The DOI matches its Crossref record; 75 of the 75 references it cites were found in a registry or an index; 129 of the 325 sources its claims cite were checked for support.
Voice-Based Conversational Agents for the Prevention and Management of Chronic and Mental Health Conditions: Systematic Literature Review
Open accessPublished version of a preprint
Abstract
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.
Overview
CoChat Summary
## TL;DR
This systematic review identified 12 studies on voice-based conversational agents (VCAs) for chronic and mental health conditions out of 7170 prescreened papers, and found that the evidence is still preliminary. The studies were all nonexperimental, highly heterogeneous, and generally focused on technology acceptance and system accuracy rather than actual clinical effectiveness.
## Key findings
- Of 7170 prescreened papers, 12 met the inclusion criteria; all 12 studies were nonexperimental.
- VCAs were used for behavioral support (n=5), health monitoring (n=3), or both (n=4), and were delivered via smartphones (n=5), tablets (n=2), smart speakers (n=3), or unspecified devices (n=2).
- Technology acceptance was assessed in 7 studies, but only 3 used validated instruments; system performance was reported in 6 studies, with 3 describing good or very good speech recognition performance and 1 reporting mediocre recognition accuracy.
- The included studies targeted a range of conditions, including cancer (n=3 VCAs), diabetes (n=2), heart failure (n=2), hearing impairment, asthma, Parkinson disease, dementia, autism, intellectual disability, and depression.
- Risk bias varied markedly, with scores ranging from 1 to 11.25 and a median of 6.36.
## Methods
- Systematic literature review following PRISMA, searching PubMed MEDLINE, Embase, PsycINFO, Scopus, and Web of Science; the search was conducted using literature available in July 2020.
- Included primary research on prevention, treatment, or management of chronic diseases or mental disorders delivered through a voice-based conversational agent, with empirical evaluation of system accuracy and/or technology acceptance.
- Two independent reviewers screened titles, abstracts, and full texts; two reviewers extracted data; disagreement was resolved with a third investigator; synthesis was narrative, and risk of bias was assessed using an adapted CONSORT-based tool.
## Limitations
- The paper states that its search strategy used nonspecific terms such as “health,” which may have retrieved many irrelevant papers; this was an explicit tradeoff to avoid missing relevant literature.
- The review covers research evidence only and does not include industry developments; this is stated in the paper.
- The authors state that their bias tool may be unsuitable because none of the included papers were experimental trials; this is a limitation stated in the paper.
- Evident from the text: the evidence base is small (12 studies), all studies were nonexperimental and mostly laboratory/short-term evaluations, so the review cannot establish actual effectiveness of VCAs in improving health outcomes.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures
Verification
Claims the paper makes
53 Valid · 6 Not valid · 39 Not checked · 73 Own results and interpretations
Some cited sources support this claim and some don’t — read each source’s verdict below.
In the paper
“Half of the studies (n=6) did not report on any system measures [59,60,62,65-67]”
§ Voice-Based Conversational Agents for the Prevention and Management of Chronic and Mental Health Conditions: Systematic Literature Review > Results > Methodology of the Included Studies
What the sources say
- Supportedcites [59] Amith, M, 2019, Stud Health Technol Inform source
The Methods (p. 4) describe participant surveys of usability, speech-interface quality, and HPV vaccine beliefs; the Results (pp. 4–6) report survey scores rather than technical system-performance measures. This supports citing the paper as one study that did not report system measures. The paper alone cannot establish the review’s aggregate count of six studies. (LLM, conf=0.82)
- Disputedcites [60] Amith, M, 2020, AMIA Jt Summits Transl Sci Proc sourceExperimental — accuracy not yet measured
The cited paper reports measures of its conversational agent rather than reporting none. Its Methods (p. 4) identify the SUS and SUISQ instruments, and its Results (pp. 4–5) report system-usability and speech-interface scores. The paper cannot establish how many of the other included studies reported measures. (LLM, conf=0.80)
Checked by a reasoning model — may be wrong.
