Evidence Intelligence™
- EvidenceScore™
- Emerging
- Score 59 · Based on 1 study
- ImpactScore™
- 50
- Neutral
- ConsistencyScore™
- unclear
- Not enough independent studies
Supporting studies: Based on 1 study
Add to Evidence TrackerDigital Health ToolsPharmacological TreatmentsPrecision Medicine
Research Summary
Analyzed using Evidence Intelligence™
AI virtual trials match real trial results for GLP-1 vs insulin in type 2 diabetes
Last updated May 16, 2026
Key finding
AI models can recreate trial results, but need more validation before use in treatment decisions.
Researchers used AI to create virtual patients and test whether they could recreate results from a real diabetes trial. They trained the AI using data from over 5,000 people with type 2 diabetes. The AI successfully matched the original trial's findings that GLP-1 drugs work better than basal insulin for blood sugar control, weight loss, and blood pressure. The AI model correctly showed GLP-1 was more effective across all three outcomes tested. When researchers used the AI to test treatments in broader groups of people beyond the original trial, they found GLP-1 wasn't always the best choice for everyone. This suggests treatment benefits may vary depending on individual characteristics.
Quick read
Study at a glance
The essential study design details in one scan.
EvidenceScore™
Moderate
Study type
non-randomized clinical trial (non-RCT or NRCT)
Follow-up
Medium-Term (3–12 mo)
Risk of bias
Some Concerns
Save research, organize studies, and quickly find important evidence again.
Plain-language summary
What this paper says
A plain-language read of the study’s main message and where it applies.
Study focus
AI models can recreate trial results, but need more validation before use in treatment decisions.
Published in
Journal Reference
Publication details and source links for this paper.
MacLellan CR, Petkov H, McKeag C, et al. Emulating real-world GLP-1 efficacy in type 2 diabetes through causal learning and virtual patients. PLOS Digit Health. 2025;4(7):e0000927. doi:10.1371/journal.pdig.0000927
Main Effects
HbA1c → ↓ (GLP-1 better than insulin)
Weight (BMI) → ↓ (GLP-1 better than insulin)
Blood pressure → ↓ (GLP-1 better than insulin)
Evidence network
How this study fits
Understand where this research contributes within the broader evidence network.
Evidence Context
This study contributes evidence to Basal insulin, GLP-1 receptor agonists and Body weight, HbA1c, Systolic blood pressure.
This study contributes evidence to
Primary intervention
Basal insulin
Primary outcomes
- Body weight
- HbA1c
- Systolic blood pressure
Primary intervention
Primary outcomes
Evidence relationships
Intervention and outcome relationships this study adds to the evidence network.
6
Evidence pairs
6
Relationships
5
Evidence topics
contributes_evidence
Editorial context
Why this study matters
See why this paper is useful beyond its individual results.
Evidence network role
This section describes how the study fits into the current evidence network. It does not determine whether an intervention works on its own.
Moderate contributionModerate confidenceNetwork score: 54
5
Related topics
6
Evidence pairs
524
Related studies
High relevance in at least one topic
Why it is useful
- Contributes to 6 evidence relationships
- Uses a randomized study design signal
- Linked to 5 direct semantic evidence topics
Topic contributions
Evidence topic
Contributes evidence
Evidence topic
Contributes evidence
Evidence topic
Contributes evidence
Evidence topic
Contributes evidence
Add related evidence to your Evidence Tracker
Save studies and evidence pages, organize your personal Evidence Tracker, and keep the research you care about in one place.
Primary evidence
Evidence relationship
GLP-1 Receptor Agonists and Body Weight
Related evidence
Evidence relationship
GLP-1 Receptor Agonists and HbA1c
Save evidence
Evidence relationship
Insulin Therapies and HbA1c
Save evidence
Core evidence
Study findings
The primary outcomes reported in this study.
