Select Page

February 22, 2022

In a pre-print article published recently, the authors used life table and survival analyses methods to compare probability of dying from COVID-19 against probability of dying from other causes using data from the province Hubei of China, of which Wuhan is the capitol city.

The authors conducted age-specific death rates in a life table analysis which they call a relative survival persepctive. They used data from Table 1 of a Verity et al publication. They employed right censoring.  They were able to compare 10-year conditional probabilities of dying across various ages.  They found at every age considered that the conditional probability of dying from COVID-19 was higher than the conditional probability of dying from other non-COVID-19 causes. It appeared to increase monotonically for those who had been infected.

The authors did not appear to use any kind of sophisticated survival analyses methods perhaps because of lacking data to do time-to-event analyses. However they then later state that they employed a Gompertz model which apparently told them that once infected by the Sar-Cov-2 virus, the probability of dying increases exponentially with age. I am not sure why they would not have presented this modeling along with their life table results.


Written by,

Usha Govindarajulu



COVID-19, survival analysis, Wuhan, biostatistics



David A. Swanson, Dudley Poston, Steven G. Krantz, Arni S.R. David Swanson DA, Poston D, Krantz SG, Rao ASRS. “The Age-Related Probability of Dying from COVID-19 among Those Infected: A Relative Survival Analysis.”

Verity, R., et al., (2020). Estimates of the severity of coronavirus disease 2019: A model-based analysis. Lancet 20 (6): 669-677.