CoChat
Sign Up
Cited by 103Mobile Health and mHealth ApplicationsDigital Mental Health InterventionsSocial Robot Interaction and HRI

The DOI matches its Crossref record; 102 of the 102 references it cites were found in a registry or an index; none of the 97 sources its claims cite is checked for support yet.

Conversational Agents as Mediating Social Actors in Chronic Disease Management Involving Health Care Professionals, Patients, and Family Members: Multisite Single-Arm Feasibility Study

Open access
Kowatsch, Tobias, Schachner, Theresa, Harperink, Samira, Barata, Filipe, Dittler, Ullrich, Xiao, Grace, Stanger, Catherine, v Wangenheim, Florian, Fleisch, Elgar, Oswald, Helmut, Möller, Alexander·2021·Journal of Medical Internet Research
DOI 10.2196/25060PMC PMC7929753

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

Background: Successful management of chronic diseases requires a trustful collaboration between health care professionals, patients, and family members. Scalable conversational agents, designed to assist health care professionals, may play a significant role in supporting this collaboration in a scalable way by reaching out to the everyday lives of patients and their family members. However, to date, it remains unclear whether conversational agents, in such a role, would be accepted and whether they can support this multistakeholder collaboration. Objective: With asthma in children representing a relevant target of chronic disease management, this study had the following objectives: (1) to describe the design of MAX, a conversational agent–delivered asthma intervention that supports health care professionals targeting child-parent teams in their everyday lives; and (2) to assess the (a) reach of MAX, (b) conversational agent–patient working alliance, (c) acceptance of MAX, (d) intervention completion rate, (e) cognitive and behavioral outcomes, and (f) human effort and responsiveness of health care professionals in primary and secondary care settings. Methods: MAX was designed to increase cognitive skills (ie, knowledge about asthma) and behavioral skills (ie, inhalation technique) in 10-15-year-olds with asthma, and enables support by a health professional and a family member. To this end, three design goals guided the development: (1) to build a conversational agent–patient working alliance; (2) to offer hybrid (human- and conversational agent–supported) ubiquitous coaching; and (3) to provide an intervention with high experiential value. An interdisciplinary team of computer scientists, asthma experts, and young patients with their parents developed the intervention collaboratively. The conversational agent communicates with health care professionals via email, with patients via a mobile chat app, and with a family member via SMS text messaging. A single-arm feasibility study in primary and secondary care settings was performed to assess MAX. Results: Results indicated an overall positive evaluation of MAX with respect to its reach (49.5%, 49/99 of recruited and eligible patient-family member teams participated), a strong patient-conversational agent working alliance, and high acceptance by all relevant stakeholders. Moreover, MAX led to improved cognitive and behavioral skills and an intervention completion rate of 75.5%. Family members supported the patients in 269 out of 275 (97.8%) coaching sessions. Most of the conversational turns (99.5%) were conducted between patients and the conversational agent as opposed to between patients and health care professionals, thus indicating the scalability of MAX. In addition, it took health care professionals less than 4 minutes to assess the inhalation technique and 3 days to deliver related feedback to the patients. Several suggestions for improvement were made. Conclusions: This study provides the first evidence that conversational agents, designed as mediating social actors involving health care professionals, patients, and family members, are not only accepted in such a “team player” role but also show potential to improve health-relevant outcomes in chronic disease management.

Overview

AI snapshot

Written from the paper’s full text.

TL;DR

In a multisite single-arm feasibility study, MAX, a conversational agent designed as a mediating social actor for pediatric asthma, was feasible and positively evaluated across patients, family members, and health care professionals. It showed high reach, strong patient-agent working alliance, 75.5% completion, improved asthma knowledge, and scalable delivery with most conversational turns handled by the agent.

Key findings

  • Reach was 49.5% (49/99 recruited and eligible patient-family member teams started interacting with MAX); intervention completion was 75.5% (37/49).
  • Patient-MAX working alliance was high across coaching sessions 2, 8, and 14 (means 6.34, 6.14, and 6.34, respectively), and acceptance ratings were above the neutral scale value of 4 (perceived usefulness 6.42, ease of use 6.75, enjoyment 6.47, control 5.53, intention to continue 5.58).
  • Asthma knowledge increased significantly from pretest to posttest (7.73 to 8.79 with baseline observation carried forward, n=48, P<.001, d=0.91; 9.43 in complete cases, n=37, P<.001, d=1.19).
  • Family members supported patients in 269/275 (97.8%) coaching sessions; 99.5% (15,078/15,152) of conversational turns were between patients and MAX versus 0.5% (74/15,152) between patients and health care professionals. Health care professionals took an average of 221 seconds to assess inhalation videos and 2.34 days to provide feedback.

