In recent months, key government studies from the Center for Disease Control and Prevention (CDC) and the Substance Abuse and Mental Health Administration (SAMHSA) have confirmed the rising scale of stimulants’ use and experienced harms.
But researchers from the RAND Corporation, a nonprofit that studies policy including drugs, warned on August 20 that data on methamphetamine are burdened by significant uncertainty. This is roused by inadequate data infrastructure and a broader political tunnel-vision on opioids’ centrality to the overdose crisis.
“Methamphetamine surveillance data give us serious pause,” wrote RAND adjunct researcher Greg Midgette and others in a report on drug consumption in the United States. “We can be confident that methamphetamine consumption is increasing, perhaps dramatically, but we cannot make any claim beyond that.”
The researchers share degrees of uncertainty for the other drugs studied—heroin, cocaine, marijuana—but not to the same extent as methamphetamine. Simply put, “national data sets do not do a good job of capturing its use,” they wrote.
One of the data sources that the RAND report references—Arrestee Drug Abuse Monitoring (ADAM), conducted by the US Department of Justice—was discontinued just before what is believed to be meth’s “first peak” in US consumption, from 2004 to 2006. The following year, ADAM was renewed—yet it “covered very few counties with substantial methamphetamine use,” noted RAND, since it “almost exclusively” captured urban areas. In 2013, ADAM was once more terminated, just as the drug seemed to be making a resurgence, as the researchers described.
The ADAM interviews were also flawed by being conducted after people had been arrested, as Midgette explained to Filter. “In a pretty traumatic moment, you’re introducing survey questions about a stigmatized subject?” he asked.
SAMHSA’s National Survey on Drug Use and Health (NSDUH) also has meth-related limitations. The survey’s authors modified how it asked about meth in 2007, after previously referring to it as a “prescription drug”—a categorization that had caused reason for concern that their results were “an underestimate of use,” wrote the authors of NSDUH’s interview specifications.
Estimates of meth use did increase after meth’s street names were added. People may “know they consumed ‘crank,'” for example, “but they didn’t know crank is meth,” said Midgette.
NSDUH misses the mark when it comes to heroin as well, according to RAND, whose researchers estimated a nearly five-times larger past-month heroin user population (2.3 million) in 2016 than NSDUH’s figure.
Even after changing its descriptions of meth, SAMHSA did not designate a stand-alone section of the survey for the drug—involving specific questions about some of its riskier forms of consumption, like injection—until 2015, two years after meth was making another comeback.
Midgette also found that NSUDH provides participants with “not a whole lot of incentive to respond,” while also potentially not “respecting privacy.”
Knowing how many people are dying for meth-related reasons is just as important as understanding how folks are using the drug. But these figures, too, are often muddled. In the CDC’s data, for example, meth-involved deaths are grouped together with other “psychostimulants with abuse potential,” making them difficult to tease apart.
“Collapsing methamphetamine together with psychostimulants might have made sense in the past,” said Midgette. “But now that’s a potentially important distinction: There might be all sorts of other trends, like with polydrug use.” And that’s an important trend not to miss, given that the majority of fatal overdoses involve more than one substance.
The failure to adequately quantify meth-related harms extends to research on harm reduction interventions for the drug. Out of more than 900 studies addressing meth use amongst queer men, only one looked at harm reduction—something that the RAND researchers called be “worrisome—particularly in a medical context.”
Midgette identified wastewater testing as a promising opportunity to improve drug use data collection. “You can see that there’s a new drug or a higher concentration of another drug than previously,” he said of this method, explaining that it’s especially useful when individual-level surveying has proved limiting.
However, data collection is “always going to be imperfect,” said Midgette. “That’s because drug use is not a stable thing.”
Photo of a glass meth pipe by Jlcoving/Wikimedia Commons