Aim: In infants with acute bronchiolitis, the precision of parental disease severity assessment is unclear. We aimed to determine if parental assessment at the time of hospitalisation predicted the use of supportive care, and subsequently determine the likelihood that the infant with acute bronchiolitis would receive supportive care.
Methods: From the Bronchiolitis ALL south-east Norway study, we included all 267, 0-12 month old, infants with acute bronchiolitis whose parents at the time of hospitalisation completed a three-item visual analogue scale (VAS) concerning Activity, Feeding and Illness. Respiratory rate, oxygen saturation (SpO2 ) and use of supportive care were recorded daily. By multivariate logistic regression analyses we included significant predictors available at hospital admission to predict the use of supportive care.
Results: The parental Activity, Feeding and Illness VAS scores significantly predicted supportive care with odds ratios of 1.23, 1.26 and 1.36, respectively. The prediction algorithm included parental Feeding and Illness scores, SpO2 , gender and age, with an area under the curve of 0.76 (95% CI 0.69, 0.81). A positive likelihood ratio of 2.1 gave the highest combined sensitivity of 81% and specificity of 61%.
Conclusion: Parental assessment at hospital admission moderately predicted supportive care treatment in infants with acute bronchiolitis.
Trial registration: ClinicalTrials.gov NCT00817466.
Keywords: Acute bronchiolitis; Disease severity; Infants; Parental assessment; Predicting supportive care.
©2018 Foundation Acta Paediatrica. Published by John Wiley & Sons Ltd.