A novel cohort analysis approach to determining the case fatality rate of COVID-19 and other infectious diseases

Charit S Narayanan

However, numerous factors pose challenges to calculating the CFR using the cohort 191 method. Firstly, the date of infection is difficult to determine since the onset-to-death 192 interval is still uncertain. This is because there are numerous stages to contracting a Additionally, the CFR varies significantly with factors such as age, smoking habits, 199 etc. If the proportions of differing-risk groups in each infected cohort varies, the CFR 200 will be mis-representative [ [9] , [10] ] 201 Finally, the risk of mortality varies by place. With any rapidly spreading contagion, 202 the samples chosen for determining the CFR are biased since the virus propagates so 203 quickly, affecting different people in different places. So far, the CFR is heavily skewed 204 by the numbers from Hubei Province, where the majority of cases originate. As the 205 virus moves to different places, the mortality will inherently be different, and must, 206 therefore, be representative of the threat posed so adequate precautions can be taken.