Results

Study population

From September 2018 to February 2021, 652 children were hospitalized with a diagnosis of bronchiolitis. Of them, 174 were not eligible for our study (Figure 1 ). A total of 268 children’s reports were investigated for the predictive factors of the LOFR; and 478 were investigated for the LOS.
The median age of our sample was 3.2 months (1.6 - 5.4), 263 children were boys (55.0 %) and the median weight was 5.9 kg (4.5 - 7.2). Briefly, 94 children had a familial history of atopy (19.7%) and 69 were born premature (14.4%). The median LOFR was 1.0 days (0.6 - 2.0) and the median LOS was 2.9 days (2.0 - 4.8). Table 1 presents the demographic and clinical characteristics of the patients, their comorbidities, the parameters at ED arrival and course of the stay (when available).

Factors associated with the LoFR and the LOS

In the univariate logistic regression model for LOFR: time delay between the first symptoms and the ED arrival (consultation time), RR ≥ 70/min or < 30/min or apnea, moderate use of accessory muscles and nutritional support requirement were significantly associated with the LOFR (E-Table 2 ). In the multiple logistic regression model, none of the investigated associations reached statistical significance, but the estimate comes with uncertainty ranging from virtually no association to a significant positive association (95% CI for the OR: 1.0 to 2.6) for the moderate use of accessory muscles, nutritional support and RR ≥ 70/min or < 30/min or apnea that could be associated with the LOFR (Table 2 ).
In the univariate logistic regression model for LOS: age, weight, consultation time, severe clinical condition, RR ≥ 70/min or < 30/min or apnea, SpO2% < 90 % or cyanosis, intense use of accessory muscles, nutritional and oxygen support requirement were significantly associated with the LOS (E-Table 2 ). In the multiple logistic regression model, the intense use of accessory muscles (OR=3.9, 95%CI 1.6 - 10.4, p=0.004), the severe clinical condition (OR=2.8, 95%CI 1.7 - 4.8, p=0.001), and other known variables, such as the O2 supplementation (OR=2.0, 95%CI 1.3 - 3.1, p=0.003) remained significantly associated with the LOS and the effect is large enough to be worthwhile (Table 2 ). For the consultation time (OR=1.2, 95%CI 1.1 - 1.3, p=0.001) and the age (OR=0.8, 95%CI 0.7 - 0.9, p<0.01), even if they remained significantly associated with the LOS, the effect may or may not be worthwhile.

In-hospital referral for ACT

Of the 478 infants included in our study, 299 (63%) were referred for ACT during their hospital stay (ACT+). Children in the ACT+ group were significantly older (median age 2.6 (1.5-5.0) vs 3.5 (1.8-5.8) years; p=0.011) and heavier (median weight 5.3 (4.1-6.8) vs. 6.20 (5.0-7.5) kg; p<0.01) compared to those in the ACT- group. Similarly, children in the ACT+ group were less severe at admission and were more often prescribed bronchodilators but required less nutritional support. The complete characteristics of both groups are presented in theTable 3 .