How Accurate Are Opioid Overdose Death Statistics?

How Accurate Are Opioid Overdose Death Statistics?

By Keri Blakinger 08/09/17

Researchers examined six years worth of death certificates to determine the accuracy of opioid-related death statistics for a new study.

Image: 
 A businesswoman drawing Statistic graph with a felt tip pen on glass board.

The overdose epidemic may be vastly underestimated, according to a new study out of the University of Virginia. 

A closer look at death certificates from 2008 to 2014 led Dr. Christopher Ruhm to the conclusion that opioid death rates could be 24% higher than previously estimated. 

“Opioid and heroin involved mortality rates were 24% and 22% greater than reported rates,” the study says. “The differences varied across states, with particularly large effects in Pennsylvania, Indiana, and Louisiana.”

The crux of the problem is that death certificates often list “unspecified drugs” as the official cause of death. But reported opioid fatality rates don’t count any of those deaths, which make up around one-fifth of overdoses. So Ruhm used demographics and existing data to estimate what percentage of unspecified deaths might be opioid-related and recalculate the numbers accordingly. 

Some states were already pretty accurate; Rhode Island, Connecticut and New Hampshire listed the specific drug about 99% of the time. But Pennsylvania listed it only about half the time. 

Thus, the Keystone State - which was ranked number 32 in ODs by the CDC - actually had the seventh-most opioid overdoses in 2014. 

Although the overall estimated numbers may be surprising, the indications as to where such deaths are most concentrated is somewhere in line with existing data.

“The corrected death rates demonstrate that opioid involved mortality was concentrated in the Mountain States, Rust Belt, and Industrial North—extending to New England—and much of the South, whereas heroin deaths were particularly high in the Northeast and Rust Belt, but less so in the South or Mountain States,” the study notes. 

The findings come just on the heels of the presidential opioid commission’s “urgent recommendation” to “declare a national emergency” to address the country’s opioid problem.

"My message to members of a presidential commission would be that getting the most accurate statistics possible is a crucial first step towards developing policies aimed at stemming the fatal drug epidemic," Ruhm told NBC. "This is particularly important when we have scarce funds to allocate and so would want to target them at the hardest hit areas."

This isn’t the first time indications have emerged hinting that the overdose crisis may be worse than numbers show. In the spring, the Centers for Disease Control and Prevention combed through Minnesota death records and came to a similar conclusion, but without offering hard numbers.

"It's quite concerning, because it means that the epidemic, which is already quite severe, could potentially be even worse," Dr. Victoria Hall, the Minnesota-based CDC field officer behind the spring study, told CNN at the time.

"While my data doesn't support a percent that we're underestimating, it puts out the question: Is there something we need to look into further?"

Please read our comment policy. - The Fix
Disqus comments
Keri.jpg

Keri Blakinger is a former drug user and current reporter living in Texas. She covers breaking news for the Houston Chronicle and previously worked for the New York Daily News and the Ithaca Times. She has written about drugs and criminal justice for the Washington Post, Salon, Quartz and more. She loves dogs and is not impressed by rodeo food. Find Keri on LinkedIn and Twitter.

Disqus comments