Excess Deaths From COVID-19 and Other Causes in the US, March 1, 2020, to January 2, 2021

A study analyzing US mortality in March-July 2020 reported a 20% increase in excess deaths, only partly explained by COVID-19. Surges in excess deaths varied in timing and duration across states and were accompanied by increased mortality from non–COVID-19 causes.¹ This study updates the analysis for the remainder of 2020.

The Supplement details the methods. A Poisson regression model used mortality data from 2014-2019 to predict US expected deaths in 2020. Observed deaths in weeks ending March 1, 2020, through January 2, 2021, were taken from provisional, unweighted death counts for the District of Columbia and 49 states, excluding North Carolina for insufficient data. Data sources included the National Center for Health Statistics and US Census Bureau. Data for 8 geographic regions were grouped into distinctive surge patterns. COVID-19 deaths included all deaths for which COVID-19 was cited as an underlying or contributing cause.

Temporal changes in mortality rates from non–COVID-19 causes (eg, Alzheimer disease/dementia, heart disease, diabetes, and 9 other grouped causes; see Supplement) were examined. Data included all deaths in which non–COVID-19 conditions were listed as the underlying cause of death (potentially including deaths for which COVID-19 was a contributing cause). The Joinpoint regression program version 4.8.0.1 (Statistical Research and Applications Branch, National Cancer Institute) was used to specify the weeks (joinpoints) when slopes changed (measured by the annual percentage change [APC]) and their statistical significance (2-sided test, .05 threshold).

Between March 1, 2020, and January 2, 2021, the US experienced 2 801 439 deaths, 22.9% more than expected, representing 522 368 excess deaths (Table). The excess death rate was higher among non-Hispanic Black (208.4 deaths per 100 000) than non-Hispanic White or Hispanic populations (157.0 and 139.8 deaths per 100 000, respectively); these groups accounted for 16.9%, 61.1%, and 16.7% of excess deaths, respectively. The US experienced 4 surge patterns: in New England and the Northeast, excess deaths surged in the spring; in the Southeast and Southwest, in the summer and early winter; in the Plains, Rocky Mountain, and far West, primarily in early winter; and in the Great Lakes, bimodally, in the spring and early winter (Figure). Excess deaths were increasing in all regions at the end of 2020. The 10 states with the highest per capita rate of excess deaths were Mississippi, New Jersey, New York, Arizona, Alabama, Louisiana, South Dakota, New Mexico, North Dakota, and Ohio. New York experienced the largest relative increase in all-cause mortality (38.1%). Deaths attributed to COVID-19 accounted for 72.4% of US excess deaths.

Joinpoint analyses revealed an increase in weekly mortality from non–COVID-19 causes, including heart disease from March 15 to April 11, 2020 (APC, 4.9 [95% CI, 0.7-9.3]), and October 11, 2020, to January 2, 2021 (APC, 1.1 [95% CI, 0.8-1.4]); Alzheimer disease/dementia from March 15 to April 11, 2020 (APC, 7.1 [95% CI, 2.4-12.0]), May 31 to August 15, 2020 (APC, 1.2 [95% CI, 0.7-1.6]), and September 6, 2020, to January 2, 2021 (APC, 1.3 [95% CI, 1.1-1.5]); and diabetes from March 8 to April 11, 2020 (APC, 6.5 [95% CI, 2.8-10.3]), May 31 to July 11, 2020 (APC, 2.6 [95% CI, 0.2-5.0]), and October 18, 2020, to January 2, 2021 (APC, 2.2 [95% CI, 1.6-2.8]).

The 22.9% increase in all-cause mortality reported here far exceeds annual increases observed in recent years (≤2.5%). The percentage of excess deaths among non-Hispanic Black individuals (16.9%) exceeded their share of the US population (12.5%), reflecting racial disparities in COVID-19 mortality. Excess deaths surged in the east in April, followed by extended summer and early winter surges concentrated in southern and western states, respectively. Many of these states weakly embraced, or discouraged, pandemic control measures and lifted restrictions earlier than other states.

Excess deaths not attributed to COVID-19 could reflect either immediate or delayed mortality from undocumented COVID-19 infection, or non–COVID-19 deaths secondary to the pandemic, such as from delayed care or behavioral health crises. Death rates from several non–COVID-19 diseases (eg, heart disease, Alzheimer disease) increased during surges. The model does not adjust directly for population aging, which could contribute to an overestimate of excess deaths. Other study limitations include reliance on provisional data, inaccurate death certificates, and modeling assumptions.