Skip to main content

INSIGHTS BLOG > Do 30% of Cannabis Users Really Have Cannabis Use Disorder?

Do 30% of Cannabis Users Really Have Cannabis Use Disorder?

Written on 27 December 2024

Ruth Fisher, PhD. by Ruth Fisher, PhD

 

Countless researchers and organizations cite a statistic that about 3 in 10 cannabis users have cannabis use disorder (CUD). For example,

  • “Approximately 3 in 10 people who use cannabis have cannabis use disorder.”[1]
  • “Of cannabis users, 30.3 % had CUD, and 52.4 % with CUD also had other substance use disorders.”[2]
  • “It is estimated that among those who use cannabis, between 9% and 30% may develop a cannabis use disorder (CUD).”[3]
  • “Chronic, heavy—every day or almost—use of cannabis products with THC is associated with developing cannabis use disorder, a type of substance use disorder. Studies have estimated that 22% to 30% of people who use cannabis have the disorder.”[4]

Is cannabis use really that problematic? To answer this question we need a better understanding of how these estimates are being derived.

The 30% prevalence of CUD estimate, like any fraction (or percent) depends on two basic factors:

  1. The numerator, which is the number of people with CUD, which then begs the question: How do you define someone who suffers from CUD?
  2. The denominator, which is the population to which the numerator is being compared, which then begs the question: Which population corresponds to the sample estimated in the numerator?

Diagnosing CUD

CUD is defined using criteria from the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V). The DSM, which is published by the American Psychiatric Association, is considered the authoritative guide for diagnosing mental health disorders in the US. The DSM provides 11 criteria for diagnosing most substance abuse disorders:[5]

  1. In the past 12 months, I spend a lot of time either (a) using alcohol or drugs, (b) in activities trying to obtain alcohol or drugs, or (c) recovering from the effects of my drinking or drug use.
  2. In the past 12 months, I found I needed greater amounts of alcohol or drugs than I use to in order to feel intoxicated or to get a desired effect, OR I got much less of an effect by using the same amount of alcohol or drugs as in the past.
  3. In the past 12 months, I experienced withdrawal symptoms when I tried to cut down or stop my drinking or drug use OR I drank alcohol or used drugs to relieve or avoid withdrawal symptoms.
  4. In the past 12 months, I have experienced strong desires, urges, or cravings to use alcohol or drugs.
  1. In the past 12 months, I often used alcohol or drugs in large amounts over longer periods of time than I intended.
  2. In the past 12 months, I often wanted or tried to cut down or control my alcohol or drug use.
  1. In the past 12 months, I gave up or reduced my involvement in important social, occupational, or recreational activities because of my alcohol or drug use.
  2. In the past 12 months, I continued to use alcohol or drugs despite knowing that it likely caused or made worse psychological or physical problems I had (for example, continued drinking or drug use knowing it was making my ulcer or depression worse).
  3. In the past 12 months, my continued alcohol or drug use resulted in my not fulfilling major obligations at work, school, or home (for example, repeated absences or poor performances at work or school; neglecting my children or home).
  4. In the past 12 months, I repeatedly used alcohol or drugs in situations that were physically hazardous (for example, driving a car or operating machinery).
  5. In the past 12 months, I continued to use alcohol or drugs despite having persistent or recurrent social or interpersonal problems caused or made worse by the effects of my drinking or drug use (e.g., arguments with friends or family about my drinking or drug use or physical fights).

A mild, moderate, or severe substance use disorder (SUD) is diagnosed, depending on the number of symptoms substance users report experiencing:

  • Mild SUD: 2 – 3 symptoms 
  • Moderate SUD: 4 – 5 symptoms 
  • Severe SUD: 6 or more symptoms 

How Reasonable Are the CUD Criteria?

Caveat: I’m not a doctor, so my assessment here is based on my personal sense of the situation.

After reading these criteria, it occurred to me to consider how well these criteria characterize my use of caffeine (coffee), in addition to my use of cannabis, at least for the first four criteria.

