Here, I analyse data from 11 publications including the large 138-patient cohort by Wang D et al, published in JAMA. I have also included demographic details of the 26 Hong Kong patients to date into these calculations.

For calculation of mortality, a problem is that many patients are still hospitalised at the time of publication, so their final outcome is unknown. For example, 61.6% of Wang D et al’s cohort still remained hospitalised at the time of publication. But, they are assumed to survive for the purposes of the study.

Based on an analysis of the published data, the overall mortality rate of patients hospitalised with 2019-nCoV pneumonia is 8.4%. This is biased in favour of survival, but I believe this is still a reasonable figure given the relatively high ICU admission rate of 27.3%.

This analysis does not include individuals with mild illness who were not hospitalised. We do not have good quality data for this group yet.