TOP 43 - Comparing hospital outcomes for common respiratory diagnoses in children with caregivers who prefer languages other than English: a retrospective review

Publication Number: 2795.TOP 43

Ritika Gupta, Baylor College of Medicine/Texas Children's Hospital, Houston, TX, United States; Michelle Lopez, Baylor College of Medicine, Houston, TX, United States; Marina Masciale, Baylor College of Medicine, Houston, TX, United States; Duc T. Nguyen, Baylor College of Medicine, Houston, TX, United States; Karla Fredricks, Baylor College of Medicine/Texas Children’s Hospital, Houston, TX, United States

Background: Studies have demonstrated that hospitalized children with caregivers who prefer languages other than English (PLOE) are more likely to have prolonged length of stay, ICU admissions, adverse hospital events, emergency department (ED) revisits, and increased cost of care compared to children whose caregivers prefer English (PEL). Despite these findings, few studies have examined whether these disparities persist after controlling for socioeconomic status (SES) and severity of illness (SOI).

Objective: We aim to compare hospital outcomes between children with caregivers who PEL and those who PLOE when children are admitted with common respiratory diagnoses. We also aim to determine whether preferred language is independently associated with adverse outcomes after adjusting for SES and SOI.

Design/Methods: We are conducting a retrospective cohort study of children ( < 18 years old) hospitalized with a primary diagnosis of bronchiolitis, asthma, or pneumonia from 2015-2025 within a large, pediatric, quaternary care system. This study was approved by the Baylor College of Medicine Institutional Review Board. Exposures include demographics (including preferred language), All Patient Refined Diagnosis Related Groups (APR-DRG) severity of illness (SOI) category, and Childhood Opportunity Index (COI), calculated based on patient census tract. Preferred language and patient address were obtained from the hospital's electronic medical record system, and clinical data were obtained from the Pediatric Health Information System (PHIS). Outcomes include 30-day readmission, 30-day ED re-presentation, average length of stay (LOS), and ICU stay. Data will be reported as frequencies and proportions for categorical variables and as median and interquartile range (IQR) for continuous variables. Differences between groups (PEL vs Spanish vs Other) will be assessed using the Chi-square or Fisher's exact tests for categorical variables and the Kruskal-Wallis test for continuous variables, as appropriate. Factors associated with binary outcomes will be evaluated using logistic regression. Factors associated with continuous outcomes will be determined with linear regression or generalized linear models. Variables for the multivariable models were selected based on clinical importance and by the Stata’s Lasso method with the cross-validation (CV) selection option.