The US Midlife Mortality Crisis Continues: Excess Cause-specific Mortality During 2020

Dana A. Glei


Am J Epidemiol. 2022;191(10):1677-1686. 

In This Article

Abstract and Introduction


Whether monthly excess mortality in the United States during 2020 varied by age and cause of death is investigated in this analysis. Based on national-level death counts and population estimates for 1999–2020, sex-specific negative binomial regression models were used to estimate monthly cause-specific excess mortality by age group during 2020. Among men, 71% non-COVID excess deaths occurred at working ages (25–64 years), but those ages accounted for only 36% of non-COVID excess deaths among women. Many excess deaths resulted from external causes (particularly among men), heart disease, diabetes, Alzheimer disease (particularly among women), and cerebrovascular disease. For men, the largest share of non-COVID excess deaths resulted from external causes, nearly 80% of which occurred at working ages. Although incorrectly classified COVID-19 deaths may explain some excess non-COVID mortality, misclassification is unlikely to explain the increase in external causes of death. Auxiliary analyses suggested that drug-related deaths may be driving the increase in external mortality, but drug overdoses were already increasing for a full year before the pandemic. The oldest Americans bore the brunt of COVID-19 deaths, but working-age Americans, particularly men, suffered substantial numbers of excess non-COVID deaths, most commonly from external causes and heart disease.


Older Americans have suffered the highest coronavirus disease 2019 (COVID-19) mortality. Yet, the pandemic may have exacerbated deaths from other causes as well, some of which could have taken a heavier toll among younger Americans. Indeed, only 38% of the excess mortality (i.e., from all causes including COVID-19) among Americans aged 25–44 years during March–July 2020 could be attributed directly to COVID-19.[1] At these younger ages, nearly two-thirds of the excess mortality associated with the pandemic appears to have resulted from causes other than COVID-19.[1] Thus, although the oldest Americans suffered most of the direct effects of the pandemic, working-age Americans may have been more heavily affected by the indirect impact of the pandemic on non-COVID mortality.

Although much less likely than older Americans to die of COVID-19, younger Americans bore the brunt of the economic consequences because their work lives were abruptly altered. Essential workers continued to be in high demand but faced additional stressors (e.g., fear of exposure to the virus; shortages of personal protective equipment). An additional threat was posed by recalcitrant individuals who resisted public health orders, particularly when enforced by low-wage workers (e.g., grocery and retail store employees). Business owners had to grapple with logistical challenges resulting from new public health regulations and the financial aftermath of reduced demand for their services. Other workers suffered sudden income decline because of job loss or reduced hours. The most fortunate workers retained their jobs with reduced exposure to SARS-CoV-2 but had to adapt to working from home and learning new technologies.

The unexpected closure of childcare centers and schools further exacerbated the conflicts between work and family life for young and midlife Americans. With no warning, parents of young children lost access to childcare services and became responsible for helping to homeschool their children while continuing to juggle work demands and the potential needs of their own parents, who were particularly vulnerable to COVID-19.

Depending on the period of coverage and methodology, prior studies have reported 16% to 23% excess mortality during the pandemic,[2–4] but only 72% to 89% of the excess deaths could be attributed directly to COVID-19.[2–5] During the period from January 26, 2020, to October 3, 2020, Americans aged 25–44 years had the largest relative increase in excess mortality, whereas those aged ≥85 years experienced a smaller increment.[6] Excess mortality peaked earlier—in April—and declined more rapidly for Americans aged ≥45 years than for their younger counterparts (aged 25–44 years), among whom excess mortality was persistently high throughout early April to early August, peaking in mid-July.[6]

It remains unclear how excess mortality from causes other than COVID-19 is distributed by cause and the extent to which it varies by sex and age. Compared with 2019, the absolute number of deaths in 2020 increased for various causes (i.e., diabetes; unintentional injuries; Alzheimer disease; stroke; influenza and pneumonia; and heart disease).[7] That same report showed virtually no change in the number of deaths as a result of cancer or kidney disease and revealed a small decline in the number of deaths from chronic lower respiratory diseases and suicide. Other evidence suggested that the number of deaths due to drug overdoses,[8] homicides,[9] and motor vehicle accidents[10] also increased during the pandemic. The extent of change in alcohol-related deaths for the United States as a whole remains unknown, but some evidence showed increased alcohol consumption, particularly among women and Americans aged 30–59 years,[11] and provisional data for Minnesota implied that alcohol-attributable deaths were 25% to 65% higher in June–December 2020 than in the same months in 2019.[12] To the author's knowledge, no US study has been published in which researchers investigated the extent to which excess mortality for specific causes of death varies by sex and age.

In this analysis, the changes in US mortality for selected causes of death during the COVID-19 pandemic up to December 31, 2020, were quantified. In addition, the models were used to evaluate the extent to which the levels of excess mortality varied by sex, age group, and across the months of 2020.