- Email: [email protected]
Cannabis use and cannabis use disorders among individuals with mental illness
Available online at www.sciencedirect.com
Comprehensive Psychiatry xx (2013) xxx – xxx www.elsevier.com/locate/comppsych
Cannabis use and cannabis use disorders among individuals with mental illness Shaul Lev-Ran a, b, c, d,⁎, Bernard Le Foll c, e, f, g , Kwame McKenzie a, b, g, h , Tony P. George g, i, j , Jürgen Rehm b, k a
Social Aetiology of Mental Illness (SAMI) CIHR Training Program, Centre for Addiction and Mental Health, Toronto, Ontario, Canada b Social and Epidemiological Research Department, Centre for Addiction and Mental Health, Toronto, Ontario, Canada c Addictions Program, Centre for Addiction and Mental Health, Toronto, Ontario, Canada d Addiction Medicine Services, Department of Psychiatry, Sheba Medical Center, Tel Hashomer, Israel e Translational Addiction Research Laboratory, Centre for Addiction and Mental Health, Toronto, Ontario, Canada f Departments of Family and Community Medicine, Pharmacology and Toxicology, University of Toronto, Toronto, Ontario, Canada g Department of Psychiatry, University of Toronto, Toronto, Ontario, Canada h Social Equity and Health Research Program, Centre for Addiction and Mental Health, Toronto, Ontario, Canada i Schizophrenia Program, Centre for Addiction and Mental Health, Toronto, Ontario, Canada j Division of Brain and Therapeutics, Department of Psychiatry, University of Toronto, Toronto, Ontario, Canada k Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada
Abstract Background: National epidemiological surveys have reported increased rates of cannabis use and cannabis use disorders (CUDs) among individuals with mental illness. However, this subject has not been sufficiently investigated, particularly given limitations in diagnostic tools used and lack of data pertaining to frequency of cannabis used. Objectives: To examine the prevalence of cannabis use and CUDs among individuals with a wide range of mental illness. Method: We analyzed data on 43,070 respondents age 18 and above from the National Epidemiologic Survey on Alcohol and Related Conditions, a nationally representative survey conducted from 2001 to 2002. Main outcome measures included rates of cannabis use by frequency (at least weekly and less than weekly use) and DSM-IV CUDs according to the number and type of axis I and axis II psychiatric diagnoses, assessed by the Alcohol Use Disorders and Associated Disabilities Interview Schedule-IV. We estimated the proportion of cannabis used by individuals with mental illness using reported daily dose and frequency of cannabis used by individuals with and without mental illness. Results: Rates of weekly cannabis use, less than weekly cannabis use and CUDs among individuals with 12-month mental illness were 4.4%, 5.4% and 4.0%, respectively, compared to 0.6%, 1.1% and 0.4%, respectively, among individuals without any 12-month mental illness (P b 0.0001 for all comparisons). The odds ratio for cannabis use among individuals with 12-month mental illness vs. respondents without any mental illness was 2.5, and the odds of having a CUD among individuals with 12-month mental illness were 3.2, after adjusting for sociodemographic variables and additional substance use disorders. Cannabis use and CUDs were particularly associated with bipolar disorder, substance use disorders and specific (anti-social, dependant and histrionic) personality disorders. Persons with a mental illness in the past 12 months represented 72% of all cannabis users and we estimated they consumed 83% of all cannabis consumed by this nationally representative sample. Conclusions: The current study provides further evidence of the strong association between cannabis use and a broad range of primary mental illness. This emphasizes the importance of proper screening for frequent cannabis use and CUDs among individuals with primary mental illness and focusing prevention and treatment efforts on this population. © 2013 Elsevier Inc. All rights reserved.
