Small sleepers, big data: leveraging big data to explore sleep-disordered breathing in infants and young children

Abstract

Study objectives: Infants represent an understudied minority in sleep-disordered breathing (SDB) research and yet the disease can have a significant impact on health over the formative years of neurocognitive development that follow. Herein we report data on SDB in this population using a big data approach.

Methods: Data were abstracted using the Cerner Health Facts database. Demographics, sleep diagnoses, comorbid medication conditions, healthcare utilization, and economic outcomes are reported.

Results: In a cohort of 68.7 million unique patients, over a 9-year period, there were 9,773 infants and young children with a diagnosis of SDB (obstructive sleep apnea [OSA], nonobstructive sleep apnea, and "other" sleep apnea) who met inclusion criteria, encompassing 17,574 encounters, and a total of 27,290 diagnoses across 62 U.S. health systems, 172 facilities, and 3 patient encounter types (inpatient, clinic, and outpatient). Thirty-nine percent were female. Thirty-nine percent were ≤1 year of age (6,429 infants), 50% were 1-2 years of age, and 11% were 2 years of age. The most common comorbid diagnoses were micrognathia, congenital airway abnormalities, gastroesophageal reflux, chronic tonsillitis/adenoiditis, and anomalies of the respiratory system. Payor mix was dominated by government-funded entities.

Conclusions: We have used a novel resource, large-scale aggregate, de-identified EHR data, to examine SDB. In this population, SDB is multifactorial, closely linked to comorbid medical conditions and may contribute to a significant burden of healthcare costs. Further research focusing on infants at highest risk for SDB can help target resources and facilitate personalized management.

Keywords: OSA; Health Facts; data science; infants; sleep-disordered breathing.