Background: Most COVID-19 patients cannot provide a clear exposure time; therefore, this study was designed to predict the progression of COVID-19 by using the definite departure time from Wuhan.Methods: In this retrospective study, all cases were selected from Northwestern China, which has the lowest population density. As our study endpoints, the incubation period was defined as the date of departure from Wuhan City to the date of symptom onset; we defined the confirmed time as the interval from symptom onset to positive results (samples from the respiratory tract). Both of them were estimated by fitting a Weibull distribution on the departure date and symptom onset. The differences among the variables were analyzed.Results: A total of 139 patients were ultimately enrolled, and ~10.1% of patients (14 patients) had no symptoms during their disease course. We estimated the median incubation period to be 4.0 days (interquartile intervals, 2.0–8.0), and the 95th percentile of the distribution was 15.0 days. Moreover, ~5.6% of patients (7 patients) experienced symptoms 2 weeks after leaving. Furthermore, the estimation median interval from symptom onset to final diagnosis was 4.0 days (interquartile intervals, 2.0–6.0), and the 95th percentile of the distribution was 12.0 days. Finally, the median hospitalization time was 16.0 days, ranging from 3.0 to 45.0 days. Univariate analysis showed that age (P = 0.021) and severity status (P = 0.001) were correlated significantly with hospitalization time.Conclusions: We provide evidence that departure time can be used to estimate the incubation and confirmed times of patients infected with COVID-19 when they leave an epidemic area.
Read full abstract