In this issue of JAMA Health Forum, Powell and Jacobson1 found that the past-year prevalence of the use of illicitly manufactured fentanyl (IMF) reported in a June 2024 online survey of individuals 18 years and older was 7.5% (95% CI, 6.2%-8.9%). They note that this number is vastly higher than the 0.3% estimate reported in the 2022 National Survey on Drug Use and Health (NSDUH). IMF prevalence did not jump 20- or 30-fold between 2022 and 2024; indeed, point estimates from the 2023 NSDUH are lower than in 2022.
We find the higher figure easier to reconcile with past estimates of the size of the heroin market2,3 and with evidence that IMF has not only largely displaced heroin from its traditional markets but also expanded beyond, for example, into counterfeit pills. Our point, however, is not to argue for one number over another.
Rather, what is most shocking is that more than a decade after opioid-related deaths began their terrible rise, our official national data collection concerning this phenomenon (apart from overdoses themselves) remains a single general population survey that long predates this crisis. Even as nearly half a million individuals in the US died over the past 5 years, the federal government did not see fit to launch a single new primary data collection effort to monitor the nature and extent of opioid supply or use. What is needed is not only more credible and reliable estimates of prevalence at different levels of intensity but also measures of quantity consumed and expenditures.