Changes in methamphetamine use, treatment admissions and overdose mortality can diverge. Comparing them can identify important questions about disparities, but separate surveillance systems do not track a single set of people from use through treatment to death.
Which three systems were compared?
Shearer and colleagues analyzed historical United States adult data for trends through 2019. The National Survey on Drug Use and Health supplied self-reported past-30-day use; the Treatment Episode Data Set–Admissions supplied treatment episodes; and CDC multiple-cause-of-death records supplied overdose deaths.
The analysis used rolling three-year averages. A rate labeled 2010 therefore included 2008–2010 information rather than only events or interviews from calendar year 2010. Smoothing helps stabilize sparse estimates, but adjacent annual points share data and should not be read as independent yearly snapshots.
Were the geographic populations identical?
No. Treatment and mortality analyses excluded the District of Columbia and eight states because treatment data were missing in one or more years. The public survey data did not provide corresponding state identifiers for the same exclusion. Sensitivity analyses examined the effect of the geographic restrictions, but the main systems still differed in coverage.
Treatment records counted admissions, so one person could appear more than once. The survey estimated people reporting use, while death records counted deaths. Dividing admission rates by use rates would not produce a verified percentage of users who received treatment.
What did the methamphetamine trends show?
Between 2010 and 2019, estimated use increased significantly in the authors’ White and Hispanic groups: 133% with a 95% confidence interval of 77%–188%, and 247% with an interval of 58%–435%, respectively. Upward use trends in the other reported racial and ethnic groups were not statistically significant. Nonsignificance should remain visible rather than being rewritten as no increase.
Treatment admissions and overdose deaths showed different patterns. The authors reported a 1,115% relative increase in the methamphetamine-labeled overdose rate among Black adults and 744% among American Indian adults. Relative increases depend strongly on the starting rate. In 2019, the reported death rates were 13.3 per 100,000 in the American Indian group and 5.5 per 100,000 in the White group; these absolute rates answer a different question from the percentage changes.
Was every “methamphetamine” death chemically specific?
No. The mortality analysis used ICD-10 T43.6, psychostimulants, which the article referred to as methamphetamine. That category is not exclusive to methamphetamine. The indicators were also nonexclusive: a death or admission could involve more than one substance. The figures do not establish methamphetamine-alone mortality.
Survey reports, admission classifications and death-certificate coding each have their own ascertainment limitations. Combining their graphs does not remove those limitations or create individual-level exposure verification.
How should racial and ethnic differences be interpreted?
The sources collected group information differently: survey self-identification, information recorded at admission, and death-certificate information often supplied by next of kin or observation. The harmonization also excluded some multiracial and missing-data categories.
These are surveillance categories, not evidence that race biologically determines methamphetamine toxicity. Differences can reflect social conditions, access to treatment, drug supply, other exposures and recording practices. The study did not isolate a single mechanism for the disparities.
What conclusions remain useful?
Divergent trends can highlight where further investigation and better service data are needed. They cannot establish an individual’s probability of dying after use, prove that treatment caused or prevented a death, or determine impairment. The value of triangulation is the broader view it provides while preserving the distinct populations and measurements behind each line.
Source
Shearer RD et al. Racial and ethnic differences in heroin, methamphetamine, and cocaine use, treatment, and mortality trends in three national data sources—United States, 2010–2019. Medical Care. 2024;62:151–160. doi:10.1097/MLR.0000000000001969. Historical observational surveillance analysis.
Leave a Reply