DSM-IV, published in 1994, did not include a cannabis withdrawal disorder diagnosis. DSM-IV-TR clearly stated the reason for the omission: “Symptoms of possible cannabis withdrawal . . . have been described in association with the use of very high doses, but their clinical significance is uncertain.”1
A 2002 Psychiatric Times article alerted the readership to new studies on marijuana withdrawal and called for additional research and reconsideration of the status of a cannabis withdrawal disorder.2 A proliferation of research followed that report, including publication of 2 review articles that documented compelling data supporting recognition of a clinically significant cannabis withdrawal disorder.3,4
Studies of diverse clinical and general population samples subsequently provided further support for the contention that abrupt abstinence from cannabis can produce a reliable and valid withdrawal disorder of clinical importance. This article briefly summarizes this literature and its implications for diagnostic nosology and clinical care. The article also describes the process and progress of the American Psychiatric Association’s renewed consideration for including cannabis withdrawal disorder in DSM-5.
Reviews of our research and that of many others conclude that abstinence from cannabis causes several withdrawal symptoms. These reviews also outline evidence for a “true” diagnosis of cannabis withdrawal disorder.3,4
Reliability and validity have been demonstrated in experimental studies. These studies have shown that individuals have the same symptoms across repeated cessation periods and, most importantly, similar symptom profiles have been demonstrated across studies conducted by different research groups and with diverse subject samples.5-9 These samples include adolescents, adults, the general population, and persons who are seeking clinical treatment to stop using cannabis.
The symptoms of cannabis withdrawal disorder follow a transient time course typical of other substance use withdrawal disorders.
Both human and animal laboratory studies provide clear evidence of the pharmacological specificity of cannabis withdrawal disorder.10-12 That is, the deprivation of delta-9-tetrahydrocannabinol (THC)—the primary active compound in cannabis related to its reinforcing properties and development of misuse and addiction—causes withdrawal symptoms. Symptoms of cannabis withdrawal can be suppressed by blind administration of oral preparations of delta-9-THC, and they are abated with smoked marijuana or oral delta-9-THC.
Laboratory and clinical survey studies of heavy cannabis users indicate that cannabis withdrawal disorder is not rare.6,8,13,14 Although general population studies show what might be perceived as low rates of cannabis withdrawal, the observed prevalence appears comparable to substances with well-established with-drawal disorders.15
Evidence has emerged from multiple sources, which indicate that cannabis withdrawal symptoms can be clinically important. Studies that compare cannabis withdrawal with tobacco withdrawal consistently report that the severity and time course of withdrawal symptoms appear comparable.16-18
A number of studies indicate that cannabis users report that they take drugs to relieve withdrawal symptoms.6,19,20 This suggests that the disorder contributes to ongoing use in those trying to quit.
A substantial proportion of adults and adolescents in treatment for cannabis dependence acknowledge moderate to severe withdrawal symptoms, and they complain that these symptoms make cessation more difficult.6,16
Those who live with cannabis users observe and complain about significant withdrawal effects in the user.13 This suggests that such symptoms are disruptive to daily living.
Subsequent articles provide further support for the validity and clinical importance of cannabis withdrawal disorder. A prospective study showed that adolescents entering treatment for cannabis use disorders experienced withdrawal symptoms that were most severe during the initial week of abstinence.21 Significant symptom reductions were observed over 4 weeks of abstinence.
Reports of cannabis withdrawal in the general population, consisting of both clinical and nonclinical samples, are consistent with the type and duration of symptoms previously reported and indicate that cannabis withdrawal disorder is not rare. Twenty-nine percent of past-year cannabis users and 44% of those using cannabis more than 3 times per week reported experiencing at least 2 withdrawal symptoms.22,23 Furthermore, 81% to 91% of adults who are heavy users of cannabis reported experiencing at least 2 withdrawal symptoms, and 49% to 76%, at least 4 withdrawal symptoms.8 Two studies of adolescents in treatment for cannabis use reported an average of 4.5 to 6.0 withdrawal symptoms.22,23
Positive correlations between the number of withdrawal symptoms reported and severity of dependence support the concurrent validity of the disorder in adolescent and adult clinical samples. Evidence that those with more severe withdrawal symptoms may have more difficulty in quitting has emerged from clinical studies of adolescents. Those studies show that withdrawal symptom severity at treatment entry predicts cannabis dependence severity 1 year later, a more chronic course of cannabis dependence, and a greater likelihood of rapid relapse.22,23
Other relevant findings include a study that indicates the majority of heavy cannabis users take action to relieve their withdrawal symptoms.8 Those actions include relapse to cannabis use or use of other drugs (eg, sedatives) to relieve cannabis withdrawal symptoms. Similarly, in a general population sample,19 8% of past-year cannabis users reported withdrawal symptoms that caused impairment or cannabis use for withdrawal relief. A laboratory study and a survey study that compared cannabis withdrawal with tobacco withdrawal reported similar levels of withdrawal discomfort for the two disorders.18,24 Participants in the survey study reported equally strong associations between withdrawal and relapse for both disorders.
