CoChat

Assessing the Effectiveness of Digital Health Behavior Strategies on Type 2 Diabetes Management: Systematic Review and Network Meta-Analysis

Open access

Published version of a preprint

The DOI matches its Crossref record; 108 of the 108 references it cites were found in a registry or an index; 46 of the 60 sources its claims cite were checked for support.

Abstract

Background: Various mobile technologies and digital health interventions (DHIs) have been developed for type 2 diabetes mellitus (T2DM) management. Strategies are crucial for ensuring the effectiveness of DHIs. However, there is currently a lack of categorization and summarization of the strategies used in DHIs for T2DM. Objective: This study aims to (1) identify and categorize the strategies used in DHIs for T2DM management; (2) assess the effectiveness of these DHI strategies; and (3) compare and rank the efficacy of different strategy combinations on glycated hemoglobin A1c (HbA1c) levels, fasting blood glucose (FBG) levels, BMI, and weight loss. Methods: Relevant randomized controlled trials (RCTs) were extracted from PubMed, Web of Science, and Scopus databases. Three rounds of screening and selection were conducted. The strategies were identified and categorized based on the principles of behavior change techniques and behavior strategies. The synthesis framework for the assessment of health IT was used to structure the evaluation of the DHI strategies qualitatively. A network meta-analysis was performed to compare the efficacy of different strategy combinations. The data quality was assessed using the Cochrane Risk of Bias tool. Results: A total of 52 RCTs were included, identifying 63 strategies categorized into 19 strategy themes. The most commonly used strategies were guide, monitor, management, and engagement. Most studies reported positive or mixed outcomes for most indicators based on the synthesis framework for the assessment of health IT. Research involving a medium or high number of strategies was found to be more effective than research involving a low number of strategies. Of 52 RCTs, 27 (52%) were included in the network meta-analysis. The strategy combination of communication, engagement, guide, and management was most effective in reducing HbA1c levels (mean difference [MD] –1.04, 95% CI –1.55 to –0.54), while the strategy combination of guide, management, and monitor was effective in reducing FBG levels (MD –0.96, 95% CI –1.86 to –0.06). The strategy combination of communication, engagement, goal setting, management, and support was most effective for BMI (MD –2.30, 95% CI –3.16 to –1.44) and weight management (MD –6.50, 95% CI –8.82 to –4.18). Conclusions: Several DHI strategy combinations were effective in reducing HbA1c levels, FBG levels, BMI, and weight in T2DM management. Health care professionals should be encouraged to apply these promising strategy combinations in DHIs during clinical care. Future research should further explore and optimize the design and implementation of strategies. Trial Registration: PROSPERO CRD42024544629; https://tinyurl.com/3zp2znxt

Overview

CoChat Summary

TL;DR

This systematic review and network meta-analysis identified 63 strategies across 19 themes in digital health interventions for type 2 diabetes and ranked strategy combinations for glycemic and weight outcomes. Combinations including communication, engagement, guide, and management were most effective for HbA1c, while guide, management, and monitor was effective for fasting blood glucose, and communication, engagement, goal setting, management, and support was most effective for BMI and weight loss.

Key findings

  • Across 52 RCTs, 63 strategies were categorized into 19 themes; the most common were monitor (40/52, 77%), guide (35/52, 67%), management (32/52, 61%), and engagement (28/52, 54%).
  • Of 52 RCTs, 27 (52%) were included in the network meta-analysis. DHI-BDGI (communication, engagement, guide, and management) was most effective for reducing HbA1c (MD –1.04, 95% CI –1.55 to –0.54); DHI-GIK (guide, management, and monitor) was effective for reducing FBG (MD –0.96, 95% CI –1.86 to –0.06); DHI-BDFIQ (communication, engagement, goal setting, management, and support) was most effective for BMI (MD –2.30, 95% CI –3.16 to –1.44) and weight loss (MD –6.50, 95% CI –8.82 to –4.18).
  • Medium- or high-strategy studies more often reported ≥10 positive or mixed SF/HIT results than low-strategy studies: high 16/18 (89%), medium 20/23 (87%), low 7/11 (64%).
  • Cost-effectiveness was the main exception: 35/52 (67%) studies either did not report relevant results or reported neutral/negative findings.

Methods

  • Systematic review and network meta-analysis following PRISMA-NMA 2020, registered with PROSPERO (CRD42024544629); searches of PubMed, Web of Science, and Scopus for English-language papers from January 1, 1999, to March 10, 2024.
  • Three rounds of screening selected RCTs of digital health interventions for adults aged ≥18 years with T2DM, compared with no intervention; outcomes included HbA1c, FBG, BMI, and weight loss, plus broader SF/HIT domains.
  • Strategies were identified using behavior change technique and behavior strategy frameworks; quality was assessed with the Cochrane Risk of Bias tool; network meta-analysis used Stata 15.1 with frequentist random-effects models, mean differences with 95% CIs, SUCRA rankings, node-splitting for inconsistency, Egger tests, and sensitivity analyses removing individual studies.