1. Introduction ⁎ Corresponding author. Centre for Addiction and Mental Health, Toronto, ON, Canada M5N2N5. Tel.: +1 416 535 8501; fax: +1 416 260 4156. E-mail address: [email protected] (S. Lev-Ran). 0010-440X/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.comppsych.2012.12.021
Cannabis is the most widely used illicit substance worldwide [1–3]. Globally, the number of people who have used cannabis at least once is estimated to be between
2
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
125 and 203 million [3], and lifetime prevalence of cannabis use among young adults in the United States (US) has been reported to be around 50% [4]. Among past-year cannabis users, prevalence of cannabis use disorders (CUDs) has been estimated to be more than 35% [5]. Previous large national epidemiological surveys have reported increased rates of cannabis use and CUDs among individuals with mental illness. Population-based data collected in the early 1980s by the Epidemiologic Catchment Area (ECA) study showed that roughly 50% of individuals with DSM-III psychiatric disorders meet criteria for lifetime diagnosis of CUDs [6]. The National Comorbidity Survey (NCS) conducted in the early 1990s reported that 90% of individuals with cannabis dependence have lifetime DSM-III psychiatric disorders and that cannabis dependence significantly increased the odds for a mood (OR = 2.2) or anxiety (OR = 2.6) disorder [7]. Additional longitudinal reports have explored the association between cannabis use and specific psychiatric disorders. Buckner and colleagues [8] reported that the odds of cannabis dependence among individuals with social anxiety disorder were almost 5 times more than among individuals without social anxiety disorder. Hayatbakhsh and colleagues [9] reported that individuals with “externalizing behaviors” during childhood and adolescence had a significantly increased risk (OR = 2.5) of having a CUD in young adulthood. Wittchen and colleagues [10] reported that in longitudinal analyses, non-cannabis substance use disorders (SUDs) and mood and anxiety disorders were associated with increased risk of cannabis use and CUDs. Nevertheless, many of these reports have not used assessment tools which clearly differentiate primary psychiatric disorders and substance-induced disorders. This is particularly important when exploring the association between mental illness and cannabis use, as it is the only way to conclude whether specific psychiatric disorders (such as mood and anxiety disorders) are associated with higher rates of cannabis use even when cannabis-induced psychiatric disorders are ruled out. In addition, though previous research implies that the association between mental health problems and cannabis use is particularly affected by the frequency of cannabis use [3,11], most reports on rates of cannabis use among individuals with mental illness have clustered all cannabis users into one category, without reporting separately on different frequencies of cannabis use and CUDs. Finally, previous reports have not reported differentially on rates of cannabis use and CUDs across a wide range of axis I and axis II psychiatric disorders. The aim of this study was to analyze rates of cannabis use and CUDs among individuals with a wide range of concurrent primary mental illness, ruling out substance-induced psychiatric disorders. Though there is currently no standard for “high” and “low” frequencies of cannabis use, we focused differentially on individuals using cannabis less than once per week and those using cannabis at least weekly. This was based on several studies using these categories of frequency of cannabis use when examining the association between
cannabis use and mental illness [11–14]. We hypothesized that individuals with concurrent primary mental illness have significantly higher rates of frequent cannabis use and CUDs compared to individuals without mental illness and that these rates further increase as the number of primary mental disorders increases. We used data from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) to examine the relationship between types of mental illness and patterns of cannabis use. The NESARC is the largest epidemiological survey to-date on mental illness and substance use, and the first and only national survey to include specific diagnoses of primary and substance-induced psychiatric disorders.
