Skip to main content
Tessa NewsLink
Paste a health news link, or browse

Source study found

Story checked

Adolescent cannabis use disorder diagnoses rise after legalization (opens in a new tab)

news-medical.net · 2026-09-15

Short answerEvidenceSource

Short answer

Mostly not supported

Mostly not supported.

One claim goes further than the study. 3 other points were not covered by the paper.

  • 1 supported
  • 1 overstated
  • 3 not covered

Checked against the study summary. The full text wasn't available, so some details couldn't be settled either way.

Share this check

Follow the evidence trail
1
2

NewsLink checks it

Mostly not supported

One claim overstates the study. One of five checks out. Three claims the study doesn't address.

  • 1 supported
  • 1 overstated
  • 3 not covered
Open claim evidence
3
Then inspect each claim

Evidence layer

Claim by claim

Each claim gets a verdict. Expand it to see the evidence directly below.

5 claims in this story

Showing all 5 claimsChoose a verdict to focus the list.

Claim 1 of 5OverstatedNew diagnoses of cannabis use disorder among adolescents in Northern California began climbing after recreational cannabis was legalized in California in 2016, reversing a steady decline in the three years prior.View evidenceHide evidence

As statedabout 6% a year before the 2016 vote and about 8% a year after

Why this verdict

The abstract-level profile supports a pre-legalization decline in EHR-documented incident adolescent CUD diagnoses and a modeled slope increase after RCL passage. However, the presentation labels this as a causal headline claim about legalization, while the paper evidence is observational interrupted time series evidence of a temporal association. The headline framing therefore outruns what the design can establish. The exact stated annualized magnitudes are also not fully reproduced in the supplied abstract profile.

1

Study evidence

Prior to RCL passage, adolescent incident CUD diagnosis rates were decreasing.aRR = 0.95 per year (95% CI 0.92–0.97)

“Population-based time-series study of adolescents aged 13 to 17 years with Kaiser Permanente Northern California healthcare system membership from January 2014 to February 2020.”
Claim 2 of 5Not coveredThe upward trend in cannabis use disorder diagnoses continued after recreational retail sales began, and by early 2020 diagnoses had returned to their 2014 level.View evidenceHide evidence

As statedmore than three years after passage; back to 2014 level by early 2020

Why this verdict

The profile supports that increases persisted through retail implementation and that the post-implementation trend continued upward but was not statistically significant. It also reports a statistically significant negative slope change at implementation. The claim that diagnoses had returned to their 2014 level by early 2020 is not available in the supplied abstract-level profile, so that part cannot be verified at this depth.

1

Study evidence

Prior to RCL passage, adolescent incident CUD diagnosis rates were decreasing.aRR = 0.95 per year (95% CI 0.92–0.97)

“Population-based time-series study of adolescents aged 13 to 17 years with Kaiser Permanente Northern California healthcare system membership from January 2014 to February 2020.”
Claim 3 of 5Not coveredThe authors say the shift may reflect changes in the broader legal and social environment around legalization passage, including reduced criminal penalties, a normative signal that cannabis is socially acceptable or low-risk, and youth-appealing advertising and higher-potency products as the retail market matured.View evidenceHide evidence

Why this verdict

The supplied abstract-level profile reports the authors’ broad interpretation that incident adolescent CUD rates appeared to increase after RCL passage and should be monitored, but it does not include the proposed mechanisms listed here, such as reduced penalties, normative signaling, youth-appealing advertising, or higher-potency products. These may be in the full discussion, but they are not verifiable from the abstract-depth evidence supplied.

1

Study evidence

Prior to RCL passage, adolescent incident CUD diagnosis rates were decreasing.aRR = 0.95 per year (95% CI 0.92–0.97)

“Population-based time-series study of adolescents aged 13 to 17 years with Kaiser Permanente Northern California healthcare system membership from January 2014 to February 2020.”
Claim 4 of 5Not coveredThe article says the study has limitations because it was restricted to adolescents insured through and seeking care at Kaiser Permanente Northern California, lacked a comparison group of unexposed adolescents, was not preregistered, and ended before the COVID-19 pandemic.View evidenceHide evidence

Why this verdict

Several limitations are supported by the profile, including restriction to Kaiser Permanente Northern California members, observational interrupted time-series vulnerability to unmeasured secular changes, and follow-up ending in February 2020. However, the abstract profile does not verify all stated limitations, especially non-preregistration and the exact framing of no unexposed comparison group. As a bundled claim about what the article says, it is only partly supported at abstract depth.

1

Study evidence

Prior to RCL passage, adolescent incident CUD diagnosis rates were decreasing.aRR = 0.95 per year (95% CI 0.92–0.97)

“Population-based time-series study of adolescents aged 13 to 17 years with Kaiser Permanente Northern California healthcare system membership from January 2014 to February 2020.”
Claim 5 of 5SupportedResearchers analyzed six years of electronic health record data from 637,000 adolescents receiving care at Kaiser Permanente Northern California using an interrupted time series design to test whether trends in new cannabis use disorder diagnoses shifted across policy periods.View evidenceHide evidence

As stated637,000 adolescents; 637,461 studied; 8,804 newly diagnosed over six years

Why this verdict

The profile supports the core methods claim: a population-based interrupted time-series study of KPNC adolescents aged 13–17, n=637,461, using EHR-derived monthly incident CUD diagnoses from January 2014 to February 2020 and Poisson regression to model RCL passage and retail-sales implementation periods. The abstract profile does not verify the ancillary total of 8,804 newly diagnosed adolescents, but the main design and cohort-size claim is supported.