Limitations

  • The paper states that non-English or nonindexed papers may have been overlooked; three rounds of screening and extraction could have introduced inconsistencies; only specific outcomes were meta-analyzed, limiting generalizability; BMI and weight loss had substantial heterogeneity; the number of original studies in meta-analyses was relatively small; and blinding was not feasible in most studies, possibly exaggerating estimated intervention effects and contributing to low-to-moderate evidence quality.
  • Evident from the text: the network meta-analysis included only 27/52 RCTs for HbA1c, 7/52 for FBG, 11/52 for BMI, and 8/52 for weight loss, so several strategy combinations rest on few studies.
  • Evident from the text: heterogeneity was high for BMI (I²=95.8%) and weight loss (I²=85.6%), and cost-effectiveness was mostly not positively reported, which limits conclusions about economic benefit.

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

Integrity signals

Registry

Crossref · 10.2196/63209

Cites

108 of 108 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Flow diagram for the search and selection of the included studies.
Figure 1 Flow diagram for the search and selection of the included studies.CC BY · PubMed Central
Figure 2 Risk of bias graph: review authors’ judgments about each risk of bias item presented as percentages across all included studies.
Figure 2 Risk of bias graph: review authors’ judgments about each risk of bias item presented as percentages across all included studies.CC BY · PubMed Central
Figure 3 Evidence network of outcome indicators: (A) evidence network for glycated hemoglobin A1c, (B) evidence network for fasting blood glucose, (C) evidence network for BMI, and (4) evidence network for weight loss. ABDEFGIKLOPQ: action planning, communication, engagement, feedback, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; ABCDEFGHIJKLNO: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, reward, and shape; ABCDEFGHIJKLMNOQR: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, restructure, reward, shape, support, and tailor; BDFGIKLOPQ: communication, engagement, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; BDFIQ: communication, engagement, goal setting, management, and support; BDGI: communication, engagement, guide, and management; BDGKNOP: communication, engagement, guide, monitor, reward, shape, and stimulate; BGKR: communication, guide, monitor, and tailor; DFGKP: engagement, goal setting, guide, monitor, and stimulate; DGIKPQR: engagement, guide, management, monitor, stimulate, support, and tailor; DHI: digital health intervention; EGIKPR: feedback, guide, management, monitor, stimulate, and tailor; EKP: feedback, monitor, and stimulate; FGIKQ: goal setting, guide, management, monitor, and support; GIK: guide, management, and monitor.
Figure 3 Evidence network of outcome indicators: (A) evidence network for glycated hemoglobin A1c, (B) evidence network for fasting blood glucose, (C) evidence network for BMI, and (4) evidence network for weight loss. ABDEFGIKLOPQ: action planning, communication, engagement, feedback, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; ABCDEFGHIJKLNO: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, reward, and shape; ABCDEFGHIJKLMNOQR: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, restructure, reward, shape, support, and tailor; BDFGIKLOPQ: communication, engagement, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; BDFIQ: communication, engagement, goal setting, management, and support; BDGI: communication, engagement, guide, and management; BDGKNOP: communication, engagement, guide, monitor, reward, shape, and stimulate; BGKR: communication, guide, monitor, and tailor; DFGKP: engagement, goal setting, guide, monitor, and stimulate; DGIKPQR: engagement, guide, management, monitor, stimulate, support, and tailor; DHI: digital health intervention; EGIKPR: feedback, guide, management, monitor, stimulate, and tailor; EKP: feedback, monitor, and stimulate; FGIKQ: goal setting, guide, management, monitor, and support; GIK: guide, management, and monitor.CC BY · PubMed Central
Figure 4 Surface under the cumulative ranking curve rank of outcome indicators: (A) surface under the cumulative ranking curve rank for glycated hemoglobin A1c, (B) surface under the cumulative ranking curve rank for fasting blood glucose, (C) surface under the cumulative ranking curve rank for BMI, and (D) surface under the cumulative ranking curve rank for weight loss. ABDEFGIKLOPQ: action planning, communication, engagement, feedback, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; ABCDEFGHIJKLNO: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, reward, and shape; ABCDEFGHIJKLMNOQR: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, restructure, reward, shape, support, and tailor; BDFGIKLOPQ: communication, engagement, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; BDFIQ: communication, engagement, goal setting, management, and support; BDGI: communication, engagement, guide, and management; BDGKNOP: communication, engagement, guide, monitor, reward, shape, and stimulate; BGKR: communication, guide, monitor, and tailor; DFGKP: engagement, goal setting, guide, monitor, and stimulate; DGIKPQR: engagement, guide, management, monitor, stimulate, support, and tailor; DHI: digital health intervention; EGIKPR: feedback, guide, management, monitor, stimulate, and tailor; EKP: feedback, monitor, and stimulate; FGIKQ: goal setting, guide, management, monitor, and support; GIK: guide, management, and monitor.
Figure 4 Surface under the cumulative ranking curve rank of outcome indicators: (A) surface under the cumulative ranking curve rank for glycated hemoglobin A1c, (B) surface under the cumulative ranking curve rank for fasting blood glucose, (C) surface under the cumulative ranking curve rank for BMI, and (D) surface under the cumulative ranking curve rank for weight loss. ABDEFGIKLOPQ: action planning, communication, engagement, feedback, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; ABCDEFGHIJKLNO: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, reward, and shape; ABCDEFGHIJKLMNOQR: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, restructure, reward, shape, support, and tailor; BDFGIKLOPQ: communication, engagement, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; BDFIQ: communication, engagement, goal setting, management, and support; BDGI: communication, engagement, guide, and management; BDGKNOP: communication, engagement, guide, monitor, reward, shape, and stimulate; BGKR: communication, guide, monitor, and tailor; DFGKP: engagement, goal setting, guide, monitor, and stimulate; DGIKPQR: engagement, guide, management, monitor, stimulate, support, and tailor; DHI: digital health intervention; EGIKPR: feedback, guide, management, monitor, stimulate, and tailor; EKP: feedback, monitor, and stimulate; FGIKQ: goal setting, guide, management, monitor, and support; GIK: guide, management, and monitor.CC BY · PubMed Central
