This visualization shows the growth in the number of cases in log-linear scale. A straight line indicates that the growth is exponential and double every X days. You can set the number of days to include in calculating the trend by using the slider.
Improve this visualization on ObservableHQ: https://observablehq.com/@yy/covid-19-spreading-trends
This visualization shows the temporal evolution of national-level confirmed cases and the number of deaths, for the countries with at least one death. You can mouse-over to see the past trajectory.
Please note that the case fatality is affected by various factors, particularly at the early stage of the outbreak. First of all, the progression from contracting the disease to death takes time. Therefore, the fatality rate would increase for a while even after the number of cases stop increasing. Also the early-stage fatality rate may either be underestimated if there are aggressive testing (many early detected cases, but not yet progressed to deaths) or overestimated (deaths occur before adequate testing can be done).
Because of strong age-related fatality rate variance, population structure (e.g. Italy has more elderly population than South Korea, although this is not enough to explain the high fatality in Italy) or the nature of early clustered outbreaks (e.g. South Korea vs. US) may significantly affect the case fatality rate, especially at the early stage.
Yet, fatality rate is an important indicator that reflects many serious scenarios that require attention, such as (i) large, undetected community spreading, (ii) extreme stress to the healthcare system (more preventable deaths), or (iii) emergence of more aggressive subtype.
Improve this visualization on ObservableHQ: https://observablehq.com/@yy/covid-19-fatality-rate