Evidence Intelligence™
- EvidenceScore™
- Emerging
- Score 59 · Based on 1 study
- ImpactScore™
- 50
- Neutral
- ConsistencyScore™
- unclear
- Not enough independent studies
Supporting studies: Based on 1 study
Add to Evidence TrackerSystolic blood pressure
Basal insulin → Systolic blood pressure
Basal insulin → Systolic blood pressure
Evidence Intelligence™
- EvidenceScore™
- Emerging
- Score 59 · Based on 1 study
- ImpactScore™
- 55
- Slightly Positive
- ConsistencyScore™
- unclear
- Not enough independent studies
Supporting studies: Based on 1 study
Add to Evidence TrackerEvidence Intelligence™
- EvidenceScore™
- Emerging
- Score 59 · Based on 1 study
- ImpactScore™
- 100
- Very Positive
- ConsistencyScore™
- unclear
- Not enough independent studies
Supporting studies: Based on 1 study
Add to Evidence TrackerEvidence Intelligence™
- EvidenceScore™
- Emerging
- Score 59 · Based on 1 study
- ImpactScore™
- 100
- Very Positive
- ConsistencyScore™
- unclear
- Not enough independent studies
Supporting studies: Based on 1 study
Add to Evidence TrackerSystolic blood pressure
GLP-1 receptor agonists → Systolic blood pressure
GLP-1 receptor agonists → Systolic blood pressure
Evidence Intelligence™
- EvidenceScore™
- Emerging
- Score 59 · Based on 1 study
- ImpactScore™
- 100
- Very Positive
- ConsistencyScore™
- unclear
- Not enough independent studies
Supporting studies: Based on 1 study
Add to Evidence TrackerEvidence Library
Build your evidence library
Save research, organize studies, and quickly find important evidence again.
Evidence Suggest
- AI models can recreate results from real diabetes trials, correctly showing GLP-1 drugs work better than insulin for blood sugar, weight, and blood pressure.
- Treatment benefits may vary by individual, with GLP-1 not always the best choice for everyone.
- This AI method needs more testing before it can be used widely for treatment decisions.
Who this applies to
People with type 2 diabetes on metformin and sulfonylurea considering GLP-1 or insulin treatment; healthcare providers choosing between treatment options; researchers testing new AI methods for drug evaluation.
Keep in Mind
This tests an AI method, not actual treatments, so it doesn't provide new clinical evidence.
Between the Lines
- Tested drug classes, not specific medications
- Virtual patients showed less variation than real people
- Only validated against one trial
- Excluded records with missing data
Connected Evidence
Explore related studies, evidence collections, and research questions.
Relationships organized using the Dediabetes Evidence Intelligence™ framework.
This study contributes to evidence on Insulin Therapies and Body Weight, Insulin Therapies and HbA1c.
Related evidence relationships
Explore in Evidence ExplorerThis study contributes to the evidence on the following intervention-outcome relationships.
Medications
Insulin Therapies → Body Weight
Medications
Evidence Intelligence™
- EvidenceScore™
- Emerging
- Score 59 · Based on 1 study
- ImpactScore™
- 50
- Neutral
- ConsistencyScore™
- unclear
- Not enough independent studies
Supporting studies: Based on 1 study
View evidenceMedications
Insulin Therapies → HbA1c
Medications
Evidence Intelligence™
- EvidenceScore™
- Emerging
- Score 59 · Based on 1 study
- ImpactScore™
- 50
- Neutral
- ConsistencyScore™
- unclear
- Not enough independent studies
Supporting studies: Based on 1 study
View evidenceMedications
Insulin Therapies → Blood Pressure
Medications
Evidence Intelligence™
- EvidenceScore™
- Emerging
- Score 59 · Based on 1 study
- ImpactScore™
- 55
- Slightly Positive
- ConsistencyScore™
- unclear
- Not enough independent studies
Supporting studies: Based on 1 study
View evidenceIncluded in these evidence collections
Curated evidence collections and hubs this study is part of.
