Last updated July 18, 2026
Key finding
Prebreakfast capillary glucose concentration was 131.1 (28.0) mg/dL in the iNCDSS group and 142.1 (35.8) mg/dL in the physician group.
This study evaluated an AI-based insulin decision support system against standard therapy by senior endocrinologists for managing Type 2 diabetes. The AI system showed comparable glycemic outcomes.
Quick read
The essential study design details in one scan.
EvidenceScore™
Moderate
Study type
RCTs
Follow-up
Short-Term (≤3 mo)
Risk of bias
Some Concerns
Save research, organize studies, and quickly find important evidence again.
Plain-language summary
A plain-language read of the study’s main message and where it applies.
Study focus
This study evaluated an AI-based insulin decision support system against standard therapy by senior endocrinologists for managing Type 2 diabetes. The AI system showed comparable glycemic outcomes.
This research highlights the potential of AI in diabetes management, offering a viable alternative to traditional physician-led insulin titration. As diabetes prevalence continues to rise, integrating AI into clinical practice could enhance patient care, improve glycemic control, and reduce the burden on healthcare professionals. The findings suggest that AI can be a valuable tool in optimizing insulin therapy, potentially leading to better patient outcomes.
The study did not include a long-term follow-up to assess sustained outcomes. Sample size and population characteristics may limit generalizability. Potential biases in participant selection were not addressed.
Published in
Publication details and source links for this paper.
Zhen Y, Yujuan F, Congling C, et al. Can an Artificial Intelligence–Based Insulin Clinical Decision Support System for Type 2 Diabetes Achieve Glycemic Outcomes That Are Not Inferior to Standard Therapy Administered by Senior Endocrinology Physicians?. JAMA Network Open. 2025;8(5):e258910. doi:10.1001/jamanetworkopen.2025.8910
The iNCDSS group achieved a mean target glucose range of 76.4%, compared to 73.6% in the physician group.
Prebreakfast capillary glucose concentration was significantly lower in the iNCDSS group (131.1 mg/dL) than in the physician group (142.1 mg/dL).
The mean daily insulin dose was lower in the iNCDSS group (27.2 units) compared to the physician group (30.2 units).
Evidence network
Understand where this research contributes within the broader evidence network.
This study contributes evidence to Insulin dosage titration by senior endocrinology physicians, Smart insulin dosage software (App group) and Incidence of hypoglycemia, Mean capillary glucose concentration measured before breakfast, Time in range, and 1 more.
This study contributes evidence to
Primary intervention
Insulin dosage titration by senior endocrinology physicians
Primary outcomes
Evidence topics
Primary intervention
Intervention and outcome relationships this study adds to the evidence network.
Editorial context
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.
2
Related topics
8
Evidence pairs
401
Related studies
Evidence topic
Contributes evidence
Evidence topic
Contributes evidence
Save studies and evidence pages, organize your personal Evidence Tracker, and keep the research you care about in one place.
Primary evidence
Evidence relationship
Related evidence
Evidence relationship
Save evidence
Evidence topic
Save evidence
Core evidence
The primary outcomes reported in this study.
Insulin dosage titration by senior endocrinology physicians → Incidence of hypoglycemia
Insulin dosage titration by senior endocrinology physicians → Incidence of hypoglycemia
Insulin dosage titration by senior endocrinology physicians → Mean capillary glucose concentration measured before breakfast
Insulin dosage titration by senior endocrinology physicians → Mean capillary glucose concentration measured before breakfast
Insulin dosage titration by senior endocrinology physicians → Time in range
Insulin dosage titration by senior endocrinology physicians → Time in range
Insulin dosage titration by senior endocrinology physicians → Total daily insulin dose
Insulin dosage titration by senior endocrinology physicians → Total daily insulin dose
Smart insulin dosage software (App group) → Incidence of hypoglycemia
Smart insulin dosage software (App group) → Incidence of hypoglycemia
Smart insulin dosage software (App group) → Mean capillary glucose concentration measured before breakfast
Smart insulin dosage software (App group) → Mean capillary glucose concentration measured before breakfast
Smart insulin dosage software (App group) → Time in range
Smart insulin dosage software (App group) → Time in range
Smart insulin dosage software (App group) → Total daily insulin dose
Smart insulin dosage software (App group) → Total daily insulin dose
Evidence Library
Save research, organize studies, and quickly find important evidence again.
Evidence Library
Save research, organize studies, and quickly find important evidence again.
Explore related studies, evidence collections, and research questions.
Relationships organized using the Dediabetes Evidence Intelligence™ framework.
This study contributes to evidence on Smart Insulin Dosage Software and Incidence of hypoglycemia, Smart Insulin Dosage Software and CGM Time in Range.
This study contributes to the evidence on the following intervention-outcome relationships.
Devices & Technology
Devices & Technology
Curated evidence collections and hubs this study is part of.
All studies measuring CGM Time in Range
Measures CGM Time in Range as a key outcome.
All studies measuring Incidence of hypoglycemia
Measures Incidence of hypoglycemia as a key outcome.
All studies on Smart Insulin Dosage Software
Contributes to Smart Insulin Dosage Software evidence base.
Latest published studies
Published within the last 2 years.
Jump to pre-filtered views in Evidence Explorer.
2 results
2 results
2 results
2 results
2 results
Generated from the study's connected evidence using Evidence Intelligence™.
Smart Insulin Dosage Software may worsen CGM Time in Range or be associated with harm.
ConsistencyScore™: Results are consistent across studies.
Ranked evidence signals
Time in range
EvidenceScore™ Moderate | EvidenceScore™ 69.0 | moderate positive | ConsistencyScore™ Consistent | 1 study
Why this answer: This answer is based on 3 supporting studies and existing graph evidence signals.
Limitations
Smart Insulin Dosage Software appears to improve Insulin Resistance.
ConsistencyScore™: Consistency cannot yet be determined from the available evidence.
Ranked evidence signals
Total daily insulin dose
EvidenceScore™ Emerging | EvidenceScore™ 59.0 | strong positive | ConsistencyScore™ Unclear | 1 study
Why this answer: This answer is based on a single supporting study.
Limitations
Current evidence does not show a clear benefit of Smart insulin dosage software (App group) for Incidence of hypoglycemia.
ConsistencyScore™: Results are consistent across studies.
Ranked evidence signals
Incidence of hypoglycemia
EvidenceScore™ Moderate | EvidenceScore™ 69.0 | neutral | ConsistencyScore™ Consistent | 1 study
Why this answer: This answer is based on a single supporting study.
Limitations
Insulin dosage titration by senior endocrinology physicians appears to improve Time in range.
ConsistencyScore™: Consistency cannot yet be determined from the available evidence.
Ranked evidence signals
Time in range
EvidenceScore™ Emerging | EvidenceScore™ 59.0 | strong positive | ConsistencyScore™ Unclear | 1 study
Why this answer: This answer is based on a single supporting study.
Limitations
Next steps
Choose a next path through related evidence topics, Evidence Explorer views, and research summaries.
Follow the topics this study contributes to.
Open broader Evidence Explorer views for this relationship.
Read related research summaries.
Focused on evidence, not advertising.
Your data is always protected.
New studies added every day.
