On March 5, the Centers for Disease Control and Prevention a published a National Vital Statistics Report on the drugs most frequently involved in overdose deaths between 2017 and 2023. It’s the first such report since 2018, and the first NVSR related to the overdose crisis since 2023.
Software programs that search through death certificates can pull information from “literal text” fields, where the medical examiner or coroner (or someone in law enforcement, depending on location) enters their notes. The three literal text fields researchers used for this NVSR pertained to cause of death, significant conditions contributing to the death and the description of how it happened.
Data from death certificates are coded using the International Classification of Diseases, 10th Revision (ICD–10), which groups overdose deaths under broad categories of drugs rather than listing the drugs individually. So in most CDC updates on overdose mortality you can’t see exactly how many deaths were associated with, for example, fentanyl—it’s just understood to represent the biggest share of ICD–10 code T40.4, the category for synthetic opioids other than methadone. Similarly, most of the “psychostimulants with abuse potential” category is represented by methamphetamine, but it doesn’t actually say that.
In 2016, the CDC’s National Vital Statistics System published a report covering the drugs most frequently involved in overdose deaths between 2010 and 2014. After the methodology for searching literal text fields evolved, in 2018 a report was published on the same topic covering 2011 to 2016. This is the first update since then, though there have been a couple of NVSR on regional drug breakdowns or focused on fentanyl specifically.
Unsurprisingly, fentanyl was the most frequently mentioned drug each year between 2017 and 2023, increasing from 38.9 percent of deaths to 68.9 percent. These data also represent fentanyl analogs, when those were listed on death certificates. Heroin, meanwhile, dropped from the second-most cited drug in 2017 to the 7th by 2023.
Data like this don’t just reflect changes in the drug supply, but also changes in patterns of drug use. Someone can have more than one substance in their system at the time that they died, but that doesn’t necessarily mean that they were consumed together.
For example, gabapentin’s prevalence on death certificates more than doubled, moving it from the 13th most commonly cited drug in 2017 to the seventh in 2021, after which it leveled off somewhat. Many people who use opioids use gabapentin or similar medications like Lyrica to potentiate the effects.
Similarly, each year the top 15 most commonly cited drugs includes several benzodiazepines, but they don’t all represent the same thing. Alprazolam—generic for Xanax—was the fifth most frequently cited drug each year during the study period, while clonazepam (Klonopin) and diazepam (Valium) also showed up each year at or around the bottom of the list. But after three years in the #15 spot, in 2023 diazepam was bumped off the list and replaced by bromazolam—a benzo that is not regulated. Bromazolam is sometimes found mixed with fentanyl, but is also pressed into counterfeit pills sold as Xanax. In 2020, the first year bromazolam appeared on death certificates, it was cited in fewer than 10 cases. In 2023, it was cited in 1,734.
While more novel synthetic drugs have been entering the supply over the past few years, this isn’t necessarily why more drugs are being cited on death certificates. These days it’s the norm for the names of specific drugs to get entered into death certificates, but that’s relatively recent. In 2023, 96 percent of overdose deaths had one or more specific drugs cited in the death certificate; back in 2011 it was only 75 percent.
Death investigation protocols vary haphazardly by jurisdiction, and ultimately a lot of what appears on a death certificate is at the discretion of whoever signs it.
And it’s become increasingly common for a death investigation to consist of looking at a body externally, getting a toxicology screen that shows drugs were in the person’s system, and deciding that those drugs were what killed them.
“For deaths in which multiple drugs are involved, it cannot be determined from the literal text analysis whether the death was caused by just one of the drugs or by a combination of some, or all, of the drugs present,” the NVSR authors wrote. “Some medical certifiers focus on a single lethal drug rather than listing multiple drugs involved in the death, while others list multiple drugs because they believe the drugs to be of equal lethality or that the interaction of all drugs mentioned is important.”
Top image (cropped) via Alabama Department of Forensic Sciences. Inset graphics via Centers for Disease Control and Prevention.