Figure 5 Forest plots of network meta-analysis results: (A) forest plots for glycated hemoglobin A1c, (B) forest plots for fasting blood glucose; (C) forest plots for BMI, and (D) forest plots for weight loss. ABDEFGIKLOPQ: action planning, communication, engagement, feedback, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; ABCDEFGHIJKLNO: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, reward, and shape; ABCDEFGHIJKLMNOQR: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, restructure, reward, shape, support, and tailor; BDFGIKLOPQ: communication, engagement, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; BDFIQ: communication, engagement, goal setting, management, and support; BDGI: communication, engagement, guide, and management; BDGKNOP: communication, engagement, guide, monitor, reward, shape, and stimulate; BGKR: communication, guide, monitor, and tailor; DFGKP: engagement, goal setting, guide, monitor, and stimulate; DGIKPQR: engagement, guide, management, monitor, stimulate, support, and tailor; DHI: digital health intervention; DL: DerSimonian and Laird method; EGIKPR: feedback, guide, management, monitor, stimulate, and tailor; EKP: feedback, monitor, and stimulate; FGIKQ: goal setting, guide, management, monitor, and support; GIK: guide, management, and monitor; IV: inverse variance method; MD: mean difference.
Figure 5 Forest plots of network meta-analysis results: (A) forest plots for glycated hemoglobin A1c, (B) forest plots for fasting blood glucose; (C) forest plots for BMI, and (D) forest plots for weight loss. ABDEFGIKLOPQ: action planning, communication, engagement, feedback, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; ABCDEFGHIJKLNO: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, reward, and shape; ABCDEFGHIJKLMNOQR: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, restructure, reward, shape, support, and tailor; BDFGIKLOPQ: communication, engagement, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; BDFIQ: communication, engagement, goal setting, management, and support; BDGI: communication, engagement, guide, and management; BDGKNOP: communication, engagement, guide, monitor, reward, shape, and stimulate; BGKR: communication, guide, monitor, and tailor; DFGKP: engagement, goal setting, guide, monitor, and stimulate; DGIKPQR: engagement, guide, management, monitor, stimulate, support, and tailor; DHI: digital health intervention; DL: DerSimonian and Laird method; EGIKPR: feedback, guide, management, monitor, stimulate, and tailor; EKP: feedback, monitor, and stimulate; FGIKQ: goal setting, guide, management, monitor, and support; GIK: guide, management, and monitor; IV: inverse variance method; MD: mean difference.CC BY · PubMed Central
Figure 6 Comparison-adjusted funnel plots of trials included in the network meta-analysis: (A) comparison-adjusted funnel plots for glycated hemoglobin A1c, (B) comparison-adjusted funnel plots for fasting blood glucose, (C) comparison-adjusted funnel plots for BMI, and (D) comparison-adjusted funnel plots for weight loss. ABDEFGIKLOPQ: action planning, communication, engagement, feedback, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; ABCDEFGHIJKLNO: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, reward, and shape; ABCDEFGHIJKLMNOQR: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, restructure, reward, shape, support, and tailor; BDFGIKLOPQ: communication, engagement, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; BDFIQ: communication, engagement, goal setting, management, and support; BDGI: communication, engagement, guide, and management; BDGKNOP: communication, engagement, guide, monitor, reward, shape, and stimulate; BGKR: communication, guide, monitor, and tailor; DFGKP: engagement, goal setting, guide, monitor, and stimulate; DGIKPQR: engagement, guide, management, monitor, stimulate, support, and tailor; DHI: digital health intervention; EGIKPR: feedback, guide, management, monitor, stimulate, and tailor; EKP: feedback, monitor, and stimulate; FGIKQ: goal setting, guide, management, monitor, and support; GIK: guide, management, and monitor.
Figure 6 Comparison-adjusted funnel plots of trials included in the network meta-analysis: (A) comparison-adjusted funnel plots for glycated hemoglobin A1c, (B) comparison-adjusted funnel plots for fasting blood glucose, (C) comparison-adjusted funnel plots for BMI, and (D) comparison-adjusted funnel plots for weight loss. ABDEFGIKLOPQ: action planning, communication, engagement, feedback, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; ABCDEFGHIJKLNO: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, reward, and shape; ABCDEFGHIJKLMNOQR: action planning, communication, cues, engagement, feedback, goal setting, guide, identity, management, model or demonstrate, monitor, prompt, restructure, reward, shape, support, and tailor; BDFGIKLOPQ: communication, engagement, goal setting, guide, management, monitor, prompt, shape, stimulate, and support; BDFIQ: communication, engagement, goal setting, management, and support; BDGI: communication, engagement, guide, and management; BDGKNOP: communication, engagement, guide, monitor, reward, shape, and stimulate; BGKR: communication, guide, monitor, and tailor; DFGKP: engagement, goal setting, guide, monitor, and stimulate; DGIKPQR: engagement, guide, management, monitor, stimulate, support, and tailor; DHI: digital health intervention; EGIKPR: feedback, guide, management, monitor, stimulate, and tailor; EKP: feedback, monitor, and stimulate; FGIKQ: goal setting, guide, management, monitor, and support; GIK: guide, management, and monitor.CC BY · PubMed Central
Figure 7 Sensitivity analyses using the method of removing individual studies separately: (A) sensitivity analyses for glycated hemoglobin A1c, (B) sensitivity analyses for fasting blood glucose, (C) sensitivity analyses for BMI, and (D) sensitivity analyses for weight loss.
Figure 7 Sensitivity analyses using the method of removing individual studies separately: (A) sensitivity analyses for glycated hemoglobin A1c, (B) sensitivity analyses for fasting blood glucose, (C) sensitivity analyses for BMI, and (D) sensitivity analyses for weight loss.CC BY · PubMed Central

