ABSTRACT
The most widely used measure of covid mortality is a headcount ratio of deaths due to covid, as captured by the case fatality rate, which is the ratio of covid deaths to covid cases. This is a relative measure of mortality, in contrast to the absolute measure of an aggregate headcount, as captured by the gross or aggregate fatality, which is just the raw (non-normalized) number of covid deaths. The present note examines two elementary principles which a measure of mortality (like one of poverty or urbanisation or unemployment) might be expected to satisfy. These are what are called the probability principle and the subgroup consistency principle respectively. Headcount ratios are found to satisfy the first principle but not the second, and aggregate headcounts to satisfy the second principle but not the first, which makes neither variety of a headcount measure satisfactory on its own, and by itself. This note advances the case of a “mixed” measure, as intermediate between ratio and aggregate measures, expressed as a geometric mean of the case fatality rate and the gross fatality. The ranking of countries by mortality is found to be a variable function of the precise mortality indicator employed.
Acknowledgement
The author acknowledges helpful comments from an anonymous referee.
Disclosure statement
The author has no potential conflict of interest to report.
Additional information
Notes on contributors
S. Subramanian
S. Subramanian is a former professor from the Madras Institute of Development Studies and a former National Fellow of the Indian Council of Social Science Research. He has research interests in aspects of social and economic measurement, collective choice theory, and development economics. He is a Fellow of the Human Development and Capability Association.