Declines in marijuana abuse and dependency were greatest among teens (37 percent decrease) and young adults (18 percent decrease) over that period. The change in marijuana abuse and dependency among adults age 26 and older was not statistically meaningful, according to the CDC.
These figures come from nearly 900,000 responses to the National Survey on Drug Use and Health (NSDUH), a massive annual federal survey of American substance use. Dependence and abuse were measured by common criteria set by the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) used by the American Psychiatric Association.
Respondents were considered dependent on marijuana if they reported "health and emotional problems associated with [marijuana] use, unsuccessful attempts to reduce use, tolerance, withdrawal, reducing other activities to use [marijuana], spending a lot of time engaging in activities related to [marijuana] use, or using [marijuana] in greater quantities or for a longer time than intended," according to the CDC.
Similarly, respondents were considered abusers of marijuana if they reported "problems at work, home, and school; problems with family or friends; physical danger; and trouble with the law because of [marijuana] use."
The study noted that abuse and dependency were relatively rare among marijuana users: only 11.9 percent of people who used marijuana in the past year met one of these criteria. That number's fallen by nearly 30 percent since 2002, when 16.7 percent of past-year marijuana users were abusers of the drug or dependent on it.
Marijuana dependence and abuse are becoming less common even as more people use marijuana, the CDC found. Past-month marijuana use is up 35 percent since 2002 among Americans age 12 and over, increasing from 6.2 percent that year to 8.4 percent in 2014. Rates of use increased among every age group except for teenagers, who saw a non-significant decrease in use over the same period.
These numbers from the CDC contradict numbers in a report last year from a team of researchers at Columbia University and elsewhere, who found that marijuana use disorders increased between 2002 and 2013. Those figures came from a different national survey which showed a much more dramatic rise in overall marijuana use than the NSDUH does. The authors of the Columbia study said in their report that the reason for the differences in use rates between the two surveys is "not clear."
Heavy marijuana use -- daily or near-daily in any given month -- did increase sharply over this period, according to the CDC. About 2 percent of Americans used marijuana daily in the past month in 2002, but 3.5 percent used daily in 2014. Again, though, teens bucked this trend: Daily or near-daily use among 12-to-17 year olds fell from 2.4 percent in 2002 to 1.6 percent in 2014.
Normally, researchers expect increases in heavy marijuana use to lead to more marijuana abuse or dependency. But the opposite seems to be happening here -- abuse and dependency are falling as heavy use becomes more common. That represents a bit of a public health puzzle.
The CDC authors posit that changes to medical marijuana law might explain some of the discrepancy. "With changes in medical marijuana laws and, in particular, state laws or policies allowing limited access to low percentages of delta-9-tetrahydrocannabinol (THC) or cannabidiol (CBD), persons who use marijuana daily for medical reasons might be using strains that pose lower risk for dependence or abuse," they write.
It's also likely that changing laws on marijuana use, and changing public attitudes toward the drug, help reduce some of the problem behaviors that lead to abuse or dependence. One of the criteria for abuse, for instance, is "trouble with the law" on account of the drug. It stands to reason that fewer marijuana users have trouble with the law when the law allows marijuana use.
The report also traces some of Americans' changing attitudes toward marijuana use. In 2002, 38.3 percent of Americans aged 12 and over said there was "great risk of harm" from using marijuana monthly. By 2014, that number fell to 26.5 percent. Conversely, the share of Americans seeing "no risk" of using marijuana monthly roughly doubled, from 10 percent in 2002 to 19.9 percent in 2014.
"This national surveillance report provides an exceptional analysis of how marijuana use and perceptions have changed over the past decade in American society," said SAMHSA’s Principal Deputy Administrator Kana Enomoto in a release. The findings are particularly timely, as voters in five states will decide whether to legalize recreational marijuana this November.
More from Wonkblog: