Is Marijuana Addictive?
Stay up to date with the latest headlines via email.
Is marijuana addictive?
The National Institute on Drug Abuse (NIDA) says it is. According to its “Marijuana Abuse” research report, “Long-term marijuana use can lead to addiction; that is, people have difficulty controlling their drug use and cannot stop even though it interferes with many aspects of their lives."
The Office of National Drug Control Policy’s abovetheinfluence.com Web site is blunter. “Marijuana is addictive, with more teens in treatment with a primary diagnosis of marijuana dependence than for all other illicit drugs combined,” it declares.
Sentiment among marijuana users and advocates is the exact opposite. While a minority of pot-smokers get high so frequently it impairs their functioning, the vast majority insist they can do it occasionally or regularly without problems.
The word “addiction” conjures up the stereotype of a heroin junkie, willing to lie, manipulate, steal, and perform cut-rate oral sex in order to avoid suffering the withdrawal—nausea, diarrhea and flu-like distress—that comes after they go without the drug for several hours. Cocaine, however, does not produce a similar physical withdrawal. So over the last generation, the concept has evolved to a more complex, subjective model.
NIDA now calls addiction “a complex illness characterized by intense and, at times, uncontrollable drug craving, along with compulsive drug seeking and use that persist even in the face of devastating consequences.” The DSM-IV, the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders, lists the criteria for drug “dependence” as tolerance, withdrawal, continuing to use despite negative psychological or physical consequences, using more than you want to, unsuccessful attempts to cut down or quit, excessive time spent procuring the drug, and withdrawal from social, work or family obligations.
Some marijuana users reach that point. “Marijuana controls our lives!” the 12-step recovery group Marijuana Anonymous exclaims on its Web site. “Our lives, our thinking, and our desires center around marijuana—scoring it, dealing it, and finding ways to stay high.”
The vast majority don’t reach that point. Among the commonly used drugs, pot is the least likely to cause dependence. The estimate most often cited, based on a NIDA-supported survey from the early 1990s, is that 9 percent of people who use marijuana will develop dependence at some point in their lives, compared with 15 percent for alcohol, 17 percent for cocaine, 23 percent for heroin, and 32 percent for tobacco.
Mitch Earleywine, author of Understanding Marijuana: A New Look at the Scientific Evidence and a psychology professor at the University at Albany in New York, disputes that 9 percent figure. If you focus on genuine problems, “instead of some manufactured diagnosis,” he says, maybe 4 to 8 percent of regular marijuana users have problems.
The two main arguments that cannabis is addictive are the number of people admitted to drug-treatment programs primarily for marijuana use and research indicating that chronic use may cause a withdrawal syndrome.
The proportion of people admitted to rehab primarily for marijuana use increased significantly from 1999 to 2009, from 13.5 percent of admissions to 18 percent of the nearly 2 million tracked by federal Substance Abuse and Mental Health Services Administration in its Treatment Episode Data Set. More than half of teenagers in rehab were there for pot.
However, according to SAMHSA figures from 2009, 56 percent of the more than 350,000 people admitted to drug treatment for marijuana were referred by the criminal-justice system, such as after an arrest or probation violation. Only 15 percent were “self-referred,” seeking rehab voluntarily. For the 282,000 heroin admissions, the proportions were exactly the opposite: 55 percent came in on their own, and only 15 percent were referred by legal authorities. For crack, 36 percent of the about 130,000 admissions were self-referred, and 29 percent sent over by the criminal-justice system.
The National Organization for the Reform of Marijuana Laws notes that the increase in pot-rehab admissions parallels “the proportional increase in marijuana arrests reported over the same period.”
Teenagers are less likely to come in without pressure, counters Alan Budney, a professor of psychiatry at the University of Arkansas for Medical Sciences Center for Addiction Research. In the adolescent clinic he runs, “only about 5 percent of kids say ‘I need help.’”
Evidence of Withdrawal
There is some evidence emerging that marijuana causes withdrawal. Over the last 15 years, several experiments have demonstrated withdrawal symptoms in mice, rats and monkeys. In a 1998 study at Virginia Commonwealth University, mice were dosed with massive amounts of THC, from 10 to 30 milligrams per kilogram of body weight, twice a day for a week, and then injected with rimonabant, a “cannabinoid antagonist” that blocks the brain’s CB1 receptors for both endogenous cannabinoids and THC. That “promptly precipitated a profound withdrawal syndrome.”
