Are We Over-Diagnosing Cannabis Use Disorder in Prescribed Medical Cannabis Patients?
| Audience | Medical cannabis patients, prescribing clinicians, addiction medicine specialists, and policymakers shaping cannabis use disorder diagnostic standards |
| Primary Topic | whether standard DSM-5 Cannabis Use Disorder criteria appropriately classify patients using medical cannabis exactly as prescribed |
| Source | Read the full source | Read PDF |
Are We Over-Diagnosing Cannabis Use Disorder in Prescribed Medical Cannabis Patients?
A new Addiction journal study of 2,054 Australians prescribed medical cannabis found that removing tolerance and withdrawal from Cannabis Use Disorder criteria, which are expected effects of following a prescription, cut apparent disorder rates among as-prescribed users by more than half. The findings raise questions about whether current diagnostic tools fit prescribed medical cannabis populations.
| Study Type | Secondary analysis of a cross-sectional consumer survey (CAMS22) |
| Sample | 2,054 Australian adults who used prescribed medical cannabis in the past 12 months |
| As-Prescribed Subgroup | 986 participants (48%) reporting cannabis use only as prescribed |
| Diagnostic Frameworks Compared | Standard 11-item DSM-5 CUD vs. proposed 9-item Prescription CUD (excludes tolerance and withdrawal) |
| Any CUD, Full Sample | 36.4% (11-item) vs 31.2% (mixed Pr-CUD/CUD framework) |
| Any CUD, As-Prescribed Subgroup | 20% (11-item) vs 9% (9-item Pr-CUD); absolute difference 11.8 points (95% CI 8.3-15.8) |
| Moderate-Severe CUD, As-Prescribed Subgroup | 5% (11-item) vs 2% (9-item Pr-CUD); absolute difference 2.9 points (95% CI 1.4-5.0) |
| Published | August 2, 2026 |
| Journal | Addiction |
| PMID | 42543176 |
| DOI | 10.1111/add.70559 |
Researchers used data from the CAMS22 survey, an anonymous online cross-sectional survey of Australian adults who reported medical cannabis use in the past year, and applied two different diagnostic frameworks to the same respondents.
The standard approach used all 11 DSM-5 Cannabis Use Disorder criteria for every participant. The alternative Prescription Cannabis Use Disorder framework used only 9 criteria, excluding tolerance and withdrawal, for the 986 participants who reported using cannabis strictly as prescribed.
Among the 986 patients using cannabis only as prescribed, standard DSM-5 criteria classified 20 percent as meeting any Cannabis Use Disorder threshold. The prescription-aware framework classified only 9 percent, an absolute reduction of 11.8 percentage points (95% CI 8.3 to 15.8; odds ratio 1.3, 95% CI 1.1 to 1.4).
For moderate-to-severe Cannabis Use Disorder, the same subgroup dropped from 5 percent under standard criteria to 2 percent under the prescription-aware framework (absolute reduction 2.9 percentage points, 95% CI 1.4 to 5.0; odds ratio 2.6, 95% CI 1.6 to 4.4).
Tolerance and withdrawal are pharmacologically expected with many prescribed medications, including opioids and benzodiazepines, and are not by themselves evidence of a substance use disorder when a patient is following a prescription.
The authors argue that applying diagnostic criteria designed for illicit cannabis use to appropriately prescribed patients may inflate apparent disorder prevalence, with downstream consequences for clinical labeling, insurance, employment, and legal proceedings.
Every major cannabinoid receptor pathway can produce tolerance with regular dosing, which is why prescribers of opioids, benzodiazepines, and other controlled substances already distinguish physical dependence from a substance use disorder in clinical documentation.
As more jurisdictions move toward prescription-based medical cannabis models, diagnostic tools originally built around unregulated, illicit use are being applied to a clinically different population, and this study is part of a growing effort to test whether that translation holds up.
This finding matches what many of us who prescribe medical cannabis see in practice. A patient who has titrated to a stable, effective dose and experiences mild tolerance or a few days of disrupted sleep if they miss doses is showing expected pharmacology, not loss of control over their use.
That said, this is one survey from one country, and it should not be read as permission to stop screening prescribed patients for genuine problematic use. The right response is to keep asking about function, cravings, and unsuccessful attempts to cut down, the criteria that actually distinguish a use disorder from expected physiologic adaptation, rather than assuming every prescribed patient with tolerance has a diagnosable condition, or dismissing every prescribed patient’s use as automatically safe.
How to Read a Diagnostic Framework Comparison Study
This study did not test a treatment. It tested two different ways of counting the same survey responses to see how diagnostic definitions change apparent disease prevalence.
That distinction matters because the result here is a measurement question, not a treatment effect question.
Four Questions for Interpreting the Findings
What Was Actually Compared?
Two scoring systems applied to the same 2,054 respondents, not two different patient populations or a treatment versus placebo.
Who Was Included?
Australian adults who self-reported prescribed medical cannabis use in the past year through an anonymous online survey, not a clinically verified or randomly sampled population.
What Changed, and What Did Not?
The proposed criteria changed which respondents met a Cannabis Use Disorder threshold. It did not change how much cannabis anyone used or their underlying clinical status.
What Would Confirm This Finding?
Replication in other countries and prescribing systems, and validation of the Prescription CUD framework against independent clinical outcomes rather than self-report alone.
The Same Study Can Mean Different Things Depending on the Question Being Asked
Scientific papers rarely answer a single question. Patients, clinicians, researchers, policymakers, and critics often read the same data differently. The perspectives below explore how this study looks through several evidence-based lenses.
