| Audience | Clinicians, addiction and mental health professionals, patients with ADHD, and evidence-focused readers trying to understand how ADHD and cannabis use overlap. |
| Primary Topic | Pooled prevalence of ADHD by sex and by substance among adults who use psychoactive substances and alcohol and are engaged with treatment services, with cannabis showing the highest subgroup rate. |
| Source | Read the full PubMed record |
Table of Contents
- 1 in 3 Cannabis Users in Treatment Had ADHD: What the New Meta-Analysis Shows
- How to Read a Prevalence Meta-Analysis Without Turning It Into a Causal Claim
- The Same Study Can Mean Different Things Depending on the Question Being Asked
- A Reason to Ask About ADHD, Not a Diagnosis
- Lower the Threshold for ADHD Screening
- The Overlap Is Real, the Direction Is Not Shown
- This Is a Treatment Sample, Not the General Public
- Prevalence Varied by Sex and Substance
- Services Should Build ADHD Screening In
- Shared Vulnerability Remains an Open Question
- What Better Evidence Still Needs
- Frequently Asked Questions
1 in 3 Cannabis Users in Treatment Had ADHD: What the New Meta-Analysis Shows
A July 10, 2026 systematic review and meta-analysis pooled 84 studies and 34,036 adults engaged with treatment services and estimated an overall ADHD prevalence of 22%. The highest subgroup was people who use cannabis at 33%. The finding is about co-occurrence in treatment populations, not proof that cannabis causes ADHD, but it strengthens the case for routine ADHD screening in people who use cannabis.
| Study Type | Systematic review and meta-analysis |
| Population | 34,036 adults aged 18 and older engaged with drug, alcohol, or health services |
| Settings | Acute hospitals, community services, and prison healthcare |
| Studies Pooled | 84 studies contributing 99 prevalence estimates |
| Overall Finding | ADHD prevalence of 22%, 95% CI 19 to 25 percent |
| Highest Subgroup | People who use cannabis at 33%, 95% CI 28 to 39 percent |
| Lowest Subgroup | Women who use cocaine at 5%, 95% CI 1 to 11 percent |
| Autism Note | Only one study reported autism spectrum disorder, so an ASD meta-analysis was not feasible |
| Heterogeneity | Significant heterogeneity was detected across studies |
| Journal | Journal of Dual Diagnosis |
| Published | July 10, 2026 |
| PMID | 42429286 |
| DOI | 10.1080/15504263.2026.2686088 |
This was a systematic review and meta-analysis, not a single new cohort. The authors combined 84 studies contributing 99 separate prevalence estimates, covering 34,036 adults who were engaged with drug, alcohol, or health services across acute hospitals, community settings, and prison healthcare.
Because the sample is drawn from people already connected to treatment or institutional care, the numbers describe a clinical population, not the general public. That distinction shapes how far the results can travel.
The overall pooled ADHD prevalence was 22%, with a 95% confidence interval of 19 to 25 percent. Within that, the highest subgroup estimate belonged to people who use cannabis at 33%, with a 95% confidence interval of 28 to 39 percent, which the authors reported as significantly higher than the overall rate.
In plain terms, roughly one in three adults who used cannabis in these treatment samples met criteria for ADHD. That is a large overlap, and it is the headline most relevant to cannabis clinicians and patients.
Cannabis was not the only elevated group. The same 33% rate appeared among women who use benzodiazepines, and men who use benzodiazepines were close behind at 31%. Among people who use opioids, prevalence was 27% for women.
At the other end, the lowest rate was 5% among women who use cocaine. These contrasts show that ADHD prevalence varied meaningfully by both substance and sex rather than being uniform.
A high co-occurrence rate does not establish direction. ADHD symptoms may increase the likelihood of substance use, substance use patterns may influence how ADHD is recognized or assessed, and shared underlying factors may drive both.
This review measured how often the two appear together in treatment populations. It was not designed to prove that cannabis causes ADHD or that ADHD alone causes cannabis use, and it should not be read that way.
The actionable message is screening. If ADHD is present in roughly a third of cannabis-using adults in these settings, then a low threshold for validated ADHD screening is reasonable, followed by structured assessment where indicated.
Recognizing ADHD can change the whole care plan, from how attention and impulsivity are addressed to how treatment adherence and relapse risk are managed. Missing it leaves a treatable condition unaddressed.
ADHD in adults is frequently underdiagnosed, and the overlap with substance use has been described for years without precise pooled numbers by substance and sex. This review adds quantitative structure to a clinical impression many providers already hold.
For cannabis medicine specifically, the finding reinforces a recurring theme: the people in front of us often carry unrecognized psychiatric comorbidity, and better screening tends to improve care more than assumptions in either direction.
What stands out to me is not a claim that cannabis creates ADHD. It is the reminder that ADHD is sitting quietly in a large share of the cannabis-using patients we already treat, and that we will only find it if we look.
The right clinical response is not alarm about cannabis. It is a lower threshold for validated ADHD screening in people who use cannabis, followed by thoughtful assessment and a care plan that treats the whole person rather than a single substance.
How to Read a Prevalence Meta-Analysis Without Turning It Into a Causal Claim
Prevalence studies are easy to misread because a large co-occurrence number feels like a cause.