2. Methods 2.1. Participants and procedure We analyzed cross-sectional data from a populationbased national representative sample, the National Epidemiological Survey of Alcohol and Related Conditions (NESARC) study (Wave 1, 2001–2002) [15] conducted by the National Institute on Alcohol Abuse and Alcoholism (NIAAA). The research protocol, including informed consent procedures, received full ethical review and approval from the U.S. Census Bureau and U.S. Office of Management and Budget. The interview was developed to advance measurement of substance use and mental disorders in largescale surveys. Face-to-face interviews were conducted with 43,093 adults (response rate, 81%), aged 18 years and older from the civilian non-institutionalized population residing in the US, including the District of Columbia, Hawaii and Alaska. The NESARC sample was weighted to adjust for probabilities of selection of a sample housing unit or housing unit equivalent, the non-response at the household and person levels, the selection of one person per household and the oversampling of African–Americans, Hispanics and young adults (ages 18–24). The weighted data were poststratified and adjusted to match the target population based on the 2000 decennial census in terms of region, age, sex, race and ethnicity [16]. Details regarding sampling, purpose and weighting have been previously published [17]. Characteristics of interviewers, training and field quality control have been described elsewhere [5]. 2.2. Assessments The Alcohol Use Disorder and Associated Disabilities Interview Schedule–DSM-IV Version (AUDADIS-IV) was used to assess substance use and psychiatric disorders [18]. The AUDADIS-IV has been reported to have excellent reliability and validity in the US [17] and internationally [19]. It includes an extensive list of symptom questions that separately operationalizes DSM-IV criteria for SUDs and additional axis I and axis II diagnoses.
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
2.3. Cannabis use and cannabis use disorders (CUDs) Cannabis use in this study referred to cannabis use in the last 12 months. Respondents were asked about frequency of cannabis use. We categorized cannabis users according to frequency of use, dividing them into “at least weekly” and “less than weekly” cannabis users. Individuals who did not provide this information (n = 23) were excluded from analyses pertaining to frequency of cannabis use. Doses of cannabis used were measured as number of joints consumed on days that cannabis was used in the last 12 months. CUDs referred to cannabis abuse or dependence in the last 12 months as defined by DSM-IV. 2.4. Mental Illness Any psychiatric disorder in this study referred to any axis I, axis II or SUD in the last 12 months. Axis I disorders included mood (major depressive disorder, bipolar I, bipolar II and dysthymia) and anxiety (panic disorder, social anxiety disorder, specific phobia, generalized anxiety disorder) disorders. In the case of all mood and anxiety disorders, we included only primary mental illness and excluded all cases of substance-induced mental disorders as diagnosed in the AUDADIS-IV. Primary (non substance-induced) mood and anxiety disorders were defined if: (1) the respondent did not use alcohol or drugs in the previous 12 months; (2) the episode(s) did not occur in the context of drug or alcohol intoxication or withdrawal; (3) the episode(s) started before initiation of drug or alcohol or (4) the episode(s) began after drug or alcohol consumption began, but persisted for more than 1 month after cessation of intoxication or withdrawal. Respondents were classified as having independent (primary) mood and anxiety disorders if none or only some of their episodes were substance-induced [20]. SUDs referred to any alcohol or drug (excluding cannabis) abuse or dependence. The substances included in this study were cocaine (including crack cocaine), heroin, hallucinogens, inhalants/solvents, sedatives, tranquilizers, opioids and amphetamines. Axis II disorders included any of the following personality disorders: paranoid, schizoid, histrionic, antisocial, obsessive–compulsive, dependent, and avoidant personality disorders. 2.5. Statistical analysis Respondents with any 12-month mental illness were examined with respect to demographic characteristics (sex, age, educational level, household income, marital status, urbanicity, race/ethnicity and region). Prevalence of weekly cannabis use, less than weekly cannabis use and CUDs among individuals with and without any primary psychiatric diagnosis in the last 12-months were calculated with crosstabulations providing 95% confidence intervals. We used the χ 2-statistic to compare differences between groups in the proportion of cannabis users and individuals with CUDs. As it has been repeatedly shown that patterns of cannabis use, as
3
