Teen Cannabis Surveys Miss Rising Use Disorder, Kaiser Study Finds

Survey-Diagnosis Divergence Widens Post-Legalization

Kaiser Permanente researchers identified a widening gap between adolescent self-reported cannabis use and clinical cannabis use disorder diagnoses in the five years following state-level legalization. The study analyzed electronic health records and survey data from adolescents aged 12-17 across California, Oregon, Washington, and Colorado. Self-reported past-month use remained stable or declined slightly between 2021 and 2026. Yet formal CUD diagnoses recorded in clinical settings increased by 18 percent over the same period.

The DSM-5-TR criteria for cannabis use disorder require at least two of eleven specified symptoms within a 12-month period—tolerance, withdrawal, impaired control, and social or occupational impairment among them. Kaiser's study found that adolescents meeting diagnostic thresholds were three times more likely to have no corresponding survey response indicating problematic use. Survey instruments designed to capture prevalence miss the clinical severity axis.

The measurement gap has direct implications for resource allocation. States rely on Youth Risk Behavior Surveillance System data and the Monitoring the Future survey to model treatment capacity, prevention budgets, and school-based intervention programs. If those instruments underestimate disorder prevalence, treatment infrastructure remains undersized.

Clinical Diagnoses Concentrate in Medicaid-Enrolled Cohorts

Sixty-eight percent of new adolescent CUD diagnoses occurred in patients enrolled in Medicaid, compared to 41 percent of survey respondents reporting weekly or daily use. This socioeconomic skew suggests that survey-based prevalence estimates, which sample broadly across income strata, may underweight the populations most likely to develop clinical disorder. Medicaid-enrolled adolescents also showed higher rates of co-occurring mental health diagnoses, including major depressive disorder and generalized anxiety disorder, complicating treatment pathways.

Survey instruments typically ask about frequency of use and perceived harm. They don't probe for DSM-aligned symptomatology such as failed quit attempts, craving intensity, or functional impairment in school or family settings. The result? A dataset optimized for prevalence tracking but poorly calibrated for disorder detection. In California, where the study cohort was largest, the survey-diagnosis gap was most pronounced in counties with the highest density of licensed retail dispensaries, raising questions about access-driven disorder risk that current surveillance tools can't resolve.

The divergence between what adolescents report in surveys and what clinicians document in diagnostic encounters suggests we're flying blind on the question of harm, particularly in communities where legal access has normalized high-potency products.

Implications for State Prevention Budgets and Federal Monitoring

Federal agencies including the Substance Abuse and Mental Health Services Administration and the Centers for Disease Control and Prevention use state-level survey data to allocate prevention grants and set national baselines for adolescent substance use. If those baselines systematically undercount disorder prevalence, states with legal markets may receive insufficient federal support for treatment infrastructure even as clinical demand rises. The Kaiser study didn't identify a causal mechanism linking legalization to increased disorder rates, but the temporal correlation and geographic concentration in legal states warrant further investigation.

California's Department of Health Care Services reported in August 2026 that adolescent admissions to outpatient substance use disorder treatment programs increased 22 percent year-over-year. Largest single-year jump since the state launched legal adult-use sales in 2018. Oregon and Washington reported similar trends. The Kaiser findings suggest that survey-based early warning systems may be lagging clinical reality by 18 to 24 months, a delay that complicates real-time policy adjustment. For full background on this story, see the CannIntel topic hub on Youth Cannabis Use Disorder.

Study authors recommended that state health departments supplement existing survey tools with periodic medical-record audits to cross-validate disorder prevalence and that federal agencies consider revising YRBSS and MTF instruments to include DSM-aligned screening questions. No legislative action is pending, but the research has circulated among staff at the Senate Caucus on International Narcotics Control and the House Energy and Commerce Subcommittee on Health.

The next signal: whether SAMHSA incorporates clinical-record cross-checks into its 2027 National Survey on Drug Use and Health methodology, expected to be finalized by December 2026. If federal surveillance tools remain unchanged, the measurement gap identified by Kaiser will persist, and states will continue to allocate prevention resources based on incomplete data. The political variable is whether public-health agencies treat this as a methodological refinement or a regulatory priority requiring statutory change.