A study published January 7 suggests that the Centers for Disease Control and Prevention (CDC) may be drastically undercounting national numbers of drug overdose deaths. The study by Dr. Troy C. Quast, an associate professor of economics at the University of South Florida College of Public Health, compared deaths in the state of Florida, as counted by the CDC’s Multiple Cause of Death database (MCOD), with data from the Florida Medical Examiners Commission. He found that the CDC’s numbers were significantly lower than those calculated by state authorities.
“The discrepancies are likely largely due to incomplete death certificates submitted to CDC’s National Center for Health Statistics,” Quast told Filter. “Prior research (including by experts at the CDC) has shown that a considerable proportion of death certificates for drug overdoses do not specify the drug or drugs that caused the overdose, and certificates that are later updated may not be filed in time to be included in the CDC’s data. This phenomenon is present in varying degrees throughout the US.”
Quast examined data from 2003-2017 for three specific drug classes (amphetamines, benzodiazepines and opioids) and for three specific drugs (methadone, cocaine and heroin). He found that the CDC undercounted over 125,800 opioid-involved deaths and over 39,300 heroin-involved deaths during this period.
The agency, according to Quast’s analysis, undercounted Florida deaths involving amphetamine by 17 percent, benzodiazepine deaths by 45 percent, methadone deaths by 45 percent, and heroin-involved deaths by 25 percent. For the three drug classes, he found, local medical examiners counted anywhere from 19 to 39 percent more deaths than the CDC did.
Having discovered these large discrepancies in Florida, Quast suggested that there may be similar gaps between federal- and state-collected data across the US. “It is likely that the number of deaths caused by specific drugs has been significantly undercounted, especially during the surge in fentanyl overdoses in the mid- to late-2010s,” he told Healio Primary Care.
C. Michael White, another researcher interviewed by Healio who is unaffiliated with the study, suggested that the CDC gathers data more quickly than counties and states, and that the accuracy of CDC findings suffers as a result. “In essence, the CDC is trading having more current data trends for some of the accuracy of their data,” he said.
Further research would clearly be needed to demonstrate whether the discrepancies seen in Florida apply to other states throughout the US, and to what extent. But if that is the case, the CDC would need to re-evaluate how it balances urgency and accuracy in gathering drug-death data.
Quast also analyzed the deaths data by other factors like age, gender, race and ethnicity. He found some differences in patterns over time and by demographics. “[But] our analysis did not identify a cause or pattern to explain the differences,” he said. “Efforts to improve the reporting of fatal overdoses may enhance our understanding of and subsequently may improve the response to the drug overdose epidemic.”
“It is important to ensure that there are not discrepancies in the extent of undercounting by group or region,” Quast continued. “For instance, if a certain racial group has a higher likelihood to have overdose deaths not included, it may lead to in inadequate response. While this issue is minor relative to the broader message that the CDC attempts to convey with these data, it could be helpful for the CDC to note that the data are incomplete.”
Data that are accurate as well as timely it vital to better inform targeted harm reduction provisions like naloxone distribution, syringe exchanges and supervised consumption sites. The consequence of underestimating what we already know is a devastating crisis of drug-related deaths could be even more lives lost.
Image from Michael Longmire via Unsplash.