CoChat

Artificial Intelligence-Based Conversational Agents for Chronic Conditions: Systematic Literature Review

Open access
Schachner, Theresa, Keller, Roman, v Wangenheim, Florian · 2020 · Journal of Medical Internet Research

Published version of a preprint

The DOI matches its Crossref record; 77 of the 77 references it cites were found in a registry or an index; 25 of the 138 sources its claims cite were checked for support.

Abstract

Background: A rising number of conversational agents or chatbots are equipped with artificial intelligence (AI) architecture. They are increasingly prevalent in health care applications such as those providing education and support to patients with chronic diseases, one of the leading causes of death in the 21st century. AI-based chatbots enable more effective and frequent interactions with such patients. Objective: The goal of this systematic literature review is to review the characteristics, health care conditions, and AI architectures of AI-based conversational agents designed specifically for chronic diseases. Methods: We conducted a systematic literature review using PubMed MEDLINE, EMBASE, PyscInfo, CINAHL, ACM Digital Library, ScienceDirect, and Web of Science. We applied a predefined search strategy using the terms “conversational agent,” “healthcare,” “artificial intelligence,” and their synonyms. We updated the search results using Google alerts, and screened reference lists for other relevant articles. We included primary research studies that involved the prevention, treatment, or rehabilitation of chronic diseases, involved a conversational agent, and included any kind of AI architecture. Two independent reviewers conducted screening and data extraction, and Cohen kappa was used to measure interrater agreement.A narrative approach was applied for data synthesis. Results: The literature search found 2052 articles, out of which 10 papers met the inclusion criteria. The small number of identified studies together with the prevalence of quasi-experimental studies (n=7) and prevailing prototype nature of the chatbots (n=7) revealed the immaturity of the field. The reported chatbots addressed a broad variety of chronic diseases (n=6), showcasing a tendency to develop specialized conversational agents for individual chronic conditions. However, there lacks comparison of these chatbots within and between chronic diseases. In addition, the reported evaluation measures were not standardized, and the addressed health goals showed a large range. Together, these study characteristics complicated comparability and open room for future research. While natural language processing represented the most used AI technique (n=7) and the majority of conversational agents allowed for multimodal interaction (n=6), the identified studies demonstrated broad heterogeneity, lack of depth of reported AI techniques and systems, and inconsistent usage of taxonomy of the underlying AI software, further aggravating comparability and generalizability of study results. Conclusions: The literature on AI-based conversational agents for chronic conditions is scarce and mostly consists of quasi-experimental studies with chatbots in prototype stage that use natural language processing and allow for multimodal user interaction. Future research could profit from evidence-based evaluation of the AI-based conversational agents and comparison thereof within and between different chronic health conditions. Besides increased comparability, the quality of chatbots developed for specific chronic conditions and their subsequent impact on the target patients could be enhanced by more structured development and standardized evaluation processes.

Overview

CoChat Summary

TL;DR

This systematic review identified only 10 studies of AI-based conversational agents for chronic conditions, indicating a scarce and immature field dominated by quasi-experimental studies and prototypes. Natural language processing was the most common AI technique and most agents allowed multimodal interaction, but evaluation measures and AI reporting were heterogeneous.

Key findings

  • From 2052 deduplicated citations, 10 papers met inclusion criteria; 7 were quasi-experimental, 2 were RCTs, and 1 was proof-of-concept. Seven chatbots were prototypes and not publicly available.
  • Natural language processing was the most used AI technique (n=7), followed by speech recognition (n=5), machine learning (n=3), natural language understanding (n=2), neural networks (n=2), and AIML (n=2). Six studies used multimodal interaction.
  • Reported technical performance was high in one study: accuracy 89%, precision 90%, sensitivity 89.9%, specificity 94.9%, F-measure 89.9%; another reported message response rates of 81%-97%. Two RCTs found chatbot interaction reduced symptoms of depression and anxiety compared with controls.
  • The agents addressed a broad variety of chronic conditions, including depression/anxiety, heart failure, asthma, Alzheimer disease, Parkinson/dementia, COPD, JIA, diabetes/glaucoma, and psoriasis. Evaluation measures were not standardized, and health goals ranged from self-care and monitoring to diagnosis, education, and coaching.

