On November 24, the Substance Abuse and Mental Health Services Administration (SAMHSA) published reports measuring behavioral health for each of the 50 states and the District of Columbia. These “barometer” reports have not appeared since 2019, and the new ones combine three years of National Survey on Drug Use and Health (NSDUH) data rather than just one. They provide more region-specific snapshots for each jurisdiction, and continue the agency’s tradition of organizing NSDUH data in kind of weird ways.
The three-year national average for “drug use disorder,” which includes marijuana, cocaine, heroin, hallucinogens, inhalants, methamphetamine or “prescription psychotherapeutic drugs,” was 9.3 percent. The average for substance use disorder, which includes all of those plus alcohol, was 17 percent.
The past-month national average for illicit drug use, a category that includes the same substances as “drug use disorder,” was 15.9 percent. For past-month marijuana use specifically, the national average was 14.5 percent.
The average for “opioid misuse” was 3.2 percent. That category is intended to represent heroin and prescription opioid analgesics—and even non-opioid analgesics—but not street-supply fentanyl.
The NSDUH did not begin including street-supply fentanyl until 2022, which is not mentioned in the description of methodology where SAMHSA explains the decision to aggregate data from 2021, 2022 and 2023. Previous state-level barometer reports have just used one annual NSDUH at a time.
Each of the 51 reports highlights the same handful of key NSDUH categories like cigarette use and illicit drug use. For example, the Arkansas barometer report shows that the state’s rate of past-month cigarette use among adults over 12 was 19.8 percent, compared to a national average of 14.7 percent. It also puts these figures in the context of the West South Central census division, as well as broken down into seven catchment areas within the state.
This non-intuitive way of organizing fentanyl data is not as overt an issue in the new barometer reports, because you wouldn’t really know fentanyl use was a factor at all.
While all the substance use-related categories are for people over age 12, the mental health categories use more specific age ranges. In addition to the categories already mentioned, the reports cover past-month binge alcohol use; past-year alcohol use disorder; past-year major depressive episodes for respondents aged 12 to 17; any mental illness within the past year for respondents over 18; serious mental illness within the past year for respondents over 18; serious thoughts of suicide within the past year for respondents aged 12 to 17; and the same for respondents over 18.
The NSDUH reaches around 70,000 respondents each year. The surveys for 2022 and 2023 (and 2024, though that survey isn’t included in the barometer reports) measured “fentanyl misuse” by grouping data for street-supply and pharmaceutical versions in the same category. They then allowed the pharmaceutical fentanyl data to count toward the totals for illicit drug use, while making the confusing choice to exclude street-supply fentanyl from the same category.
This non-intuitive way of organizing data for fentanyl use is not as overt an issue in the new barometer reports, because without looking at the original NSDUH you wouldn’t really know fentanyl use was a factor at all. The only reference to fentanyl, street-supply or pharmaceutical, is a single line noting that the category of “opioid misuse does not include use of illegally made fentanyl.”
The larger issue with NSDUH data is that they purport to represent how substance use and mental illness are currently affecting United States adults, but the survey is not offered to the populations most affected by either of those things. The survey is only offered to people with fixed addresses, a definition that does not include prisons or jails, or any other relevant institutions like residential treatment centers or psychiatric facilities.
Image (cropped) via Substance Abuse and Mental Health Services Administration
