Multi-Agent ADR Prediction with Human–Machine Decision Comparison
Official title: Multi-Agent Collaborative Framework for Adverse Drug Reaction Prediction: Evidence-Based Verification and Human-Machine Decision Comparative Study
Run by Peking University Third Hospital · NCT07808866 · Data updated 2026-11-12
A real-world data study tests a collaborative AI model to forecast adverse drug reactions and compares its outputs with physicians to see if early warning signals can be improved.
Quick facts
| Study type | Observational |
|---|---|
| Sex | All |
| Enrollment | 20000 (estimated) |
| Sponsor | Peking University Third Hospital Academic / other |
| Locations | 2 sites (Beijing, Beijing Municipality and 1 other locations) |
| Status | Recruiting |
| Started | June 19, 2026 |
| Main results expected | June 19, 2027 |
| Study ends | December 19, 2027 |
| Registry record updated | September 9, 2026 |
| Trial ID | NCT07808866 on ClinicalTrials.gov |
Is this trial a fit for you?
✓Likely a good fit
Ideal candidates are individuals whose records include definite ADR events with complete documentation across the specified drug categories.
✕Probably not a fit
Patients with incomplete records or ADRs not clearly documented may not benefit from the model.
A plain-English reading of the study's rules. Only the study team can confirm whether you qualify.
What this trial studies
This observational study builds a multi-agent predictive model using de-identified medical records to forecast adverse drug reactions. It first analyzes retrospective desensitization cases to train the model, then prospectively enrolls patients to evaluate performance and compare AI predictions to expert assessments in a blinded manner. No new clinical interventions are performed, and participation carries no additional risk beyond existing data use. The study focuses on drugs across several categories, including anti-infectives, cardiovascular, anti-tumor, CNS, and digestive agents. Outcomes center on prediction accuracy and alignment with clinician judgments.
Why it matters
Potential benefit: If successful, the model could enhance early ADR detection and support safer clinical decision-making.
How similar studies have gone: Prior work shows promise for AI-assisted ADR prediction, but this exact multi-agent, human–machine comparison approach is relatively novel.
Who can join: the full rules
Show full inclusion / exclusion criteria
Inclusion Criteria: * Cases shall involve drug categories including anti-infectives, cardiovascular agents, anti-tumor drugs, central nervous system drugs, digestive system drugs, etc. Each case must contain at least one definite adverse drug reaction (ADR) event, with complete supporting documentation (medical history, medication history, ADR occurrence process, and clinical outcome). Exclusion Criteria: * Cases with incomplete supporting documentation lacking medical history, medication history, ADR occurrence process or clinical outcome. * Cases only with suspected or possible ADRs without definite clinical confirmation.
Where it's running
-
Peking University Third Hospital
Beijing, Beijing Municipality, China Recruiting -
Peking University Third Hospital
Beijing, China Not yet recruiting
Study contacts
- Study coordinator: Weilong Zhang
- Email: zhangwl2012@126.com
- Phone: 010-82266782
Similar studies
Other studies for the same condition or treatment. Recruiting studies are listed first.
-
Adaptive Decision Support for Addiction Treatment: Testing a Care Tool for ED Doctors to Help Buprenorphine Use
We’re testing a refined decision-support tool in emergency departments to help doctors prescribe buprenorphine to adults with moderate to severe opioid use disorder.
-
Ofloxacin blood levels and side effects in older adults with bone and joint infections.
This project will test whether higher blood levels of ofloxacin on day 3 cause more side effects in people over 65 being treated for bone and joint infections.
-
Surgery With Prehabilitation, Direct Surgery, or Rehab for Hip Impingement
This study tests three approaches for femoroacetabular impingement to see which helps hip pain and function best: surgery with prehab, surgery without prehab, or rehabilitation alone.
-
Disposable noninvasive skin closure to help healing and reduce scarring after plastic surgery
This study tests a disposable noninvasive skin closure device on one side of a plastic surgery incision to see if it helps wounds heal better and results in less scarring.
-
Systemic Sclerosis Patient Cohort: Longitudinal Observation at Peking University Third Hospital
This observational study follows people with systemic sclerosis to understand how the disease progresses and to identify markers and potential targets for treatment.
-
Motor imagery–enhanced rehab after ACL reconstruction
This study tests whether mentally practicing movement can improve knee function and brain activity after ACL reconstruction, in comparison with standard rehab.
How to take part
- Review the eligibility criteria above with your treating physician.
- Check the official listing on ClinicalTrials.gov for the latest contact details and recruitment status.
- Contact the study coordinator or principal investigator to ask for pre-screening. Pre-screening is free and never obligates you to enroll.