well as prevalence of CUDs vary greatly between men and women [5,21], these analyses were also conducted separately by gender. We used logistic regression to analyze the odds for any 12-month mental illness among different demographic groups, and to analyze the odds for cannabis use and CUDs among individuals with 12-month mental illnesses, while controlling for sociodemographic variables (age, sex, household income, region, educational level, marital status and urbanicity) and any non-cannabis SUDs. In order to examine the effect of age on the relationship between cannabis use and mental illness, we calculated the odds for 12-month cannabis use among individuals with and without mental illness by age group and controlling for other sociodemographic variables. We used the t-test to compare daily dose of cannabis used (measured as number of joints used on days cannabis is consumed) among individuals with and without 12-month mental illness. We used the Mantel– Haenszel χ 2 test for trend to compare rates of cannabis use according to the number of 12-month mental diagnoses. In order to correct for multiple testing, we used an adjusted p-value according to the maximum number of categories included in an analysis (ie, 8 categories of axis I diagnoses). Based on the Bonferroni correction, we calculated that in order to maintain 95% level of confidence we use an adjusted p value of p b 0.00625. Based on the formula presented by Lasser et al. [22], we estimated the proportion of all cannabis used in the US that was consumed by persons with mental illness using the following calculation: ðM ÞðD1 ÞðF1Þ=½ðN ÞðD 2 ÞðF2Þ þ ðM ÞðD1 ÞðF1Þ where M = the percentage of current cannabis users with mental illness; D1 = the mean dose of joints per day consumed by current cannabis users with mental illness; F1 = the mean number of days in the last year in which cannabis was consumed by current cannabis users with mental illness; N = the percentage of current cannabis users without mental illness; D2 = the mean dose of joints per day consumed by current cannabis users without mental illness and F2 = the mean number of days in the last year in which cannabis was consumed by current cannabis users without mental illness. Though this formulation was developed for cigarette smoking, in which standardization is more readily possible, it has merit when dealing with cannabis as well. One of the major challenges in cannabis research is assessing intensity of cannabis used and both dose and frequency have been proposed as proxy terms for the intensity of use [23]. Moreover, substance use clinical trials frequently use daily dose of substances as the primary outcome measure [24]. Accordingly, despite potential errors in estimation, this formulation is useful as it combines daily dose and frequency. We used the same formulation to calculate the portion of cannabis used among occasional and regular cannabis users with mental illness and among individuals with SUDs.
4
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
All results were based on population weighted data to allow conclusions representative for the US. To accurately estimate variances taking the NESARC sample design components into account, analyses were conducted with Software for Survey Data Analysis (SUDAAN) Version 10 [25], a software program that uses Taylor series linearization to make adjustments for the NESARC's sample design characteristics.
3. Results 3.1. Sociodemographic characteristics and association between 12-month mental illness and concurrent cannabis use and CUDs The prevalence of cannabis use and of CUDs among individuals with 12-month mental illness according to demographics is presented in Table 1. The population prevalence of 12-month cannabis use was 4.1%. In the general population, 2.4% reported less than weekly cannabis
use, 1.7% reported at least weekly cannabis use and 1.5% had a CUD. The majority of cannabis users (56.4%) were in the young (18–29) age group: among individuals with 12month mental illness, 59.2% of cannabis users were in the young age group vs 49.2% of cannabis users among individuals without 12-month mental illness. The prevalence of cannabis use and CUDs among individuals with 12-month mental illness was 9.9% and 4.0%, respectively, compared to 1.6% and 0.4% among individuals without any mental illness in the last 12 months. Individuals with 12-month mental illness represented 72.2% of all individuals who used cannabis in the last 12 months and 81.8% of individuals with 12-month CUDs. Prevalence of mental illness was significantly higher among individuals using cannabis at least weekly (77.2%) compared to individuals using cannabis less than weekly (68.7%; p b 0.01). Respondents with any axis I, axis II or SUD had elevated rates of both weekly cannabis use, less than weekly cannabis use and CUDs in the last 12 months (Table 2). Rates of cannabis use and CUDs were particularly elevated among individuals with Bipolar I
Table 1 Prevalence and logistic regression analyses of 12-month mental illness by selected sociodemographic characteristics. Variable Sex Men (ref) Women Age 18–29 (ref) 30–44 45–64 65+ Education less than high school (ref) high school some college or higher Household income $0–$19,999 (ref) $20,000–$39,999 $40,000–$59,000 $60,000 or above Marital status Married (ref) separated/divorced/widowed never married Urbanicity Urban (ref) Rural Race/ethnicity White (ref) Black American Indian Asian Hispanic Region Northeast (ref) Midwest South West Abbreviations: OR = Odds ratio, ref = reference.