Demonstration of cannabis withdrawal disorder and evidence of its role in relapse suggest that medications targeting the disorder could help patients achieve abstinence. Multiple human laboratory studies and a handful of small clinical trials that evaluated potential pharmacotherapy approaches have appeared in the literature.25 Most have targeted pharmacological mechanisms different from those of cannabis, which could provide withdrawal symptom relief, reduce the desire for or liking of cannabis, or reduce conditions that might trigger cannabis use (eg, depressed or anxious mood, sleep difficulties, cravings). Although still in its early stages, this approach has yielded mostly negative findings or has provided only weak signals of potential efficacy.
More promising findings have emerged from treatment approaches that target the cannabinoid (CB1) receptor. Delta-9-THC is a partial agonist of the CB1 receptor, and the reinforcing effects of cannabis result from its activation. Multiple laboratory studies have demonstrated that the efficacy with which dronabinol (the oral preparation of THC) suppresses or attenuates withdrawal symptoms is dose-dependent. Most recently, the combination of dronabinol and lofexidine (an a2-adrenergic receptor agonist) was shown to diminish withdrawal severity and to reduce the likelihood of relapse in a laboratory analogue model of relapse.26
CB1 receptor antagonism, which would putatively suppress the reinforcing effects of cannabis smoking, has also been evaluated. Initial laboratory findings with rimonabant, a medication that was used for the treatment of obesity outside the United States, are promising. However, this medication was removed from the European market and production was suspended because of safety concerns associated with depression and suicide.
Research on pharmacological approaches to treating patients with cannabis use disorders has yet to provide robust clinical efficacy data for any specific medication. This area of investigation is relatively new, however. It is hoped that rapid advances in our understanding of the neurobiology of cannabis and the endogenous cannabinoid system as well as the surge of pharmacotherapeutic studies will identify effective medications.
How the DSM-5 Work Group is addressing the issue
One can readily argue that the accumulated body of research on cannabis withdrawal strongly supports inclusion of cannabis withdrawal as a disorder in DSM-5. Indeed, the DSM-5 Work Group on substance use disorders recently presented a draft of diagnostic criteria for this withdrawal disorder for public comment.27
After identifying cannabis withdrawal as a topic to be addressed, a small subcommittee was assigned to review the literature and make recommendations to the full committee. A white paper that included tentative diagnostic criteria for discussion by the entire Work Group was developed and distributed. After modifications, the white paper was distributed to professional colleagues who had published on cannabis withdrawal for their comments and suggestions. The subcommittee then drafted the criteria posted for public comment.
Because no optimal, objective algorithm or guidelines exist for determining which specific symptoms to include or how many symptoms should be required for the diagnosis, the Work Group took a conservative approach to decision making. We presumed that if research suggests the need, we could add symptoms at a future date and that this would be preferable to removing symptoms from the list. Selecting items with evidence from both controlled laboratory studies and clinical studies appeared to be a rational approach.
Symptoms or signs with some data that suggest their validity (but not with a high degree of confidence) could be mentioned as possible symptoms of cannabis withdrawal in the DSM-5 text rather than listed as formal criteria. This would alert clinicians to such symptoms and allow them to consider their importance for individual patients. It is hoped that inclusion in the text would prompt future research to examine their potential validity and significance more thoroughly. My colleagues and I also considered the algorithms used for other established withdrawal disorders and again sought to integrate suggestions from the feedback provided by outside experts. The aforementioned processes and committee consensus guided the Work Group’s current proposal.
The following signs and symptoms were put forth to be included in the criteria for cannabis withdrawal disorder: irritability, anger, or increased aggression; nervousness or anxiety; sleep difficulty; decreased appetite or weight loss; restlessness; depressed mood; and physical discomfort (eg, stomach pain, shakiness/tremors, sweating, fever, chills, headache). The working proposal requires 3 or more of these symptoms to make the diagnosis (Table). The text would further describe these items and discuss other symptoms that have been observed less consistently but may be important manifestations of cannabis withdrawal.
Cannabis withdrawal disorder symptoms:
a working proposal for DSM-5
a working proposal for DSM-5
Three or more of the following symptoms would be required for a diagnosis of cannabis withdrawal syndrome:
• Irritability, anger, or increased aggression
• Nervousness or anxiety
• Sleep difficulty (insomnia)
• Decreased appetite or weight loss
• Depressed mood
• At least 1 of the following physical symptoms causing significant discomfort: stomach pain, shakiness/tremors, sweating, fever, chills, or headache
The DSM text describing this disorder would also discuss the following symptoms as having been observed in studies, with more research needed to determine their validity or significance:
• Disturbing/strange dreams
• Difficulty in concentrating
Studies have clearly demonstrated that valid abstinence effects follow cessation of heavy cannabis use and that these effects reflect a true withdrawal disorder. Recent research with clinical populations supports the external and ecological validity of the disorder and its clinical importance. Such findings further legitimize cannabis as a drug with significant addictive potential and consequences.
Clinical studies indicate that cannabis use disorders account for a significant proportion of substance use treatment admissions. Unfortunately, these disorders are difficult to treat: failed quit attempts and relapse are much more common than successful recovery. Recognition of cannabis withdrawal disorder will provide alternative clinical and therapeutic targets that may enhance the effectiveness of current treatment regimens for patients who are trying to quit.