with among the highest numbers of patients with COVID-19 in the country. We selected HCWs, including physicians, nurses, laboratory medical technologists, radiological technologists, and pharmacists, who worked in departments in which they had contact with patients with COVID-19, including emergency departments, general internal medicine departments, respiratory medicine departments, infectious disease departments, general wards, and intensive care units. An explanation of the sample size calculation appears in the eAppendix in the Supplement. This study was approved by the institutional review board of St Luke's International Hospital in Tokyo, Japan. A letter of informed consent was distributed to the participants via email, and completion of the questionnaire implied their consent. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.The web-based survey was generated using SurveyMonkey, a cloud-based survey development application. The survey solicited responses regarding participants' demographic characteristics (age and gender), professional history (job category and years of experience), working environment characteristics (mean weekly working hours, days off per month, and hours of sleep per day), types of anxiety perceived, changes compared with before the pandemic, and types of support needed.The primary outcome was the prevalence of burnout among frontline HCWs in departments with direct contact with patients with COVID-19, using the validated Japanese version of the Maslach Burnout Inventory-General Survey, 3 which is currently considered the criterion standard for measuring burnout. This 16-item questionnaire contains 3 subscales that evaluate what are considered the 3 major domains of burnout, ie, emotional exhaustion, cynicism (ie, depersonalization), and professional efficacy (ie, personal accomplishment). High levels of exhaustion (>3.5) plus either high cynicism (>3.5) or low professional efficacy (<2.5) were selected as the primary criteria for burnout.We first compared the baseline characteristics of those who did and did not have burnout by using the χ 2 difference test for categorical variables and the Mann-Whitney U test for continuous variables. Given the limited information available on confounding by potential risk factors for burnout, we used logistic regression analyses to evaluate whether any factors, such as types of anxiety perceived, changes compared with the prepandemic period, and types of support needed, were significant by integrating data on the participants' backgrounds into the model. All analyses