Disparities in Predicted and Parent-Reported Need for Pediatric Subspecialty Care

Abstract Objective Parent-reported need for subspecialty care has the potential to be influenced by factors such as access to primary care, supplier-induced demand, and parent health literacy. We sought to model predicted likelihood of need for subspecialty care based on objective clinical factors, and to compare this "predicted need" to parent-reported need and actual receipt of subspecialty care across sociodemographic groups. Methods The 2022-23 National Survey of Children's Health captures parent-reported need for subspecialty care. We modeled the association between parent-reported need for subspecialty care and clinical factors (child age, health status, presence and severity of specific conditions) and used this model to generate "predicted need." We compared weighted percentages of predicted need, parent-reported need, and actual receipt by family poverty level, child race, payer, and family language. Results Of 90,620 unweighted children included, predicted need for subspecialty care was highest for those with lowest family income (<100% federal poverty level, FPL) at 17.2% (95% CI: 16.6-17.7%) and lowest for those with highest family income (≥400% FPL) at 15.1% (95% CI: 14.8-15.3%). In contrast, parent-reported need for subspecialty care was lowest for those with lowest family income (12.8%, 95% CI: 11.6-14.0%) and highest for those with the highest family income (17.1%, 95% CI: 16.5-17.8%). Variation in these measures was also observed by payer, family language, and child race and ethnicity. Conclusion Gaps between predicted need and parent-reported need for subspecialty care were larger for populations underserved by the US health care system. Predicted need for subspecialty care may serve as an alternative means of estimating underlying population need for subspecialty care.