How to stop smoking weed

Most people who want to stop smoking weed have already tried. Of the 290+ clients we have worked with across 12 countries over 7 years, 98.4% had already tried and failed to stop before they came to us. For weed the attempts tend to look alike: a tolerance break that ends the day it was meant to, a rule about evenings only, a switch to edibles or vapes, a bag thrown out on a Sunday and bought again on Wednesday. The research shows how common it is for those attempts to break down. In a study of 82 people who had stayed off weed for at least two weeks during outpatient treatment, 71% smoked again within six months, and 71% of those who smoked again went back to heavier use¹. In a set of laboratory studies with daily smokers who were not trying to stop, 49% smoked again on the very first day weed was available to them².

The usual explanation for those failures is a lack of willpower. That explanation is wrong in a specific and useful way. Willpower, in the sense of deliberate control by the prefrontal cortex, is real, but it is the faculty most easily taken offline by the conditions in which most smoking happens: stress, tiredness, poor sleep and the presence of something that has been linked to smoking before^3-5. Weed adds two problems of its own. The first is withdrawal. Daily smokers who stop usually feel worse for two to three weeks, with irritability, anxiety, low mood and broken sleep, and the easiest way to feel better is to smoke^{6, 7}. The second is that weed usually does a job. It ends the working day, it helps people fall asleep, it takes the edge off stress. When it stops, the job is still there. So the decision to stop is made by a rested person in the morning and has to be carried out by a tired, edgy person at eleven at night who can't sleep. People are poor at predicting how different they will feel in that second state⁸.

Modern addiction research explains what is happening underneath. Koob and Volkow describe addiction as a cycle of three stages, each run by different brain systems: using and its reward, in which dopamine teaches the habit and turns it into routine; withdrawal and low mood, in which the reward system responds less and the brain's stress systems take a bigger role; and preoccupation, in which craving rises and the prefrontal cortex can't hold it back⁹. Weed has its own version of each stage. THC, the main active compound in cannabis, works on the brain's own cannabinoid system, which helps regulate stress, sleep, appetite and reward. Regular use turns that system's receptors down, which produces tolerance and withdrawal. It disrupts sleep. It becomes the way a day ends. And in much of the world it is now legal, sold on the high street and many times stronger than it was a generation ago, so the triggers for it are hard to avoid.

It is also far more common as a daily habit than most people realise. In the United States, survey data covering 1979 to 2022 show that the share of people who use cannabis daily or almost daily rose fifteenfold between 1992 and 2022. In 1992 there were ten times as many daily or near-daily drinkers as daily or near-daily cannabis users. In 2022, for the first time, there were more daily cannabis users than daily drinkers: 17.7 million against 14.7 million¹⁰. Daily use is the pattern that carries most of the risk, and it is the pattern this paper is mostly about.

The rest of this paper takes the questions people ask about stopping weed one at a time. Section 2 answers whether weed is addictive and whether it is safe to stop all at once. Sections 3 and 4 explain why weed is hard to stop: what it does to the reward, stress and habit systems, and why it ends up in every part of the day. Sections 5 to 9 cover what happens after you stop: withdrawal, sleep and dreams, how long THC stays in the body, how long the brain takes to recover, and whether memory and motivation come back. Sections 10 to 12 deal with tolerance breaks and cutting down, with stopping when the people around you still smoke, and with what to do if you smoke again. Section 13 sets out what the Behavior Control System targets for someone stopping weed, day by day, and Sections 14 and 15 report our results, the limits of this paper and when to see a doctor. Where the program is described, we say what each day targets and why. We do not publish how.

Weed can be addictive, and the risk rises with how often you use it and how strong it is. In a nationally representative US survey, 30.6% of people who had used cannabis in the past year met the criteria for a cannabis use disorder in 2012 to 2013. Over the decade before, the share of all adults who used cannabis more than doubled, and the share with a cannabis use disorder rose from 1.5% to 2.9%¹². A review in Nature Reviews Disease Primers puts cannabis use disorder at about 10% of the 193 million people who use cannabis worldwide¹³. Over a lifetime the odds look lower, because most people who try weed never use it often: in the same national survey data, the cumulative probability that a cannabis user would ever become dependent was 8.9%, compared with 22.7% for alcohol and 67.5% for nicotine¹⁴. Those figures describe everyone who has ever tried it. For someone who already smokes every day, the relevant question is not whether weed can be addictive but how strong the habit has become, and the fact that you are reading this paper is usually the answer.

Two things raise the risk most: frequency and strength. In a survey of adults in the UK, frequent use of high-potency cannabis predicted more severe dependence, especially in young people, while use of lower-potency types did not¹⁵. A systematic review of 20 studies found that higher-potency cannabis was associated with a higher risk of cannabis use disorder and of psychosis¹⁶. Section 3 describes how much stronger cannabis has become.

