182 - Association Between the Child Opportunity Index 3.0 and Critical Asthma in Children

Publication Number: 1171.182

Kiersten Flodman, Yale School of Medicine, New Haven, CT, United States; Lauren Flagg, Yale-New Haven Children's Hospital, New Haven, CT, United States; Sarah McCollum, Yale School of Medicine, New Haven, CT, United States; Veronika Shabanova, Yale School of Medicine, New Haven, CT, United States; E. Vincent S. Faustino, Yale School of Medicine, New Haven, CT, United States; Laura Chen, Yale School of Medicine, New Haven, CT, United States; Nadir Ijaz, Yale School of Medicine, New Haven, CT, United States

Background: Social drivers of health (SDOH) influence disparities in health care utilization and outcomes. The Child Opportunity Index 3.0 (COI 3.0) measures SDOH by census tract. Lower quintiles of COI 3.0 indicate less access to neighborhood factors supporting healthy child development. Existing research links COI 3.0 to pediatric intensive care unit (PICU) admissions, recurrent emergency visits, and rehospitalizations for asthma. It is unclear whether a new definition of critical asthma, based on treatment with adjunctive medications beyond albuterol and corticosteroids, is also associated with COI 3.0 (Figure 1).

Objective: We aimed to evaluate the association between census tract-level COI 3.0 and critical asthma exacerbations in hospitalized children.

Design/Methods: In this retrospective cohort study, we identified children < 18 years old admitted to a tertiary children's hospital from 1/1/2017-12/31/2021 with an admission or principal diagnosis of asthma in the Pediatric Health Information System (PHIS). For each child, we linked demographics and outcomes in PHIS with COI 3.0 based on recent home address in the electronic health record. Primary outcome was incidence of critical asthma exacerbation, defined by using adjunctive medications (ipratropium, terbutaline, aminophylline, magnesium, epinephrine, ketamine). Secondary outcomes were PICU admission, length of stay (LOS), and mechanical ventilation. We used a marginal Poisson regression adjusted for race/ethnicity and insurance payor to estimate adjusted risk ratios (aRR; 95% confidence interval, CI) of the outcomes per COI 3.0 quintile. Robust standard errors accounted for patients with multiple encounters. We used the very low quintile for base comparison.

Results: We identified 1,153 hospitalizations from 902 unique patients (Table 1). COI 3.0 quintile was very low in 553 (48%), low in 251 (21.8%), moderate in 181 (15.7%), high in 89 (7.7%), and very high in 79 (6.9%) hospitalizations. We did not find an association between critical asthma and COI 3.0 (aRR by quintile: low 1.10 (0.96, 1.27), moderate 1.00 (0.82, 1.21), high 0.96 (0.74, 1.25), very high 1.05 (0.79, 1.39)). Secondary outcomes were notable for lowest PICU admission rate in the moderate quintile (aRR by quintile: low 0.98 (0.93, 1.03), moderate 0.92 (0.87, 0.98), high 1.01 (0.93, 1.09), very high 1.02 (0.93, 1.12)) (Table 2).

Conclusion(s): In this study, critical asthma exacerbations were not associated with COI 3.0. It is possible that COI 3.0 is associated with more proximal outcomes such as hospital admission. Future work should examine this possibility, accounting for other confounders.

Figure 1. Directed acyclic graph
Directed acyclic graph (DAG) showing hypothesized relationships between the study exposure and outcomes. This DAG was generated based on previously published models describing the relationship between social drivers of health and asthma outcomes. There may be additional possible relationships between variables that are not shown here.

Table 1. Child characteristics by Child Opportunity Index 3.0 quintiles (n=1153 hospitalizations)
Patient demographics are summarized for all hospitalizations. Of note, individual patients may have more than one hospitalization in the study period. Critical asthma medications administered are also shown, by COI 3.0 quintile.

Table 2. Hospitalization outcomes by Child Opportunity Index 3.0 quintiles (n=1153 hospitalizations)
Marginal Poisson regression with robust standard errors was performed for the primary outcome of critical asthma exacerbation, defined by administration of adjunctive medications, and the secondary outcomes. Risk ratios were adjusted for race/ethnicity and insurance payor. The very low quintile is used as the base comparison group for all regressions.