Methods

  • Single-arm feasibility study conducted in two home care settings and four secondary care settings in the German-speaking part of Switzerland; recruitment occurred January-April 2019.
  • Participants were 10- to 15-year-old German-speaking patients with asthma who had smartphone access and a German-speaking family member able to receive SMS messages and support the patient every second intervention day; 99 patients were approached and 49 started interacting with MAX.
  • Outcomes included reach, patient-conversational agent working alliance, acceptance, intervention completion, cognitive and behavioral skills, and human effort/responsiveness; analyses included descriptive statistics, paired-sample t tests for asthma knowledge, and qualitative feedback.

Limitations

  • The paper states that this was a feasibility study with a limited sample, so results are not representative and must be interpreted with caution; only health care professionals from four hospitals and two cantonal patient organizations participated, leaving unclear whether MAX would work similarly with general practitioners.
  • The paper states that qualitative open coding was performed by one author, social support was self-reported and incentivized and thus likely overestimated, and the MAX interface was not integrated into hospital or patient-organization information systems, likely overestimating efforts; only costs and efforts were reported, so no cost-effectiveness implications can be drawn.
  • Evident from the text: the single-arm design without a control group limits causal conclusions about intervention effects, and the small, exploratory sample constrains generalizability.

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

Integrity signals

Registry

Crossref · 10.2196/25060

Cites

102 of 102 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

0 of 97 cited sources checked · 0 supported · 0 flagged

Show all 47 claims

  • The quote only partly states this claim. 2
  • Probably not a substantial claim. 32
  • It may state two separate claims as one. 29
55 of 55

  • The paper relies on this without support.

    Children often face problems with technological solutions when they are not integrated into existing health care systems and do not allow for explicit support by health care professionals or family members.

    “However, and in addition to shortcomings of asthma management related to cognitive and behavioral skills, children often face problems with such technological solutions when they are not integrated into existing health care systems and do not allow for explicit support by health care professionals or family members.”

  • The paper relies on this without support.

    Without a dedicated party or mediator, it often becomes a challenge to integrate all relevant stakeholders (health care professionals, family members, and patients) into the disease management process.

    “Without a dedicated party or mediator, it often becomes a challenge to integrate all of these relevant stakeholders (ie, health care professionals, family members, and the patients themselves) into the disease management process.”

  • The paper relies on this without support.

    Conversational agents can act as mediating social actors, taking over a significant amount of intervention delivery in a scalable manner and coordinating communication among health care professionals, family members, and patients if required.

    “Conversational agents, as part of such interventions, can act as mediating social actors; that is, they take over not only a significant amount of the intervention delivery in a scalable manner but also coordinate the communication among health care professionals, family members, and patients if required.”

  • The authors state this as an assumption.

    The four distinct intervention components enabled an experiential learning cycle and were assumed to influence the outcomes of the intervention.

    “Four distinct intervention components as depicted in Figure 1 enabled an experiential learning cycle [66], and were assumed to influence the outcomes of the intervention”

  • The authors state this as an assumption.

    The need to self-regulate one’s experiences and actions (as an important predictor of engagement) is also true for digital interventions.

    “Assuming that the need to self-regulate one’s experiences and actions (as an important predictor of engagement [70], as posited by self-determination theory) is also true for digital interventions”

  • The authors state this as an assumption.

    There is a positive/negative reinforcement link that connects the outcomes of the conceptual model with the perceived characteristics of the app and working alliance with the conversational agent.

    “Finally, we assume that there is a positive/negative reinforcement link that connects the outcomes of the conceptual model with the perceived characteristics of the app and working alliance with the conversational agent.”

  • The paper relies on this without support.

    Specific cognitive skills required for asthma management include knowledge about asthma triggers and the importance of medication inhalation adherence, as well as behavioral skills such as the application of correct inhalation techniques.

    “For asthma, specific cognitive skills required for disease management include knowledge about asthma triggers and the importance of medication inhalation adherence, as well as behavioral skills such as the application of correct inhalation techniques.”

  • The paper relies on this without support.

    The predefined evaluation criteria used by health care professionals validly assess inhalation technique.

    “These criteria were developed by health care professionals of the Swiss Lung Association and the participating pediatric pneumologists.”

  • The paper relies on this without support.

    The health care professionals' assessments of inhalation technique are reliable.

    “The inhalation technique of each patient (ie, behavioral skill) was systematically assessed by the patient’s responsible health care professional with the help of predefined evaluation criteria”

  • The paper relies on this without support.

    The single-item measures used to assess technology acceptance are sufficient to validly measure the constructs.

    “Single-item measures were used to reduce the burden of the intervention and answers were anchored on 7-point Likert scales ranging from “strongly disagree” (1) to “strongly agree” (7).”

  • The paper relies on this without support.

    The observed improvements in cognitive and behavioral skills are attributable to the MAX intervention rather than other factors.

    “We have shown that conversational agents framed as digital assistants of health care professionals have the potential to improve cognitive and behavioral skills in chronic disease management, with asthma in children as one specific example.”