  1. Time Spent: Considering how much time I spend shopping for coffee beans, grinding them at home before brewing a pot of coffee, waiting in line at coffeeshops to buy coffee when I’m not at home, and drinking my coffee, this criterion definitely applies to my use of caffeine. As for cannabis, anyone who has to go out of their way to buy cannabis (you can‘t pick it up at the grocery store during your regular shopping trips), and/or anyone who rolls their own joints, or otherwise preps their cannabis for use, probably also experiences this criterion as it relates to cannabis.
  2. Tolerance: I have definitely built up a tolerance to caffeine, since it takes several cups before I get the jitters. As for cannabis, when I first started using it, it took several occasions before I felt anything. Studies report that a significant portion of new users, perhaps almost half, have the same experience.[6] Since I — and many others — technically developed a tolerance between my first uses of THC and the use in which I finally felt the effects, does that make this criterion relevant for me? More generally, anyone who uses THC at least semi-regularly has probably built up at least some tolerance, in which case this criterion applies to them.
  3. Withdrawal: I get a caffeine headache if I go more than 24 hours or so without coffee. The best way to get rid of my headache (Tylenol doesn’t help) is simply to drink some coffee. So my coffee habit qualifies me for this SUD criterion. As for cannabis, the Cleveland Clinic cites a study reporting that, “47% of people who used marijuana regularly experienced withdrawal symptoms,” meaning almost half of regular users satisfy this criterion for SUD. The Cleveland Clinic goes on to note that the most common symptoms of cannabis withdrawal include “anger, irritability and aggression; nervousness and anxiety; restlessness; decreased appetite, which may lead to weight loss; depressed mood; insomnia; and disturbing dreams and nightmares,” and that these symptoms may last up to two weeks in most cases, or up to three weeks in heavier users.[7] Most people who go on a diet probably experience these same symptoms. In other words, these symptoms are common, relatively benign, and self-limiting.
  4. Cravings: I love my coffee; I enjoy the experience of drinking it, and I enjoy the uplifting feeling it provides. I look forward to it every morning, and there are times I long to have some later in the day. Similarly, I use cannabis to help me sleep. Especially after not having had access to it for a few nights, I long for it to help me get a good night’s sleep, so I can wake up feeling refreshed. I’m sure many cannabis users also look forward to the effects cannabis provides them. So I qualify as meeting this criterion for both caffeine and cannabis.

I don’t consider myself a heavy coffee drinker (1 – 2 cups of coffee most days) and certainly not a heavy cannabis user (1 dose many nights), and neither my caffeine use nor my cannabis use even comes close to causing any dysfunction in my life. Yet, based on these criteria, I would probably be diagnosed as having a mild to moderate caffeine use disorder, and I may even be diagnosed as having a mild cannabis use disorder.

I agree that criteria 5 and 6 start to touch on problematic use. Someone who’s using more of a substance than they would like is as least starting to have some trouble. And certainly criteria 7 – 11 suggests substance use is causing nontrivial dysfunction in people’s lives. 

In short, I think most would agree that while people who qualify for a mild CUD diagnosis may occasionally suffer some discomfort, they shouldn’t be consider as having a “use disorder.” At the same time, most would probably also agree that people classified as having a moderate CUD may indeed be considered to have some amount of dysfunction, while people classified as having severe CUD almost certainly do suffer from some amount of disorder.

Definition of the Population of Interest

The reported 30% reported prevalence of CUD is taken relative to “cannabis users.” The term “cannabis user” is slippery, having no uniform definition. Is a cannabis user someone who has ever used cannabis? Someone who uses cannabis monthly? Weekly? Daily? Does dose matter? What about profile of ingredients – is someone who uses only CBD the same as someone who uses THC? It’s not hard to quickly realize how difficult it is to pin this down. 

In the studies cited at the beginning of this article, researchers vary on how they define cannabis user and thus the denominator of the fraction (or percentage) being estimated. Definitions include the number of people who reported using cannabis

  • Ever
  • In the past 3 years
  • In the past year

The criteria used to classify cannabis users as having a CUD generally use a 12 month criterion (as seen in the list of criteria presented above), in which case the appropriate denominator should also be based on a 12 month criterion. But again, I don’t think lumping together people with a wide variety of use patterns – frequency, dose size, sample composition – during the course of a year as being in anyway meaningful.

National Survey on Drug Use and Health (NSDUH)

Perhaps the most widely cited data on the subject of cannabis use and abuse is the National Survey on Drug Use and Health (NSDUH), which is conducted every year by the Substance Abuse and Mental Health Services Administration (SAMHSA). According to the NSDUH, the portion of the 12+ US population in 2023 that reported[8] (see Figure 1):

  • Ever using cannabis is about 47%, 
  • Using cannabis in the past year is about 22%, and 
  • Using cannabis in the past month is about 15%. 