Dependence also takes time to show itself, and it tends to resolve over years rather than months. Among people in the national survey who did become dependent on cannabis, half had done so about five years after they first used it. Remission is common in the long run: the lifetime probability of remission from cannabis dependence was 97.2%, but half of those cases took about six years from the start of dependence^{14, 17}. Six years is a long time to lose to a habit that most people expect to stop on their own.

2.1Is it safe to stop all at once?

For most people, yes. Cannabis withdrawal is real and common, but it is mainly a syndrome of mood and behavior, of light to moderate intensity, and it can usually be handled without a hospital stay¹⁸. That is different from alcohol and from benzodiazepines, where stopping suddenly after heavy daily use can cause seizures. In a meta-analysis of 47 studies covering 23,518 people who used cannabis regularly or were dependent on it, 47% had cannabis withdrawal syndrome when they stopped. The rate was 17% in general population samples, 54% in outpatient samples and 87% in inpatient samples, and it was higher in daily users and in people who also used tobacco or other drugs¹⁹. Withdrawal is the subject of Section 5. The point here is that it is usually unpleasant rather than dangerous, and that it is most dangerous to the attempt, not to the body: in one study of adults who had tried to stop, 70.4% of those who had been through withdrawal had used cannabis to relieve it⁷.

2.2Who should talk to a doctor first

Some people should talk to a doctor before they stop. The first group is anyone who uses cannabis for a medical condition. Stopping can bring the condition back, and a doctor needs to plan what replaces it. The second is anyone who also drinks heavily every day or takes benzodiazepines or other sleeping or anxiety medication. Alcohol and benzodiazepine withdrawal can be dangerous, and there is a specific risk of swapping one for the other: in a study of 28 daily cannabis users who stopped for 13 days, those with a past alcohol problem increased their drinking by 52% while they were off cannabis, while those without one did not²⁰. In another study of adults who had tried to stop, 41.5% had used alcohol to relieve cannabis withdrawal⁷.

The third group is anyone who has had psychosis, hallucinations or severe paranoia, or has a family history of schizophrenia or bipolar disorder. In a study of first-episode psychosis across 11 sites in Europe and Brazil, daily cannabis use was associated with roughly three times the odds of a psychotic disorder compared with never using, and daily use of high-potency cannabis with nearly five times the odds²¹. A meta-analysis of 10 studies covering 66,816 people found the same dose pattern: the heaviest users had 3.9 times the odds of a psychotic outcome compared with people who didn't use, although the authors note that a causal link can't be firmly established²². In a Danish national register study, 47.4% of people diagnosed with a cannabis-induced psychosis between 1994 and 2014 were later diagnosed with schizophrenia or bipolar disorder²³. Someone with that history needs psychiatric care alongside any attempt to stop. The fourth group is anyone with depression, especially with thoughts of self-harm: heavy cannabis use is associated with a higher risk of developing depression²⁴, and the low mood of early withdrawal can make an existing depression feel worse for a few weeks.

Finally, anyone who has repeated bouts of severe nausea and vomiting while using cannabis should see a doctor. Cannabinoid hyperemesis syndrome is a cycle of vomiting linked to regular cannabis use. In a systematic review of 183 articles, everyone described had a history of regular cannabis use, 97.4% used it at least weekly, 92.3% took hot baths or showers for relief, and symptoms resolved after stopping cannabis in 96.8%; the authors conclude that stopping appears to be the best treatment²⁵. Severe vomiting causes dehydration, which needs medical care.

2.3What a doctor can do

There is no approved medication for stopping cannabis. A Cochrane review of 21 randomized trials with 1,755 participants found that none of the medications tested, including THC preparations, antidepressants, anticonvulsants, buspirone and N-acetylcysteine, made abstinence at the end of treatment more likely than placebo²⁶. A doctor can still help in three ways: by checking for and treating anything the cannabis was covering, such as anxiety, depression, pain or insomnia; by managing any alcohol or medication use safely; and by treating severe withdrawal symptoms, especially sleep, where short-term medical help has been tested in the laboratory²⁷. Talking treatments have better evidence. A Cochrane review of 23 randomized trials with 4,045 participants found that psychosocial treatment reduced how often people used and how dependent they were, compared with minimal treatment, and that an intensive combination of motivational interviewing, cognitive behavioral therapy and rewards for abstinence had the most consistent support. Rates of lasting abstinence were low across all of them²⁸. Our program does not prescribe or recommend medication. It works alongside whatever medical care a person has, and it never replaces it.

Because weed doesn't only feel good. It borrows a system your brain uses every day to calm down, and daily use changes that system so that being without weed feels worse than normal. Self-control at work runs on clear goals, outside structure and a rested prefrontal cortex. Smoking happens at the end of the day, when the structure has gone, the prefrontal cortex is tired, and the urge has been learned so well that it runs before you've decided anything^{3, 29}. Being good at your job doesn't protect you, because the part of you that is good at your job isn't the part that reaches for the grinder at nine at night.

