Assessment of Social Vulnerability in Pediatric Head and Neck Cancer Care and Prognosis in the United States.

ABSTRACT

Importance Prior investigations in social determinants of health (SDoH) in pediatric head and neck cancer (HNC) have only considered a narrow scope of HNCs, SDoH, and geography while lacking inquiry into the interrelational association of SDoH with disparities in clinical pediatric HNC. Objectives: To evaluate the association of SDoH with disparities in HNC among children and adolescents and to assess which specific aspects of SDoH are most associated with disparities in dynamic and regional sociodemographic contexts. Design, Setting, and Participants: This retrospective cohort study included data about patients (aged ≤19 years) with pediatric HNC who were diagnosed from 1975 to 2017 from the Surveillance, Epidemiology, and End Results Program (SEER) database. Data were analyzed from October 2021 to October 2022. Exposures Overall social vulnerability and its subcomponent contributions from 15 SDoH variables, grouped into socioeconomic status (SES; poverty, unemployment, income level, and high school diploma status), minority and language status (ML; minoritized racial and ethnic group and proficiency with English), household composition (HH; household members aged ≥65 and ≤17 years, disability status, single-parent status), and housing and transportation (HT; multiunit structure, mobile homes, crowding, no vehicle, group quarters). These were ranked and scored across all US counties. Main Outcomes and Measures: Regression trends were performed in continuous measures of surveillance and survival period and in discrete measures of advanced staging and surgery receipt. Results: A total of 37 043 patients (20 729 [55.9%] aged 10-19 years; 18 603 [50.2%] male patients; 22 430 [60.6%] White patients) with 30 different HNCs in SEER had significant relative decreases in the surveillance period, ranging from 23.9% for malignant melanomas (mean [SD] duration, lowest vs highest vulnerability 170 [128] months to 129 [88] months) to 41.9% for non-Hodgkin lymphomas (mean [SD] duration, lowest vs highest vulnerability 216 [142] months vs 127 [94] months). SES followed by ML and HT vulnerabilities were associated with these overall trends per relative-difference magnitudes (eg, SES for ependymomas and choroid plexus tumors mean [SD] duration, lowest vs highest vulnerability 114 [113] months vs 86 [84] months; P < .001). Differences in mean survival time were observed with increasing social vulnerability, ranging from 11.3% for ependymomas and choroid plexus tumors (mean [SD] survival, lowest vs highest vulnerability 46 [46] months to 41 [48] months; P = .43) to 61.4% for gliomas not otherwise specified (NOS) (mean [SD] survival, lowest vs highest vulnerability 44 [84] months to 17 [28] months; P < .001), with ML vulnerability followed by SES, HH, and HT being significantly associated with decreased survival (eg, ML for gliomas NOS mean [SD] survival, lowest vs highest vulnerability 42 [84] months vs 19 [35] months; P < .001). Increased odds of advanced staging with non-Hodgkin lymphoma (OR, 1.21; 95% CI, 1.02-1.45) and retinoblastomas (OR, 1.31; 95% CI, 1.14-1.50) and decreased odds of surgery receipt for melanomas (OR, 0.79; 95% CI, 0.69-0.91) and rhabdomyosarcomas (OR, 0.90; 95% CI, 0.83-0.98) were associated with increasing overall social vulnerability. Conclusions and Relevance In this cohort study of patients with pediatric HNC, significant decreases in receipt of care and survival time were observed with increasing SDoH vulnerability.