12-month mental illness (N = 12,659) prevalence, % (95% CI)
OR
30.0 (28.8–31.3) 29.6 (28.3–30.9)
1.0 0.98 (0.93–1.04)
38.4 (36.5–40.4) 32.2 (30.6–33.8) 28.1 (26.8–29.4) 16.8 (15.7–17.8)
1.0 0.76 (0.71–0.81) 0.63 (0.58–0.67) 0.32 (0.29–0.35)
22.6 (20.5–24.9) 30.4 (29.1–31.8) 30.1 (28.9–31.4)
1.0 1.50 (1.32–1.70) 1.48 (1.30–1.68)
31.4 (29.9–32.9) 31.0 (29.3–32.7) 29.7 (28.3–31.1) 28.0 (26.7–29.4)
1.0 0.98 (0.90–1.07) 0.92 (0.86–0.99) 0.85 (0.78–0.92)
26.4 (25.2–27.6) 32.21 (30.6–33.7) 37.7 (36.0–39.4)
1.0 1.32 (1.34–1.41) 1.69 (1.59–1.79)
30.8 (28.6–33.1) 29.3 (28.3–30.4)
1.0 0.93 (0.84–1.04)
30.5 (29.4–31.6) 29.5 (27.9–31.2) 41.6 (37.1–46.3) 19.9 (16.9–23.2) 27.1 (25.3–28.9)
1.0 0.95 (0.89–1.03) 1.63 (1.35–1.95) 0.56 (0.47–0.68) 0.85 (0.77–0.93)
27.9 (25.6–30.3) 32.6 (29.8–35.6) 27.8 (26.5–29.2) 31.6 (28.7–34.6)
1.0 1.25 (1.05–1.49) 1.0 (0.87–1.14) 1.19 (1.00–1.43)
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
5
Table 2 Prevalence of cannabis use and cannabis use disorders according to 12-month mental illness status (n = 43,093).
Total No mental illness Men Women Any mental illness Men Women Any axis I disorder a Men Women Any SUD b Men Women Any axis II disorder c Men Women
Less than weekly, % (95% CI)
At least weekly, % (95% CI)
CUD, % (95% CI)
2.4 (2.2–2.6) 1.1 (0.9–1.2) 1.4 (1.1–1.6) 0.8 (0.6–1.0) 5.4 (4.9–6.0)⁎ 7.2 (6.3–8.2)⁎ 3.8 (3.3–4.4)⁎ 4.4 (3.8–5.1)⁎ 6.5 (5.2–7.9)⁎ 3.3 (2.7–3.9)⁎ 12.3 (11.0–13.7)⁎ 12.7 (11.0–14.5)⁎ 11.5 (9.4–13.8)⁎ 5.0 (4.3–5.8)⁎ 6.2 (5.0–7.5)⁎ 3.8 (3.1–4.7)⁎
1.7 (1.5–1.9) 0.6 (0.5–0.7) 0.8 (0.6–1.0) 0.3 (0.2–0.5) 4.4 (3.9–5.0)⁎ 6.6 (5.7–7.5)⁎ 2.4 (2.0–3.0)⁎ 3.8 (3.2–4.5)⁎ 7.1 (5.7–8.8)⁎ 2.0 (1.6–2.6)⁎ 10.4 (9.2–11.8)⁎ 11.6 (10.0–13.5)⁎ 7.7 (6.0–9.8)⁎ 5.3 (4.5–6.2)⁎ 7.4 (6.2–9.0)⁎ 3.1 (2.5–4.0)⁎
1.5 (1.3–1.6) 0.4 (0.3–0.5) 0.6 (0.4–0.7) 0.2 (0.1–0.3) 4.0 (3.6–4.5)⁎ 6.0 (5.3–6.9)⁎ 2.1 (1.7–2.5)⁎ 3.5 (3.0–4.2)⁎ 6.4 (5.0–8.1)⁎ 2.0 (1.6–2.5)⁎ 10.2 (9.1–11.4)⁎ 11.3 (9.8–13.0)⁎ 7.1 (6.1–9.8)⁎ 4.8 (4.1–5.5)⁎ 6.8 (5.7–8.2)⁎ 2.6 (2.1–3.4)⁎
Abbreviations: CUD = cannabis use disorders. a Any mood (Bipolar I, Bipolar II, Major Depressive Disorder, dysthymia) or anxiety (panic, social phobia, specific phobia, Generalized Anxiety Disorder) disorder. b Any alcohol or drug use disorder (excluding cannabis). c Any antisocial, avoidant, dependent, paranoid, schizoid, histrionic personality disorder. ⁎ Significantly different from respondents without any psychiatric disorder, p ≤ 0.0001.