3.1What THC does in the brain

The brain makes its own cannabis-like chemicals, called endocannabinoids, and has receptors for them throughout, the most common being the CB1 receptor. This system works like a dimmer switch on other signals: it turns down stress responses, helps regulate sleep, appetite, pain and memory, and shapes how rewarding things feel. Stress changes it in both directions. Morena and colleagues' review describes how stress lowers levels of one endocannabinoid, anandamide, which helps set off the stress response and anxiety, and raises another, which helps end the stress response; chronic stress reliably reduces CB1 receptors in almost every brain region studied³⁰.

THC acts on the same CB1 receptors, more strongly and for longer than the brain's own chemicals do. That is why the first hit after a hard day works so well: it does in minutes what the brain's own system is struggling to do. THC also acts on the reward system. In a brain imaging study, THC caused dopamine release in the human striatum, the region at the centre of reward and habit learning³¹. Dopamine there is less about pleasure than about learning what to want: it fires most strongly when a reward is better than expected and moves, with repetition, to the things that predict the reward^{32, 33}. The grinder, the lighter, the balcony, the end of the evening's work: each becomes a signal that something good is about to happen.

3.2What daily smoking does to it

The brain adapts to a flood of THC by turning its receptors down. In a brain imaging study of 30 men who smoked cannabis every day and 28 who did not, the daily smokers had about 20% fewer available CB1 receptors in the outer layers of the brain, the cortex, with a 21% reduction in the anterior cingulate, a region involved in attention and self-control. The longer they had smoked, the lower their receptor levels³⁴. A second study, of 11 cannabis-dependent men and 19 controls, found receptor availability 15% lower in almost all brain regions³⁵. Section 8 shows how quickly these receptors come back.

Daily use also seems to dampen the reward system itself. In a study using brain imaging, people who used cannabis had lower dopamine synthesis capacity in the striatum than controls, the difference was driven by those who met criteria for abuse or dependence, and the more they used, the lower it was³⁶. In another, 24 people with a cannabis use disorder and 24 controls were given a drug that raises dopamine. The cannabis users had markedly blunted responses in brain, body and mood, and in the reward centre the blunting was linked to higher scores for negative emotion and to more severe addiction³⁷. A review of THC and the dopamine system concludes that a single dose of THC increases dopamine release, while long-term use is associated with a blunted dopamine system³⁸. In plain terms: the daily smoker's brain has adjusted to weed being there, and when it is gone, ordinary life feels duller than it should for a while.

3.3The weed is stronger than it used to be

The weed people smoke today is not the weed of thirty years ago. The University of Mississippi's potency monitoring program, which tests cannabis seized by US authorities, found that the average THC content of cannabis plant material rose from about 4% in 1995 to about 12% in 2014, while its CBD content, the compound that may soften some of THC's effects, fell, so that THC went from 14 times the CBD content to about 80 times³⁹. Over the next decade the average rose from 8.9% in 2008 to 17.1% in 2017, and the concentrates used in vapes and dabs rose from 6.7% to 55.7% THC¹¹. A meta-analysis of 12 studies from seven countries, covering cannabis samples from 1970 to 2017, found that THC in herbal cannabis rose by 0.29 percentage points a year and in resin by 0.57, while CBD did not change⁴⁰.

Strength matters for stopping because it matters for dependence. Daily use of high-potency cannabis predicts more severe dependence¹⁵, and higher potency is associated with a higher risk of cannabis use disorder¹⁶. A habit built on 17% weed and concentrates is a bigger adjustment for the brain to make than one built on the weed of the 1990s, and the withdrawal that follows can be correspondingly harder. It is one reason that people who stopped easily at twenty find it much harder at thirty-five.

3.4From a choice to a habit

With repetition, behavior moves from deliberate, goal-directed control to habit, a process that runs on its own once the trigger is present²⁹. Wood and Rünger describe habits as responses that are set off by the context in which they have been repeated, which is why they persist even when the person's goals have changed⁴¹. Stress speeds this shift: stressed people fall back on habits rather than on goal-directed choices⁴. In a daily smoker, the habit is linked to particular times, places, objects and moods. When 38 regular cannabis users who had not smoked for three days held a cannabis-related object during brain scanning, reward-related regions including the ventral tegmental area, the anterior cingulate, the insula and the amygdala responded more strongly than to a neutral object⁴². A meta-analysis of cue studies found that regular cannabis users react moderately to cannabis triggers, and that their attention locks onto them much more strongly than their reported craving would suggest⁴³. The habit is often ahead of the person: by the time they notice the urge, attention has already gone to the weed.