Figure 1

Based in the DSM-V criteria presented above, 6.8% of the total population, or 31.0% of past year cannabis users, has a cannabis use disorder. This is where the often cited “3 in 10 cannabis users suffer from a CUD” estimate comes from. The data suggest this estimate has been stable at least over the past three years.

If we partition people determined as suffering from a CUD into mild, moderate, and severe categories, we see that 

  • 3.7% of the population, or 17.2% of past year cannabis users, suffers from a mild CUD, 
  • 1.8% of the population, or 8.2% of past year cannabis users, suffers from a moderate CUD, and 
  • 1.2% of the population, or 5.6% of past year cannabis users, suffers from a severe CUD.

In other words, over half of the people allegedly suffering from a CUD have what would be considered a mild CUD (17.2% of the 31.0%). As we saw, a mild form of CUD is not what I (and probably most people) would probably call a “use disorder.” In fact, I (and probably many others) consider what’s determined to be a moderate CUD as being more of an inconvenience than a disorder. Setting aside the portion classified as suffering from a mild CUD yields much more modest estimates of CUD from the NSDUH data of 3.0% of the total population or 13.8% of past year cannabis users. 

We can take it a step further by considering that cannabis’s morally controversial nature, together with the fact that many people could suffer negative consequences from admitting to using it, lead many people to under-report cannabis use. This underreporting problem may lead both the numerator (people classified as having CUD) and the denominator (people using cannabis in the past year) to be underestimated. If we assume that casual users are likely to report no use, while heavy users are likely to report less intense use, it’s likely the denominator would be more underestimated than the numerator. So then if we assume that the estimates of past year cannabis use are underestimated by 10% (a very conservative estimate), then the estimates of the prevalence of moderate and severe CUD as portion of past year cannabis users further drops from 13.8% (1 in 7.2 users) to 12.6% (1 in 8.0 users).

Taking it to a greater extreme, one could ask: What’s the difference between someone who has used cannabis once in the past 12 months and someone who has ever used cannabis? I would argue not much, if anything. In that case, and if the point is to understand the propensity for “cannabis users” to develop CUD, wouldn’t a more meaningful measure be the portion of people with CUD relative to those who have ever used cannabis? Again, focusing only on people with moderate and severe CUD, the prevalence of CUD among people who have ever reported using cannabis is 6.4% (1 in 15.6 users).

Figure 2

In Conclusion

Based on a closer examination of the criteria used to diagnose CUD – and assuming all the other data and aspects of the methodologies used are accurate – I think more reasonable estimates of the prevalence of CUD is 14% of past year cannabis users, rather than the commonly reported figure of 30%. Taking it a step further, let's assume the data are accurate but ask a more meaningful question: What portion of people reporting they have ever used cannabis qualify as having a moderate or severe CUD during the past 12 months? In this case, the prevalence of CUD is about 6% of ever users.

 

References

[1] Understanding Your Risk for Cannabis Use Disorder. CDC. 2024. https://www.cdc.gov/cannabis/health-effects/cannabis-use-disorder.html

[2] Choi NG, Moore J, and Choi BY. Cannabis use disorder and substance use treatment among U.S. adults. Journal of Substance Use and Addiction Treatment. 2024. https://www.sciencedirect.com/science/article/abs/pii/S294987592400198X

[3] Winters KC et al. Interventions for cannabis use disorder. Curr Opin Psychol. 2020. https://pmc.ncbi.nlm.nih.gov/articles/PMC8175010/

[4] Cannabis (Marijuana). National Institute on Drug Abuse. 2024. https://nida.nih.gov/research-topics/cannabis-marijuana#cannabis-addictive

[5] https://arc.psych.wisc.edu/self-report/dsm-5-checklist-dsm5/

[6] Novak W. High Culture: Marijuana in the Lives of Americans. The Psychedelic Library. (n.d.) http://www.psychedelic-library.org/high_culture2.htm

[7] Marijuana (Weed) Withdrawal. Cleveland Clinic. 2024. https://my.clevelandclinic.org/health/diseases/marijuana-weed-withdrawal

[8] 2022-nsduh-detailed-tables.pdf, 2023-nsduh-detailed-tables.pdf