The mice got “very dramatic withdrawal” symptoms such as “wet-dog shakes,” says Dale Deutsch, a biochemistry professor at Stony Brook University and former head of the International Cannabinoid Research Society. A 2010 study from the University of Texas Health Science Center found that rhesus monkeys given 1 mg/kg of THC twice a day developed head shaking and sleep disturbances after being given rimonabant, and that those symptoms eased after the monkeys were again given THC.
Earleywine dismisses those studies as “artificial.” The rimonabant, he notes, “tosses the THC off every cell in their body at once. That’s so unlike what happens in humans.” Even so, he adds, the shakiness described as “withdrawal” in mice lasted only 15 minutes.
“They never were able to show there was any clear-cut withdrawal until rimonabant,” says Deutsch. “In real life, however, the THC hangs around a long time, and you do not get precipitated withdrawal unless you give an antagonist.”
THC has a much longer half-life than other drugs. Cocaine is half gone from the body within 90 minutes. Heroin, even after being metabolized to morphine, takes two to seven hours. But as THC is absorbed by the body’s fatty tissues and membranes, its half-life is at least three to four days, and has been estimated at 12 days in chronic marijuana users. (That is also why drug tests are more likely to detect marijuana than cocaine or heroin.)
The mice also got doses of THC far above what humans would normally consume. For a 154-pound human, 10 mg/kg twice a day would be like smoking a half-ounce of 10 percent-THC pot in one day—with none of the THC going up in smoke. The minimum effective dose of THC in humans, notes Deutsch, is 10 micrograms per kilo of body weight.
The reason for such large doses, says Aron Lichtman, a pharmacology and toxicology professor at Virginia Commonwealth University who coauthored a 2002 study on cannabis withdrawal in rodents, is that rats and mice metabolize drugs very quickly. They have “very efficient livers,” he says.
It’s normal to give rodents 10 times the equivalent human dose in drug experiments, he explains; it’s also done in studies of cocaine and morphine. Humans in marijuana studies also get very high doses, he adds.
Rimonabant is used, he explains, because “spontaneous withdrawal”—symptoms that develop merely from stopping the drug—is hard to reproduce in rodents. It has been observed in monkeys, he says. One study found them more active at night 24-48 hours after cessation of THC.
However, Deutsch says, rimonabant may cause problems on its own, because it interferes with the body’s endogenous cannabinoid system. A French drug company sold it as an appetite suppressant, but it was taken off the European market in 2008 because it increased the risk of depression. The U.S. Food and Drug Administration rejected it for similar reasons in 2007. In the 2010 University of Texas study, the control group of monkeys that were given rimonabant but no THC also developed head shaking and tachycardia.
Symptoms of Withdrawal
The evidence for cannabis withdrawal in humans is based on inpatient studies involving abstinence after sustained consumption of oral THC or smoked cannabis, and also on symptoms either described in interviews or checked on lists by people in rehab and in studies of outpatients trying to quit.
The main complaints, says Alan Budney, are irritability, sleeplessness, anger, restlessness, and “not feeling right.” They also include appetite loss and, less frequently, depression and nausea. They peak after two or three days of abstinence and last a week or two, he wrote in 2008.
“There’s no doubt that it exists. We see it all the time in clinics,” he says. When he first published his studies, he says, he got phone calls from Marijuana Anonymous thanking him for confirming what its participants had been talking about in meetings.
The concept is “still somewhat controversial,” says Lichtman. Earleywine believes it is “very misleading” to use the word withdrawal. “There’s obviously some mild discomfort in humans who quit suddenly,” he says, “but the connotations of the word ‘withdrawal’ make people think of heroin.” He also finds it absurd that loss of appetite is considered a symptom of marijuana withdrawal, given that “it’s a drug so notorious for enhancing appetite that it’s got its own expression—the munchies.”
The DSM-IV did not include cannabis withdrawal, on the grounds that its symptoms’ “clinical significance is uncertain.” Budney is trying to get it listed in the forthcoming DSM-V.
“We think it’s important enough,” he says. He adds that he’s not on a crusade against marijuana: “I’m out there helping people who want to quit.”
Whatever cannabis withdrawal is, it’s definitely much milder than detoxing from heroin or alcohol. Stopping drinking can cause fatal seizures in severe alcoholics, notes Carl Hart, a neuroscientist at Columbia University College of Physicians and Surgeons. He calls cannabis withdrawal “unpleasant but not threatening.”
“It’s certainly not like heroin withdrawal,” says Budney. But he adds that the “milder symptoms are what make people go back and smoke.”
