(856) NeRDS: Neighborhood Resources and Disease Severity in Pediatric Critical Care

Introduction: Sociodemographic factors have been implicated in disparate health outcomes for critically ill children. Recently, the impact of neighborhood-level factors on health disparities has been reported but is understudied in pediatric critical care. To address this gap, we examined the association between neighborhood-level indices and disease severity on presentation to the PICU.

Methods: We conducted a retrospective ecological study on critically ill children admitted to Lurie Children’s Hospital from Jan 1, 2013 to Dec 31, 2019. Using a Poisson regression with robust errors, we measured the association between composite area-based categorical indices (Child Opportunity Index [COI]; Social Vulnerability Index [SVI]; Chicago-based Neighborhood Disorder Index [NDI]) and the peak Pediatric Logistic Organ Dysfunction 2 Score (PELOD-2) in the first 72-hours of PICU admission (as peak PELOD-2 in the acute phase is a more robust predictor of organ dysfunction burden and mortality than other clinical severity of illness indices). We adjusted for race, ethnicity, age, sex, complex chronic conditions, and payor except when race, ethnicity, and payor were included in the index (for SVI). We also evaluated state- and metropolitan-normed indices and restricted analyses to the Chicagoland cohort.

Results: There were 8298 encounters, most from very low COI (28%), very high SVI (29%), and very high NDI (27%) neighborhoods; and predominantly male (55%), White (38%), non-Hispanic (65%), and publicly insured (70%) patients. The median peak PELOD-2 score was 3 (IQR 2-5). There was no association between COI categories and peak PELOD-2. Very high SVI and high NDI were associated with peak PELOD-2 (p < 0.05) in adjusted analyses. In the Chicagoland cohort, very high SVI was associated with peak PELOD-2 (p < 0.05) in adjusted analyses. Using state and metropolitan-normed indices, there was no association in the entire cohort or when restricted to the Chicagoland cohort in adjusted analyses.

Conclusions: Critically ill children from disinvested neighborhoods were more likely to have a higher burden of peak organ dysfunction than children from resourced neighborhoods. Further studies should assess for causal links between neighborhood indices and PICU severity, and examine the impact of urbanicity.