disorder (Table 3) and specific (anti-social, and histrionic) personality disorders (Table 4). The relationship between cannabis use and mental illness persisted when we controlled for sociodemographic variables. Compared with respondents without 12-month mental illness, those with any 12-month mental illness were significantly more likely to use cannabis in the last 12 months (OR = 5.2, 95% CI = 4.5–5.9). When additionally controlling for any non-cannabis SUD, the increased odds for cannabis use among individuals with mental illness were retained (OR = 2.5, 95% CI = 2.1–3.0). Adjusting for sociodemographic variables, the odds of having a CUD were 8.0 (95% CI = 6.2–10.3) among individuals with 12-month mental illness compared to those without 12-month mental illness. After additional adjustment for non-cannabis SUDs, the odds ratio for having a CUD among individuals with mental illness compared to those without mental illness was 3.2 (95% CI = 2.2–4.6). We calculated the odds for 12-month cannabis use among individuals with and without mental illness by age group and controlling for other sociodemographic variables. In the 18– 29 year age group, individuals with any 12-month mental illness were significantly more likely to use cannabis in the last 12 months (OR = 5.8, 95% CI = 4.8–7.1) compared to those without mental illness. The association was further retained in other age groups: in the 30–44 and the 45– 64 year age group, individuals with any 12-month mental illness were significantly more likely to use cannabis in the last 12 months (OR = 8.2 (95% CI = 5.6–12.0) and 7.4 (95% CI = 4.6–11.9), respectively) compared to those without any mental illness. In the 65+ age group, this association was not
retained (OR = 5.0, 95% CI = 0.5–48.3), though it should be noted that the prevalence of cannabis use in this age group was generally low (0.04%). 3.2. Daily dose of cannabis used The mean number of joints consumed (on days that cannabis was used) by individuals with 12-month mental illness was 2.2, compared to 1.6 among individuals without any mental illness in the last 12 months (p b 0.0001). Among individuals with 12-month prevalence of mental illness, the mean number of joints among individuals using cannabis less than weekly was 1.4 joints per day, compared to 3.1 joints per day among individuals using cannabis at least weekly (p b 0.0001). 3.3. Cannabis use and multiple psychiatric diagnoses Persons with multiple 12-month psychiatric diagnoses had higher rates of weekly and less than weekly cannabis use than persons with only 1 DSM-IV diagnosis (p b 0.0001, Fig. 1). 3.4. Proportion of cannabis used by individuals with mental illness We estimated that persons with a 12-month mental illness consumed 83% of all cannabis consumed in the US. Among individuals with mental illness, 2.4% of the cannabis was used by individuals using cannabis less than weekly, and 97.6% was used by individuals using cannabis at least weekly. Sixty percent of all cannabis was consumed by individuals with SUDs. Of these, 46% used cannabis at least
6
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
Table 3 Prevalence of cannabis use and cannabis use disorders according to specific 12-month DSM-IV Axis I diagnoses (n = 43,093).
