STATS ARTICLES 2003
Research Raises Questions on Gateway Theory of Drug Use
January 30, 2003
Recent research opens new questions about marijuana as a gateway drug
A persistent problem in social science research is the difficulty of determining whether one factor or event actually causes another. The best way to determine whether something is causal is a controlled experiment - where in one condition, the factor is present and in another, it isn't.
But you can't do this when you want to study teenagers and marijuana to see whether those who take up pot-smoking somehow become more likely to develop other illicit drug addictions. Most ethicists - not to mention, parents - would object to an experiment that subjected one group of kids to marijuana at a young age in order to compare them with a control group who were kept marijuana free.
This has lead to endless wrangling over the "gateway" theory, in which one side cites statistics such as "kids who smoke pot are 85 times more likely to use cocaine than those who don't" and the other mentions that a similar statistic could link bicycle riders and Hell's Angels, but that doesn't mean we should ban bikes to keep kids from joining motorcycle gangs.
Now, a new study published in the Journal of the American Medical Association seems to give ammunition to those who support the "pot leads to heroin addiction" position. Researchers found that in a group of both identical and fraternal twins, if one twin tried pot before age 17 and his or her sibling did not, the first twin was 2.1 to 5.2 times more likely to develop a drug problem, either with cocaine, heroin or other substances. Not 85 times, but still significant.
What does this actually mean? There have been numerous studies which suggest that early initiation of any drug use is linked with significantly higher addiction risk - however, early use is typically considered to start at 12 or younger, not 16. If age sixteen is "early" pot initiation, then at least 40% of American teens are early users, according to the latest survey by the Substance Abuse and Mental Health Services Administration (SAMHSA). The average age of first marijuana use is 17, and only 9% of first time marijuana smokers have not previously tried alcohol and/or cigarettes.
Other studies typically labeled early users as those who tried marijuana at age 13 or lower - and these studies have found that the vast majority of such users have problems like mental illness, severe family dysfunction, poverty, histories of child abuse and other trauma which are much more likely to be the root of any later addiction than early exposure to a particular substance.
In the twins study, then, we don't know if the younger users had some experience that lead to him trying pot before their siblings that also predisposed him to higher addiction risk.
Further, the new study finds extraordinarily high rates of drug problems in all of the twins. For example, in both twin groups, whether they used marijuana before or after age 17, 36.2% of the sample had met the criteria for alcoholism (technically called alcohol dependence) at some point in their lives and 25.9% had marijuana dependence.
This is a lot higher than the lifetime alcoholism rate for the American population - 13.8% according to the Epidemiological Catchment Survey (ECA). This may slightly underestimate true rates, since older people surveyed may not recall periods of dependence accurately, but it is nonetheless a major difference.
Moreover, in 2001, 5.8% of people aged 18-25 - the age group with the highest rate of alcohol dependence - met the criteria for alcoholism in the year prior to survey, according to SAMHSA. Four point four percent of the population has lifetime cannabis dependence (ECA), and 3.3% of those aged 18-25 where dependent in the prior year to 2001 (SAMHSA). In other words, with rates of dependence in the twins study group so much higher than those in the American population, it is hard to generalize from them.
But it is safe to say that population rates of drug dependence do not support the notion that addiction rates are elevated enough to reflect 40% of teens having a 2-5 fold increased risk. Only 2.5% of the population over 12 is currently dependent on any illicit drug.
So what to make of the gateway theory? The Institute of Medicine of the National Academy of Sciences, which was asked by Congress to look at the potential harms which could result from legalized medical marijuana said that marijuana use, "does not appear to be a gateway drug to the extent that it is a cause, or even a significant predictor of serious drug abuse." Canadian and British and older U.S. government reports have similarly found little support for the notion.
An editorial published in JAMA with the latest study notes that animal research could answer some questions about whether there is a pharmacological mechanism which attracts pot users to harder drugs, and notes that the authors of the new study mention some such data. The editorial doesn't mention, however, that in the next sentence, the authors diminish the relevance of animal data by noting that the doses of pot used to make rodents more susceptible to other drugs are equivalent to much higher than those commonly taken by young people. And neither the article nor the editorial note that lab rats and mice actually hate marijuana, and, unlike other drugs, will normally avoid it, not work to get more. This makes the rodent model unlikely to be of much relevance tp human behavior.
The fact that marijuana smokers are more likely to use other drugs than non-smokers can be accounted for much more neatly by a model recently developed by the Rand Institute. Basically, Rand shows that an underlying preference for consciousness alteration happens to be coupled with marijuana being the easiest illegal drug to get.
We have to use Occam's razor when trying to determine causation - there's no need to invoke some mystical "brain change" that leads a minority of pot smokers to progress to other drugs. Just as people who like dark chocolate might want to try white chocolate, or just as art lovers may enjoy both Picasso and Van Gogh, people who are interested in drugs are interested in drugs.
As with everything in life, the true obsessives are a minority. After all, though 34% of the population over 12 reports having tried marijuana, only 1.4% has ever even tried, let alone become addicted to, heroin.