Estimating the cure rate and case fatality rate of the ongoing epidemic COVID-19

ABSTRACT

The COVID-19 outbreak has been associated with over 250000 confirmed infections and over 11000 confirmed deaths worldwide. The epidemic poses a huge challenge to global public health security. Cure rate (CR)/case fatality rate (CFR) is an important data to assess the severity of an epidemic disease and an important consideration to evaluate the therapeutic effect. The current calculation of CR or CFR for COVID-19 epidemic is based on formula number of cure or deaths/number of diagnosed. This calculation method was called crude CR/CFR. However, there is no more scientific method to calculate the CR and CFR of the epidemic due to the changing data of the epidemic, especially the large increase of the number of diagnosed and hospitalized. The estimated CFR calculated by the basic formula is called crude CFR. Considering a large number of hospitalized patients, we revised the calculation method for CR, which is the ratio of cumulative discharges on a given day to the sum of cumulative discharges on a given day and cumulative deaths before j days is used to estimate the CR. In addition, the CFR can be estimated accordingly. According to statistical theory, if the COVID-19 epidemic event satisfies the statistical random event hypothesis, the estimated daily CR and CFR rate should be basically stable without considering the discovery of future special drugs and special medical methods. Therefore, the j value is selected with the minimum variance or coefficient of variation of the daily estimate of CR as the selection index, and the best value of j is determined by reference to clinical observation statistics. According to the results, the coefficient of variation of the estimated CR of Wuhan, Hubei, China, Hubei excluding Wuhan and China excluding Hubei reached the minimum value at j=8,10,10,10 and 12, respectively, indicating that this is the least difference in the estimated value of each day. When j=9, the CR of COVID-19 was estimated to be about 95.8% in China, while the CFR was estimated to be about 4.2%, which is lower than that of the SARS epidemic in mainland China in 2003;the CR in Wuhan was estimated to be about 94.5%, while the CFR in Wuhan was estimated to be about 5.5%. Because of the large number of cases and low cure rate in Wuhan, the cure rate in Hubei Province is lower than the national average. The CR of Hubei excluding Wuhan was similar to the national average. The CR of China excluding Hubei Province is the highest, which may be related to the factors of more imported cases, different age structure of imported cases, relatively few patients and better treatment conditions in the other provinces. This method of estimating the CR optimized the calculation method of crude CR, considering the influence of the population still in hospital on the valuation. So the comparatively accurate CR/CFR predictions for the ongoing epidemics are conducive to decision making regarding epidemic prevention and control as well as evaluating the curative effects of availed treatments.