How we die now.

Several years ago I installed WeCroak, an app that pushes five quotes about death and mortality to you each day. It is inspired by a Bhutanese folk saying: to be a happy person one must contemplate death five times daily. I don’t know if WeCroak has made me happier, but I think it’s healthy to reflect on the fact that we will all die as it forces us to think about how we are spending our finite time on earth. Looking at how we die as a population should force us to reflect on our values, progress and how we invest in our health and communities.

Century over century, human progress towards longevity has been inexorable. We keep living longer, healthier lives… except when wars and epidemics intervene. In the United States, the steady upward march of life expectancy held until the 2010s, when we saw both a national flattening and alarming regional declines, as the opioid epidemic swept much of the country and rising obesity and diabetes slowed the long decline in deaths from heart disease to a crawl. Cancer deaths kept falling, but not enough to offset these trends; 2018 and 2019 showed a slight recovery, and then COVID-19 hit.

Our experience with the COVID-19 pandemic and debate over the response has politicized public health like never before. While gun deaths have long been political flash points in America (with fights over whether they can be considered a public health issue), in the 2020s debates over our health - both physical and mental - have consumed media attention and possibly swung elections.

The COVID-19 spike and the return to “normal”

By the most conservative measure - fatalities where COVID-19 was recorded as the underlying cause of death - the novel virus killed 1,035,634 people from 2020 to 2024, a staggering number by any measure. Even those who would minimize the disease or litigate our response as an overreaction have to confront a drop in life expectancy unmatched since World War II, and a spike in deaths with no parallel since the 1918 flu.

One thing we understand about COVID is that it was most likely to kill the elderly, the obese and people with other pre-existing conditions. But even if we want to challenge the COVID death count and reallocate some of the 1M+ deaths to other underlying causes, the top-line spike doesn’t go away, nor does the “all other causes” represented by the blue line in this chart. Even non-COVID deaths still rose substantially. In 2021, excluding deaths with COVID as the underlying cause, both women and men died at a rate 8.3% higher than 2019. Many of those deaths were cases where COVID was also listed as a cause of death, but it is not the full picture.

What accounted for all the non-COVID excess deaths in 2021?

Our COVID narratives often start and end with the lockdowns, the puzzles, the sourdough starters and Zoom gatherings of 2020. By 2021 we were rolling out vaccines, returning to the office and moving and changing jobs at a high rate. But 2021 was our deadliest year. COVID deaths peaked in 2021, and an additional 192,000 more people died of something else compared to 2019 as isolation, rising crime, deferred care, and the strain of the disease on people already compromised took their toll. The changes in death rates in 2021 compared to 2019 across leading causes are shown below, but the top contributors were:

• 60K more deaths from injury and despair — overdoses +36K (+55%), homicide +7K (+36%), motor vehicle +8K (+20%), alcoholic liver disease +9K (+37%)
• 74K more cardiometabolic deaths — heart disease +36.5K (+5.5%), diabetes +15.6K (+18%), stroke +12.9K (+9%), hypertension +6.3K (+17%), kidney disease +2.8K
• Another 54K from old age and certification — the residual “all other diseases” +42.8K (+12%), malnutrition +5.9K (+53%), Parkinson’s +3.2K
• Those were offset by drops in deaths from chronic lower respiratory disease (−14.6K) and flu and pneumonia (−7.9K). You might recall there was no flu season that year.

Where things stand against 2019

By 2024 - the last year for which full data is available - the death rate for men had returned to 2019 levels while women were still 2% above. The good news: we saw double-digit drops in lung cancer and chronic lower respiratory disease. But deaths of despair and those resulting from unhealthy lifestyles (liver diseases, hypertensive heart disease) ran well above 2019 rates. Note: the surge in malnutrition rates is largely understood to be standardization in recording old-age deaths.

Overdoses and other deaths of despair are falling, but not for everyone and not everywhere

Overdose deaths, most of them opioids and fueled by the fentanyl crisis, peaked in 2022 and have been falling steadily since then, though less so, curiously, for the oldest population.

Homicides have fallen sharply since 2021, with deaths among the youngest groups, aged 15–24 and 25–34, falling the most. Rates for those cohorts are now below 2018 levels.

The rise in deaths from car accidents is one of the most vexing legacies of the pandemic. Rates across all age groups spiked in 2021 and 2022, and have not returned to 2018 or 2019 levels in most groups.

Suicide

When I started this work, I expected the national concern over mental health to translate to rising suicide rates. Suicide rates fell in 2020, then rose in 2021 and 2022, but now sit roughly at 2018 and 2019 levels. But the picture, by age group (seen above), by gender and state by state, is more muddied. Suicide deaths are increasingly by firearm.

Nationally, progress is uneven

We are a nation of 50 states and diverse lifestyles and polarized political views. All of the issues discussed here - COVID, homicide, overdoses - are national in scope but the reality is experienced locally. And the reality is that recovery - if that’s what it is - has been felt differently. Broadly, rural areas and Western states are seeing slower recoveries in homicide and overdose death rates than the Northeast.