20-year study found adolescent self-harm independently predicted weekly cannabis use at age 35, even after controlling for mental health and behavior problems
Among all the adverse outcomes associated with adolescent self-harm, weekly cannabis use at age 35 was the only one that remained independently associated after accounting for adolescent mental health, substance use, and antisocial behavior.
Quick Facts
What This Study Found
In a population-based cohort of 1,671 Australians followed from adolescence to age 35, those who self-harmed during adolescence (8% of the cohort) had significantly worse outcomes across multiple domains: social disadvantage, mental health, and substance use.
Most of these associations were explained by concurrent adolescent mental health disorders and substance use. However, weekly cannabis use at age 35 uniquely survived all adjustments (adjusted OR 2.27, 95% CI 1.09-4.69). Even after controlling for adolescent depression, anxiety, antisocial behavior, and substance use, adolescent self-harm still independently predicted heavy cannabis use two decades later.
Other notable associations before full adjustment included daily smoking (OR 2.00), cannabis dependence (OR 2.67), and financial hardship (OR 1.88).
Key Numbers
Sample: 1,671 analyzed (135 self-harmers, 1,536 controls). Weekly cannabis use at 35: OR 3.18 after mental health adjustment, OR 2.27 after full adjustment (95% CI 1.09-4.69). Cannabis dependence: OR 2.67 before adjustment. Daily smoking: OR 2.00. Self-harm prevalence: 8%.
How They Did This
Prospective cohort study using the Victorian Adolescent Health Cohort Study (Australia). A stratified random sample of 1,943 adolescents from 44 schools was recruited starting in 1992 and followed through 2014. Self-harm was assessed across four adolescent waves (mean age 15.9). Outcomes at age 35 were analyzed using progressively adjusted logistic regression models.
Why This Research Matters
The finding that weekly cannabis use was the only outcome that remained independently associated with adolescent self-harm after exhaustive adjustment is striking. It suggests a specific pathway from self-harm to heavy cannabis use that operates independently of the mental health and behavioral problems typically blamed for both.
The Bigger Picture
This study reframes the relationship between self-harm and substance use. Rather than both being symptoms of underlying mental illness, the independent persistence of the cannabis association suggests that self-harm in adolescence may specifically channel into cannabis use through pathways not captured by depression, anxiety, or conduct disorder diagnoses.
What This Study Doesn't Tell Us
Self-harm was self-reported. Weekly cannabis use at age 35 was a relatively small outcome, and the confidence interval for the adjusted association was wide (1.09-4.69). The study cannot determine the mechanism linking self-harm to later cannabis use. Attrition over 20 years of follow-up may introduce bias.
Questions This Raises
- ?What pathway connects adolescent self-harm to heavy cannabis use two decades later?
- ?Is cannabis used as self-medication for the psychological sequelae of self-harm?
- ?Would early intervention for self-harming adolescents reduce their later cannabis use?
Trust & Context
- Key Stat:
- Weekly cannabis use at 35 was 2.3x more likely in those who self-harmed as teens, even after controlling for mental health
- Evidence Grade:
- Population-based longitudinal cohort with 20 years of follow-up and progressive statistical adjustment. Strong design published in The Lancet Child & Adolescent Health.
- Study Age:
- Published in 2017. The unique persistence of the self-harm-cannabis association continues to be explored in developmental research.
- Original Title:
- 20-year outcomes in adolescents who self-harm: a population-based cohort study.
- Published In:
- The Lancet. Child & adolescent health, 1(3), 195-202 (2017)
- Authors:
- Borschmann, Rohan(2), Becker, Denise, Coffey, Carolyn(2), Spry, Elizabeth, Moreno-Betancur, Margarita, Moran, Paul, Patton, George C
- Database ID:
- RTHC-01340
Evidence Hierarchy
Follows a group of people over time to track how outcomes develop.
What do these levels mean? →Frequently Asked Questions
Why was cannabis use the only outcome that stayed significant?
Most negative outcomes associated with adolescent self-harm (depression, anxiety, other substance use) were explained by concurrent mental health problems during adolescence. Cannabis use at age 35 persisted even after accounting for all of these factors, suggesting a distinct pathway from self-harm to long-term cannabis use.
Does self-harm cause cannabis addiction?
The study cannot determine causation, only that the association persists after extensive statistical adjustment. Possible explanations include cannabis as self-medication, shared neurobiological vulnerabilities, or unknown third factors not captured in the study.
Read More on RethinkTHC
Cite This Study
https://rethinkthc.com/research/RTHC-01340APA
Borschmann, Rohan; Becker, Denise; Coffey, Carolyn; Spry, Elizabeth; Moreno-Betancur, Margarita; Moran, Paul; Patton, George C. (2017). 20-year outcomes in adolescents who self-harm: a population-based cohort study.. The Lancet. Child & adolescent health, 1(3), 195-202. https://doi.org/10.1016/S2352-4642(17)30007-X
MLA
Borschmann, Rohan, et al. "20-year outcomes in adolescents who self-harm: a population-based cohort study.." The Lancet. Child & adolescent health, 2017. https://doi.org/10.1016/S2352-4642(17)30007-X
RethinkTHC
RethinkTHC Research Database. "20-year outcomes in adolescents who self-harm: a population-..." RTHC-01340. Retrieved from https://rethinkthc.com/research/borschmann-2017-20year-outcomes-in-adolescents
Access the Original Study
Study data sourced from PubMed, a service of the U.S. National Library of Medicine, National Institutes of Health.
This study breakdown was produced by the RethinkTHC research team. We analyze and report published research findings without making health recommendations. All interpretations are based solely on the published abstract and study data.