Excess Mortality in the U.S. in 2020

Data and Metrics

The oldest Americans bore the brunt of COVID-19 mortality, but working-age Americans suffered the vast majority of excess non-COVID deaths, most commonly resulting from external causes.

Today we are thrilled to host Nerdy Guest Dana Glei, PhD, discussing her research on “excess mortality,” including an increase in drug-related mortality during the pandemic.

Dr. Glei completed her PhD in sociology from Princeton University specializing in demography (overlapping with Nerdy Girls Lauren and Jenn!). As her life partner likes to say, “Demographers study people being born, people moving, and people dying. Dana focuses on the people dying part.” She works for Georgetown University as a Senior Research Investigator, as well as with the Human Mortality Database project (www.mortality.org) at the University of California-Berkeley and other clients. She is a member of the Committee on Population (CPOP) for the National Academy of Sciences.

Currently, her research focuses on disparities in health and how risk-taking behavior such as smoking and drug abuse effect mortality.

1.) What is excess mortality? Was most excess mortality during 2020 due to COVID-19?

DAG: “Excess mortality” means mortality rates that are higher than we would expect based on historical trends. During the current pandemic, the number of deaths has been much higher than it was in the months and years before the pandemic. Prior studies have estimated there was 16-23% excess mortality during the pandemic in the U.S. Most of those excess deaths were listed as COVID-19, but other COVID-19 deaths were likely incorrectly assigned to another cause, especially early in the pandemic when testing was limited. The existing published estimates (from NCHS and Woolf) for the total number of excess deaths during 2020 range from 428K-523K. ALL the estimates of excess mortality in the US are substantially higher than the # of deaths from COVID-19.
Scientists have estimated that the percentage of excess deaths that can be directly linked to COVID-19 ranges from 72% to 89%. However, among younger Americans (aged 25-44), that percentage is much lower (38%). That is, nearly 2/3 of excess deaths among younger Americans appear to have resulted from causes other than COVID-19.

2.) So excess mortality from non-COVID-19 causes differed by age?

DAG: Yes. The oldest Americans were much more likely to die of COVID-19 than younger people. But if you look at excess deaths from other causes, the vast majority were at working ages (25-64). Injury-related deaths (also called external causes) accounted for the largest share of non-COVID excess deaths, nearly 80% of which occurred among working age adults. While some of the excess non-COVID deaths could be COVID-19 deaths that were incorrectly assigned to some other cause such as influenza or pneumonia, that is unlikely to explain the increase in mortality from external causes. Instead, these deaths are likely to be an indirect result of economic distress, the disruption of normal life, and heightened uncertainty related to the pandemic.

3.) Which external causes of death contributed most to 2020 excess mortality?

DAG: It looks like the rise in injury-related deaths may be driven primarily by drug-related mortality. Mortality rates from drug overdoses showed the biggest percentage increase, peaking at 24% higher than expected in May 2020. There was also an increase in homicides (peaking at 19% higher than expected in August), which might also be related to the drug epidemic.
Early in the pandemic, many predicted that suicides would increase due to social isolation and other impacts of the pandemic on mental health. Contrary to expectation, suicides declined from March-October 2020 (we don’t have data yet beyond October 2020 because of the time it takes to investigate and register suicides).

Mortality rates from motor vehicle accidents were also lower than expected in March (-19%), April (-30%), and May (-16%). The available numbers from more recent months are not significantly different from expected.

Thus, it looks like drug-related mortality and homicides were the biggest culprits of excess mortality from external causes. The social and economic upheaval resulting from the pandemic may have exacerbated the drug epidemic, but that problem began well before the pandemic. In fact, drug overdoses increased every month after February 2019; that is, they were increasing for a full year before the pandemic was declared. Even if it were possible to vaccinate everyone and eradicate the virus, that will not solve the US drug epidemic. Unfortunately, there is no forthcoming vaccine for addiction.

A recent article in the Scientific American argued that the drug epidemic was exacerbated by state and federal policies that cracked down on opioid prescribing without offering addicts any treatment and without providing chronic pain sufferers any good alternative to relieve their pain (see link below). They reduced the supply of prescription opioids, but it was replaced with street heroin—which was much cheaper—and then later by fentanyl and other synthetic opioids—which are much more potent and easier to hide. As a result, drug-related mortality has skyrocketed.
One strategy to reduce drug-related mortality is to make drug treatment widely accessible without imposing a stigma. Many overdose deaths have been averted by use of medication that can reverse an opioid overdose (i.e., Naloxone), but timing is of the essence. Fentanyl can kill a person in 5-15 minutes whereas heroin can take hours. More importantly, Naloxone cannot prevent deaths resulting from worrisome increases in the use of methamphetamine, cocaine, and drugs used to treat anxiety/insomnia (i.e., benzodiazepines).

We also need to do more than simply treat people after they are already addicted. To effectively combat the drug epidemic, we must address the underlying root issues that make people vulnerable to substance abuse, whether that is mental illness, physical pain, poor job prospects for Americans with little education and few job skills, weak networks of social support, homelessness, structural racism, or other factors. Indeed, many of these problems are intertwined. Trying to deal with them one at a time would be like playing Whack-a-Mole. In the same way that cutting off the supply of Oxycontin and other prescription opioids resulted in a huge increase in use of street heroin, it would be futile to distribute Naxolone widely without helping people who have mental disorders, chronic pain, and/or difficulty making a living. We need to avoid Whack-a-Mole policies. Instead, policies need to address the cluster of social and economic factors that make life difficult for so many Americans. If people have work that gives their life meaning and provides enough money for them to afford housing, food, health care, and some of the amenities that make life worth living, that might solve not only the drug epidemic, but also many related problems. The drug epidemic is only a symptom of larger underlying issues.


During 2020, US mortality rates were up to 20% higher than in prior years. For older people (age 65+), the vast majority of the excess deaths were due to COVD-19. Among working-age Americans (25-64), the largest share of excess deaths resulted from external causes, probably driven by drug overdoses and homicide.


We’re Overlooking A Major Culprit in the Opioid Crisis

San Francisco Contends With a Different Sort of Epidemic: Drug Deaths

Link to Original FB Post