Any mood disorder Men Women Bipolar I Disorder Men Women Bipolar 2 Disorder Men Women Major depression Men Women Dysthymia Men Women Any anxiety disorder Men Women Panic disorder Men Women Social phobia Men Women Specific phobia Men Women GAD Men Women
Less than weekly, % (95% CI)
At least weekly, % (95% CI)
CUD, % (95% CI)
5.8 (4.8–6.9)⁎ 7.9 (6.0–10.2)⁎ 4.6 (3.7–5.7)⁎ 8.4 (6.3–11.0)⁎ 9.8 (6.6–14.1)⁎ 7.3 (5.0–10.5)⁎ 7.6 (4.8–12.0)⁎ 5.6 (2.5–12.1) 9.0 (5.0–15.7) 4.8 (3.7–6.2)⁎ 7.8 (5.4–11.3)⁎ 3.3 (2.4–4.5)⁎ 3.8 (2.5–5.8) 4.9 (2..6–9.0) 3.3 (1.9–5.9) 4.0 (3.4–4.8)⁎ 6.2 (4.8–8.0)⁎ 2.9 (2.3–3.7)⁎ 4.4 (3.1–6.3)⁎ 7.3 (4.4–11.8) 3.3 (1.9–5.4) 5.7 (4.2–7.7)⁎ 9.5 (6.3–14.0)⁎⁎ 3.5 (2.3–5.3)⁎⁎ 3.4 (2.7–4.4)⁎ 5.0 (3.5–7.2)⁎⁎ 2.7 (2.0–3.7)⁎ 4.5 (3.1–6.4)⁎⁎ 5.9 (3.3–10.5) 3.9 (2.4–6.2)
4.9 (4.1–5.9)⁎ 8.9 (7.0–11.3)⁎ 2.7 (2.1–3.6)⁎ 9.6 (7.2–12.7)⁎ 15.5 (10.7–22.1)⁎ 5.2 (3.3–8.0)⁎ 4.8 (2.8–8.3)⁎ 7.6 (3.6–15.1) 3.0 (1.3–6.7) 3.4 (2.4–4.2)⁎ 6.2 (4.2–9.0)⁎ 1.8 (1.2–2.8)⁎⁎ 6.0 (4.3–8.4)⁎ 10.8 (6.9–16.3)⁎⁎ 3.8 (2.3–6.1)⁎⁎ 3.6 (2.9–4.5)⁎ 7.0 (5.2–9.3)⁎ 2.0 (1.5–2.7)⁎ 5.9 (4.1–8.4)⁎ 11.9 (7.3–18.8) 3.5 (2.2–5.4)⁎⁎ 3.4 (2.4–4.9)⁎ 4.8 (2.9–7.8) 2.6 (1.5–4.5) 3.8 (2.9–4.9)⁎ 7.9 (5.5–11.0)⁎ 2.0 (1.3–2.9)⁎ 5.8 (4.0–8.3)⁎ 12.3 (7.1–19.5) 3.2 (2.0–5.1)⁎⁎
4.9 (4.1–5.9)⁎ 6.3 (6.5–10.6)⁎ 3.0 (2.3–3.9)⁎ 9.4 (7.0–12.5)⁎ 14.6 (10.2–20.3)⁎ 5.5 (3.5–8.6)⁎ 6.1 (3.6–10.0)⁎ 8.8 (4.3–16.8) 4.3 (2.1–8.4) 3.1 (2.3–4.0)⁎ 5.2 (3.6–7.5)⁎ 2.0 (1.3–3.1)⁎⁎ 5.7 (3.9–8.4)⁎ 8.7 (5.0–14.6)⁎⁎ 4.4 (2.7–7.1) 3.2 (2.5–3.9)⁎ 5.7 (4.2–7.7)⁎ 2.0 (1.5–2.6)⁎ 5.0 (3.6–7.1)⁎ 8.8 (5.4–14.2)⁎⁎ 3.5 (2.2–5.6)⁎ 3.9 (2.8–5.4)⁎ 5.0 (3.0–8.4) 3.2 (2.0–5.1)⁎⁎ 2.9 (2.2–3.9)⁎ 5.6 (3.6–8.5)⁎⁎ 1.8 (1.2–2.6)⁎ 5.5 (3.7–8.3)⁎⁎ 10.1 (5.5–17.9) 3.6 (2.3–5.8)⁎⁎
Abbreviations: CUD = cannabis use disorders. ⁎ Significantly different from respondents without any psychiatric disorder, p ≤ 0.0001. ⁎⁎ Significantly different from respondents without any psychiatric disorder, p b 0.001.
weekly; we estimated they consumed 97.5% of all cannabis consumed by individuals with SUDs.
