What cannabis evidence can say

Which parent-facing statements about cannabis prevalence and change are supported by official state estimates?

SAMHSA NSDUH pooled 2022–2023 and 2023–2024 state estimates and official change tables; New Jersey, Maryland and Ohio; ages 12–17 and 18+. Selected results include 24 estimates and 12 comparisons across past-month and past-year use.

Parse estimates and Bayesian credible intervals; use official change tests that account for shared 2023 observations; apply a structured claim rubric and distinguish retail dates from other policy events. Six past-month comparisons are primary.

Existing verification.log records eight passing tests. verification.json records 1,120 estimate rows cross-checked against 560 official comparison rows, plus source checks and 17 generated claims.

None of the six primary past-month comparisons meets nominal p < .05. Ohio adolescent prevalence is 6.48% (95% Bayesian credible interval 5.22–8.03%) versus 7.13%; official change p=0.342.

Non-significance does not prove no change or no policy effect. Pooled windows and short Ohio post-sales coverage cannot isolate a retail effect; 18+ is not 21+ and state estimates are not county estimates. Treatment measures in this release are 2024-only and excluded. Ohio’s August 6, 2024 retail date remains flagged because its primary announcement was inaccessible (404).

Official-table parsers, source checks, structured claim rules, uncertainty charts, policy timeline, aggregate tables and a self-contained offline review report.

Trace the Ohio estimate and p value; explain percentage points and overlapping periods; audit claim wording; resolve the Ohio primary-source date gap or retain it explicitly. Log only completed personal contributions.

Ask whether families or educators actually encounter these claims and which explanation format would be useful. If invited and approved, offer a small sourced resource for an existing parent-information session or educator resource collection. No school or family has requested or agreed to use it.

What already exists

The offline report, 17 structured claims, uncertainty charts and claim-review brief provide sourced explanations and examples of unsupported inferences.

Functional adaptation still needed

A future authorized adaptation would turn selected claims into a short accessible family-facing Q&A or educator handout with source dates, links, plain-language uncertainty and a clearly bounded state-level scope. Resolve or visibly retain the Ohio primary-date gap. Review reading level, print usability and update responsibility; obtain school approval before use in a school setting. This is an evidence-explanation resource, not individualized medical advice.

Adoption and outcome evidence

Adoption: actual approved inclusion in a resource collection, lesson or information session and subsequent reuse. Outcome: a short voluntary before/after exercise on distinguishing prevalence, unresolved change and causal claims; record sample size and limitations without attributing changes solely to the resource. Downloads, views and praise are not evidence that understanding improved.

The user supplied or authorized the project goal and scope. AI substantially produced the initial code, tests, analysis execution, charts and documentation. This case-study draft is also AI-written. No completed independent user coding, source audit, beneficiary review, adoption or measured impact is recorded. Passing software checks does not establish subject-matter validity. Update this wording only with evidence of work actually performed.