Expert Commentary

‘Deaths of despair’: Research on opioid crisis origins and the link between minimum wages and suicide reduction

Recent research details the rise of OxyContin in the U.S. and how three pieces of paper spared some parts of the country from the worst of the opioid crisis.

deaths of despair
(Miro Polca / Unsplash)

Late last year, the National Center for Health Statistics released provisional data showing a large increase in drug overdose deaths from April 2020 to April 2021.

The center estimates 100,306 drug overdose deaths in the U.S. over that time, a 29% rise from 78,056 overdose deaths over the same period one year prior.

Drug overdose deaths are one part of what Princeton University economists Anne Case and Angus Deaton term “deaths of despair.” Liver disease from alcohol use and suicide are the other contributors.

Case and Deaton first described the contours of deaths of despair in the U.S. in an influential 2015 paper and a follow-up paper in 2017. Their fundamental assertion is the rise in deaths from drugs, alcohol and self-harm was spurred by decades of decline in economic opportunity for middle-aged people with relatively low levels of education.

The bulk of those additional 20,000-plus deaths during the first throes of the COVID-19 pandemic were related to overdoses of fentanyl, the synthetic opioid 50 to 100 times more potent than morphine, according to the data. The figures include accidental and intentional overdoses.

“Many of these deaths also involved other drugs, such as methamphetamine and cocaine, with which fentanyl-related substances may be mixed — often without the knowledge of the people who take them,” Nora Volkow, director of the National Institute on Drug Abuse, testified during a December 2021 Congressional hearing.

Total deaths from cirrhosis and other liver ailments caused by alcohol use have steadily risen since the late 1990s, peaking at 29,505 in 2020, the most recent year available. The death rate from liver disease related to heavy drinking reached 9 per 100,000 people in 2020, up from 4.3 per 100,000 in 1999.

Suicide has likewise trended upward over the past two decades, but fell from a high of 48,312 deaths in 2018 to 45,940 last year. The suicide rate remained elevated in 2020 at 13.9 per 100,000 people, up from 10.5 per 100,000 in 1999.

“The increase in deaths of despair was almost all among those without a bachelor’s degree,” Case and Deaton write in their March 2020 book on the topic. “Those with a four-year degree are mostly exempt; it is those without the degree who are at risk.”

For some population subgroups, the rise of deaths of despair in the U.S. has been substantial, even before the pandemic. In their 2017 paper, “Mortality and Morbidity in the 21st Century,” Case and Deaton note the mortality rate for white people aged 50 to 54 with only a high school diploma was nearly one-third lower in 1999 than for Black people in the same age group with any level of education.

By 2015, the situation had flipped: using the same categories, the mortality rate was nearly one-third higher for white people than for Black people.

“Ultimately, we see our story as about the collapse of the white working class after its heyday in the early 1970s, and the pathologies that accompany that decline,” they write.

If you’re feeling suicidal, please talk to somebody. You can reach the National Suicide Prevention Lifeline at 1-800-273-8255; the Trans Lifeline at 877-565-8860; or the Trevor Project at 866-488-7386. Text “START” to the Crisis Text Line at 741741. If you don’t like the phone, consider using the Lifeline Chat. The Veterans Crisis Line connects service members and veterans in crisis, as well as their family members and friends, with qualified Department of Veteran’s Affairs responders through a confidential toll-free hotline, online chat, or text messaging service. Dial 1-800-273-8255 and Press 1 to talk to someone or send a text message to 838255 to connect with a VA responder. You can also start a confidential online chat session at Veterans Crisis Chat. This text was adapted from the American Association of Suicideology’s reporting recommendations and the National Institute of Mental Health.

Recent research suggests deaths of despair not exclusively a white, working class problem

Subsequent research from Case and Deaton and others suggests deaths of despair are not exclusively a white, working-class problem. Steven Woolf, a professor of family medicine and population health at Virginia Commonwealth University, and Heidi Schoomaker, a student at Eastern Virginia Medical School, Norfolk, explore the geographic and racial contours of deaths of despair in a November 2019 paper in the Journal of the American Medical Association, building on a 2018 paper by Woolf and others.

Woolf and Schoomaker find, for example, that from 1999 to 2017, rural, white populations in the U.S. experienced an increase in midlife drug overdose deaths of 749% — from 4 deaths per 100,000 people to nearly 34 deaths per 100,000. But midlife drug overdose deaths rose sharply for white people living in metropolitan counties, too — a 531% increase over the same period. The drug overdose death rate among that group moved from 6.7 deaths per 100,000 to 42.5 deaths per 100,000.

deaths of despair
Abandoned warehouse in Loveland, Colorado. (Tara Evans / Unsplash)

The highest increase in drug overdose deaths for middle-aged white people happened in the suburbs of cities with more than 1 million residents. In those suburbs, the overdose death rate for midlife white people jumped 858%, from 4.7 deaths per 100,000 to 45.2 deaths per 100,000. Midlife drug overdose death rates for Hispanic people also peaked highest in the suburbs of large cities, according to the paper.

