knowledge-based advice by providing assistance on posology, interactions, indications, and adverse reactions 2) assesses symptoms in case of out-of-prescription medicine inteka intention, and generates a medical report for the registered healthcare staff 3) To motivate the user through interactive dialogue and cues, trying to increase medication adherence. Storage of medical prescriptions with posology; Delivery information a pharmacological property; Delivery suggestion in respose to a the patient asking for permission to take a medicine; In response of permission asking, it also assesses symptoms associated with the medicine and with deterioration of heart failure and provides suitable advice accordingly; Generatinof timestamped assessment answers as medical report available for sharing by email or text message  ASR error rate: Average score insertion rate is (6.0 ± 2.3)%, Average score deletion of (3.7 ± 1.0)%; SRT measurement accuracy: significant bias of 0.6 dB ASR error rate: Average score insertion rate is (6.0 ± 2.3)%, Average score deletion of (3.7 ± 1.0)%; SRT measurement accuracy: significant bias of 0.6 dB; SRT bias: paired-sample t-test, p = 0.002, M = −7.4±0.9 dB SNR with smart speaker, M = −8.0 ± 0.9 dB SNR with the starndard MST) Basic greeting: all IPAs could acquire and maintain some context information from the user like the location but these services couldn´t provide any suggestion to the user, which would be a good point to continue the conversation as evaluated in the follow-up parameter; Email: some IPAs like Google Assistant and Microsoft Cortana only could write an email, Amazon Alexa could read and reply, while the Apple Siri could read and write an email. The follow-up parameter in this activity obtained bad results in every IPA with the highlight went to Amazon Alexa for suggesting the user to reply after reading an email to the sender, with predefined messages; Social network: all IPAs could let the user to manage their calendar and, except for Google Assistant, the possibility to send or read a message in a social network like Twitter or Facebook. The highlight in this activity went to the Amazon Alexa that allows the user to check the birthday calendars of their friends, this could be useful to letting the elderly people upkeep the birthdays dates of their relative ones. The followup in this activity also obtained bad results, with the IPA not suggesting anything to the user; Social game: present in almost all IPAs except the Apple Siri. The social games presented were mostly in quizzes format. The context data acquire from the IPA were practically none, but the IPA frequently had more initiative to follow-up in their dialog with the user suggesting some clues to help the user or giving other related games to play.