That second target rests on research on how rewards compete. Bickel and colleagues describe addiction as, in part, a problem of valuing the immediate reward too highly and the delayed one too little, and argue that treatment should strengthen the value of delayed, effortful rewards⁴⁴. Animal research finds that an enriched environment, with more to do, explore and learn, protects against and reduces drug seeking⁴⁵. The new pursuit is not a distraction. It is a competing source of reward, built in the same hours as the habit.

Most daily smokers don't smoke at random. They smoke at the same points every day, and each of those points has a reason: the first one to get going or to take the edge off the morning, one at the end of work to switch off, one after an argument or a bad call, one in the evening because the evening is empty without it, and one to get to sleep. The reasons feel different, but they work the same way. Weed has become the tool for a job, and every time it does that job at that time and in that place, the link gets stronger⁴¹.

4.1Smoking to cope

People smoke weed for different reasons, and the reason matters. A meta-analysis of 48 studies with 11,274 participants looked at five motives: to cope with bad feelings, to enhance good ones, to be social, to fit in, and to expand awareness. When all five were considered together, coping was one of only two motives that independently predicted problems from cannabis use⁴⁶. Smoking to feel better is the motive most closely tied to the habit getting out of hand, and it is the one that most daily smokers recognise when they are honest about the evening joint.

Stress sets the urge off directly. In a laboratory study, 87 people who were dependent on cannabis went through either a standard social stress test or a neutral task and were then shown cannabis triggers. Both the stress and the cannabis triggers raised craving⁴⁷. Over time a second loop forms on top of the first. Once someone is dependent, part of what weed relieves is the irritability, restlessness and low mood of early withdrawal, which starts within a day or two of the last use⁶. In a study of 104 adults who had seriously tried to stop, withdrawal symptoms prompted people to act to relieve them, including by using cannabis, which the authors describe as negative reinforcement: the habit is kept going by the relief of feeling less bad⁴⁸. Koob and Le Moal describe this shift, from using for pleasure to using to escape a low state the drug itself helped create, as central to addiction⁴⁹.

4.2The first one of the day and the last one

Two points in the day deserve particular attention. The first is the morning. Screening tests for problem cannabis use, such as the Cannabis Abuse Screening Test, ask how often a person smokes before midday and how often they smoke alone, because both mark a habit that has moved from social use to a private daily routine⁵⁰. For some people the morning joint is how the day starts. For others it is the evening one that does the damage to the morning. One of our clients described it this way: “Normally it was a struggle waking up. You can imagine if you have these bad habits in the evening that in the morning you wake up and it takes an hour, 2 hours to kickstart the day, to get yourself in the mood to do stuff.”

The second is the last one before bed. Many daily smokers no longer remember how to fall asleep without weed, and that is the hardest job to take away. THC can shorten the time it takes to fall asleep, but a review of the research concludes it may impair the quality of sleep over the long term⁵¹. When the evening joint stops, sleep is usually the first thing to break, and broken sleep is one of the strongest predictors of smoking again. Section 6 covers it in detail. Here the point is that the bedtime joint is often the one people least want to give up, because it is the one that seems to protect something they need.

4.3Habits that travel together

Weed rarely travels alone. Many people roll it with tobacco, smoke a cigarette with it or straight after, drink with it, or pair it with the phone, porn or food. Agrawal and colleagues' review of cannabis and tobacco describes shared genetic risk, shared routes of use and shared smoking triggers that keep both going, similar withdrawal syndromes, and early evidence that stopping both together predicts better results⁵². In a within-person study of 12 people who used both, withdrawal from cannabis alone and from tobacco alone was of similar size, and stopping both at once was more severe for a short time⁵³. The cost of keeping one going is high: in laboratory studies of daily cannabis smokers, those who also smoked cigarettes were about 19 times more likely to smoke cannabis again when it became available². Our clients describe the same thing. Two of them who came to us for weed came with other habits as well, and one described all three of his as one struggle: “These three things had a huge influence on my life. Not specifically the act of doing it, but the struggle I was going through for years and years and years trying to quit these bad habits.”

The order follows from the research. Stable routines are what habits run on, and a new routine placed in the same time and setting as the old habit can take over its context⁴¹. Light is the main signal that sets the body clock, and morning light in particular anchors sleep timing⁵⁴. Circadian disruption and poor sleep are linked to substance use and relapse across drugs⁵⁵. Building those anchors first means the first evening without weed is not an empty evening.

Most daily smokers feel withdrawal within one to three days of their last smoke, feel worst somewhere between day two and day six, and feel most symptoms fade within two weeks. In the study that first mapped the time course, 18 daily cannabis users were followed through five days of smoking as usual and 45 days without it. Withdrawal effects usually began between days 1 and 3, peaked between days 2 and 6, and most lasted 4 to 14 days, with a size and time course comparable to tobacco withdrawal⁶. A later review puts the usual length of the syndrome at up to three weeks¹⁸. Some symptoms last longer. In a study of chronic users who stopped for 28 days under supervision, symptoms were most pronounced in the first 10 days, but some were still present at the end of the 28 days⁵⁶.

