Men aged 18 or 19 years who are heavy cannabis users are 40% more likely to die by age 60 than their peers with no history of cannabis use, a new longitudinal study shows.
"This is the first study to show that mortality rate among heavy cannabis users in adolescence is significantly higher than that of individuals who never used cannabis," first author Edison Manrique-Garcia, MD, PhD, of the Karolinska Institutet, Stockholm, Sweden, told
Medscape Medical News.
Death from injury, either accidental or deliberate, were the top causes of death in heavy cannabis users, Dr Manrique-Garcia noted.
The study was
published online April 22 in the
American Journal of Psychiatry.
Surprising Results?
Investigators examined the records of Swedish men (most aged 18 to 19 years) entering compulsory military training in 1969-1970 and identified deaths between the time of training entry and 2011. In the cohort of 45,375 individuals, 3918 (8.6%) died during the 42-year follow-up period. Of those who died, 651 (17%) reported using cannabis at the time of conscription.
For ever-cannabis users as well as heavy users (50 or more times in adolescence), age at time of death was earlier than for nonusers. The difference was statistically significant for both categories of users compared with never-users (log-rank test,
P < .001).
After adjustment for potential confounders, persons who were heavy cannabis users at age 18-19 had a significantly higher risk for early death (hazard ratio, 1.4; 95% confidence interval [CI], 1.1 - 1.8) relative to those without this history. The only cause-of-death category for which there was a dose-dependent relationship was injury undetermined whether accidentally or purposely inflicted (
P for trend < .01), the researchers report.
There was a threefold to fourfold greater risk for premature death among those with a diagnosis of psychosis, but the risk did not differ between cannabis users and nonusers. "No interaction was observed between cannabis use and diagnosis of psychotic disorders with regard to mortality," the researchers write.
"This study shows that use of cannabis can have implications later in life, but further research is needed to clarify the pathway towards mortality," said Dr Manrique-Garcia.
The authors acknowledge that these results "might seem surprising" in light of the fact that a study published in 1990 involving the same cohort failed to find an association between cannabis use and increased risk for death. However, the current analysis had longer follow-up, and the cohort members had reached an age at which the harmful somatic effects of cannabis use were more likely to be apparent, they note.
It is possible that heavy users of cannabis at age 18 to 19 continued to use cannabis during follow-up and that this had the strongest effect on mortality down the line. About 10% of people who ever use cannabis and up to half of those who use it daily will become dependent and continue using it despite experiencing problems, the authors note.
Experts Weigh In
Commenting on the findings for
Medscape Medical News, Wayne Hall, PhD, professor director and inaugural chair, Centre for Youth Substance Abuse Research, the University of Queensland, Australia, noted that the increased relative risk for premature death in regular users of cannabis reported in this study is "very modest, which is what the prior literature would suggest, and we cannot be sure that the elevation in risk is attributable to cannabis use because, as the authors acknowledge, they had a limited ability to control for the effects of other variables that could explain the increased risk, such as alcohol and tobacco use, both of which are higher in regular cannabis users."
Richard Saitz, MD, MPH, chair and professor, Department of Community Health Sciences, Boston University School of Public Health, Massachusetts, told
Medscape Medical News the study is "noteworthy because it is large and has very long-term follow-up. These are the kinds of studies that tell us the risks of things like alcohol and various foodstuffs, because we rarely have better designs like randomized trials. So we rely on this type of study, particularly to identify harms."
Dr Saitz, who was not involved in the study, said the main finding that marijuana is associated with an increased risk for early death is "certainly plausible and raises concern about the current rising tide of public opinion that is a belief that marijuana is completely safe. No doubt the study will be criticized heavily by those who believe that uncritically."
The main limitation of the study, Dr Saitz said, is that marijuana could be associated with other risks for death that occurred or accrued after age 18-19, such as risky behavior that could lead to injury, heavy alcohol use, other drug use, and especially smoking.
"In future studies, it will be helpful to have repeated assessments of those risks and to see what the risks are for various causes of death," he said.
It is also worth noting, Dr Saitz said, that the risk for death was for having used cannabis in adolescence more than 50 times.
"We really don't know how that translates into use after age 18-19. It could be the marijuana use that is the risk, because heavy use then could be associated with continued use. But it may also be a marker for increased risk of death that plays out through associated exposures and behaviors," he explained.
"If people are concerned about their health and health risks, then objectively, the best response to this report would be to take it as a cautionary report ― that marijuana may be associated with increased mortality.
"One can ask oneself how good the data need to be when there is a risk of serious harm. Usually we require less evidence to warn regarding harm than we do regarding benefit. To be more certain, randomized trials would be needed. There is unlikely to be one ever that looks at this length of follow-up, but having shorter-term, larger trials for possible medical indications would likely inform the question," Dr Saitz concluded.
Funding for the study was provided by the Swedish Council for Working Life and Social Research and the Stockholm County Council. The authors, Dr Hall, and Dr Saitz have disclosed no relevant financial relationships.
Am J Psychiatry. Published online April 22, 2016.
Abstract
. 