Methods

  • Systematic literature review following PRISMA, searching PubMed MEDLINE, EMBASE, PyscInfo, CINAHL, ACM Digital Library, ScienceDirect, and Web of Science in February 2020, with Google alerts and reference-list screening through April 2020.
  • Inclusion required primary research on prevention, treatment, or rehabilitation of chronic diseases; a conversational agent; and any AI architecture. Two independent reviewers screened titles, abstracts, and full texts, calculated Cohen kappa, and resolved disagreements by consensus.
  • Data were extracted into a 30-column spreadsheet and synthesized narratively. Risk of methodological bias was assessed with an adapted Maher et al CONSORT-based tool, with Cohen kappa of 79% between raters.

Limitations

  • The paper states key limitations: heterogeneity and relatively small number of included studies, prevalence of quasi-experimental studies, and no meta-analysis. It also states that risk of bias varied extensively, reducing trust in findings from high-risk studies, and that a different taxonomy could have led to omission of relevant agents.
  • The paper notes that many search results were conference abstracts omitted by eligibility criteria, and that study selection may still omit relevant AI-based conversational agents if another taxonomy were applied.
  • Evident from the text: only 10 studies were included, 50% were conducted in North America, none in Africa, and 90% of research locations were in Western cultures, limiting generalizability. The two RCTs both disclosed a relevant conflict of interest, and the review reports broad heterogeneity and lack of depth in reported AI techniques and systems.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.2196/20701

Cites

77 of 77 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram of included studies. Search updates were conducted until April 2020, with no additional papers being identified for inclusion. IRR: interrater reliability.
Figure 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram of included studies. Search updates were conducted until April 2020, with no additional papers being identified for inclusion. IRR: interrater reliability.CC BY · PubMed Central

Verification

Claims the paper makes

13 Valid · 1 Not valid · 46 Not checked · 85 Own results and interpretations

70 more claims · 2 with their cited sources checked

Create a free account to see all 145 claims and to verify the 113 references still open.

See all claims — free account
  • It may state two separate claims as one. 37
  • Probably not a substantial claim. 43

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

4 more claims about this paper

Create a free account to see every claim other papers make about this one.

See all claims — free account

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

3 of 3

Citations

Cites

Show all 77 references

Research works printed but not linked: 4

  1. RadziwillNBentonMEvaluating quality of chatbots and intelligent conversational agents20172020-08-30https://arxiv.org/pdf/1704.04579
  2. Masche, J, 2017, Advanced Computational Methods for Knowledge Engineering, 212
  3. CuiLHuangSWeiFTanCDuanCZhouMSuperAgent: a customer service chatbot for e-commerce websites2020-08-30https://www.aclweb.org/anthology/P17-4017.pdf
  4. Ireland, D, 2016, Stud Health Technol Inform, 227, 55

Cited by

See all 267

Metadata

Authors
Schachner, T., Keller, R., & v Wangenheim, F.
Affiliations
ETH Zurich (CH)
Corresponding author
Schachner, Theresa
Venue
Journal of Medical Internet Research
Published
2020-09-14
Volume
22
Issue
9
Pages
e20701
ISSN
1438-8871 (ISSN-L)
OpenAlex
W3046201587
Cited by
267
References
75
Citations by year
2026: 59 · 2025: 60 · 2024: 41 · 2023: 54 · 2022: 30 · 2021: 22 · 2020: 1
FWCI
2.03
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • Journal of Medical Internet Research (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Repository for Publications and Research Data (ETH Zurich) (submittedVersion, cc-by)
  • ETH Zürich Research Collection (cc-by)
Funders
National Research Foundation
Topics
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Health Sciences › Health Professions › General Health Professions › Mobile Health and mHealth Applications
  • Social Sciences › Psychology › Applied Psychology › Digital Mental Health Interventions
Keywords
chatbots, artificial intelligence, chronic disease, natural language processing, multimodal interaction, patient support, health education, AI architecture, evaluation metrics, digital health interventions, rehabilitation, disease prevention and treatment, research gaps, generalizability of study results
MeSH terms
Artificial Intelligence, Chronic Disease, Communication, Humans, Natural Language Processing

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 2026-10-04, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-04 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).

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

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.