5.1What it feels like

Cannabis withdrawal is mostly a syndrome of mood and behavior, with some physical symptoms⁵⁷. In Budney's study, the effects that followed a withdrawal pattern were aggression, anger, anxiety, decreased appetite, weight loss, irritability, restlessness, shakiness, sleep problems and stomach pain⁶. In a national US survey of frequent cannabis users, 44.3% reported two or more withdrawal symptoms, and the symptoms fell into two groups: one of weakness, sleeping too much and slowed movement, and one of anxiety, restlessness, depression and insomnia⁵⁸. Among adults trying to stop on their own, 95.5% reported at least one withdrawal symptom during their attempt, with a median of nine⁷. Women report stronger withdrawal than men, including physical symptoms such as nausea and stomach pain¹⁸.

5.2How many people get it

About half of people who use cannabis regularly get withdrawal when they stop. In the meta-analysis by Bahji and colleagues of 47 studies and 23,518 people, the pooled rate was 47%. It was 17% in general population samples, 54% in people treated as outpatients and 87% in people treated as inpatients, and it was higher among daily users and those who also used tobacco or had other substance problems¹⁹. The pattern is what you would expect from a syndrome driven by dose: the more and the longer someone has smoked, the more their brain has adapted, and the more there is to readjust.

5.3Why withdrawal decides so many attempts

Withdrawal matters less for what it does to the body than for what it does to the attempt. In a study of 49 cannabis-dependent adults who stopped for two weeks under monitoring, the more withdrawal interfered with their daily lives, the more likely they were to smoke during the attempt, and the more they were smoking a month later⁶⁰. Among adults who had tried to stop on their own, 70.4% of those who had been through withdrawal had used cannabis to relieve it, and many had used alcohol or tobacco for the same purpose⁷. In laboratory studies, the daily smokers with more negative mood and more disrupted sleep during withdrawal were more likely to have a severe relapse when cannabis was offered again². The first two weeks are when the attempt is most fragile, and the reason is usually not a lack of resolve. It is that the person feels worse than they have felt in years and knows exactly what would fix it.

That points to what helps. If withdrawal is mostly irritability, anxiety, low mood and bad sleep, then the person who has already rebuilt their sleep, built some daily exercise and learned a few ways to bring their stress down will feel it less and cope with it better. A small pilot study of 12 cannabis-dependent adults who were not trying to stop found that ten sessions of moderate treadmill exercise over two weeks cut their use from an average of 5.9 joints a day to 2.8 during the exercise period, and reduced craving⁶¹. It was a small, uncontrolled study, and it needs larger trials. But it fits with wider research on exercise as a treatment for drug use⁶².

Slow, structured breathing has been shown to lower negative mood and stress markers in healthy adults⁶³, and naming an emotion in words reduces the amygdala's response to it⁶⁴. Neither is a treatment for withdrawal on its own. They are tools for the irritable, anxious moments that are most likely to end an attempt in its first month.

Because your brain has been sleeping on THC, and it has to relearn how to sleep without it. Daily THC suppresses REM sleep, the stage in which most vivid dreaming happens. When the THC stops, REM comes back with extra force, a rebound, and the dreams that come with it are intense, strange and often about smoking. At the same time, falling asleep takes longer and you wake more in the night. For most people this is worst in the first one to two weeks and settles within about a month, though the research on exactly how long is thin^{18, 65}.

6.1What weed does to sleep while you smoke

THC can make it quicker to fall asleep, which is why so many people use it at night. A review of cannabis and sleep concludes that THC may shorten the time it takes to fall asleep but could impair sleep quality over the long term⁵¹. The best-known laboratory work on this is old but clear. In studies by Feinberg and colleagues, experienced users given THC or cannabis extract every day had fewer of the rapid eye movements that mark REM sleep, with some tolerance to the effect. When the THC stopped abruptly, eye movement density during REM rose to extremely high levels and REM periods lengthened⁶⁶. In plain terms, the daily smoker is not getting a normal amount of dream sleep, and the brain keeps count.

6.2The first nights without it

Sleep studies of people stopping cannabis show the same picture from several angles. In a laboratory study of 20 daily cannabis users who alternated between smoking as usual and three days without, abstinence brought less total sleep, lower sleep efficiency, a longer time to fall asleep, an earlier and larger share of REM sleep, and worse sleep quality as people rated it themselves²⁷. In a study of 18 heavy users who stopped abruptly on an inpatient unit, sleep was recorded on nights 1, 2, 7, 8 and 13. Across those two weeks total sleep time and sleep efficiency fell, time awake after first falling asleep rose, and the amount of REM sleep declined from its early high⁵⁹. Strange, vivid dreams tend to peak later than the other symptoms, around day 4 in an inpatient study¹⁸. A systematic review of 36 studies concluded that sleep is frequently disrupted during cannabis withdrawal, though differences in how studies measured it make the exact pattern hard to pin down⁶⁵.

