Cannabis Use Disorder Risk, Internet Use, and Depression: What a Quebec Study Found
| Audience | Adults who use cannabis, clinicians, behavioral-health professionals, and readers interested in cannabis use disorder, mood symptoms, and digital behavior |
| Primary Topic | problematic internet use and depression among adults at high risk for cannabis use disorder |
| Source | Read the full source |
Cannabis Use Disorder Risk, Internet Use, and Depression: What a Quebec Study Found
A cross-sectional study of 874 Quebec adults at high risk for cannabis use disorder found that problematic internet use was associated with moderate-to-severe depressive symptoms after adjustment for measured demographic, substance-use, and anxiety factors. The result supports broader clinical screening, but it does not establish which condition came first or whether one caused the other.
| Study Type | Cross-sectional secondary analysis of a population-based online survey |
| Population | Quebec adults using recreational cannabis at least weekly and screening at high risk for cannabis use disorder |
| Participants | 874 adults |
| Recruitment | April 18 to May 15, 2024 |
| Cannabis Risk Measure | Cannabis Abuse Screening Test score of 7 or higher |
| Depression Measure | PHQ-8 score of 10 or higher |
| Internet Use Measure | Internet Addiction Test score of 50 or higher |
| Problematic Internet Use | 10.6% of participants |
| Moderate-to-Severe Depressive Symptoms | 26.5% of participants |
| Adjusted Association | OR 2.19, 95% CI 1.15 to 4.19; P = .018 |
| Key Boundary | Cross-sectional association, not causation or temporal sequence |
| Published | July 29, 2026 |
| PMID | 42577019 |
| DOI | 10.1016/j.abrep.2026.100730 |
Moderate-to-severe depressive symptoms were reported by 58.1% of participants with problematic internet use and 22.8% of those without it. After adjustment, problematic internet use remained associated with depressive symptoms, with an odds ratio of 2.19 and a 95% confidence interval from 1.15 to 4.19.
The model included measured demographic characteristics, substance-use indicators, and anxiety symptoms. Adjustment makes the association more informative, but it cannot remove unmeasured confounding or create temporal information that the survey did not collect.
Participants entered this analysis because they scored 7 or higher on the Cannabis Abuse Screening Test. Depressive symptoms were defined with the PHQ-8, and problematic internet use with the Internet Addiction Test.
These validated instruments are useful for population research and clinical case-finding. They do not substitute for a diagnostic interview, so the paper concerns elevated risk and symptom burden rather than confirmed cannabis use disorder, major depressive disorder, or a formally established internet-use disorder.
Participants with problematic internet use also had higher anxiety scores and more frequent problematic alcohol and other-substance use. This pattern argues against interpreting the internet-depression association as an isolated two-variable relationship.
For clinical care, the result supports a broader history that includes cannabis frequency and consequences, mood, anxiety, sleep, alcohol, other drugs, and how online behavior affects daily function. The study did not test whether any specific screening or treatment strategy improves outcomes.
A single survey cannot determine whether depressive symptoms contributed to dysregulated internet use, whether problematic internet use worsened mood, whether both reflected shared stressors, or whether other factors produced the observed association.
The authors discuss possible shared pathways such as coping, emotion regulation, sleep, and social context, but these were not established as mediators in this analysis. Longitudinal and intervention studies are needed before prevention or treatment claims can be made.
Cannabis use disorder can coexist with depression, anxiety, alcohol use, other drug use, sleep disruption, and functional problems. Digital behavior may be one more clinically relevant domain when patients describe loss of control or impairment.
The practical value of this paper is not a new causal theory. It is a reminder that focused screening can miss a wider pattern of distress and coping behaviors that may need coordinated assessment.
I would not use this paper to blame screens, cannabis, or depression for the others. I would use it as a prompt to ask a better set of questions. When cannabis use feels hard to control, mood, anxiety, sleep, alcohol, other substances, and digital habits all deserve attention.
The most useful next step is clinical clarification. Screening results can open a conversation, but diagnosis, risk assessment, and treatment planning require context, duration, functional impact, and the patient’s goals.
How to Interpret This Cannabis, Internet Use, and Depression Study
This paper identifies a clinically relevant cluster of screening results in one population.
Its value lies in broader assessment, not in assigning cause or prescribing a single solution.
A Four-Step Reading Frame
Evidence type
This is a cross-sectional observational analysis, so it can describe association but not temporal sequence or causation.
Population
Participants were Quebec adults who used recreational cannabis at least weekly and screened at high risk for cannabis use disorder.
Outcome meaning
The study used validated screening thresholds for depressive symptoms and problematic internet use, not diagnostic interviews.
Clinical boundary
The result supports broader assessment but does not prove that changing cannabis use or internet use will improve depression.
Eight Ways to Read the Cannabis, Internet Use, and Depression Signal
Clinical, methodological, behavioral, safety, equity, and research perspectives
An Association Is Not a Personal Verdict
If cannabis use, low mood, and online behavior all feel difficult to control, this study suggests that the overlap is worth discussing. It does not show that one problem caused the others, and a screening score is not a diagnosis.
