Pediatric Resident University of Oklahoma Children's Hospital The University of Oklahoma Health Sciences Center Oklahoma City, Oklahoma, United States
Background: Type 1 diabetes mellitus (T1DM) is one of the most common chronic diseases of childhood, with approximately 1 in 300 people in the US receiving a diagnosis by age 18. Among this population, diabetic ketoacidosis (DKA) remains the leading cause of morbidity and mortality. Previous studies examining predictive factors for DKA admission have found relationships with age, sex, race, insurance, and location.
Objective: This study seeks to identify predictive factors of admission for DKA among patients in Oklahoma served by OU Children’s Hospital.
Design/Methods: The study is a retrospective chart review of admissions for DKA to OU Children’s Hospital from January 2015 through December 2019. Factors studied: admission date, type of diabetes, status of diabetes (new-onset vs. known-diagnosis), age, sex, insurance, and zip code +4-digit code. Each zip code +4-digit code was cross-referenced with the University of Wisconsin School of Medicine and Public Health’s Neighborhood Atlas + Area Deprivation Index (ADI), a proxy for socioeconomic status to the neighborhood level. Relationships were analyzed using the chi-square and Student’s t-test.
Results: A total of 486 admissions of 349 patients were studied. 68 patients (19.5% total patients) had multiple admissions, accounting for 205 admissions (42.2% total admissions). The median interval until readmission was 192 days; 70% of readmitted patients were readmitted by 1 year, and 90% of readmitted patients were readmitted by 2 years. Patients with multiple admissions: 57% female, 43% male; patients with single admissions: 54% male, 46% female. Patients whose first admission was with a known-diagnosis of diabetes had a significant difference in readmission rate compared to patients whose first admission was for new-onset diabetes (46/169 vs. 22/180, respectively; p = 0.0011, chi-square). The median ADI national percentile for readmitted patients was significantly different than that for patients who were not readmitted (75.8 vs. 61.4, respectively; p = 0.0091, Student’s t-test). The median ADI state decile for readmitted patients was significantly different than that for patients who were not readmitted (6.9 vs. 4.8, respectively; p = 0.0073, Student’s t-test). Conclusion(s): A patient’s ADI is a significant risk factor for DKA admission. I intend to perform a future study to explore if additional resources (CGM technology, more frequent visits, virtual visits, etc.) can reduce rates of DKA admission among these higher-risk patients.
Authors/Institutions: Chris T. Sebastian, The University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, United States; David Sparling, The University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, United States; Jeanie B. Tryggestad, The University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, United States
