rethinkTHC Search
Menu
Study breakdown

When Average Cannabis Use Goes Up Among Teens, Heavy Use Goes Up Too

Cross SectionalStrong evidence
The takeaway

Adolescent cannabis use in Sweden follows the total consumption model: when average use rises across the population, heavy use rises proportionally.

Policy researchers, public health professionals, and anyone interested in the population dynamics of cannabis use.

1 point rise in average use = 1.8 point rise in heavy users

What the researchers found

The distribution of cannabis use frequency among Swedish adolescents remained remarkably stable over 33 years, and increases in average use were consistently associated with proportional increases in high-frequency users.

Why it matters

This has major implications for cannabis policy. If the total consumption model applies, then any policy that increases average cannabis use among adolescents will inevitably increase the number of heavy users, not just casual experimenters.

The numbers in context

A 1 percentage point increase in average frequency of use was associated with a 1.794 percentage point increase in high-frequency users among 9th graders (p < 0.001). The elasticity for the median user (P50) was 0.914.

How the study worked

Researchers analyzed annual surveys of Swedish 9th-grade (n=180,059) and high school students (n=80,925) from 1990 to 2023 using Lorenz curves, Gini coefficients, and regression analysis to test whether cannabis use follows the total consumption model.

What this study cannot tell us

Based on Swedish adolescents only, where cannabis use levels are relatively low compared to North America. Self-reported frequency may underestimate actual use. The model may not apply in settings with very different use patterns.

How to read the evidence

Published in Addiction with over 260,000 respondents across 33 years of consistent survey methodology.

When this study was published

2026 study analyzing trends through 2023.

The bigger picture

The total consumption model is well established for alcohol but had never been tested for cannabis. Finding that it applies suggests population-level prevention strategies that reduce average use could be more effective than targeting only heavy users.

Questions still open

  • Does this model hold in countries with legal cannabis markets and higher baseline use? Could targeted interventions for heavy users still be effective even if the population-level relationship holds?

Common questions

What is the total consumption model?
A theory from alcohol research stating that when average consumption in a population rises, all levels of use rise proportionally, including the most harmful heavy use.
What does this mean for cannabis legalization?
It suggests that policies increasing average cannabis use among teens would also increase the number of heavy teen users, making population-level prevention important.

Read the original research

Does the total consumption model apply to cannabis use?

Addiction (Abingdon, England)

Citation

Norström, Thor; Leifman, Håkan. (2026). Does the total consumption model apply to cannabis use?. Addiction (Abingdon, England). https://doi.org/10.1111/add.70353

Explore the wider topic