For Black people, the overdose death rate increased the most in cities with less than 250,000 people — but among Black people aged 55 to 64, those living in large cities experienced the biggest increase. Suicides were up the most for Black and white adults in rural counties. For Hispanic people and American Indians, suicides rose the most in metropolitan areas.   

“The causes of economic despair may be more nuanced; perceptions and frustrated expectations may matter as much as absolute income or net worth,” Woolf and Schoomaker write in their 2019 paper. The rate of increase of deaths of despair, they note, was generally lower on the coasts and in states with strong economic growth, such as Texas.

The crumbling economic conditions that Case and Deaton point to, particularly the declining availability of well-paying jobs for people without a college degree, fall under what economists call demand-side factors. The idea is that despair stemming from a lack of good jobs has led people to demand drugs and alcohol, and has led to self-harm.

As economists, Case and Deaton naturally focus on the financial difficulties individuals and families face. But in their 2020 book and academic writings, they also explore myriad social changes that come into play. As they put it in their book, “we see the decline in wages as slowly undermining all aspects of people’s lives.”

Wages adjusted for inflation have lost ground for white men without a college degree since 1979, according to Case and Deaton. Plus, “the wage decline has come with job decline — from better jobs to worse jobs — with many leaving the labor force altogether because the worse jobs are unattractive, because there are few jobs at all, or because they cannot easily move, or some combination of these reasons,” they write.

The supply-side narrative centers on the supply of opioids rising in the U.S. at the stroke of a doctor’s pen. With more opioids available, more people were exposed to them, meaning more people became addicted. Journalists have covered the supply side angle in books and articles, but it’s less explored in academic research.

“Although there is general agreement that the causes of the crisis include a combination of supply- and demand-side factors, and interactions between them, there is less consensus regarding the relative importance of each,” writes Temple University health and labor economist Catherine Maclean and co-authors in a National Bureau of Economic Research working paper published November 2020.

Case and Deaton, in their own National Bureau of Economic Research working paper from September 2021, offer that demand for opioids came first. “Pain and despair created a baseline demand for opioids, but the escalation of addiction came from pharma and its political enablers,” they write.

That’s not to say there is no supply-side research.

Marketing an opioid crisis

The rise of drug overdoses in the U.S. is, in part, a story of marketing. Marketing is the process by which a company connects a product to a market. A market is a group of people who want — or need — to buy a product.

Successful marketing is achieved through the act of selling. Selling is about convincing consumers that a product is worth their money. The proposition becomes more of a slam dunk if someone else, such as a health insurance company, is paying for it. Selling is yet more effective if trusted messengers, such as doctors, vouch for the product.

Here, the drug is oxycodone. The product is OxyContin. The company is Purdue Frederick.

OxyContin hit the pain medication market in 1996, with promotion and sales “managed by the company’s marketing arm, Purdue Pharma, launched in the nation’s best-known corporate tax haven — Delaware,” writes journalist Beth Macy in her 2018 book, Dopesick: Dealers, Doctors and the Drug Company that Addicted America. “Purdue Pharma touted the safety of its new opioid-delivery system everywhere its merchants went.”

One recent academic paper is among the first to use real-world evidence to explore supply factors that have fueled the opioid crisis. The August 2021 research in the Quarterly Journal of Economics finely traces how Purdue Pharma marketed OxyContin to American doctors and patients for decades, and how three pieces of paper saved parts of the country from being quite so hard hit by opioids.

When Purdue Pharma introduced OxyContin in the mid-1990s, there were five states requiring that every prescription have three copies. The prescribing doctor kept one copy, the pharmacy kept another and sent the third to a state drug-monitoring program.

Prescriptions, in other words, had to be filled out in triplicate.

California, Idaho, Illinois, New York, and Texas were triplicate states in 1996, accounting for roughly one-third of the nation’s adult population at the time. The Department of Justice classifies oxycodone as a schedule II drug, meaning it has a “high potential for abuse which may lead to severe psychological or physical dependence.”

“An earlier line of academic research found these triplicate programs had a really chilling effect in the prescribing of these schedule II drugs,” says Abby Alpert, one of the paper’s authors and an assistant professor of health care management at the University of Pennsylvania. “That suggested, first of all, doctors didn’t want their prescribing monitored by the government. And it was also just an additional hassle for doctors.”

deaths of despair
(Karine Germain / Unsplash)

Purdue Pharma took notice as it conducted research into marketing OxyContin as a viable pain management option for patients without cancer. Oxycodone would be delivered via a time-released pill, with the drug entering the bloodstream over 12 hours.

