Daily Teenage Cannabis Use Cuts High School Completion Odds by 63% — Multi-Cohort Evidence

Three Cohorts, One Analysis

This wasn't a single study. It was a meta-individual analysis — integrating raw, participant-level data from three of the most rigorous longitudinal cohorts in the Southern Hemisphere.

The design's power came from its depth of confound control. The analysis adjusted for up to 53 covariates — including other drug use, family dysfunction, mental health, socioeconomic status, parental education, and childhood behavioral problems. The goal was to strip away everything that might independently predict both cannabis use and poor outcomes, leaving only the association between cannabis itself and what happened next.

What They Found

Seven outcomes. Every one showed a dose-response relationship with adolescent cannabis frequency.

The numbers were stark for daily users. But the dose-response pattern — more cannabis, worse outcomes, in a graded fashion — was what gave the findings their weight. Less-than-monthly users showed small or non-significant associations. Monthly users showed intermediate effects. Weekly users showed clear effects. Daily users showed dramatic effects. That gradient is what you'd expect if the relationship were at least partly causal, not just a product of shared risk factors.

The 53 Covariates

The most common criticism of observational cannabis studies is confounding: kids who use cannabis daily also tend to have other risk factors (family instability, mental health problems, poverty, other substance use) that independently predict poor outcomes. Maybe cannabis is just a marker for a troubled life, not a cause of the trouble.

Silins addressed this more thoroughly than almost any prior study. The 53 covariates included:

After all 53 adjustments, the associations remained. Daily cannabis use before age 17 still predicted a 63% reduction in odds of completing high school. The confounders didn't explain it away.

But "not explained by 53 covariates" is not the same as "causal." There could be unmeasured confounders. The relationship might be partially causal and partially driven by shared vulnerability factors — which is what most researchers now believe.

Why This Study Matters

The Silins study moved the conversation from "does cannabis affect brain scans?" to "does cannabis affect lives?" Brain imaging studies like Albaugh's cortical thinning work show structural changes. Cognitive studies like Scott's meta-analysis show test score differences. But this study showed outcomes that every parent, teacher, and teenager can understand: diplomas, degrees, dependence, and despair.

The multi-cohort design made it especially hard to dismiss. If the same pattern appears independently in three different samples from two different countries, it's unlikely to be a quirk of one dataset.

For parents navigating conversations about cannabis with their teenagers, this study provides something rare: numbers that connect directly to life outcomes rather than abstract neuroscience. Not "your prefrontal cortex is thinning" — but "daily use before 17 means you're less than half as likely to finish high school." That message lands differently in a conversation with a 15-year-old.

And for young adults who used heavily as teenagers and are now dealing with educational setbacks, the study is honest about the risk while the broader literature is honest about recovery. Cognitive function recovers. Brain chemistry normalizes. Educational pathways remain open. The dose-response works in both directions: stopping or reducing use is associated with improved trajectories.

Frequently Asked Questions

Does this prove cannabis causes these outcomes?

It's among the strongest observational evidence available, controlling for 53 potential confounders across three independent cohorts. But observational studies cannot definitively prove causation. The relationship is likely partly causal and partly driven by shared risk factors. What's clear is that the association persists even after extensive confound control and follows a dose-response pattern — both hallmarks of likely causation.

What about occasional teenage use?

The study showed a dose-response: less-than-monthly users had small or non-significant associations with adverse outcomes. The dramatic effects (60%+ reductions in school completion) were concentrated in daily users. Occasional, experimental use before age 17 carried much less risk than daily use — though it still conferred elevated odds of later cannabis dependence.

If I used daily as a teen, am I doomed?

No. The study reports group-level averages, not individual destinies. Many daily adolescent users completed high school, earned degrees, and avoided dependence. The odds are worse, not zero. And other evidence shows that cognitive function recovers with abstinence and that brain chemistry normalizes. Educational pathways remain open at any age.

Why does age matter so much?

The adolescent brain is still developing — particularly the prefrontal cortex, which governs planning, impulse control, and decision-making. Cannabis disrupts this development in ways that adult brains, which have completed maturation, are more resilient to. The Silins findings are specific to use before age 17, during the most active period of prefrontal development.