MINNEAPOLIS — Referrals to pediatric rheumatology are often of poor quality, contain inadequate information, or are based almost solely on a single data point that is often not very informative on its own, such as a positive antinuclear antibody (ANA) test result, according to research presented at the American College of Rheumatology’s Pediatric Rheumatology Symposium (PRSYM) 2026. Results of three research posters at the meeting suggested a need for education for nonrheumatologists, particularly primary care physicians (PCPs), in when pediatric rheumatology referrals are truly indicated and how to appropriately document that.

“What [often] comes to us in the documentation is there’s no close physical exam findings, [and] the history is very spotty,” Barbara Ostrov, MD, a pediatric rheumatologist at Albany Medical Center and chief of service at Bernard & Millie Duker Children’s Hospital, both in Albany, New York, told Medscape Medical News. She oversaw a group of medical students at Albany Medical College whose research involved examining the sources and quality of referrals to pediatric rheumatology in a pair of research posters, one led by Ritika Patel and another led by Freddy Murgida.

“Poorly constructed referrals frequently lack essential history, examination findings, or diagnostic clarity, delaying appropriate care and straining resources,” they wrote. There are only about 450 pediatric rheumatologists in the US — a workforce that is already projected to decline by 16% by 2030 — and past research has shown that a large proportion of referrals to these specialists do not result in a rheumatologic diagnosis.

A recent increase in pediatric rheumatology referrals with nonspecific symptoms and test results — rather than clear-cut signs of an inflammatory or autoimmune condition — is increasing the time that pediatric rheumatologists spend on nonrheumatologic conditions, they wrote.

Referrals Leading to a Rheumatologic Diagnosis

The researchers conducted a retrospective chart review of all new pediatric rheumatology referrals from 2021 to 2023 at their institution and then assessed the training and specialty of the referring providers and the characteristics of the patients referred. After excluding duplicate records, internal referrals, and referrals from patients with a previous diagnosis, they identified 1088 new referrals over those 3 years.

Most referrals (65%) were for female patients, with 68% in children aged 11 years or older, 20% in children aged 6-10 years, and 12% in children aged 5 years or younger. More than half the referrals were for joint pain (57%), followed by an ANA test (18%), abnormal labs (11%), and a rash (11%), the latter usually with family history and digit discoloration.

Just over half of the referrals (55.3%) led to a nonrheumatologic diagnosis, and 5.5% were lost to follow-up. The other 40.2% resulted in a definitive rheumatologic diagnosis. The most common final diagnosis was aches and pains (19%), followed by hypermobility (13%), juvenile idiopathic arthritis (9%), primary Raynaud (7%), amplified pain syndrome (6%), and an autoinflammatory diagnosis (5%).

Most of the referrals (78%) came from physicians while the remainder came from advanced practice providers (APPs). The vast majority came from pediatrics (74%), followed by family medicine (11%) and orthopedics (8%). More referrals from physicians (41.9%) than APPs (34.1%) resulted in a rheumatologic diagnosis (P = .034).

A similar proportion of rheumatologic diagnoses resulted from pediatricians’ referrals (39.1%) as from family physicians (41%), and just over half the orthopedic referrals (54.7%) led to a rheumatologic diagnosis (P = .0455 compared with PCPs’ referrals).

Grading the Referrals

The researchers then graded the referrals as “good,” “fair,” or “poor” across five domains: clarity, appropriateness, clinical history, physical exam, and lab or radiographic testing. About half the referrals (52%) scored as “poor” in at least one domain, particularly in including relevant test results (29%), physical exam notes (24%), and clinical history (18%).

Referrals rated as “good” in their appropriateness were significantly more likely to result in a rheumatologic diagnosis (P < .00001), and referrals from physicians were significantly more likely than those from APPs to be rated as “good” for clarity (P < .05).

Overall, the referrals had highly variable quality with frequent omission of essential clinical details, which “leads to inefficiencies, delayed diagnoses, and unnecessary utilization of limited subspecialty resources,” the researchers wrote. The higher proportion of nonrheumatologic diagnoses from PCPs and APPs may be due to less training or exposure to rheumatologic diagnoses, they hypothesized, and their results overall suggest a need for targeted education in “proper interpretation of tests, clinical history, symptom presentation, and exam findings” as it relates to suspicion of rheumatologic conditions.

Challenges With ANA-Based Referrals

Another research poster looked specifically at one of the more common reasons for pediatric rheumatology referrals — positive ANA tests. Most ANA-based referrals don’t lead to a rheumatologic or autoimmune disease diagnosis, but autoantibodies can precede clinical autoimmunity that presents later on.

“This is where our job gets really hard, really fast because someone gets a test and gets a positive result and then is told that their child might have lupus,” Josh Umland, DO, a pediatric rheumatology fellow at the University of California, Los Angeles, told Medscape Medical News. The time between receiving that result and seeing a pediatric rheumatologist can be fraught with anxiety, he said, so it’s worth considering how necessarily appropriate all those ANA-based referrals are.

Umland and his colleagues, therefore, examined the long-term outcomes of ANA-based referrals at their institution. They used electronic health record data in a retrospective manual chart review of all new patient referrals from July 2015 to June 2017. Of 720 referrals during that time, 178 (24.7%) were ANA-based, and 10 of those (5.6%) resulted in a rheumatologic diagnosis at the first visit.

Following the remaining patients for an average 76 months through November 2025, an additional 11 patients (6.2%) developed a rheumatologic or autoimmune diagnosis during follow-up. The other 88.2% remained disease-free.

Those 11 cases included one case of Crohn’s disease, one case of immune thrombocytopenia, one case of Hashimoto thyroiditis, one case of rheumatoid arthritis, three cases of psoriatic arthritis, two cases of undifferentiated connective tissue disease (one classified as hypermobile Ehlers-Danlos syndrome), and two cases of lupus (one with optic neuritis).

“These findings reaffirm and extend prior literature: Most children referred for elevated ANA results remain healthy,” Umland and his colleagues concluded while noting that positive ANA results continue to be a large source of referrals despite the low rate of subsequent autoimmune diagnoses. “There is room for improvement in educating nonrheumatologists in reasons for obtaining ANA testing, especially in light of the workforce shortage,” they wrote.

“If there’s signs of inflammation, that’s where the ANA should come in,” Ostrov said. “Not every kid who has a viral illness or just aches and pains should have an ANA done, and that happens a lot. You have to listen to the history and examine the patients.”

Elizabeth Murray, MD, a rheumatology fellow at Nationwide Children’s Hospital and The Ohio State University, both in Columbus, Ohio, agreed that ANA results on their own are not necessarily very informative.

“I generally advise people that other lab work is so much more helpful than ANA,” Murray told Medscape Medical News. That could include doing a complete blood count to check for cytopenia or a urinalysis to check for proteinuria, she said.

Umland and his colleagues further noted that additional research would be helpful in understanding the “potential prognostication of positive ANA testing.”

None of the research received external funding. Umland, Murray, Patel, Murgida, and Ostrov had no disclosures.

Tara Haelle is a science/health journalist based in Dallas.