“Unless there is hard data to suggest otherwise, we do not feel that any further research of OxyContin for non-cancer pain would be appropriate in the triplicate states,” company representatives concluded following a focus group with doctors held prior to launch, according to a batch of documents Alpert and her co-authors obtained using open records requests to the U.S. Drug Enforcement Agency. “In our judgment, the data from Texas seems to be very convincing relative to the attitudes of ‘triplicate’ doctors toward Class II narcotics, and unless there is reason to believe this could be different in another market (i.e., California, New York) than [sic] the findings from the Houston groups should be considered valid for all markets.”

Yearly overdose death rates were slightly higher in triplicate states in the decade before OxyContin’s 1996 arrival, “but this flips within a few years of the launch,” Alpert and her co-authors write in the paper, “Origins of the Opioid Crisis and its Enduring Impacts.”

While the drug overdose death rate rose in all states from 1996 to 2005 compared with the prior decade, the increase was relatively limited in triplicate states. By 2000, OxyContin distribution per person was 2.5 times higher in states without the triplicate prescription program compared with triplicate states.

“These large and statistically significant differences [in OxyContin distribution] persist through 2016,” the authors write, even though triplicate programs had folded in the five states by 2004. If the states without a triplicate program had had a triplicate program, they would have experienced 34% fewer drug overdose deaths and 45% fewer opioid overdose deaths from 1996 to 2017, the authors estimate.

Regulatory failures and the need for an ‘enormous investment’ in treatment

Pain management is a real concern for doctors, particularly specialists who treat cancer and chronic conditions. OxyContin, however, was heavily marketed to primary care physicians, “who may not have been adequately trained in pain management,” according to a 2003 report from the federal Government Accountability Office.

Purdue Pharma spent $700,000 on OxyContin advertising in 1996, according to the report. Purdue’s strategy, along with other companies selling similar drugs, included advertising in medical journals. By 2001, Purdue’s OxyContin advertising budget reached $4.1 million and OxyContin sales topped $1 billion, making it the “most frequently prescribed brand-name narcotic medication for treating moderate-to-severe pain in the United States,” according to the government report.

Around the same time, news outlets began reporting on OxyContin misuse, including people crushing tablets and snorting the powder to get an immediate high, bypassing the controlled release. By the time Purdue in 2010 introduced a new OxyContin formula, which prevented rapid release of the drug when crushed, it was too late. Many of those addicted “substituted from OxyContin to heroin,” Alpert and her co-authors write.

Over the years, members of the Sackler family, which bought Purdue Frederick in 1952, directed the company to market to doctors who prescribed OxyContin at high rates, “ignored and worked around safeguards intended to reduce prescription opioid misuse, and promoted false narratives about their products to steer patients away from safer alternatives and deflect blame toward people struggling with addiction,” U.S. Rep. Carolyn Maloney said during a Dec. 2020 Congressional hearing.

“This is not just the fact that Purdue Pharma was marketing using false claims but also the aggressiveness and amount of advertising that also could have been maybe monitored or regulated more extensively,” Alpert says.

Food and Drug Administration regulatory failures occurred from the start, according to an August 2020 commentary in the American Medical Association Journal of Ethics by Andrew Kolodny, medical director of opioid policy research at Brandeis University.

President Franklin Roosevelt signed the Food, Drug and Cosmetic Act in 1938, strengthening the government’s ability to regulate the contents of food, cosmetics and drugs and barring companies from making false claims about drugs.

If the Food and Drug Administration had enforced the law, it would have labelled oxycodone only for specific, very painful conditions, according to Kolodny. The initial OxyContin label instead “featured a broad indication, allowing Purdue to promote the drug’s use for common conditions for which opioids are more likely to harm than help, such as low-back pain and fibromyalgia,” he writes.

Kolodny adds that while fewer doctors today are prescribing opioids, “overprescribing is still a problem. According to a recently published report, more than 2.9 million people initiated opioid use in December 2017. The FDA’s continued approval of new opioids exacerbates this problem.”

As Alpert notes, the U.S. is well beyond the brink of reversing the opioid crisis. “Really, now, the solution is an enormous investment in substance abuse treatment,” she says.

Schoomaker, now a third-year medical student treating patients in the Norfolk, Virginia area, recalls working in an emergency department in a rural part of the state and seeing opioid overdose cases nearly every shift. In Norfolk, she has quickly observed a truism of medical treatment in the U.S.: Many patients may not see a doctor until their condition has substantially progressed. In that sense, the medical system is downstream from the economic and social factors that affect health. One way to help reduce deaths of despair could involve moving the role of the medical practitioner further upstream, Schoomaker says. It’s not an easy proposition, she acknowledges, nor is it likely to be one-size-fits all.

