Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 36 of the 36 references it cites were found in a registry or an index; 17 of the 43 sources its claims cite were checked for support.

My Diabetes Coach, a Mobile App–Based Interactive Conversational Agent to Support Type 2 Diabetes Self-Management: Randomized Effectiveness-Implementation Trial

Open access

Published version of a preprint

Abstract

Background: Delivering self-management support to people with type 2 diabetes mellitus is essential to reduce the health system burden and to empower people with the skills, knowledge, and confidence needed to take an active role in managing their own health. Objective: This study aims to evaluate the adoption, use, and effectiveness of the My Diabetes Coach (MDC) program, an app-based interactive embodied conversational agent, Laura, designed to support diabetes self-management in the home setting over 12 months. Methods: This randomized controlled trial evaluated both the implementation and effectiveness of the MDC program. Adults with type 2 diabetes in Australia were recruited and randomized to the intervention arm (MDC) or the control arm (usual care). Program use was tracked over 12 months. Coprimary outcomes included changes in glycated hemoglobin (HbA1c) and health-related quality of life (HRQoL). Data were assessed at baseline and at 6 and 12 months, and analyzed using linear mixed-effects regression models. Results: A total of 187 adults with type 2 diabetes (mean 57 years, SD 10 years; 41.7% women) were recruited and randomly allocated to the intervention (n=93) and control (n=94) arms. MDC program users (92/93 participants) completed 1942 chats with Laura, averaging 243 min (SD 212) per person over 12 months. Compared with baseline, the mean estimated HbA1c decreased in both arms at 12 months (intervention: 0.33% and control: 0.20%), but the net differences between the two arms in change of HbA1c (−0.04%, 95% CI −0.45 to 0.36; P=.83) was not statistically significant. At 12 months, HRQoL utility scores improved in the intervention arm, compared with the control arm (between-arm difference: 0.04, 95% CI 0.00 to 0.07; P=.04). Conclusions: The MDC program was successfully adopted and used by individuals with type 2 diabetes and significantly improved the users’ HRQoL. These findings suggest the potential for wider implementation of technology-enabled conversation-based programs for supporting diabetes self-management. Future studies should focus on strategies to maintain program usage and HbA1c improvement. Trial Registration: Australia New Zealand Clinical Trials Registry (ACTRN) 12614001229662; https://anzctr.org.au/Trial/Registration/TrialReview.aspx?ACTRN=12614001229662

Overview

CoChat Summary

TL;DR

In a 12-month Australian randomized trial of 187 adults with type 2 diabetes, the My Diabetes Coach app-based conversational agent was widely adopted and significantly improved health-related quality of life versus usual care, but did not significantly improve HbA1c. Program use declined over time, and the HbA1c between-arm difference was not statistically significant.

Key findings

  • Among 187 randomized adults (93 intervention, 94 control), 92/93 intervention participants completed at least one chat with Laura; they completed 1942 chats total, averaging 243 min (SD 212) per person over 12 months, and 83/93 uploaded glucose data.
  • HRQoL (AQoL-8D utility) improved significantly in the intervention arm versus control at 12 months: between-arm difference 0.04 (95% CI 0.00-0.07; P=.04); intervention baseline-to-12-month change 0.04 (95% CI 0.01-0.06; P=.007); per-protocol between-arm difference 0.06 (95% CI 0.02-0.09; P=.003).
  • HbA1c decreased in both arms at 12 months (intervention −0.33%, control −0.28%), but the between-arm difference was not significant (−0.04%, 95% CI −0.45 to 0.36; P=.83).
  • A dose-response relationship was observed for HRQoL: participants completing >24 chats had greater AQoL-8D improvement than those completing <7 chats (between-arm difference 0.09, 95% CI 0.02-0.17; P=.02), with no significant HbA1c difference (−0.33%, 95% CI −1.11 to 0.44; P=.40).