Those symptoms are likely confined to a minority of potheads. Hart says they appear in a “select group of heavy, heavy users”; some, but not all, people who smoke several joints a day will have sleep disturbances and become irritable when they quit. Budney estimates that half of heavy users and “not a high percentage” of moderate users will suffer withdrawal.
Dr. Jeff Hergenretter, a Sebastopol, California, physician who has been seeing mainly medical-marijuana patients for the past 12 years, says the number is insignificant. Less than 1 percent of his 2,000 patients, he says have described any withdrawal symptoms, “and it’s short-lived, mild, lasting a day or two.” The vast majority have “no withdrawal symptoms whatsoever. No discomfort, no dysphoria, no nothing.”
Marijuana withdrawal is most comparable to quitting cigarettes, says Budney. In a study he published in 2008, tobacco smokers and daily pot-smokers who were trying to quit reported equal levels of discomfort in telephone interviews. However, 44 percent of the cigarette-smokers were using nicotine-replacement therapies such as patches.
That “definitely would warp results,” he agrees, but “we just had to take who we had.” Nevertheless, he says, the people using nicotine replacement also reported being irritable, and an “accumulation across these studies” confirms his theory.
“Most people think it’s akin to coffee craving. I know that’s true in my case,” says Fred Gardner, editor of O’Shaughnessy’s, a California-based magazine devoted to medical-marijuana research.
“Caffeine withdrawal may be in the DSM this time,” says Budney. While coffee can improve people’s work functioning, he explains, those who stop drinking it can suffer acute headaches.
“I don’t get withdrawal. I just miss smoking it,” says one veteran pothead, an artist who can’t afford to buy as much since he and his wife lost their day jobs in the Great Recession. But another longtime toker, a college professor, says, “when I don’t smoke for a couple days, I get headaches.”
Scientists have not yet been able to explain why some people will get intoxicated compulsively and destructively, while others can take the same drug and not develop such problems. Starting use at a younger age and having had a traumatic childhood definitely increase the risks, but there are large and poorly understood variations in individuals’ personalities and neurochemistry. For similar reasons, it is hard to predict how depressed people will respond to various medications.
“It’s just like how people respond to meds. Not everything is pharmacological,” says Budney. “People are vulnerable. Some are more vulnerable than others.”
“That you have some sort of withdrawal really isn’t important,” Carl Hart avers. If treating withdrawal was all that was needed to get people to stop using a drug, he explains, it would be easy. The issue is more about how people use the drug than about the pharmacology of it.
“Whether or not our addiction is psychological, physical, or both, matters little,” Marijuana Anonymous says. “When it comes to the use of marijuana, we have lost the power of choice.”
The neurological basis for cannabis withdrawal is “still an open question,” says Lichtman. A Spanish researcher recently found that THC inhibits a brain-messenger chemical called cyclic AMP, and that quantities of cyclic AMP “overshoot” in the cerebellum during withdrawal. However, he cautions, this is just a correlation, not clear proof of changes in the brain.
Only a small percentage of marijuana users become dependent, he says, but “given the huge numbers of people smoking marijuana, even if [dependence] is a low amount, that’s still a huge number of people.”
Marijuana “doesn’t meet what I think are stringent definitions of addiction. Habit-forming it is, addicting it isn’t,” Dr. Hergenretter responds. “I have one patient out of 2,000 who describes himself as addicted to cannabis. He says that every time he gets some, he smokes it all and wants more. He’s really an exception. He’s one of the rare cases who describes a dysphoria the next day if he doesn’t smoke.”
The 9-percent dependence figure “doesn’t meet reality in any real way,” Hergenretter continues. While users develop tolerance, he explains, it’s very common for them never to increase how much or how often they smoke, and most are not “socially damaged by need for marijuana,” not compelled to use it to the point of “stealing their neighbors’ TV.”
With more than 800,000 people a year arrested for marijuana, the definition of “dependence” has been politicized, Mitch Earleywine says. “The distinction between ‘psychological’ and ‘physical’ addiction is completely invented,” he argues. “When later editions of the DSM came out, psychological was added primarily because cannabis wasn't creating withdrawal symptoms, and prohibitionists wanted there to still be some sort of cannabis dependence. So dependence suddenly became two types: ‘with physical’ and ‘without physical features.’ Essentially, if you had enough problems but didn't show tolerance or withdrawal, you could still get a dependence diagnosis.”
One of the main criteria for defining problematic drug use is continuing to get high even in the face of negative consequences. But if getting arrested is one of those consequences, responds Fred Gardner, “that’s a function of prohibition.”
Many marijuana users see risking arrest to grow, buy and smoke the herb as defying an unjust law, not a symptom of addiction.