A useful conversation focuses on what has changed, how long it has lasted, and what is affecting sleep, work, relationships, safety, or daily routines. The goal is not to assign blame to cannabis or screens. It is to identify the full pattern and decide what kind of support fits.
Expand the History Beyond Cannabis Frequency
The study population was defined by weekly cannabis use and an elevated Cannabis Abuse Screening Test score, but the clinically relevant pattern extended into depression, anxiety, alcohol, other substances, and digital behavior. Frequency alone would not capture that complexity.
When cannabis-related impairment is suspected, assessment can include mood, anxiety, sleep, suicidality when indicated, alcohol and other drugs, loss of control, withdrawal, role function, and the function of online activity. The paper supports case-finding, not a new diagnostic shortcut or a specific treatment protocol.
Cross-Sectional Data Cannot Establish Direction
All key variables were collected in the same survey period. That design can estimate how strongly screening results occur together, but it cannot determine whether problematic internet use preceded depressive symptoms, followed them, or developed alongside shared risks.
Multivariable adjustment reduced some measured confounding, including anxiety and other substance-use indicators. It did not address every possible factor, and backward variable selection can add uncertainty. The adjusted odds ratio should therefore be read as a refined association, not as an estimate of causal effect.
Validated Tools Still Have Boundaries
The CAST, PHQ-8, and Internet Addiction Test provide structured ways to identify elevated risk or symptom burden. They are useful in research and can help clinicians decide when a deeper conversation is warranted.
None of these scores alone confirms a clinical diagnosis. Interpretation depends on duration, impairment, context, differential diagnosis, and patient priorities. The internet-use construct is especially broad and may combine gaming, social media, gambling, information seeking, or other behaviors with different meanings. Screening should open inquiry rather than close it.
Function Matters More Than Raw Screen Time
The study focused on loss of control and functional effects captured by the Internet Addiction Test, not simply the number of hours spent online. That distinction matters because work, education, social support, and health care can all require substantial digital activity.
A more useful history asks whether online behavior displaces sleep, movement, responsibilities, or in-person relationships, and whether it is used to manage distress. The paper did not identify a particular platform or online activity as the driver of the association, so platform-specific claims would go beyond the evidence.
Depressive Symptoms Require Their Own Risk Assessment
More than one quarter of participants reported moderate-to-severe depressive symptoms, and the proportion was higher among those with problematic internet use. A population statistic cannot determine any individual’s level of risk, but it supports taking mood symptoms seriously in cannabis-related care.
Clinical assessment may need to address functional decline, hopelessness, sleep, substance interactions, and self-harm risk when indicated. The study used the PHQ-8, which does not include the PHQ-9 self-harm item, so it cannot describe suicidality. That important boundary should remain explicit.
Quebec Context Shapes Generalizability
Participants came from a stratified Quebec web panel in a legal adult-use cannabis environment. Cultural norms, health care access, product availability, digital habits, language, and local prevention programs may shape both exposure and reporting.
The sampling strategy strengthens relevance within the target Quebec population, but it does not guarantee that the same prevalence or association applies elsewhere. Replication across regions and clinical settings is important, especially where cannabis laws, stigma, broadband access, and behavioral-health services differ.
Longitudinal Studies Must Test Sequence and Change
The next useful studies should follow people over time and measure cannabis use patterns, cannabis-related impairment, specific digital activities, mood, anxiety, sleep, and life stress repeatedly. That design could clarify which changes tend to precede others.
Intervention research is also necessary. It should test whether treating depression, cannabis use disorder, problematic digital behavior, or shared mechanisms improves outcomes across domains. Until then, it is premature to present screen reduction, cannabis reduction, or any single therapy as the evidence-based answer to this observed association.
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
Frequently Asked Questions
What did the Quebec study examine?
It examined whether problematic internet use was associated with moderate-to-severe depressive symptoms among adults who used cannabis weekly and screened at high risk for cannabis use disorder.
How many people participated?
The analysis included 874 adults living in Quebec, Canada.
How common was problematic internet use?
Problematic internet use was identified in 10.6% of participants using the Internet Addiction Test threshold specified by the study.
How common were moderate-to-severe depressive symptoms?
A PHQ-8 score of 10 or higher was present in 26.5% of participants.
What was the adjusted association?
Problematic internet use was associated with higher odds of moderate-to-severe depressive symptoms, with an adjusted odds ratio of 2.19 and a 95% confidence interval from 1.15 to 4.19.
Did the study prove that internet use causes depression?
No. The cross-sectional design measured variables at one point in time and cannot establish cause or temporal order.
Did every participant have a cannabis use disorder diagnosis?
No. Participants met a screening threshold indicating high risk for cannabis use disorder, but the study did not use a diagnostic clinical interview.
Did cannabis cause the association?
The study cannot answer that question. It focused on people at high risk for cannabis use disorder and did not establish cannabis as the cause of problematic internet use or depressive symptoms.
What can clinicians reasonably take from the findings?
The findings support asking about mood, anxiety, sleep, alcohol, other substance use, cannabis-related impairment, and digital behavior when one of these concerns is present.
What research is needed next?
Longitudinal studies are needed to determine temporal order, and intervention studies are needed to test whether changing any of these behaviors improves clinical outcomes.
