To explore the epidemiological characteristics of pediatric influenza in Chengdu from 2017 to 2022 and the application of pediatric early warning score (PEWS) in prediction of the severity of illness in children.

From January 2017 to September 2022, a total of 6 785 pediatric influenza cases were selected from 4 medical institutions, including Chengdu Second People′s Hospital, West China Second University Hospital, Sichuan University, Chengdu Qingbaijiang District People′s Hospital, and Lianxin Community Health Center, Jinjiang District, Chengdu. Took retrospective analysis method, according to the severity of illness, they were divided into three groups: mild group (n=6 141, ordinary cases), severe group (n=448, severe cases) and critical group (n=196, critical cases). The clinical data of children were analyzed retrospectively. Chi-square test was used to compare the incidence rate of severe and critical cases between influenza A and influenza B children. One-way ANOVA and Bonferroni method were used to overall and pairwise compare the PEWS among the three groups. The receiver operating characteristic (ROC) curve was used to analyze the optimal threshold score of PEWS in predicting the occurrence of severe and critical influenza in children. All procedures in this study were approved by the Ethics Committee of Chengdu Second People′s Hospital (Approval No. 202212). Informed consents were obtained from guardians of all children.

① Before the implementation of non-pharmaceutical interventions (NPIs) for preventing and control the coronavirus disease 2019 (COVID-19) (from Janurary 2017 to September 2019), children were mainly infected with influenza A virus in winter and spring. It mainly affected infants, toddlers, and preschoolers, the total proportion was 75.7% (4 280/5 654), the proportion of severe and critical cases was 10.1%. After implementation of NPIs (from January 2020 to September 2022), children were mainly infected with influenza B virus, and influenza A and B viruses were alternately infected with small epidemics. School-age and adolescents were primarily affected, with the total proportion of 66.1% (748/1 131), the proportion of severe and critical cases was 6.3%. ②The PEWS of children in mild group, severe group and critical group were (0.27±0.57) score, (2.44±0.64) score and (5.16±1.17) score, respectively, the PEWS of children in severe and critical group were higher than those of in mild group, respectively, and the differences were statistically significant (t=5.46, 35.92; both with P<0.001). ③The incidence rate of severe and critical cases in influenza A children was 9.9% (472/4 757), which was higher than that in influenza B children (8.5%, 172/2 028), and the difference was statistically significant (χ²=19.88, P<0.001). Among severe and critical cases, the PEWS of influenza A children was (4.33±0.55) score, which was higher than that of (3.58±0.25) score in influenza B children, and the difference was statistically significant (t=4.74, P<0.001). ④The area under the ROC curve (ROC-AUC) of PEWS predicting severe influenza in influenza children was 0.916±0.041, and the optimal threshold score was PEWS with 2.50 score; The ROC-AUC of PEWS predicting critical influenza in severe influenza children was 0.959±0.017, the optimal threshold score was PEWS with 5.21 score.

After the implementation of NPIs, the pediatric influenza in Chengdu showed a low level epidemic, with alternating of influenza A and B virus infection. The PEWS can be used to predict the severity of influenza in children.