1

Study evidence

Prior to RCL passage, adolescent incident CUD diagnosis rates were decreasing.aRR = 0.95 per year (95% CI 0.92–0.97)

“Population-based time-series study of adolescents aged 13 to 17 years with Kaiser Permanente Northern California healthcare system membership from January 2014 to February 2020.”
Then look for missing context

Context layer

What the story left out

Important study details the story did not include.

  • At retail implementation, the paper reports a statistically significant negative slope change, and the subsequent upward post-implementation trend was not statistically significant.

    The story says the upward trend continued after retail sales began, but it does not clearly convey the negative slope change at implementation or that the post-implementation increase was not statistically significant, both of which materially qualify the retail-sales-period interpretation.

    From Interrupted time series (Poisson regression) with segmented policy timepoints and 6-month lags

  • The policy exposure dates were modeled with fixed 6-month lags, and results depend on that modeling choice.

    The supplied story claims do not mention the modeled 6-month lag for RCL passage and retail-sales implementation, even though the abstract profile identifies it as part of the exposure specification and a limitation.

    From Interrupted time series (Poisson regression) with segmented policy timepoints and 6-month lags

7 things the story did carry across
  • The study is an observational interrupted time-series analysis of monthly EHR-derived incident CUD diagnoses among KPNC adolescents aged 13–17 from January 2014 to February 2020.
  • The outcome is EHR-documented incident CUD diagnosis, not a direct population measure of cannabis use or true disorder prevalence.
  • Before RCL passage, adolescent incident CUD diagnosis rates were decreasing; after the modeled RCL-passage timepoint, the slope increased.
  • The authors emphasize monitoring CUD diagnoses or disorder-level outcomes in addition to cannabis use prevalence/frequency when evaluating legalization effects.
  • Generalizability may be limited because the study population consists of members of a single integrated health system in Northern California.
  • As an observational interrupted time series, the design is vulnerable to unmeasured confounding or concurrent secular changes affecting trends.
  • The study period ended in February 2020, limiting evidence about later trends, including the COVID-19 period.
Then read the study layer

Study layer

Study at a glance

Scan the study first. Expand only the parts you want to inspect.

Pieces of work

1

Evidence read

study summary

Lead result

secondary data

1Lead resultsecondary dataEstimate whether adolescent incident cannabis use disorder (CUD) diagnosis rates in Northern California changed after California recreational cannabis legalization (RCL) passage and after retail sales implementation, using an interrupted time series.Interrupted time series (Poisson regression) with segmented policy timepoints and 6-month lagsExpand

In plain English

Population-based interrupted time-series study of adolescents (13–17 years) enrolled in Kaiser Permanente Northern California (n = 637,461; mean age 14.07, SD 1.43; 50.93% male; 66.45% non-White) using monthly counts of incident cannabis use disorder (CUD) diagnoses from electronic health records between January 2014 and February 2020. Exposure indicators were California recreational cannabis legalization (RCL) passage (9 Nov 2016) and retail sales implementation (1 Jan 2018), each modeled with a 6-month lag. Interrupted time series analyses used Poisson regression to estimate adjusted rate ratios (aRRs) for pre-existing trend and slope changes at the two policy timepoints. Before RCL passage, CUD rates were decreasing (aRR = 0.95 per year, 95% CI 0.92–0.97). At the time of RCL passage (+6-month lag) there was an increase in slope (adjusted slope change aRR = 1.26, 95% CI 1.14–1.38). At implementation (+6 months) there was a negative slope change (adjusted slope change aRR = 0.86, 95% CI 0.76–0.98); following implementation the trend continued to increase but was no longer statistically significant (adjusted trend aRR = 1.02, 95% CI 0.96–1.09). Authors conclude incident adolescent CUD rates appeared to increase following RCL passage, with increases persisting through retail implementation and up to three years post-passage.

Key findings

  • Prior to RCL passage, adolescent incident CUD diagnosis rates were decreasing.aRR = 0.95 per year (95% CI 0.92–0.97)
  • At the time of RCL passage (modeled with a 6-month lag), there was an increase in the slope of incident CUD diagnoses.Adjusted slope change aRR = 1.26 (95% CI 1.14–1.38)
“Population-based time-series study of adolescents aged 13 to 17 years with Kaiser Permanente Northern California healthcare system membership from January 2014 to February 2020.”
What this piece can’t prove
  • Outcome is limited to EHR-documented incident CUD diagnoses (clinical detection/recording), not direct measurement of underlying cannabis use or disorder prevalence in the full population.
  • Study population is members of a single health system (Kaiser Permanente Northern California), which may limit generalizability.
  • Policy exposures modeled with a fixed 6-month lag for each timepoint as specified; results depend on this modeling choice.
  • Study period ends February 2020 (up to ~3 years post-passage).

1 further detail could not be confirmed from the summary.

Finally, the search trail

Method layer

NewsLink found the paper. Tessa takes you deeper.

NewsLink checks the story. Tessa is where you inspect the paper, authors, evidence, and research context.

Papers considered

The selected paper, plus nearby candidates.

PubMed, Europe PMC, Crossref · 15 candidate papers

And 9 more candidates considered.