JAMA Health Forum. 2026 May 01. 7(5):
e261199
Importance: The heterogeneous population of children and youth with special health care needs (SHCN) has suboptimal health indicators. Subpopulations may have distinct health needs and indicators.
Objective: To identify distinct subpopulations of children and youth with SHCN and examine associations with policy-relevant health indicators.
Design, Setting, and Participants: Using pooled data from the nationally representative National Survey of Children's Health, latent class analysis identified subpopulations of children and youth with SHCN aged 0 to 17 years from 2019 to 2021. Weighted multivariable logistic regressions examined associations between class membership and 7 health indicators, adjusting for demographic and household characteristics. Data were analyzed from September 2023 to February 2026.
Exposure: Fourteen indicators reflecting health-related needs, functional limitations, and health care use were selected from the national survey by dual review.
Main Outcomes and Measures: Health indicators representing the 10 domains for a healthy life conceptual framework were selected by dual review. Seven indicators rated by a national expert panel as potentially modifiable and high impact were included.
Results: The National Survey of Children's Health had a sample size of 29 433 in 2019, 42 777 in 2020, and 50 892 in 2021. The estimated male population of children and youth with SHCN was 56% (n = 10 106 838), and the estimated female population was 44% (n = 7 920 925). A total of 19% (n = 3 332 877) were aged 0 to 5 years, 36% (n = 6 476 250) were aged 6 to 11 years, and 46% (n = 8 218 637) were aged 12 to 17 years. Among a weighted population of 18 027 763 children and youth with SHCN, 4 distinct classes were observed: low intensity (67%; n = 12 083 605), developmental and behavioral impacts (14%; n = 2 542 398), dynamic (13%; n = 2 349 486), and complex (6%; n = 1 053 274). Classes differed significantly in demographic characteristics and health indicator associations. The low-intensity class exhibited few health-related needs, little functional limitation, and minimal health care use. The developmental and behavioral impacts class exhibited predominantly functional limitations, while the dynamic class exhibited primarily health care use. The complex class exhibited substantial health-related needs, functional limitations, and health care use. Compared with the low-intensity class, the other classes had significantly lower odds of experiencing positive health indicators (household always able to provide basics: adjusted odds ratio [aOR], 1.51; 95% CI, 1.32-1.70; family demonstrates resilience: aOR, 1.31; 95% CI, 1.15-1.49; child received care when needed: aOR, 3.02; 95% CI, 2.44-3.74; child had adequate and continuous insurance: aOR, 1.94; 95% CI, 1.72-2.18; insurance covers mental and behavior health: aOR, 2.47; 95% CI, 2.19-2.77; no difficulty accessing specialists: aOR, 2.24; 95% CI, 1.87-2.70; no difficulties finding mental health treatment: aOR, 1.93; 95% CI, 1.62-2.30).
Conclusions and Relevance: This study identified 4 distinct subpopulations of children and youth with SHCN. The unique sociodemographic and health indicator associations for latent classes suggest that policies and programs may require tailoring to maximize efficiency and effectiveness across this otherwise diverse and heterogeneous pediatric population.