Canonical Source · peer reviewed study

COVID-19 Death Determination Methods, Minnesota, USA, 2020–2022

This page documents how this source is reused across Claim by Claim. A source can support, challenge, qualify or contextualize different claims; appearing here does not imply endorsement of every linked proposition.

Publisher: Emerging Infectious Diseases
Year: 2024
DOI: 10.3201/eid3007.231522
Source type: peer reviewed study
Source level: Primary source
Knowledge backlinks: 1

Open original source ↗ Back to Evidence Library Explore connections →

Evidence context

How this source is used

Evidence roles describe what this source helps establish inside each Investigation. They are local to that use, not permanent labels attached to the publication itself.

Investigation

Was the U.S. COVID-19 death toll substantially inflated by 'with COVID' deaths?

Challenges

A Minnesota comparison of death certificates with a more intensive standardized mortality definition found over 90% agreement for most groups, with undercounting observed more often than overcounting.

Minnesota's 14,004-death analytic comparison recorded 13,108 concordant COVID deaths, 483 case-defined COVID deaths absent from certificates and 413 COVID-attributed certificates ruled out. This is not a national misclassification fraction.

Qualifies

The Minnesota analysis also identified both ruled-out deaths that mentioned COVID-19 and confirmed COVID-19 deaths omitted from death certificates, demonstrating real bidirectional classification error.

Supports caution about individual certificates without supporting large net national overcount.

Subject families

Health & medicine