Tracking all cause of death and estimating excess mortality during the COVID-19 pandemic: statistical and computational tools

Abstract

This was a presentation of statistical methodology I developed to track all cause of death mortality over time and estimate the excess mortality during the COVID-19 pandemic. The raw data are counts of all cause deaths over a given period (e.g., a month) as they varies with calendar time. Data from the five years pre pandemic was used to estimate the (counter-factual) number of deaths during the pandemic perios had the pandemic not occured. The difference between the recorded all cause of death mortality and the expected all cause of death mortality was defined to be the excess deaths. The talk discusses the estimation and the quantification of uncertainty. It was presented to the World Health Organization Head Quarters HQ Technical Advisory Group.