A mathematician explains mortality rate and more. Mortality from COVID-19 may end up settling between 0.5% and 2% of people who have symptoms, a mathematician told the New York Times. Caveat: Those numbers are based on the best available data, and the best available data is still very incomplete. The prediction also includes assumptions such as how many unreported cases of the disease are out there. Mathematician Adam Kurcharski, who studies the spread of disease outbreaks at the London School of Hygiene & Tropical Medicine, made the estimate to a New York Times reporter in an article in which he explained why the mortality numbers in the early stages of a disease are so fuzzy. One major reason, he said, is that many countries may be missing large numbers of cases. Iran, where the death toll has topped 100, is an example of a country where fatalities started before officials began looking for the disease. The United States, where the death toll stands at 12 and the number of confirmed cases at just over 230, has also seen delays in testing, meaning there are likely many infections that have gone unreported. Another reason is that dividing the current number of known cases by the current number of deaths isn't accurate, Kurcharski added; people who die have been sick for two or three weeks. The delay between infection and death means the true case-fatality rate won't be known until well into the epidemic. "Ideally, we would monitor a large group of people from the point at which they develop symptoms until they later die or recover," Kurcharski told the Times. To understand the delay, Kurcharski said, imagine a disease with a case-mortality rate of 1%. When the first person dies of the disease in the hospital, you can assume that three weeks ago, when he or she became ill, there were about 100 cases of the illness circulating. After three weeks of undetected circulation, there will certainly be more cases out there: Perhaps 500, if the case numbers doubled each week. Source