Verification

Claims the paper makes

22 Valid · 0 Not valid · 2 Not checked · 101 Own results and interpretations

62 more claims

Create a free account to see all 125 claims and to verify the 1 reference still open.

See all claims — free account
  • Probably not a substantial claim. 30
  • It may state two separate claims as one. 35
  • The quote only partly states this claim. 1

Create a free account to see these claims.

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.

4 of 4

Citations

Cites

Show all 108 references

Research works printed but not linked: 2

  1. MaglianoDJBoykoEJIDF Diabetes Atlas 10th Edition Scientific CommitteeIDF diabetes atlas: 10th editionInternational Diabetes Federation20212025-01-24https://diabetesatlas.org/idfawp/resource-files/2021/07/IDF_Atlas_10th_Edition_2021.pdf
  2. Zhou, P, 2014, Int J Clin Exp Pathol, 7, 8848

Cited by

Metadata

Authors
Li, M., Liu, S., Yu, B., Li, N., Lyu, A., Yang, H., He, H., Zhang, N., Ma, J., Sun, M., Du, H., & Gao, R.
Affiliations
  • Xi'an Jiaotong University (CN)
  • Beijing University of Chinese Medicine (CN)
Corresponding author
Gao, Rui
Venue
Journal of Medical Internet Research
Published
2025-02-14
Volume
27
Pages
e63209
ISSN
1438-8871 (ISSN-L)
OpenAlex
W4405462628
Cited by
12
References
108
Citations by year
2026: 4 · 2025: 8
FWCI
1.08
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)
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, digital health interventions, mobile health, behaviour change techniques, behavioral strategies, HbA1c, fasting blood glucose, body mass index, weight loss, strategy combination, patient engagement, goal setting, diabetes self-management education, glycemic control, patient support, communication strategies, monitoring strategies
MeSH terms
Blood Glucose, Diabetes Mellitus, Type 2, Glycated Hemoglobin, Humans, Health Behavior, Randomized Controlled Trials as Topic, Telemedicine

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 2026-10-10, 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).

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

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.