Am J Epidemiol. 2023 Apr 7:kwad081. doi: 10.1093/aje/kwad081. Online ahead of print.
ABSTRACT
Monitoring morbidity and mortality in resurgences of respiratory infections has been underpinned with the epidemic of COVID-19 and poses significant challenges. For example, case fatality rates and deaths attributed to specific respiratory pathogens are known to suffer from significant biases undermining their comparability through time and space. As a result, it is difficult to evaluate the protective effect of public health interventions or quantify the impact of a resurgence to the general population through direct recording of COVID-19 related deaths. To overcome these limitations, it has been proposed that more robust less biased metrics, such as the all-cause deaths, can be used to monitor the effect of an epidemic over a population and over time. More specifically, metrics of excess mortality over time, which have been used for influenza surveillance in the past, are increasingly considered important for COVID-19 surveillance. Here, we discuss excess mortality surveillance focusing on standardised single-point and standardised cumulative metrics that allow comparability of excess mortality through space and time. We explain why z-score allows for comparison of excess mortality between countries and different periods, while cumulative z-score allows assessment of excess mortality over long periods. Our commentary re-iterates the importance of standardised statistics of excess mortality for COVID-19 surveillance as we move towards co-existence with SARS-CoV-2 that will allow drawing conclusions from best practices in different health systems and different periods.
PMID:37026399 | DOI:10.1093/aje/kwad081
