The interpretation of COVID-19 in cause-of-death statistics: a matter of causality

Abstract

Background: Mortality is an important indicator for estimating the impact of the COVID-19 pandemic. However, different registrations provide different figures and the question is how to interpret the number of COVID-19 deaths reported. Objective: To study the role of COVID-19 in dying in order to explain the representation of COVID-19 in cause-of-death statistics. Methods: Analysis of all death certificates mentioning COVID-19 in the Dutch cause-of-death registry during the pandemic (n=51,181). The role of COVID-19 as cause of death was studied by the way it was reported on death certificates. A calculation of odds ratios was performed for studying associations between COVID-19 and other reported causes of death.Results: In 24% of the cases COVID-19 was the only cause of death mentioned on a death certificate. In 76% of the cases, one or more other diseases played a role in dying. Three patterns emerged: COVI9 associated with 1. neurodegenerative disorders, 2. chronic respiratory disorders, and 3. metabolic disorders. Of all death certificates mentioning the diseases, COVID-19 was the start of the causal chain leading to death in 45.2% of the cases, while COVID-19 was selected for cause-of-death statistics by special World Health Organization WHO instructions in 93.9% of the cases. Conclusions: Cause-of-death statistics overestimate the role of COVI9 as underlying cause of death. In a majority of the deceased cases, there is an association of COVID-19 with other diseases not captured by cause-of-death statistics reporting (only) one cause of death per deceased. ti-causal approach is needed to evaluate the pandemic and inform health policy.

Keywords

• covid-19
• underlying cause of death
• death certificate
• cause-of-death statistics
• health policy