Using an AI tool to prioritize specialty referrals for primary care patients
Official title: Efficacy of an Artificial Intelligence Algorithm for Gatekeeping in Referrals From Primary Care to Specialized Care: a Randomized Controlled Trial
Run by Hospital de Clinicas de Porto Alegre · NCT07019116 · Data updated 2026-11-12
This project will test whether an AI tool can sort and prioritize specialty-care referrals so primary care patients with chronic conditions get more timely and appropriate specialist access.
Quick facts
| Study type | Interventional |
|---|---|
| Sex | All |
| Enrollment | 934 (estimated) |
| Sponsor | Hospital de Clinicas de Porto Alegre Academic / other |
| Locations | 1 site (Porto Alegre, Rio Grande do Sul) |
| Status | Recruiting |
| Started | November 15, 2025 |
| Main results expected | December 2028 |
| Study ends | December 2029 |
| Registry record updated | February 20, 2026 |
| Trial ID | NCT07019116 on ClinicalTrials.gov |
Is this trial a fit for you?
✓Likely a good fit
This applies to referrals from selected specialty waitlists submitted by primary care clinics across municipalities in Rio Grande do Sul, mainly for patients with chronic conditions.
✕Probably not a fit
Referrals that include attachments or previous rounds of discussion, or any referral where the AI gives under 80% confidence, are excluded and would not benefit from the tool in this project.
A plain-English reading of the study's rules. Only the study team can confirm whether you qualify.
What this trial studies
Investigators will randomly implement an AI triage algorithm across selected specialty waitlists in the electronic referral system used in Rio Grande do Sul, Brazil, and compare outcomes to standard gatekeeping. Eligible referrals come from municipal primary care clinics for selected specialties; referrals with attachments, prior discussion rounds, or AI confidence below 80% are excluded. The trial will measure how the AI affects referral authorization patterns, wait times, specialist visit appropriateness, and backlog size. Implementation is centered at the Central de Regulação Ambulatorial in Porto Alegre and coordinated with the state health department.
Why it matters
Potential benefit: If successful, the AI could speed care for high-risk patients, reduce unnecessary specialist visits, and shrink the referral backlog.
How similar studies have gone: Some AI triage tools have shown promise in other referral and emergency-triage settings, but randomized implementation of gatekeeping AI across a large public referral system is relatively novel.
Who can join: the full rules
Show full inclusion / exclusion criteria
Inclusion Criteria: * All referrals from a given specialty (waitlist) will be eligible. * Specialties will be selected following Rio Grande do Sul Health Department priorities. Exclusion Criteria: * Referrals that the AI algorithm can not evaluate. These include referrals with attachments (further information in image or PDF files) and referrals with previous rounds of discussion. * Referrals in which the algorithm has low confidence in the decision (i.e., informed data lead to a decision with a probability below 80%) will not be included in the study.
Where it's running
-
Central de Regulação Ambulatorial
Porto Alegre, Rio Grande do Sul, Brazil Recruiting
Study contacts
- Principal investigator: Dimitris V. Rados, Ph.D. — TelessaúdeRS
- Study coordinator: Dimitris V Rados, Ph.D.
- Email: drados@hcpa.edu.br
- Phone: +555133082092
Similar studies
Other studies for the same condition or treatment. Recruiting studies are listed first.
-
Suicide prevention in primary care clinics in Nepal
This project will try a package of tools and training to help primary care providers in Bagmati Province, Nepal, better find and manage patients at risk of suicide.
-
Explicit 15-minute arrival-time messages to help patients arrive on time
This project will test whether showing an explicit arrival time (15 minutes before the scheduled appointment) in MyChart and automated texts helps primary care patients arrive on time for in-person visits.
-
Patient-centered clinical decision support for recommended screenings in primary care
This project will test a patient-centered computer tool that helps adults aged 18–74 and their primary care doctors identify which preventive screenings are recommended and support shared decision-making.
-
Reducing unnecessary antibiotics for urinary tract infections by improving urine collection
This study will test whether a bilingual (English and Spanish) animated video and pictorial flyer can help adults in safety‑net primary care give cleaner urine samples to reduce contaminated cultures and unnecessary antibiotic prescriptions.
-
Making AI triage more accurate for patients contacting their GP
This project will test whether updated AI in the Patchs online consultation system can make triage decisions more accurate and fair for patients contacting GP practices in England.
-
Project CARE: Hereditary cancer risk checks in primary care
This project will test whether sending primary care patients a chatbot link can identify people at high hereditary cancer risk and connect them to genetic counseling and testing.
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.