Diabetes Mellitus, Type 2NCT07888855
Impact of AI-Based Breath Ammonia Kidney Function Results on Clinical Pharmacist Medication Recommendations
An observational study of Diabetes Mellitus, Type 2 and Diabetic Nephropathies, run by Shanghai Pudong Hospital at 1 site in China.
Trial summary
This retrospective study evaluates whether providing clinicians with artificial intelligence (AI)-based kidney function results from a breath ammonia test can improve the quality of medication recommendations made by clinical pharmacists. Read more about this study
- Who is it for?
- Ages 18 years and older
- Conditions the study is for
- Diabetes Mellitus, Type 2, Diabetic Nephropathies
- How many people
- 380 participants
- Where
- 1 siteChina
- Study dates
- 9 Oct 2026to 31 Aug 2028
- Phase
- Observational
Key requirements
Who can take part, as listed by the sponsor. Only the study team can confirm whether you are eligible, so take this list to your doctor.
- Ages 18 years and older
- All sexes
- No healthy volunteers
- Adult, Older adult
Study population: The study population consists of historical inpatient cases who previously underwent breath ammonia testing and received AI-based kidney function classification results at Shanghai Pudong Hospital. Eligible cases had complete standardized clinical records from the breath test hospitalization, with serum creatinine and eGFR results available within 7 days of the breath test serving as an independent reference standard. These de-identified cases represent patients with diabetes and/or kidney disease in real-world inpatient settings, used to evaluate the impact of AI-assisted breath ammonia information on clinical pharmacist medication recommendations.
Who may be able to join
- Historical cases with completed breath test and AI kidney function classification results.
- Complete standardized case records from the breath test hospitalization.
- Serum creatinine and eGFR results available within 7 days before or after the breath test, serving as an independent reference standard.
- Case data sufficient to support the pharmacist in making medication recommendations.
Who cannot join
- Missing standardized case data preventing independent medication evaluation.
- Breath test and reference blood test interval exceeding 7 days.
- Case in acute kidney injury (AKI) state.
- Cases that cannot maintain information isolation between Condition A and Condition B.
Trial locations and contacts
1 site in 1 country. Contact the study team or a site near you to ask about taking part.
China1 site
Shanghai Pudong Hospital
Shanghai
Jingru Dr. Gong, 800-18121329275, 50074@shpdhoa.org
About this study
Official title: A Real World Study of Remote Continuous Noninvasive Monitoring and Diagnosis of Diabetic Kidney Disease Using a Two Dimensional Heterostructure Based Wearable NH₃ Sensor
This retrospective study evaluates whether providing clinicians with artificial intelligence (AI)-based kidney function results from a breath ammonia test can improve the quality of medication recommendations made by clinical pharmacists. Using historical inpatient cases from Shanghai Pudong Hospital, each case was evaluated under two information conditions: one with standard clinical information only, and one with the additional AI-based breath test kidney function results. Independent experts blindly scored the medication recommendations. The study aims to determine whether the breath test AI information adds meaningful clinical value to pharmacist decision-making.
- Conditions
- Diabetes Mellitus, Type 2, Diabetic Nephropathies
- Keywords
- clinical pharmacist, kidney function assessment, breath ammonia, medication recommendation
- Medical subject headings
- Diabetes Mellitus, Type 2, Diabetic Nephropathies
What participants receive
What participants receive. In a randomised study each person is assigned to one group by chance.
| Study group | Role | What they receive |
|---|---|---|
| Condition A - Standard Information | De-identified standardized case summaries provided to clinical pharmacists without real serum creatinine and eGFR results and without breath test AI output. Approximately 380 cases. | |
| Condition B - Breath Test AI-Assisted Information | Identical standardized case summaries as Condition A, with the addition of breath test AI kidney function results (estimated renal function, CKD stage probability, and confidence), while real serum creatinine and eGFR remain blinded. Approximately 380 cases. |
What is being measured
The primary outcome is the main question the study is designed to answer.
Decision Score Difference Between Information Conditions
Adjusted mean difference in the composite medication recommendation score (0-40 scale) between Condition A and Condition B for the same cases, with 95% confidence interval.
Measured: up to 4 weeks
Secondary outcomes (5)
Proportion of Correct Decisions
Proportion of cases where the medication recommendation is rated correct and reasonable, compared between Condition A and Condition B.
Measured: up to 4 weeks
Recommendation Consistency Rate
Proportion of pharmacist-recommended treatment plans consistent with gold standard and expert consensus.
Measured: up to 4 weeks
Unreasonable Recommendation Rate
Proportion of cases containing principle errors (e.g., recommending nephrotoxic drugs despite declined renal function, dosing errors).
Measured: up to 4 weeks
Decision Confidence
Difference in self-reported decision confidence scores (0-10 scale) between Condition A and Condition B.
Measured: up to 4 weeks
Decision Time
Difference in time taken to complete medication recommendations between Condition A and Condition B.
Measured: up to 4 weeks
Timeline
From the start of the study to its completion; the mark is today. Estimated dates are the sponsor’s plan.
- First posted22 Oct 2026
- Study start9 Oct 2026 (estimated)
- Primary completion30 Oct 2026 (estimated)
- Study completion31 Aug 2028 (estimated)
Resources
- Official record on ClinicalTrials.govThe registry entry this page is built from, with its full change history.
- Other trials for Diabetes Mellitus, Type 2Studies of the same condition, open and under way.
- What the trial phases meanHow a treatment moves from first tests in people to approval.
More details
For doctors and researchers: how the study is designed, who runs it, and its registry IDs.
Study design
- Study type
- Observational
- Observational model
- Cohort
- Time perspective
- Retrospective
- Enrolment
- 380 participants (estimated)
Sponsor and investigators
- Lead sponsor
- Shanghai Pudong Hospital
- Sponsor type
- University, hospital or other
- Collaborators
- Sichuan Jishengtang Pharmaceutical Co., Ltd.
- Responsible party
- Principal investigator, Jingru Gong, Shanghai Pudong Hospital
IDs, dates and publications
- ClinicalTrials.gov ID
- NCT07888855
- Sponsor's protocol ID
- 2025-IIT-027-2
- Other IDs
- 2025-IIT-027-E02
- Status last verified
- Sep 2026
- First submitted
- 30 Sep 2026
- Last updated
- 22 Oct 2026
- Results posted
- No
- Data monitoring committee
- No
- FDA-regulated drug
- No
- FDA-regulated device
- No
- Individual data sharing
- Yes: De-identified standardized case summaries and AI-assisted decision outcome data will be made available for research purposes, consistent with the study's data protection and confidentiality requirements.