- Not checked — no full text of the sourcecites [62] source
- Not matchedcites [65] Ireland, D, 2016, Stud Health Technol Inform
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Disputedcites [66] sourceExperimental — accuracy not yet measured
The Evaluation Method section lists chatbot quality and system usability among kBot’s evaluation criteria. The Evaluation Result section reports scores for quality metrics and system usability (SUS), contrary to the attribution that this study reported no system measures. The paper does not address the citing review’s count of six studies. (LLM, conf=0.86)
Checked by a reasoning model — may be wrong.
- Not checked — no full text of the sourcecites [67] source
plainn = 6- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“On the same note, conversational agents have been proven a valuable tool to deliver digital health interventions [7-9].”
§ Voice-Based Conversational Agents for the Prevention and Management of Chronic and Mental Health Conditions: Systematic Literature Review > Introduction > Background
What the sources say
- Partly supportedcites [7] sourceExperimental — accuracy not yet measured
The Results describe conversational agents used for health tasks, including self-care, and report one efficacy trial that reduced depression symptoms. But the Conclusions say efficacy was rarely evaluated and the evidence remains limited, so the review does not establish that conversational agents have been proven valuable for digital health interventions generally. Page numbers were not available in our copy. (LLM, conf=0.87)
- Partly supportedcites [8] sourceExperimental — accuracy not yet measured
The Introduction (page unavailable) says conversational agents have demonstrated benefits and are used to deliver digital health interventions. The Discussion’s Principal Findings (page unavailable) reports generally positive results but notes that evidence on health outcomes came from only two studies and that most studies were quasi-experimental. “Proven” overstates the strength of the evidence this review reports. (LLM, conf=0.83)
- Partly supportedcites [9] sourceExperimental — accuracy not yet measured
The Results (Conversational Agent Evaluation; page unavailable) report positive outcomes for some conversational-agent interventions, including improvements in food consumption, physical activity, and diabetes measures. But the Discussion (Principal Findings; page unavailable) says most studies were case studies and notes gaps in evaluating adverse effects. The source supports their potential value for delivering digital health interventions, but “proven” overstates the evidence. (LLM, conf=0.88)
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“whereas 1 study [60] described better usability of its VCA for human papillomavirus (HPV) in comparison with industry standards (ie, SUS score of mean 72/100)”
§ Voice-Based Conversational Agents for the Prevention and Management of Chronic and Mental Health Conditions: Systematic Literature Review > Results > Main Findings
What the sources say
- Supportedcites [60] Amith, M, 2020, AMIA Jt Summits Transl Sci Proc source
The Results (p. 4) compare the agent with an industry mean SUS score of 72 and describe its score as on par with or slightly above that benchmark. The Discussion (p. 6) gives the agent’s score as 77 versus 72, while noting 94% confidence in the comparison. (LLM, conf=0.90)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“A total of 3 studies [60,66,67] reported technology acceptance through a System Usability Survey (SUS).”
§ Voice-Based Conversational Agents for the Prevention and Management of Chronic and Mental Health Conditions: Systematic Literature Review > Results > Main Findings
What the sources say
- Supportedcites [60] Amith, M, 2020, AMIA Jt Summits Transl Sci Proc source
The Methods (p. 4) identify the System Usability Survey (SUS) as one of the instruments used, and the Results (p. 4) report a SUS score for the conversational agent. This supports citing the paper as one study that assessed usability with SUS; the total of three studies comes from the citing document, not this source alone. (LLM, conf=0.91)
- Partly supportedcites [66] sourceExperimental — accuracy not yet measured
The Evaluation Method identifies technology acceptance and system usability as separate criteria. The Evaluation Result reports technology acceptance from the first questionnaire and SUS scores from a second questionnaire; this paper did not measure technology acceptance through SUS. The cited paper also cannot establish the citing sentence’s total of three studies. (LLM, conf=0.88)
- Not checked — no full text of the sourcecites [67] source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“whereas 3 [63,68,69] reported good or very good speech recognition performance”
§ Voice-Based Conversational Agents for the Prevention and Management of Chronic and Mental Health Conditions: Systematic Literature Review > Results > Main Findings
What the sources say
- Supportedcites [63] source
The Speech Recognition Results (p. 4) report that nine of fourteen participants had checkup word error rates below 21%, and the Discussion (p. 5) calls the speech-recognition performance “quite encouraging.” This supports describing this paper as one reporting good performance, although results were poor for the other five participants. The count of three studies belongs to the citing document, not to this source alone. (LLM, conf=0.86)
- Not checked — no full text of the sourcecites [68] source
- Not checked — no full text of the sourcecites [69] source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
49 more claims · 12 with their cited sources checked
Create a free account to see all 171 claims and every source verdict.