  • The paper relies on this without support.

    The MAX intervention can be delivered in a scalable way into the everyday life of patients and their family members.

    “We have demonstrated that conversational agents can take over the role of a mediating social actor in a complex health care setting with various stakeholders, and deliver a digital health intervention in a scalable way into the everyday life of patients and their family members.”

  • The paper relies on this without support.

    The MAX project team's cost estimates are accurate.

    “In case none of these additional intervention components or studies is considered, estimates of the MAX project team indicate that the development of the current MAX intervention into a “product” would cost another US $100,000.”
13 of 13

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?

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.

  • The same participant count is assigned two different percentages

    Minor

    Table 1 labels 12 nonfinishers as 25% in one row and 24% in another, although both rows use the same total of 49 participants.

  • Working-alliance response scale has conflicting endpoints

    Minor

    A scale with integer responses from 1 through 8 has eight points, not seven.

2 of 2

Citations

Cites

Flagged first, then by citations

Show all 102 references

Research works printed but not linked: 9

  1. 10.1043/1543-2165(2006)130[447:apap]2.0.co;2 · DOI
  2. Collins, LM, 2018, Optimization of Behavioral, Biobehavioral, and Biomedical Interventions
  3. Kolb, DA, 1984, Experience as the source of learning and development
  4. BoatengGSelsLKuppensPLüscherJScholzUKowatschTEmotion elicitation and capture among real couples in the lab2020ACM CHI Conference on Human Factors in Computing Systems2020Honolulu (cancelled)New York, NYAssociation for Computing Machinery
  5. NielsenJLandauerTKModel of the finding of usability problems1993ACM Conference on Human Factors in Computing Systems (CHI)1993Netherlands206213
  6. Schnyder, M, 2014, Begleitbericht Zur Entwicklung eines Lernmittels für Asthmakinder im Alter von 7 bis 12 Jahren
  7. 10.1145/3290607.3312853 · DOI · Paper page
  8. 10.2307/25148660 · DOI · Paper page
  9. HeilbronFCEscorciaVGhanemBNieblesJCActivityNet: A large-scale video benchmark for human activity understanding20152015 IEEE Conference on Computer Vision and Pattern RecognitionJune 7-12, 2015Boston, MA961970

Cited by

Sorted by citations

Items 101–103 of 103

Related

See all

Metadata

Authors
Kowatsch, T., Schachner, T., Harperink, S., Barata, F., Dittler, U., Xiao, G., Stanger, C., v Wangenheim, F., Fleisch, E., Oswald, H., & Möller, A.
Affiliations
  • University of St.Gallen (CH)
  • ETH Zurich (CH)
  • Furtwangen University (DE)
  • Johns Hopkins University (US)
  • Dartmouth College (US)
  • Kantonsspital Winterthur (CH)
Corresponding author
Kowatsch, Tobias
ORCID
Venue
Journal of Medical Internet Research
Publisher
JMIR Publications Inc.
Published
2021-02-17
Volume
23
Issue
2
Pages
e25060
ISSN
1438-8871 (ISSN-L)
DOI
10.2196/25060
PMC
PMC7929753
OpenAlex
W3122673691
Semantic Scholar
85bb70a2f7f6a60e2b1aba483a3e298c4e2f1dd1
Cited by
103
References
102
Citations by year
2026: 15 · 2025: 11 · 2024: 23 · 2023: 20 · 2022: 29 · 2021: 5
FWCI
21.63
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)
  • Zurich Open Repository and Archive (University of Zurich) (acceptedVersion, cc-by)
  • Repository of Futwangen University of Applied Sciences (Furtwangen University) (submittedVersion, cc-by)
  • Repository for Publications and Research Data (ETH Zurich) (submittedVersion, cc-by)
  • Alexandria (UniSG) (University of St.Gallen) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Zurich Open Repository and Archive (University of Zurich) (submittedVersion, cc-by)
  • ETH Zürich Research Collection (cc-by)
  • Zurich Open Repository and Archive (University of Zurich)
Funders
  • National Institute on Drug Abuse (P30DA029926)
  • National Institute of Diabetes and Digestive and Kidney Diseases (R01DK124428)
Topics
  • Health Sciences › Health Professions › General Health Professions › Mobile Health and mHealth Applications
  • Social Sciences › Psychology › Applied Psychology › Digital Mental Health Interventions
  • Social Sciences › Psychology › Social Psychology › Social Robot Interaction and HRI
Keywords
chatbots, chronic disease management, childhood asthma, asthma self-management, therapeutic relationship, healthcare workers, inhaler technique, text messaging, intervention acceptance, feasibility, primary and secondary care, patient-centered design, family involvement in care, patient knowledge and skills
MeSH terms
Chronic Disease, Communication, Family, Feasibility Studies, Female, Health Personnel, Humans, Male, Patients

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.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

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

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