Children with overweight or obesity were more likely to be evaluated for obstructive sleep apnea (OSA) if they were seen at academic practices or if they had any of a half dozen demographic or clinical factors, according to a research poster presented at the Pediatric Academic Societies (PAS) 2026 Meeting in Boston.

The factors that were significantly associated with undergoing an OSA evaluation included having English as a primary language, having public insurance, being male, being aged 2-5 years, being non-Hispanic White individual, and having a high BMI, the researchers found.

“These missed opportunities, which appear to disproportionately impact some historically marginalized subgroups, call for system-level interventions to ensure all eligible children receive appropriate screening and evaluation,” senior author Mona Sharifi, MD, MPH, an associate professor of pediatrics and biostatistics at Yale School of Public Health, New Haven, Connecticut, told Medscape Medical News. “Nudges and reminders within the electronic health record for clinicians and patient portals for families have potential to close these gaps.”

Most surprising of the findings to lead author Antony Lin, MPH, was that it was the youngest children and those with public instead of private insurance who had greater odds of receiving an evaluation. He wasn’t the only one startled by this finding.

“Nearly all of the clinicians who asked me questions about the poster were surprised by the higher odds of evaluations in the children aged 2-5 years compared with older children,” said Lin, a recent graduate of Yale School of Public Health. While there is no way to determine why that is the case from the data, he said, based on a review of the literature, it could have something to do with rapid growth in adenoids and tonsils in this age range and parents being more likely to report symptoms like frequent snoring at younger ages.

Lin noted the importance of clear guidelines and clinical decision support tools to aid primary care physicians in determining when to do an OSA assessment.

Estimates of the prevalence of OSA in US children with obesity range from 13% to 59%, yet pediatric OSA evaluation is inconsistent despite recommendations from the American Academy of Pediatrics and the American Academy of Sleep Medicine, the authors noted in their background information.

Because the evidence on OSA evaluation in primary care is limited for children with overweight/obesity, the researchers conducted a study to quantify it in three US health systems. The systems involved in Improving Pediatric Obesity Practice Using Prompts trial included one each located in the Northeast, the South, and the Midwest.

The researchers analyzed electronic medical records for children aged 2-18 years who had a BMI at the 85th percentile or higher and were seen between 2023 and 2025 at any of the 81 primary care practices. They looked at prevalence of OSA evaluation based on the presence of an OSA diagnosis, a polysomnogram order, or a specialist referral.

The researchers calculated associations between an evaluation and the clinic setting, primary language of the family, insurance status, sex, race/ethnicity, age, and BMI category, with adjustments for other independent variables and the health system.

In a population of 119,639 pediatric patients, half had overweight (50.7%), a third had class I obesity (32.3%), 11.2% had class II obesity, and 5.7% had class III obesity. About a quarter of the patients were aged 2-5 years (23.6%), whereas 36.9% were aged 6-11 years and 39.6% were aged 12-18 years.

Most of the patients were non-Hispanic White (29.8%) or non-Hispanic Black (33.6%) individuals, and about a fifth were Hispanic/Latino individuals (22.5%). Only 3.2% were Asian, 1% were non-Hispanic multi-racial, and 8.3% were of another race/ethnicity.

Overall, 20,593 patients (17.2%) of the sample analyzed in the study had evidence in their charts of an OSA evaluation.

While 18.3% of English-speaking patients received an evaluation, only 11.8% of primarily Spanish-speaking patients and 12.8% of patients with another primary language other than English received one (adjusted odds ratio [aOR], 0.74 for Spanish and 0.8 for other languages).

Odds of an evaluation were about 50% greater for children seen at an academic clinical setting (18.8%) than a nonacademic one (16.4%; aOR, 1.51). Patients with private insurance (16.4%) were less likely than those with public insurance (18.6%) to receive an evaluation (aOR, 0.87).

Although non-Hispanic White patients had the highest prevalence of evaluations (17.3%), the lower prevalence for Hispanic/Latino patients (14.6%; aOR, 1) and higher prevalence in non-Hispanic multiracial patients (18.5%; aOR, 0.9) did not translate to significantly different odds of an evaluation. A higher prevalence of evaluations seen in non-Hispanic Black patients (19.6%; aOR, 0.91) was just barely significant, but odds were significantly lower for assessment in Asian patients (14.7%; aOR, 0.76) and patients of other races/ethnicities (14.8%; aOR, 0.79).

Men were modestly more likely than women (18.1% vs 16.3%) to receive an evaluation (aOR, 0.90), and prevalence was highest in children aged 2-5 years (21.6%).

Patients with class III obesity (29.5%) had over twice the odds of undergoing an evaluation compared with those whose BMI fell within the overweight category (aOR, 2.85). Similarly, prevalence and odds of an OSA evaluation were significantly higher for children with class II obesity (20.7%; aOR, 1.71) or class I obesity (17%; aOR, 1.19).

Deepa Rastogi, MBBS, MS, chief of the Division of Respiratory and Sleep Medicine at Children’s Hospital at Montefiore Einstein in New York City, was not involved in the study but told Medscape Medical News that she found it notable that the rate of assessment for OSA was so low in this population.

“The differences by race and language are also noteworthy, highlighting an area of healthcare equity that needs to be addressed,” Rastogi said. “OSA is a disease which, if not addressed, can affect several different organs, including the heart. Therefore, it is pertinent that the providers ask for its symptoms routinely in their patient population.”

The research was funded by the National Institutes of Health. Sharifi, Lin, and Rastogi had no disclosures.

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