- Probably not a substantial claim. 69
- It may state two separate claims as one. 79
- The quote only partly states this claim. 1
Citations
Cites
A Coefficient of Agreement for Nominal Scales
Cohen, J. (1960) · Educational and Psychological Measurement · doi:10.1177/001316446002000104 · Cited by 42,398 · cited 1×
“The interpretation of the Cohen kappa coefficient was based on the categories developed by Douglas Altman: 0.00-0.20 (poor), 0.21-0.40 (fair), 0.41-0.60 (moderate), 0.61-0.80 (good), and 0.81-1.00 (very good) [56,57].”
InfluentialPRISMA-P Group, Moher, D., Shamseer, L., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., Shekelle, P., & Stewart, L. A. (2015) · Systematic Reviews · doi:10.1186/2046-4053-4-1 · Cited by 26,223 · cited 1×
“This study is compliant with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) checklist [48] (an overview of the study protocol is given in Multimedia Appendix 1 [49-55]).”
Venkatesh, V., Thong, J. Y. L., & Xu, X. (2012) · MIS Quarterly · doi:10.2307/41410412 · Cited by 15,756 · cited 2×
“However, this aspect has an important influence on the technology acceptance [71].”
CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials
the CONSORT Group, Schulz, K. F., Altman, D. G., & Moher, D. (2010) · BMC Medicine · doi:10.1186/1741-7015-8-18 · Cited by 13,371
Shamseer, L., Moher, D., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., Shekelle, P., Stewart, L. A., & the PRISMA-P Group (2015) · BMJ · doi:10.1136/bmj.g7647 · Cited by 13,120 · cited 1×
“This study is compliant with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) checklist [48] (an overview of the study protocol is given in Multimedia Appendix 1 [49-55]).”
Practical Statistics for Medical Research
Altman, D. G. (1990) · doi:10.1201/9780429258589 · Cited by 10,668
CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials
Schulz, K. F., Altman, D. G., Moher, D., & for the CONSORT Group (2010) · BMJ · doi:10.1136/bmj.c332 · Cited by 9,329
Relation of the therapeutic alliance with outcome and other variables: a meta-analytic review
Martin, D. J., Garske, J. P., & Davis, M. K. (2000) · Journal of Consulting and Clinical Psychology · doi:10.1037//0022-006x.68.3.438 · Cited by 3,355 · cited 1×
“Just as text-based [12,13] and embodied [14] conversational agents, VCAs have the potential to form an alliance [15] or rapport [16] with the patient through conversation, which is beneficial to treatment outcomes [17-19].”
The Technology Acceptance Model: Its past and its future in health care
Holden, R. J., & Karsh, B.-T. (2010) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2009.07.002 · Cited by 2,669 · cited 1×
“Thus, we include evidence from both journals and conference papers and provide an overview of aspects affecting technology adoption, that is, system and user performance, ease of use, and attitude toward the target health behavior [47].”
CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials
Schulz, K. F., Altman, D. G., Moher, D., & for the CONSORT Group (2010) · PLoS Medicine · doi:10.1371/journal.pmed.1000251 · Cited by 2,296
Research works printed but not linked: 7
- 2020, World Health Statistics 2020: monitoring health for the SDGs, sustainable development goals, 1
- 2019, Document number: WHO/MSD/MER/19.3
- 10.1136/bmj.g7647 · DOI · Paper page
- Amith, M, 2019, Stud Health Technol Inform, 257, 17
- Amith, M, 2020, AMIA Jt Summits Transl Sci Proc, 2020, 43
- Ireland, D, 2016, Stud Health Technol Inform, 227, 55
- Moorthy, AE, 2014, Information and Knowledge in Applications and Services, 324
Cited by
Mariani, M. M., Hashemi, N., & Wirtz, J. (2023) · Journal of Business Research · doi:10.1016/j.jbusres.2023.113838 · Cited by 343
Roles, Users, Benefits, and Limitations of Chatbots in Health Care: Rapid Review (Preprint)
Laymouna, M., Ma, Y., Lessard, D., Schuster, T., Engler, K., & Lebouché, B. (2024) · doi:10.2196/preprints.56930 · Cited by 302
Roles, Users, Benefits, and Limitations of Chatbots in Health Care: Rapid Review
Laymouna, M., Ma, Y., Lessard, D., Schuster, T., Engler, K., & Lebouché, B. (2024) · Journal of Medical Internet Research · doi:10.2196/56930 · Cited by 305
“…in mental health support for various age groups, reflecting the pressing need for accessible mental health services highlighted by others [4,8,12-17, Furthermore, chatbots have emerged as tools for reducing stigma [12,265], linking users to health services [266-268], and protecting sensitive…”
Huq, S. M., Maskeliūnas, R., & Damaševičius, R. (2022) · Disability and Rehabilitation Assistive Technology · doi:10.1080/17483107.2022.2146768 · Cited by 106
Imran, M., Zaman, U., Imran, I., Imtiaz, J., Fayaz, M., & Gwak, J. (2021) · Electronics · doi:10.3390/electronics10202501 · Cited by 91
“Mobile application has been introduced based on machine learning algorithms by university students from the DY Patil Institute, Mumbai, India, to detect COVID-19 patients [220].”
Method
Metadata
- Authors
- Bérubé, C., Schachner, T., Keller, R., Fleisch, E., v Wangenheim, F., Barata, F., & Kowatsch, T.
- Affiliations
- ETH Zurich (CH)
- Singapore-ETH Centre (SG)
- University of St.Gallen (CH)
- National University of Singapore (SG)
- Corresponding author
- Bérubé, Caterina
- ORCID
- Venue
- Journal of Medical Internet Research
- Publisher
- JMIR Publications Inc.
- Published
- 2021-03-29
- Volume
- 23
- Issue
- 3
- Pages
- e25933
- ISSN
- 1438-8871 (ISSN-L)
- PMC
- PMC8042539
- OpenAlex
- W3135619344
- Semantic Scholar
- f96d2964c795a9a9ff0de4b46abb73310566597c
- Cited by
- 133
- References
- 74
- Citations by year
- 2026: 22 · 2025: 33 · 2024: 34 · 2023: 15 · 2022: 18 · 2021: 11
- FWCI
- 3.13
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- Locations
- Journal of Medical Internet Research (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Repository for Publications and Research Data (ETH Zurich) (submittedVersion, cc-by)
- National University of Singapore (submittedVersion, cc-by)
- Alexandria (UniSG) (University of St.Gallen) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- ETH Zürich Research Collection (cc-by)
- Funders
- National Research Foundation
- Topics
- Social Sciences › Psychology › Applied Psychology › Digital Mental Health Interventions
- Health Sciences › Health Professions › General Health Professions › Mobile Health and mHealth Applications
- Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Telemedicine and Telehealth Implementation
- Keywords
- voice assistants, chronic disease prevention, mental health, health behavior change, technology acceptance, speech recognition accuracy, health monitoring, behavioral support, health intervention evaluation, risk of bias, cancer, diabetes mellitus, heart failure, depression, dementia, autism spectrum disorder, hearing loss, Parkinson's disease, asthma, intellectual disability, smart speakers, smartphones, tablets, risk of bias assessment tools
- MeSH terms
- Smartphone, Asthma, Communication, Humans, Mental Health, Health Behavior
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-09-30, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-09-30 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Voice-Based Conversational Agents for the Prevention and Management of Chronic and Mental Health Conditions: Systematic Literature Review. https://app.cochat.ai/papers/10.2196/25933
- The paper
- Bérubé, C., Schachner, T., Keller, R., Fleisch, E., v Wangenheim, F., Barata, F., & Kowatsch, T. (2021). Voice-Based Conversational Agents for the Prevention and Management of Chronic and Mental Health Conditions: Systematic Literature Review. Journal of Medical Internet Research. https://doi.org/10.2196/25933
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.
