Respiratory syncytial virus testing during bronchiolitis episodes of care in an integrated health care delivery system: A retrospective cohort study

Abstract

Background: Bronchiolitis has the highest incidence rate of any lower respiratory infection among infants and children <2 years of age. Respiratory syncytial virus (RSV) is the most common etiology of bronchiolitis. The American Academy of Pediatrics does not recommend routine RSV testing for infants and children with bronchiolitis. The clinical predictors of RSV testing are unknown. Objectives: The aims of this study were to identify the rates and predictors of RSV testing during bronchiolitis and to explore the relationship between RSV test results and antibiotic treatment among infants and children aged <2 years. Methods: A retrospective study was conducted of 123,264 infants ≥32 weeks' gestational age (GA) who were born at 1 of 6 Northern California Kaiser Permanente Medical Center Program hospitals between 1996 and 2004. A bronchiolitis episode of care (EOC) was defined as ≥1 medical encounters with a bronchiolitis diagnosis code followed by 14 clear days without a bronchiolitis-related medical encounter. Descriptive statistics were used to report the frequency of tests, and logistic regression was used to assess the effect of hospitalization, chronologic age, gestational age, and season on the frequency of testing for RSV. Rapid direct fluorescent antibody testing was performed during the study. Results: The birth cohort was 51.2% male and 42.7% white, 20.8% Hispanic, 20.3% Asian, 8.4% African American, and 7.9% other. Of 23,748 bronchiolitis EOCs, 4969 (20.9%) had ≥1 test for RSV. Overall, 44.2% of all tests were positive for RSV. Physicians ordered RSV tests in 30.4% and 26.7% of bronchiolitis EOCs for infants born at 32 to 33 and 34 to 36 weeks' GA, respectively, compared with 17.9% of bronchiolitis EOCs for infants born at ≥41 weeks' GA. Bronchiolitis hospitalization, younger chronologic age, prematurity, and RSV season were associated with RSV testing in a multivariate model controlling for other variables, with an adjusted odds ratio (AOR) of 28.55 (95% CI, 24.99–36.62) for hospitalization status; AOR of 6.89 (95% CI, 5.19–9.15) for chronologic age <1 month; AOR of 0.85 (95% CI, 0.76–0.95) for GA >41 weeks; and AOR of 2.48 (95% CI, 2.24–2.74) for RSV season (December–March). Among hospitalized infants who were tested and had a diagnostic code suggesting treatment with antibiotics, use of antibiotics was significantly lower among those with a positive RSV test (63.4%) than those with a negative RSV test (75.5%) (χ 2 test; P < 0.001). Conclusions: Approximately 20% of these children with bronchiolitis EOCs were tested for RSV; of those tested, about half were positive. In this integrated health care system, hospitalization with bronchiolitis, chronologic age, gestational age <37 weeks, neonatal oxygen exposure, and bronchiolitis EOC during the RSV season were the factors associated with testing for RSV.