6.3Why sleep decides whether you smoke again

Sleep is not a side issue. It is one of the strongest links between withdrawal and smoking again. In a study of 55 cannabis-dependent military veterans making a self-guided attempt to stop, those who slept poorly before they stopped were more likely to smoke again within the first two days⁶⁷. In laboratory studies, daily smokers whose sleep was more disrupted during withdrawal were more likely to relapse severely². Bolla and colleagues conclude that poor sleep could contribute to treatment failure in heavy users and that sleep is a target for treatment in its own right⁵⁹.

Part of the reason is what lost sleep does to the brain's control systems. Sleep deprivation weakens the prefrontal cortex's control over the amygdala, making emotional reactions stronger⁶⁸, and in a brain imaging study a single night without sleep reduced the availability of dopamine D2 receptors in the ventral striatum, a change of the kind linked to impulsive reward-seeking⁶⁹. A person who has slept four broken hours is more irritable, more anxious and less able to hold a plan, on the same days that withdrawal is making them irritable and anxious already. More broadly, disrupted body clocks and poor sleep are linked to substance use and relapse across drugs⁵⁵.

Longer than most people expect, if you smoke every day. THC dissolves in fat, and in a regular smoker it builds up in the body's fat stores and leaks back into the blood for weeks after the last smoke. In a study of 30 men who smoked cannabis daily and then stayed on a secure research unit, THC was still detectable in the blood of 2 of the 5 who stayed for 30 days, and its main breakdown product was still detectable in 4 of those 5⁷⁰. In an earlier study of 86 chronic users under strict supervision, urine tests stayed positive for an average of 27 days, and in some people for 46 consecutive days⁷¹. For an occasional user the time is much shorter. Being detectable is not the same as being high: the amounts that leak out of fat are small. But it does mean that a heavy smoker's body is still clearing THC through most of the first month.

7.1Where THC goes

Smoked THC reaches the blood within minutes: in one study, blood levels peaked at about nine minutes, before the last puffs had even been taken. From the blood it moves quickly into the brain and into fat, because THC dissolves far more readily in fat than in water. Huestis's review of human cannabinoid pharmacokinetics describes how, with prolonged exposure, THC concentrates in human fat and is retained for long periods, and how its release back from fat into the blood is the slowest step in clearing it from the body. In chronic users, a terminal half-life of about 4.1 days has been reported for THC in plasma, meaning it takes about four days for the remaining amount to halve, and then four more to halve again⁷².

7.2How long it can be found

The two best-supervised studies give a consistent picture. In Bergamaschi's study, 27 of the 30 daily smokers had THC in their blood on admission. It fell gradually: only 1 of the 11 still on the unit at 26 days had cleared it, and 5% of participants still had levels of at least 1 microgram per litre after 12 days⁷⁰. In Ellis's study of urine, excretion was clearly two-phase: a faster early fall that did not continue, followed by a long, slow tail. During that tail most people had runs of negative tests lasting an average of three days, followed by at least one positive one⁷¹. The practical point for someone who has just stopped is that a positive test weeks later does not mean they have smoked, and a negative one does not mean the body has finished clearing it.

7.3Can stored THC affect you after you stop?

Possibly, a little, and mostly in ways that are hard to feel. In rats given repeated THC, a day without food or a dose of the stress hormone ACTH raised blood THC by releasing it from fat; the authors called this reintoxication and noted that whether it has functional effects in people was unknown⁷³. In people, 14 regular cannabis users who did 35 minutes of exercise on a stationary bike had a small but significant rise in blood THC afterwards, larger in people with a higher body mass index; a short fast did not have the same effect⁷⁴. Bergamaschi and colleagues note that THC lingering in the blood for a month fits the time course of the subtle thinking problems that persist in the first weeks after stopping⁷⁰. None of this is a reason to avoid exercise, which helps with sleep, mood and craving. It is a reason to expect the first month to feel unfinished: the body is still clearing what it stored.

Faster than most people fear, in the parts that are best measured. In brain imaging studies, the CB1 receptors that daily smoking turns down began to recover within two days of stopping and were back to normal levels in most of the brain within about four weeks^{34, 35}. That time frame matches the length of withdrawal, which a review suggests it may help explain¹⁸. Thinking and memory follow a similar course, covered in Section 9. What does not reset in four weeks is the habit: the learned link between weed and the times, places, people and moods where you smoked. That part is slower, and it is what most of the 90 days are for.

