Which Came First, the Opioids or the Despair?

Dark days in America.

Photographer: Spencer Platt/Getty Images

Anne Case and Angus Deaton's 2015 article on rising mortality among middle-aged white Americans — and the 2017 follow-up that attributed this rise to an increase in suicides, drug overdoses and alcohol-related deaths among those without college educations — was among those rare academic papers that changed public debate. "Deaths of despair," the name that the wife-and-husband team of Princeton University economists gave the phenomenon, entered the popular lexicon, and their economic explanation for these deaths resonated:

We propose a preliminary but plausible story in which cumulative disadvantage from one birth cohort to the next – in the labor market, in marriage and child outcomes, and in health – is triggered by progressively worsening labor market opportunities at the time of entry for whites with low levels of education.

Blockbuster findings like this inevitably generate a lot of questioning and re-examination, and the Case-Deaton results have certainly done that. A big team of epidemiologists, most from the National Cancer Institute, examined death-certificate data and found results similar to Case and Deaton's but with some key differences: Drug overdoses were a much bigger cause of increased mortality than suicides or alcohol, mortality increases weren't restricted to the middle-aged, and American Indians and Alaska natives have seen even bigger mortality increases than whites. Columbia University statistics professor Andrew Gelman and political scientist Jonathan Auerbach sliced the age data more finely and found that much of the increase in mortality rates among 45-to-54-year-old whites was due to a shift in the composition of that age cohort, which as the last of the baby boomers aged into their 50s began to skew heavily toward the older end — although even in their adjusted version the middle-aged white mortality rate hasn't declined as it has for other groups in and outside of the U.S. And in a study out this summer that hasn't gotten much attention yet, University of Colorado sociologists/demographers Ryan K. Masters, Andrea M. Tilstra and Daniel H. Simon also adjust more precisely for age and find that while suicide and alcohol-related death rates haven't gone up much, drug overdoses really have.

QuickTake Heroin and Opioids