Predictive Factors for the Diagnosis of Coronavirus Disease 2019

[ABSTRACT] [Objective] : We evaluated patients suspected or confirmed to have coronavirus disease 2019 (COVID-19) to determine predictive factors for the diagnosis of COVID-19. [Methods] : We conducted a retrospective cohort study at The Jikei University Hospital, Tokyo, Japan. This study included adult patients who underwent medical examination for suspected or confirmed COVID-19 in April and May 2020. We analyzed the clinical characteristics, blood test results, and findings of computed tomography of the chest from the medical record system of the hospital. [Results] : Of the 267 patients included in this study, 27 were found to be positive for COVID-19 on reverse transcription polymerase chain reaction testing for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Of the patients, 128 (47.9%) were men, and the median age was 47 years (interquartile range, 34.5-65). Twenty-two (8.2%) patients had a history of close contact with a COVID-19 patient. The most common symptoms were fever, general malaise, and cough. Multivariate analysis with the logistic regression model revealed that close contact, fever for 4 or more days, dysgeusia, and dysosmia were independent predictive factors for reverse transcription polymerase chain reaction test results being positive for SARS-CoV-2. [Conclusion] : Patients who have had close contact with a COVID-19 patient, fever for 4 days or more, dysgeusia, or dysosmia should undergo diagnostic testing for SARS-CoV-2.