8.1The receptors come back

Two imaging studies have tracked the receptors directly. Hirvonen and colleagues scanned 30 daily cannabis smokers and 28 controls, found about 20% fewer available CB1 receptors in the smokers' cortex, and then rescanned 14 of the smokers after 26 days, on average, of continuously monitored abstinence on a secure research unit (range 13 to 32 days). Receptor levels had increased specifically in the regions that had been reduced, and were back to normal in most of them. The exception was the hippocampus, a region central to memory, which did not show the same reversal in that time³⁴. D'Souza and colleagues scanned 11 cannabis-dependent men at baseline and after 2 and 28 days of monitored abstinence. At baseline their receptor availability was 15% lower than in 19 controls. After just two days, the difference was no longer statistically significant, and people with lower receptor levels at day 2 had more withdrawal symptoms. At 28 days there was no significant difference from controls³⁵.

8.2The reward system and mood

The dopamine findings in Section 3 come from people who were still using, and there is less direct evidence on how fast they recover. What the research does show is that the blunted dopamine response in heavy users was linked to negative emotion and to the severity of the addiction³⁷, and that lower dopamine synthesis tracked how much people used³⁶. Both fit what people describe in the first weeks after stopping: ordinary pleasures feel muted, and boredom is hard to bear. Both also point to what helps: rewards that take effort and come from real activity, rebuilt deliberately rather than waited for. Lasting changes in the brain's reward and habit circuits after long drug use are thought to underlie the risk of relapse that outlasts withdrawal⁹.

8.3The habit takes longer

Receptor recovery is the brain undoing its adaptation to THC. The habit is something else: learning. Stopping does not erase what was learned. Bouton's work on extinction shows that when a learned response stops being rewarded, the original learning is not deleted. A new, competing memory is laid down, and it is tied to the context in which it was learned, which is why the old response can return in a new setting, after time passes or after a single reminder⁷⁵. A weed habit learned on a particular sofa, with particular friends, at the end of particular kinds of day, will still be there when those return, long after the receptors have recovered. That is why the program's second and third months work on triggers, thoughts and identity rather than on the body.

For most adults, memory and concentration come back within weeks of stopping. The best evidence suggests that the memory problems seen in heavy smokers are mostly residual effects of recent use and withdrawal, not permanent damage. In studies that required people to have been off cannabis for more than three days, the difference between users and non-users was small and not statistically significant⁷⁶. In studies that required at least 25 days, there was no significant difference in any area of thinking measured⁷⁷. Motivation is more complicated. Weed reduces the willingness to work for rewards while you're under its effects, and the first weeks after stopping often feel flat. But studies of dependent users who were not high have found their motivation largely intact.

9.1Memory

The clearest study followed people through a supervised month without cannabis. Pope and colleagues tested 63 current heavy users, who had smoked at least 5,000 times and were smoking daily, 45 former heavy users and 72 people who had barely used it, on days 0, 1, 7 and 28 of a 28-day washout monitored by urine tests. On days 0, 1 and 7 the current heavy users scored significantly worse at recalling word lists, and the deficit tracked how much cannabis was in their urine at the start. By day 28 there were virtually no significant differences between the groups on any test, and no link between lifetime cannabis use and test scores⁷⁸. A randomized trial in 88 regular users aged 16 to 25 assigned 62 to four weeks of abstinence and 26 to keep using as usual. The abstinent group's verbal learning improved, and the improvement came mostly in the first week⁷⁹.

Two meta-analyses fit the same picture. Scott and colleagues pooled 69 studies of 2,152 young cannabis users and 6,575 comparison participants. Frequent or heavy use was associated with a small reduction in thinking ability overall, an effect size of -0.25. In the 15 studies that required more than 72 hours of abstinence, the effect shrank to -0.08 and was no longer distinguishable from zero, against -0.30 in studies with shorter or no abstinence⁷⁶. Schreiner and Dunn found no significant effect on global performance or any of eight areas of thinking in the 13 studies that tested users after at least 25 days of abstinence, and concluded that negative residual effects are limited to the first 25 days⁷⁷.

9.2The exception: starting young and smoking for decades

The reassuring findings above come mostly from adults and from studies over weeks or months. The most important caution comes from the Dunedin study, which followed 1,037 people in New Zealand from birth to age 38 and tested their thinking at 13, before any cannabis use, and again at 38. People who used cannabis persistently showed a decline across many areas of thinking, even after accounting for education, and people who knew them noticed more problems with memory and attention. The decline was concentrated in those who started in adolescence, it was larger the more persistent the use, and stopping did not fully restore functioning in the adolescent-onset users⁸⁰. The hippocampus finding in Section 8 points the same way³⁴. For someone who started young and smoked for decades, most of the recovery in the first month is still real, but some of the loss may not come back, which is one more reason to stop now rather than later.