A Prescription Does Not Automatically Mean or Rule Out a Use Disorder
If you use medical cannabis exactly as prescribed and notice you need more to get the same effect, or feel different if you miss a dose, that alone does not mean you have a use disorder under this proposed framework.
It is still worth discussing openly with your prescriber, since only a clinician who knows your full picture can tell whether your pattern of use reflects expected tolerance or something that needs a closer look.
Consider Whether Tolerance and Withdrawal Are Doing Fair Diagnostic Work
For patients using cannabis strictly as prescribed, tolerance and withdrawal may reflect expected pharmacology rather than loss of control, similar to how these criteria are already interpreted for other prescribed controlled substances.
This does not mean abandoning screening. Criteria like unsuccessful attempts to cut down, craving, and continued use despite functional harm remain informative regardless of prescription status.
Physical Dependence Is Not the Same as Addiction
Family members may worry when they see a loved one needing a higher cannabis dose over time or feeling unwell without it.
This study is a reminder that tolerance and withdrawal are common with many prescribed medications and are not, by themselves, proof of addiction.
One Country, One Survey, One Proposed Framework
This is a single cross-sectional, self-reported survey from Australia, and the Prescription CUD framework it tests has not been independently validated or adopted into diagnostic manuals.
Self-classification of using only as prescribed cannot be verified against pharmacy records within this design, which leaves room for misclassification in either direction.
Diagnostic Definitions Have Real Consequences
A Cannabis Use Disorder diagnosis can affect insurance coverage, employment, custody proceedings, and legal outcomes, so how these criteria are defined is not a purely academic question.
The authors argue that policy and insurance systems should account for the difference between prescribed, monitored cannabis use and unregulated illicit use when applying diagnostic labels.
A Within-Sample Comparison, Not a New Population Study
By applying two scoring frameworks to the same respondents, the authors isolated the specific effect of removing tolerance and withdrawal criteria, which is methodologically useful.
The next step for this research is external validation: does the Prescription CUD framework track meaningfully with functional outcomes, treatment-seeking, or clinician-rated severity, not just self-reported symptom counts.
Echoes a Long-Standing Debate in Prescribed Controlled Substances
Addiction medicine has long distinguished physical dependence from substance use disorder for opioids and benzodiazepines prescribed appropriately for pain, anxiety, or other indications.
This study extends that same clinical logic to medical cannabis, arguing that a prescription-aware diagnostic lens may reduce misclassification without eliminating the need for careful screening.
Validation Against Outcomes Is the Next Step
Future work should test whether the Prescription CUD framework predicts real clinical outcomes, such as treatment-seeking, functional impairment, or dose escalation beyond prescribed parameters, better than standard DSM-5 criteria.
Studies outside Australia, in prescribing systems such as the United States, will also be needed before any framework change could be broadly adopted.
Join the Conversation
Have a question about how this applies to your situation? Ask Dr. Caplan
Want to discuss this topic with other patients and caregivers? Join the forum discussion
When a new paper overlaps with earlier CED Clinic coverage, we preserve the chain instead of hiding the overlap. These links point to older related posts so readers can compare what is new, what is repeated, and how the evidence has moved.
A BMJ cohort study of 606,434 veterans found GLP-1 drugs reduced cannabis use disorder risk by 14 percent.
SAMHSA’s 2026 advisory instructs physicians to evaluate cannabis use based on clinical signs of harm, not use alone.
A survey of cannabis users found significant overlap between medical and recreational use patterns.
Frequently Asked Questions
What did this study examine?
It examined whether removing tolerance and withdrawal from standard DSM-5 Cannabis Use Disorder criteria changes apparent disorder prevalence among 2,054 Australian adults prescribed medical cannabis.
What is Prescription Cannabis Use Disorder?
It is a proposed 9-item diagnostic framework that uses the same DSM-5 criteria as standard Cannabis Use Disorder but excludes tolerance and withdrawal, intended for patients using cannabis strictly as prescribed.
How much did the diagnosis rate change?
Among the 986 patients using cannabis only as prescribed, any Cannabis Use Disorder fell from 20 percent under standard 11-item DSM-5 criteria to 9 percent under the 9-item Prescription CUD framework.
What about more severe disorder?
Moderate-to-severe Cannabis Use Disorder in the same as-prescribed subgroup fell from 5 percent to 2 percent using the prescription-aware framework.
Where and when was this published?
The study was published in the journal Addiction on August 2, 2026, with PMID 42543176 and DOI 10.1111/add.70559.
Who conducted the study?
Researchers Nicholas Lintzeris and Llewellyn Mills, affiliated with the University of Sydney and South Eastern Sydney Local Health District in Australia, using data from the CAMS22 survey.
Does this mean tolerance never matters clinically?
No. Tolerance and withdrawal can still be clinically important. The study argues they should be weighed differently when a patient is using cannabis strictly as prescribed rather than automatically counted as disorder criteria.
Is the Prescription CUD framework an official diagnosis?
Not yet. It is a proposed research framework and has not been adopted into DSM-5 or validated against independent clinical outcomes.
What are the main limitations?
The study relies on a single, self-reported, anonymous cross-sectional survey from Australia, so findings may not generalize to other countries or prescribing systems, and using only as prescribed could not be independently verified.
What should patients using prescribed medical cannabis take from this?
A prescription does not automatically confirm or rule out a use disorder. Patients should discuss any changes in tolerance, dose, or perceived need for cannabis openly with their prescriber.