A better reading separates who was studied, what was measured, and what the design can and cannot support.
Four questions worth asking before you simplify the result
Who was studied?
Adults already engaged with drug, alcohol, or health services in hospitals, community settings, and prisons, not a general-population sample of cannabis users.
What was measured?
The co-occurrence of an ADHD diagnosis or assessment with substance use, reported as prevalence by substance and by sex.
What did the design allow?
Cross-sectional prevalence estimates can show overlap but cannot establish which condition came first or whether one causes the other.
What action is justified now?
A lower threshold for validated ADHD screening in people who use cannabis is justified. Causal statements about cannabis and ADHD are not.
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 Reason to Ask About ADHD, Not a Diagnosis
If you use cannabis and have struggled with attention, organization, or impulsivity, this study is a reason to raise ADHD with a clinician. It does not mean cannabis caused anything, and it does not diagnose you.
ADHD is treatable, and recognizing it can change how the rest of your care is planned.
Lower the Threshold for ADHD Screening
With roughly one in three cannabis-using adults in these settings meeting ADHD criteria, a validated screening tool is a reasonable default rather than an exception.
Recognizing ADHD can reshape adherence, impulsivity management, and relapse planning in ways that treating a substance alone will not.
The Overlap Is Real, the Direction Is Not Shown
A skeptical reader should hold two facts together: the co-occurrence is substantial, and the cross-sectional design cannot say which came first.
Significant heterogeneity and varied ADHD assessment methods also widen the uncertainty behind any single subgroup number.
This Is a Treatment Sample, Not the General Public
Everyone counted here was already engaged with services in hospitals, community settings, or prisons. That selection concentrates comorbidity and inflates prevalence relative to the general population.
The 33% figure describes cannabis users in these clinical settings, not cannabis users everywhere.
Prevalence Varied by Sex and Substance
The review reported ADHD prevalence by sex, with striking contrasts such as 33% among women who use benzodiazepines and only 5% among women who use cocaine.
These differences argue against treating people who use substances as a single uniform group.
Services Should Build ADHD Screening In
If ADHD is this common in treatment populations, then routine, validated screening pathways in addiction and mental health services are a sensible system-level response.
Under-recognition is a service design problem as much as an individual clinical one.
The overlap is consistent with several explanations, including self-management of ADHD symptoms, shared genetic or environmental vulnerability, and effects of substance use on ADHD recognition.
This review does not adjudicate between them, and readers should resist collapsing the possibilities into one tidy story.
What Better Evidence Still Needs
Stronger evidence would come from longitudinal designs, general-population sampling, standardized ADHD assessment, and analyses that separate cannabis from co-occurring substances.
Until then, this review is best used to prompt screening and sharpen questions, not to declare cause.
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.
CED coverage of a study on rising cannabis use disorder among adolescents and the treatment delays that follow. It highlights how access barriers can leave co-occurring conditions unaddressed. The piece is useful background for understanding comorbidity in cannabis-using patients.
CED analysis of a BMJ cohort study examining cannabis use disorder outcomes. It illustrates how large datasets are used to study CUD and its correlates. The framing pairs well with a prevalence meta-analysis of comorbid conditions.
CED coverage connecting psychiatric vulnerability and cannabis use patterns in young adults. It reinforces how mental health conditions and cannabis use frequently overlap. This is a natural companion to ADHD and substance use comorbidity.
Frequently Asked Questions
What did this study actually measure?
It pooled 84 studies covering 34,036 adults engaged with treatment or health services and estimated how common ADHD is among people who use various substances, broken down by substance and by sex.
What was the headline cannabis finding?
People who use cannabis had the highest ADHD prevalence of any subgroup at 33%, with a 95% confidence interval of 28 to 39 percent, compared with an overall pooled rate of 22%.
Does this mean cannabis causes ADHD?
No. This was a cross-sectional prevalence meta-analysis. It shows that ADHD and cannabis use often co-occur in treatment populations, but it cannot establish that one causes the other.
Does it mean one in three of all cannabis users has ADHD?
No. The 33% figure applies to cannabis users engaged with treatment, hospital, or prison health services, not to cannabis users in the general population.
Why is the treatment-setting detail so important?
People already in services tend to carry more concentrated comorbidity, so prevalence in this group is expected to be higher than in the broader public. Setting shapes the number.
Were other substances also linked to high ADHD prevalence?
Yes. Benzodiazepine use was associated with high rates, including 33% among women and 31% among men, while the lowest rate was 5% among women who use cocaine.
How strong is this evidence?
It is a systematic review and meta-analysis, which is a strong design for pooling prevalence, but significant heterogeneity, varied ADHD assessment methods, and the cross-sectional data limit precision and prevent causal claims.
What is the practical takeaway for clinicians?
Use a low threshold for validated ADHD screening in adults who use cannabis, then complete structured assessment where indicated, because unrecognized ADHD can affect adherence, impulsivity, and relapse risk.
Did the study look at autism spectrum disorder?
It intended to, but only one included study reported autism spectrum disorder, so a meaningful ASD meta-analysis was not feasible.
What kind of research would strengthen confidence here?
Longitudinal designs, general-population sampling, standardized ADHD assessment, and analyses that separate cannabis from co-occurring substances would clarify direction and improve precision.
