Variability in the diagnostic labeling of nonbacterial lower respiratory tract infections: a multicenter study of children who presented to the emergency department.

OBJECTIVE: The diagnostic labeling of presumed nonbacterial lower respiratory tract infection is unclear. Our objective was to identify patterns of specific diagnoses and treatments that were given to children who presented with lower respiratory tract infection to US academic emergency departments.
METHODS: Data were collected on all children who were aged <2 years and had lower respiratory tract infection symptoms during a similar 2- to 3-week winter period at 4 pairs of emergency departments from the same state or region. The children were identified by using relevant International Classification of Diseases, Ninth Revision, Clinical Modification codes in the primary diagnosis field. Data were collected by using standardized chart review forms for the index emergency department visit and also for 1 month before through 1 year after the index visit.
RESULTS: Among the 928 children who presented with lower respiratory tract infection symptoms, 676 (73%) were younger than 12 months and 624 (67%) had a primary diagnosis of bronchiolitis. When comparing the assigned diagnoses between emergency department pairs, bronchiolitis was the more common diagnosis at certain hospitals, whereas asthma, cough, and wheeze were more frequent at others. Independent predictors of corticosteroid treatment were visiting specific emergency departments, older age, an asthma diagnosis (compared with bronchiolitis), documented history of wheezing, observed wheezing during the index visit, eosinophil values >4%, previous use of corticosteroids, and parental history of asthma.
CONCLUSIONS: For children who are age <2 years and present to an emergency department with lower respiratory tract infection symptoms, there is large variability in the assigned diagnosis. Children who present to emergency departments that more commonly diagnose lower respiratory tract infection as "asthma" are more likely to receive corticosteroids. As clinicians, we need to develop evidence- and outcome-based definitions for lower respiratory tract infections to guide diagnosis and treatment better.