Blood Pressure Evidence Hub
All studies measuring Blood Pressure
1 studies
outcome_hubMeasures Blood Pressure as a key outcome.
Body Weight Evidence Hub
All studies measuring Body Weight
1 studies
outcome_hubMeasures Body Weight as a key outcome.
HbA1c Evidence Hub
All studies measuring HbA1c
1 studies
outcome_hubMeasures HbA1c as a key outcome.
Insulin Therapies Evidence Hub
All studies on Insulin Therapies
1 studies
intervention_hubContributes to Insulin Therapies evidence base.
Explore more in Evidence Explorer
Jump to pre-filtered views in Evidence Explorer.
All studies on Basal insulin and Body weight
Intervention: Basal insulinOutcome: Body weight
1 results
All studies on Basal insulin and HbA1c
Intervention: Basal insulinOutcome: HbA1c
1 results
All studies on Basal insulin
Intervention: Basal insulin
1 results
All studies measuring Body weight
Outcome: Body weight
1 results
All studies measuring HbA1c
Outcome: HbA1c
1 results
Questions answered by this study
Generated from the study's connected evidence using Evidence Intelligence™.
Does GLP-1 Receptor Agonists affect body weight?
Strong Evidence
GLP-1 Receptor Agonists may improve Body Weight.
ConsistencyScore™: Results are consistent across studies.
Ranked evidence signals
- 1
Body weight
EvidenceScore™ Emerging | EvidenceScore™ 59.0 | strong positive | ConsistencyScore™ Unclear | 1 study
Why this answer: This answer is based on 26 supporting studies with consistent results and a positive effect signal.
Limitations
- Population details are unavailable.
Does GLP-1 Receptor Agonists improve HbA1c?
Strong Evidence
GLP-1 Receptor Agonists appears to improve HbA1c.
ConsistencyScore™: Results are consistent across studies.
Ranked evidence signals
- 1
HbA1c
EvidenceScore™ Emerging | EvidenceScore™ 59.0 | strong positive | ConsistencyScore™ Unclear | 1 study
Why this answer: This answer is based on 27 supporting studies with consistent results and a positive effect signal.
Limitations
- Population details are unavailable.
Does Insulin Therapies improve HbA1c?
Strong Evidence
Insulin Therapies may improve HbA1c.
ConsistencyScore™: Results are generally consistent across studies.
Ranked evidence signals
- 1
HbA1c
EvidenceScore™ Emerging | EvidenceScore™ 59.0 | neutral | ConsistencyScore™ Unclear | 1 study
Why this answer: This answer is based on 7 supporting studies with generally consistent results and a positive effect signal.
Limitations
- Population details are unavailable.
Does GLP-1 Receptor Agonists improve blood pressure?
Strong Evidence
GLP-1 Receptor Agonists may improve Blood Pressure.
ConsistencyScore™: Results are consistent across studies.
Ranked evidence signals
- 1
Systolic blood pressure
EvidenceScore™ Emerging | EvidenceScore™ 59.0 | strong positive | ConsistencyScore™ Unclear | 1 study
Why this answer: This answer is based on 9 supporting studies with consistent results and a positive effect signal.
Limitations
- Population details are unavailable.
Next steps
Continue your research
Choose a next path through related evidence topics, Evidence Explorer views, and research summaries.
Evidence topics
Follow the topics this study contributes to.
Explore in Evidence Explorer
Open broader Evidence Explorer views for this relationship.
Related research
Read related research summaries.
Newer diabetes drugs show stronger weight and blood sugar benefits in prediabetesSMS Program Improves Transition from Insulin to GLP-1 RA in DiabetesAutomated insulin delivery improves glycemic control in type 2 diabetesDietary supplements may lower diabetes risk in prediabetesEmpagliflozin reduces body fluid volume in type 2 diabetes patients
No ads. No tracking.
Focused on evidence, not advertising.
Secure & private
Your data is always protected.
Always up to date
New studies added every day.