The best findings in science aren’t the ones that make you go “cool!”, they’re the ones that make you go “huh?”
A study in the Proceedings of the National Academy of Sciences reported a strange and unexpected finding. By looking at data from the CDC, researchers were able to evaluate mortality rates per 100,000 individuals, and compare this between ethnic groups. While there’s generally been a decrease in all-cause mortality, they found an increase in the mortality of middle-aged white non-Hispanic men and women in the US between 1999 and 2013 (solid red line below). This finding was unique to middle-aged White Americans – data from other countries also reported a drop in death rates.
So what makes White, middle-aged, non-Hispanic Whites unique?
The researchers started to explore potential reasons behind this change. They looked at two major factors: the cause of death, and the level of education by those involved. On cause of death, they found a dramatic increase in the number of deaths due to poisonings, suicides and chronic liver diseases. They hypothesize that one potential reason for the increase in deaths is because of increased opioid painkiller use, known to be a leading cause of drug-related death in America. They also found that those without a high school education had higher mortality than those with high school or higher education.
They then extrapolate from this, suggesting that economic insecurity that preceded the financial collapse might be responsible. The combination of widening income inequality and a lack of growth of real wages would have impacted those without a high school education more than others, and could have driven this finding. The authors argue that an increase in diagnosed and treated pain might be responsible for these findings, and support this with data from the Behavioral Risk Factor Surveillance System (BRFSS) and Medicare that show high rates of opioid use. They suggest several reasons for this finding, including hypothesizing that non-Hispanic whites have “lost the narrative of their lives.”
A concern I have with the paper is the lack of information about their data – both where they got it from, and on what data they are drawing such strong conclusions. There isn’t much of a methods section, and their results are pretty brief. As an example, you can’t see the actual percentage changes by year. While writing this piece, I’d really like to be able to tell you what the specific percentage changes are; that information isn’t there. This might be a stylistic/word count/journal issue, but it’s one that makes it difficult for me to interpret the findings and draw any conclusions. Andrew Gelman goes into more detail on his blog about this issue, which includes a response by the authors. Spoiler: they conclude the effect is real, even after controlling for cohort and age effects, and even when you break it down into single year groups rather than 10 year age groups.
As a result, despite concerns we may have, there’s something going on here. If the theory is economic hardship, then it’s interesting that this was seen among Whites, but not Blacks or Hispanic Americans, the groups that traditionally bear the brunt of economic hardships. There’s something unique happening here to middle aged, White Americans, and, quite frankly, we have no idea what it is or what could be driving it.
Case, Anne, and Angus Deaton. “Rising morbidity and mortality in midlife among white non-Hispanic Americans in the 21st century.” Proceedings of the National Academy of Sciences (2015): 201518393. Available online here: http://www.pnas.org/content/early/2015/10/29/1518393112.abstract
Help us do science!
PLOS Public Health Perspectives has teamed up with researcher Paige Brown Jarreau to create a survey of our readers. By participating, you’ll be helping us to improve Public Health Perspectives and contributing to SCIENCE on blog readership. You will also get FREE science art from Paige’s Photography for participating, as well as a chance to win prizes such as t-shirts and one of 100 $50 Amazon vouchers! It should only take 10-15 minutes to complete.
You can find the survey here: http://bit.ly/mysciblogreaders.
This research is important, because it will give us insight into who is reading the blog and why. More broadly, it will provide new knowledge on the culture on science blogging and communication. We are very excited to see the results of the survey, and hope that as many PLOS Blogs readers as possible will complete the survey. Who knows, you might even win a prize!