9.3Motivation

The idea that weed makes people unmotivated is half right. In a well-controlled laboratory study, a dose of cannabis reduced people's willingness to choose high-effort tasks for money, compared with placebo: a short-lived amotivational state while under its effects. The same study found that cannabis-dependent people tested while not intoxicated chose effortful tasks as often as controls did, though they learned from rewards less well⁸¹. A later study of adults and adolescents who used cannabis three to four days a week found no more apathy and no difference in effort for rewards, wanting or liking than in non-users⁸². A review of cannabis and behavior describes the evidence on lasting amotivation as mixed and calls for better studies⁸³.

For a daily smoker, though, the distinction between acute and lasting effects is academic, because a daily smoker is under acute effects most of the time. If someone smokes from the evening into the night and again in the morning, the short-lived amotivational state is most of their waking life. When they stop, they have to go through withdrawal's flat, irritable weeks first. Then the motivation that was being taken each day can return. One of our clients described the change this way: “Focus on work has been through the roof because my mind doesn't get disturbed anymore with wasteless thoughts, so now I'm using my mind to actually think about my business, about how I can improve on my life.”

Executive functions such as working memory, attention control and flexible thinking can be improved with practice, and they are among the first things to suffer under stress, poor sleep and poor fitness⁵. The training also has a second job: progress the client can see is evidence, against the fear that weed has done permanent damage, that their mind is coming back.

A tolerance break does what its name says: it lowers your tolerance, because the receptors that daily smoking turned down start recovering within two days and are close to normal within about four weeks^{34, 35}. What it doesn't do is change why you smoke. The triggers, the evening, the sleep problem and the stress are all waiting on the day the break ends, and so is a brain that will now feel the first joint more strongly. Many breaks also end at the worst possible time. A one-week break ends just after the peak of withdrawal, and a two-week break ends while sleep is still disturbed. Cutting down can work for people with milder problems, but people with more severe problems are more likely to aim for abstinence, and for a daily smoker who has already tried rules and limits, stopping completely is usually the more reliable choice.

10.1What the research on goals shows

In a trial of 291 adults seeking help for cannabis, those with more problems and more symptoms of dependence were more likely to choose abstinence as their goal. People tended to reach outcomes consistent with the goal they chose, and their confidence that they could reach it predicted whether they did, for both moderation and abstinence⁸⁴. That is a fair result for moderation. But it also shows who moderation suits: people whose problem is milder. Someone who smokes every day, has tried limits before and is reading this paper is usually in the other group.

There are three reasons controlled use is hard for a daily smoker. First, each session keeps the receptors turned down, so withdrawal never gets the chance to finish, and every day off is a day of mild withdrawal pressing for the next session. Second, each session reactivates the learned link between weed and its triggers. Extinction research shows that a learned response returns most readily when the old reward comes back, even briefly⁷⁵. Third, a limit has to be enforced at the moment self-control is weakest: late in the evening, tired, possibly already high. Early lapses after stopping are also strongly associated with going back to heavier use: in the outpatient study described in Section 1, 71% of people who smoked again went on to relapse to heavier use, and earlier lapses were more strongly linked to relapse¹.

10.2Switching products

Switching from joints to edibles, vapes or CBD-heavy strains changes the route, not the habit. Concentrates for vaping averaged 55.7% THC in 2017, against 17.1% for plant material, so a switch to vaping can raise the dose rather than lower it¹¹. CBD on its own does not get people high, but for someone dependent on THC it keeps the ritual, the product and the shop in their routine. And no medication, including THC preparations given as a substitute, has been shown to make abstinence more likely²⁶.

10.3Tobacco and alcohol

The habits weed travels with need a decision of their own. For tobacco the evidence leans toward stopping both: a review of cannabis and tobacco reports early evidence that stopping both together predicts better results⁵², and in laboratory studies daily cannabis smokers who also smoked cigarettes were about 19 times more likely to relapse². Stopping both at once makes withdrawal more severe for a short time⁵³, which is one more reason to rebuild sleep and routine first. For alcohol the risk is substitution. People with a past alcohol problem drank 52% more while off cannabis in one study²⁰, and 41.5% of adults trying to stop on their own used alcohol to ease cannabis withdrawal⁷. Anyone who drinks heavily every day should see a doctor before stopping alcohol, because alcohol withdrawal can be dangerous.

10.4The thoughts that argue for one more

Weed comes with an unusually persuasive set of permission thoughts, because some of them are partly true. It's legal here. It's a plant. It helps me sleep. It's less harmful than drinking. I'll just have one at the weekend. I've been good all week. Each of these is a reason to smoke dressed up as a fact, and each tends to arrive at the moment the urge does. Trying not to think them makes them come back: in Wegner's classic studies, people told not to think about a white bear thought about it more, especially once they were allowed to again⁸⁵. Approaches that teach people to notice a thought, name it and choose an action anyway have a stronger record⁸⁶. People also predict badly, in a calm state, how much an urge will push them later⁸, which is why an answer prepared in advance holds up better than one made up on the spot.