PlainHealth

Health data guide

Deaths of Despair: Suicide and Overdose in the US, 1999-2017

While the nation's two biggest killers became less deadly, suicide and drug-overdose deaths climbed sharply. Here is what the CDC NCHS data shows, and why it matters.

Key Takeaway

Between 1999 and 2017, the age-adjusted suicide rate rose about 40% and the unintentional-injury rate (driven by drug overdoses) rose about 41% - even as heart disease fell 35% and stroke fell 40%. These rising "deaths of despair" reshaped the bottom half of the leading-cause list and helped push US life expectancy down for three straight years.

Most of the story in the 1999-2017 mortality data is a story of progress. Heart disease, stroke, and cancer, the three largest causes of death, all saw their age-adjusted rates fall substantially as treatment and prevention improved. But underneath that progress, a smaller set of causes moved sharply the other way.

Suicide and drug-related unintentional injuries both rose by roughly 40% in age-adjusted terms over the period. Because these causes start from a smaller base than heart disease or cancer, the increase is easy to miss in headline numbers, but it was large enough to bend the national life-expectancy curve downward, a reversal the United States had not experienced outside of wartime or pandemic for generations.

The rise of overdose deaths

In the leading-cause rollup, the overdose epidemic shows up inside the unintentional-injuries chapter (ICD-10 V01-X59), which also contains traffic crashes and falls. Traffic deaths were roughly flat to declining over this period, so the chapter's ~41% rise was driven overwhelmingly by drug poisoning.

That rise came in waves. The first, through the 2000s, was led by prescription opioids. The second, beginning around 2010, saw many people move to heroin. The third, accelerating from about 2013, was the arrival of illicit fentanyl, which made the same dose far more lethal. The 2017 endpoint of this dataset captures the first two waves fully and the early part of the third.

Suicide: the quieter crisis

Suicide is the smallest of the ten tracked causes by national count, but it rose almost every year in the dataset, climbing from roughly 11.8 to 16.5 deaths per 100,000 in age-adjusted terms. The increase was broad rather than concentrated in any single group, and it was steep enough to keep suicide near the top of the list of fastest-rising causes for the entire period.

State-level variation is large. Mountain West and rural states tend to report the highest suicide rates, a pattern visible on the highest suicide rate ranking and on individual state profiles.

Why medical progress did not reach these causes

The contrast with heart disease is instructive. Cardiovascular deaths fell because there were specific, scalable interventions: blood-pressure and cholesterol drugs, smoking cessation, and faster emergency cardiac care. Deaths of despair have no equivalent single lever. Their drivers are partly social and economic, partly tied to the drug supply, and they respond to policy, treatment access, and community conditions rather than to one medical breakthrough.

That is the core insight behind the deaths-of-despair framing: it asks readers to see suicide, overdose, and alcohol-related deaths as a connected pattern rather than as isolated tragedies, and to notice that the country was capable of dramatic progress against some causes while losing ground on others at the same time.

Explore the data

Frequently asked questions

What are "deaths of despair"?

The term, coined by economists Anne Case and Angus Deaton, groups together deaths from suicide, drug overdose, and alcohol-related liver disease. The idea is that these causes share underlying drivers, economic dislocation, social isolation, and declining opportunity, rather than a single medical pathology. In the CDC NCHS leading-cause data, the clearest signals are the suicide chapter (X60-X84) and the unintentional injuries chapter (V01-X59), which is dominated by drug overdoses.

How much did suicide rates rise from 1999 to 2017?

The national age-adjusted suicide rate rose roughly 40% between 1999 and 2017, from about 11.8 to 16.5 per 100,000. Unlike heart disease and cancer, which improved steadily over the same period, suicide moved in the wrong direction almost every year, making it one of only three leading causes whose age-adjusted rate increased.

Are drug overdoses counted as unintentional injuries?

Most are. In the leading-cause rollup, the large majority of drug-overdose deaths fall under unintentional injuries (V01-X59), alongside motor-vehicle crashes, falls, and poisonings. The age-adjusted rate for this whole chapter rose about 41% from 1999 to 2017, and overdoses, not traffic deaths, drove almost all of that increase.

Why did these deaths rise while heart disease and cancer fell?

Cardiovascular and cancer deaths fell because of medical progress: statins, blood-pressure control, smoking declines, earlier screening, and better treatment. Deaths of despair have different roots that medicine alone does not fix, the prescription-opioid wave of the 2000s, the spread of heroin and then illicit fentanyl, and broader social and economic strain. The two trends are largely independent, which is why one set of causes could improve sharply while another worsened.

Did these trends affect US life expectancy?

Yes. The rise in overdose and suicide deaths was large enough to contribute to a decline in overall US life expectancy in 2015, 2016, and 2017, three consecutive years of decline, something not seen for decades. It is a rare example of a non-pandemic public-health reversal visible directly in the leading-cause data.

Does this data include the fentanyl and pandemic era?

Only partly. The PlainHealth series ends in 2017, which captures the prescription-opioid and early-heroin waves and the start of the fentanyl wave, but not its peak or the further surge during 2020-2021. For the most recent overdose and suicide figures, consult CDC WONDER directly.