O-009 Healthcare Disparities in Hospitalized Children and Adolescents with Crohn’s Disease: Is Race Associated with the Need for Readmissions?

Healthcare disparities account for a large portion of mortality and morbidity in children, yet they remain understudied in pediatrics, particularly in pediatric Crohn’s disease (CD). Research suggests that race, which is 1 factor associated with healthcare disparities, may influence the distribution, phenotype and treatment of CD. However, it is unclear to what extent racial differences exist, and whether any differences are due to intrinsic biologic differences between races, differences in access and treatment, or both. Current CD treatment is focused on achieving improvement in “hard” outcomes such as rates of hospitalization and surgery. However, the effects of race on hospital admissions in children with CD are unknown. Therefore, we sought to compare a cohort of Black to White children with CD to determine the extent race is associated with differences in readmissions in the United States (US). The Pediatric Health Information System (PHIS) is an administrative database containing comprehensive inpatient data from 44 children’s hospitals in the US. We extracted data for a cohort of patients less than or equal to 21 years of age hospitalized with a diagnosis of CD (January 1, 2004–June 30, 2012). Patients were excluded if data was missing for race, or if the encounter had the same admission and discharge date. A White cohort was matched 2:1 based on hospital to a Black cohort. Demographics were summarized using standard descriptive statistics. Wilcoxon 2-sample test was used to assess length of stay and number of readmissions. We evaluated the time from initial hospital discharge to readmission for each race using Kaplan-Meier analysis with Log-rank test. Multivariate logistic regression was used to assess race difference on outcomes at several time intervals, while adjusting for age, gender, region, medications, payor status, and income (all reported OR are adjusted). Due to multiple comparisons, a P-value of <0.005 was considered statistically significant. There were 4377 total patients (33% Black, 67% White) (Demographics Table 1). Black children had longer LOS (6.8 ± 7.1, median:5) than White (6.3 ± 8.9, median:4) (P < 0.001) and an overall higher frequency of readmissions (1.4 ± 2.6) than White (0.9 ± 1.9) (P < 0.001). There was no significant difference in early readmission (<30 days) between Black and White children, however Black children had more late readmissions (30 days–12 months; 28% versus 23%; P < 0.001) than White children. Kaplan-Meier analysis for time to the first readmission yielded similar results (P = 0.009, Fig. 1). After adjusting for each variable in the multivariate logistic regression analyses, factors predictive of overall readmission included use of steroids or TPN during the initial hospitalization (OR = 1.27, P = 0.002; OR = 1.35, P = 0.001, respectively), patients with “other” insurance payor status (OR = 1.60, P < 0.001), and race (Black patients OR = 1.27, P = 0.003). This study supports that there are disparities in hospital readmissions related to race. It is unclear whether this is due to disparities in care or phenotypic differences in disease between racial groups. The difference in late readmissions could suggest worse intrinsic disease, adherence, access or treatment disparities.