“Those I have talked to who have much more experience say that education is really one of the key places we can be working with our patients, in describing better what their disease is, helping those folks to make informed decisions with their care and spending the time to do that,” Schoomaker says. “But that’s a challenge when you have 20 minutes to see a patient with a million chronic diseases — and you have a waiting room of several other patients who need to be seen that day.”

Research: Raising the wage floor could reduce deaths of despair

There is a small but promising line of emerging research on how economic policy could reduce deaths of despair, specifically deaths stemming from intentional self-harm not involving drugs.

Research in the September 2020 issue of the Journal of Health Economics links a 10% increase in a state’s minimum wage to 2.7% fewer suicides among adults with a high school diploma or less. The same increase in federal tax credits for low-income earners yields a 3% reduction in suicides among adults with less education. Those credits can vary substantially based on specific income levels and number of children.

deaths of despair
(Pepi Stojanovski / Unsplash)

The authors did not identify minimum wage effects on drug- and alcohol-related deaths over the period studied, 1999 to 2017, and they did not observe major suicide reductions for those with at least a bachelor’s degree. The study, “Can Labor Market Policies Reduce Deaths of Despair?” covers 46 states plus Washington D.C. — the authors dropped Georgia, Oklahoma, Rhode Island and South Dakota due to incomplete data.

“Regardless of the root causes, there are some tangible policy levers we can use to try to improve population health,” says Chris Lowenstein, one of the authors and a doctoral student in health policy and economics at the University of California, Berkeley.

Likewise, the authors of a January 2020 paper in the Journal of Epidemiology and Community Health look at data from all 50 states and Washington, D.C., covering 1990 to 2015, and associate a $1 minimum wage increase with a 3.4% to 5.9% decrease in the suicide rate for adults aged 18 to 64 with a high school education or less.

Deaths of despair as a concept has resonated across the media and academic landscapes. Thousands of academic papers have mentioned the phrase, according to the 2021 Case and Deaton working paper.

Understanding deaths of despair extends not just to pinpointing the relative importance of drug supply and demand factors, but also to what “despair” means.

“Despair is a concept in common use, not a well-defined diagnostic category, let alone one with a clinically validated measure,” Case and Deaton write in their working paper.

National surveys on health and wellbeing do not ask about despair, per se, Case and Deaton write. Some surveys ask about pain. Case and Deaton cannot find “any study that documents falling levels of pain.”

A finer understanding of how people experience despair could unlock more policy solutions.

“I think that’s a good thing for people to be arguing about,” Lowenstein says. “Anyone who claims to have their finger on what exactly ‘despair’ is, is probably going to get some pushback.”

Further reading

Origins of the Opioid Crisis and its Enduring Impacts
Abby Alpert, William Evans, Ethan Lieber and David Powell. Quarterly Journal of Economics, November 2021.

The Great Divide: Education Despair and Death
Anne Case and Angus Deaton. National Bureau of Economic Research working paper, September 2021.

Economic Studies on the Opioid Crisis: A Review
Johanna Catherine Maclean, Justine Mallatt, Christopher Ruhm and Kosali Simon. National Bureau of Economic Research working paper, November 2020.

How FDA Failures Contributed to the Opioid Crisis
Andrew Kolodny. American Medical Journal of Ethics, August 2020.

Life Expectancy and Mortality Rates in the United States, 1959-2017
Steven Woolf and Heidi Schoomaker. The Journal of the American Medical Association, November 2019.

Changes in Midlife Death Rates across Racial and Ethnic Groups in the United States: Systematic Analysis of Vital Statistics
Woolf, et. al. The BMJ, August 2018.

Mortality and Morbidity in the 21st Century
Case and Deaton. Brookings Papers on Economic Activity, October 2017.

Age-Aggregation Bias in Mortality Trends
Andrew Gelman and Jonathan Auerbach. (Letter) Proceedings of the National Academy of Sciences, February 2016.

Increased Mortality for White Middle-Aged Americans not Fully Explained by Causes Suggested
Christopher Schmid. (Letter) Proceedings of the National Academy of Sciences, February 2016.

Rising Morbidity and Mortality in Midlife among White Non-Hispanic Americans in the 21st Century
Case and Deaton. Proceedings of the National Academy of Sciences, December 2015.

COVID, race and life expectancy

The pandemic has decreased life expectancy across the board by race and ethnicity, according to estimates from the Centers for Disease Control and Prevention.

Life expectancy for Hispanic people was 81.8 years at birth in 2019. That year, life expectancy at birth for non-Hispanic Black people was 74.7 years and 78.8 years for white people. The life expectancy gap between white and Black people in the U.S. had been narrowing steadily for decades, shrinking from 8.3 years in 1950 to 4.1 years in 2019. In 2020, life expectancy dropped to 78.8 years for Hispanic people, 77.6 for white people and 71.8 for Black people. The white-Black life expectancy gap grew to 7 years.

About The Author