68-6

59/104

so it is important to know the clinical features in child populations. Moreover, in clinical practice, it is important to distinguish children with potential COVID-19 from other diseases as an infection control strategy in outpatient clinics, where chil-dren without COVID-19 and their more vulnerable parents and elderly relatives may gather. However, there are few studies that compare child patients with and without COVID-19 for the purpose of developing infection control strategies.We conducted a study to understand the clinical features of children with COVID-19 and compare them with outpatients without COVID-19. The study was conducted in Toshima hospital, which is the second largest hospital among 10 medical insti-tutions for type 2 infectious disease with dedicated beds for infectious disease in Tokyo, and the largest hospital with pediatric COVID-19 patients.Material and MethodsWe extracted data from medical records between March 25th 2020 and October 31st 2020 for all patients who attended the pediatric ward diag-nosed with COVID-19 based on COVID-19 tests, children who had close contact with someone with COVID-19 who were recommended for COVID-19 testing by a public health center, and children who received a negative COVID-19 test from their attending pediatrician. Data collected included date of arrival, age, sex, diagnosis, information about contact with COVID-19, and clinical symptoms including fever, cough, rhinorrhea, pneumonia and diarrhea. Among clin-ical symptoms, pneumonia was defined as an abnormal chest X ray finding or clinician diagnosis if no chest X ray was performed. Children were divided into 3 groups: COVID-19 patients, close contacts who have COVID-19 patients in their family or close community such as school or nursery and were recommended for a COVID-19 test, and COVID-19-negative patients who had been suspected to have COVID-19. We used ANOVA to compare continuous variables (age) and chi-square tests to compare categorical variables (sex, contact with COVID-19 and clinical symptoms) between these groups. We calculated the sensitivity and specificity of symptoms including fever, cough, both fever and cough, and either fever or cough as a method for distinguishing the groups by symptoms. Stata software version 15.1 was used. The study was approved by the ethical committee of Toshima Hospital (No 2-28). Informed consent was difficult to obtain from the participants and the use of the data were informed to the guardians using documents with the opt-out approach.A total of 108 patients were included in this study. There were 30 patients with a positive result for COVID-19. PCR (polymerase chain reaction) tests were conducted in 28 of these patients, a LAMP (Loop-mediated Isothermal Amplification) test in one, and an antigen test in one patient. There were 25 patients who had contact with COVID-19 and performed screening test as close contact patients but were negative on SARS-CoV-2 PCR, and 51 patients who were suspected to be COVID-19 positive by their attending pediatrician but confirmed negative on testing. We excluded 2 patients from this study because they were born from COVID-19 positive or suspected-positive mother.Results of comparisons between three groups are shown in Table 1. There was no significant difference in age or sex between the groups. Compared to the children diagnosed with COVID- 19 or those who had contact with COVID-19, those who had negative SARS-CoV-2 results had signifi-cantly less contact with COVID-19 patients. Those with negative SARS-CoV-2 PCR results with COVID-19 had significantly more symptoms of cough and fever. Among children with COVID-19, there were 12 children without symptoms and among children with close contact, there were 13 children without symptom. There was no COVID- 19 child who met criteria of moderate or severe symptom.Diagnoses for children with suspected COVID-19 but negative SARS-CoV-2 PCR results were upper respiratory tract infection (17), pneumonia (7), asthmatic bronchiolitis (5), febrile seizure (5), Kawasaki disease (3), lymphadenitis (3), gastro-enteritis (2), urinary tract infection (2), appendi-citis (1), subacute necrotizing lymphadenitis (2), fever of unknown origin (1), sun stroke (1), poor milk feeding (1), psychologic disorder who had prolonged fever for 1 weeks but finally diagnosed 623Results