rethinkTHC Search
Menu
Study breakdown

Cannabis use disorder falls on a spectrum rather than being an all-or-nothing condition

Cross SectionalStrong evidence
The takeaway

Taxometric analysis of two independent samples confirms that cannabis use disorder has a dimensional structure, meaning it exists on a continuum of severity rather than as a distinct category.

Clinicians diagnosing cannabis use disorder, researchers studying addiction taxonomy, and policymakers interested in how cannabis problems are classified.

What the researchers found

Using three taxometric procedures across a large community sample (N=3,623) and a clinical inpatient sample (N=621), cannabis use disorder consistently showed dimensional rather than categorical structure, with mean CCFIs well below the 0.50 threshold for taxonicity.

Why it matters

This has direct implications for how cannabis addiction is diagnosed. The findings support the DSM-5 dimensional approach (mild/moderate/severe) over the ICD-11 binary model (dependence yes/no).

The numbers in context

Community sample CCFIs: MAMBAC=0.48, MAXEIG=0.30, L-Mode=0.43, mean=0.40. Clinical sample CCFIs: MAMBAC=0.09, MAXEIG=0.21, L-Mode=0.26, mean=0.19. Values below 0.50 indicate dimensional structure.

How the study worked

Three taxometric procedures (MAMBAC, MAXEIG, L-Mode) applied to DSM-5 CUD criteria in the nationally representative NESARC-III community sample and a clinical inpatient substance use treatment sample from Ontario, Canada.

What this study cannot tell us

Taxometric methods require specific distributional assumptions. The clinical sample was exclusively from inpatient treatment, representing the more severe end of the spectrum.

How to read the evidence

Two independent samples with three complementary statistical methods all converging on the same conclusion provides strong evidence.

When this study was published

2025 publication.

The bigger picture

The debate over whether addiction is categorical versus dimensional has major implications for treatment, policy, and how individuals understand their own relationship with cannabis. This study adds evidence that problematic cannabis use is better understood as a gradient.

Questions still open

  • Should the ICD-11 revise its categorical approach to cannabis dependence based on accumulating dimensional evidence?
  • How should treatment intensity be calibrated along the severity continuum?

Common questions

What does "dimensional" mean in this context?
A dimensional structure means cannabis use disorder exists along a spectrum of severity, like blood pressure or depression, rather than being an all-or-nothing condition.
How does this affect diagnosis?
The DSM-5 already uses a dimensional approach, classifying cannabis use disorder as mild (2-3 symptoms), moderate (4-5), or severe (6+). This study supports that approach over the ICD-11 binary dependence diagnosis.

Read the original research

Taxonicity of cannabis use disorder: Findings from a large community sample and an inpatient clinical sample.

Journal of psychopathology and clinical science, 134(7), 790-798

Citation

Garber, Molly L; Taisir, Radia; Costello, Jean; MacKillop, James. (2025). Taxonicity of cannabis use disorder: Findings from a large community sample and an inpatient clinical sample.. Journal of psychopathology and clinical science, 134(7), 790-798. https://doi.org/10.1037/abn0001031

Explore the wider topic