Methods

  • Two-arm, open-label randomized controlled trial in Australia; adults aged ≥18 years with type 2 diabetes, registered with the National Diabetes Service Scheme <10 years, with basic English and an internet-enabled smart device; recruitment June 2016–April 2017; 2x4 block randomization via REDCap to intervention or usual care.
  • Intervention: My Diabetes Coach app with embodied conversational agent Laura, modules on glucose monitoring, eating, activity, medication, and foot care, plus printed guide, website, optional Bluetooth glucose meter, and coordinator calls; control received usual care and a quarterly newsletter.
  • Coprimary outcomes were HbA1c and HRQoL (AQoL-8D) at baseline, 6, and 12 months; linear mixed-effects regression models were used for intention-to-treat, adjusted, and per-protocol analyses in Stata 15.1.

Limitations

  • Stated in the paper: participants and their GPs could not be blinded due to the intervention’s nature, so self-report bias and Hawthorne effects may exist; completed assessments were higher in the control arm than the intervention arm; subgroup analyses were underpowered and multiple testing increased false-positive risk.
  • Evident from the text: baseline HbA1c was not an inclusion criterion and 52.4% of participants were already in the recommended target range (<7% or 53 mmol/mol), likely limiting detectable HbA1c benefit.
  • Evident from the text: program use declined over 12 months (monthly chat users fell from 87% to 15%; glucose recorders from 78% to 23%), and the control arm also reduced HbA1c, which the authors suggest may reflect Hawthorne effects or other self-management.

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

Integrity signals

Registry

Crossref · 10.2196/20322

Cites

36 of 36 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Enrollment, randomization, and follow-up of study participants.
Figure 1 Enrollment, randomization, and follow-up of study participants.CC BY · PubMed Central

Verification

Claims the paper makes

8 Valid · 2 Not valid · 13 Not checked · 80 Own results and interpretations

48 more claims

Create a free account to see all 103 claims and to verify the 26 references still open.

See all claims — free account
  • Probably not a substantial claim. 25
  • It may state two separate claims as one. 25
  • The quote only partly states this claim. 4

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?

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

Research works printed but not linked: 1

  1. Crosby, R, 2013, Health Behavior Theory for Public Health: Principles, Foundations, and Applications, 163

Cited by

See all 183

Metadata

Authors
Gong, E., Baptista, S., Russell, A., Scuffham, P., Riddell, M., Speight, J., Bird, D., Williams, E., Lotfaliany, M., & Oldenburg, B.
Affiliations
  • The University of Melbourne (AU)
  • The University of Queensland (AU)
  • Griffith University (AU)
  • UNSW Sydney (AU)
  • Deakin University (AU)
  • University of Surrey (GB)
Corresponding author
Gong, Enying
Venue
Journal of Medical Internet Research
Published
2020-11-05
Volume
22
Issue
11
Pages
e20322
ISSN
1438-8871 (ISSN-L)
OpenAlex
W3047309370
Cited by
159
References
36
Citations by year
2026: 32 · 2025: 36 · 2024: 29 · 2023: 26 · 2022: 22 · 2021: 14
FWCI
16.18
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)
  • Minerva Access (University of Melbourne) (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Figshare (submittedVersion, cc-by)
  • Griffith Research Online (submittedVersion, other-oa)
Funders
  • Bupa Foundation
  • Diabetes Australia
  • Deakin University
  • Roche Diabetes Care
  • Diabetes Victoria
  • Medical Research Council
  • National Health and Medical Research Council (1170937)
Topics
  • Health Sciences › Health Professions › General Health Professions › Mobile Health and mHealth Applications
  • Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Diabetes Management and Education
  • Social Sciences › Psychology › Applied Psychology › Digital Mental Health Interventions
Keywords
type 2 diabetes, diabetes self-management, mobile health apps, chatbots, HbA1c, health-related quality of life, implementation effectiveness, digital health interventions, health behavior change
MeSH terms
Aged, Aged, 80 and over, Communication, Diabetes Mellitus, Type 2, Female, Glycated Hemoglobin, Humans, Male, Middle Aged, Mobile Applications

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 2026-10-05, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-05 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

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

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.