4. Discussion We found that persons with mental illness are almost 10 times as likely to use cannabis weekly or suffer from a CUD, and more than 5 times as likely to use cannabis less than weekly, compared to individuals without mental illness. Though increased rates of cannabis use among individuals with mental illness are well-known, our findings add to previous findings in the following: (1) we included only individuals with primary (e.g., not substance-induced) mood and anxiety disorders based on DSM-IV diagnostic criteria; (2) we differentiated cannabis users by frequency (3) we report on a wide range of primary axis I and axis II disorders and (4) we estimated percentage of cannabis consumed by individuals with mental illness based on frequency of use and reported daily doses, as an additional measure of the association between mental illness and cannabis use. The unique data collected in the NESARC pertaining to cannabis use and mental illness allowed for these analyses.
Among non-SUD axis I mental disorders, CUDs and atleast weekly cannabis use were particularly prevalent among individuals with bipolar I disorder, with approximately 10% of individuals fulfilling criteria of a CUD or reporting using cannabis at least once per week. This is lower than rates of cannabis related problems cited in various studies among individuals with bipolar disorder, which are as high as 50% [26]. These differences in prevalence rates may be accounted for by both criteria and information variance [27]. Differences in instruments used (for example AUDADIS-IV in the NESARC vs. the National Institute of Mental Health Diagnostic Interview Schedule (DIS) and Composite International Diagnostic Interview (CIDI) in the ECA and NCS, respectively) have been reported to affect prevalence rates in population-based studies [28]. It has been further suggested that respondents may be less forthcoming in surveys conducted by government agencies to reveal information pertaining to substance use. Nevertheless, this would not seem to specifically affect individuals with bipolar disorder; hence, the increased rate of frequent cannabis use and CUDs among individuals with bipolar disorder relative to individuals with other psychiatric disorders should be noted.
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
7
Table 4 Prevalence of cannabis use and cannabis use disorders according to specific 12-month DSM-IV Axis II diagnoses (n = 43,093).
Antisocial PD Men Women Avoidant PD Men Women Dependent PD Men Women Obsessive–Compulsive PD Men Women Paranoid PD Men Women Schizoid PD Men Women Histrionic PD Men Women
Less than weekly, % (95% CI)
At least weekly, % (95% CI)
CUD, % (95% CI)
10.1 (8.1–12.5)⁎ 9.8 (7.3–13.0)⁎ 10.8 (7.6–15.2)⁎ 5.7 (4.1–8.0)⁎ 9.7 (6.2–14.7)⁎⁎ 3.2 (1.9–5.4) 7.1 (3.5–13.9) 11.4 (3.9–28.9) 4.7 (2.0–10.8) 3.7 (2.9–4.7)⁎ 4.8 (3.4–6.6)⁎⁎ 2.6 (1.9–2.7)⁎⁎ 5.7 (4.5–7.2)⁎ 6.9 (4.6–10.3)⁎⁎ 4.9 (3.6–6.7)⁎ 3.9 (2.8–5.5)⁎ 5.0 (3.2–7.8) 2.9 (1.7–5.0) 9.0 (7.1–11.5)⁎ 10.2 (7.2–14.3)⁎ 7.9 (5.4–11.5)⁎
13.4 (11.2–15.9)⁎ 13.6 (11.0–16.7)⁎ 12.7 (9.4–17.0)⁎ 5.3 (3.6–7.7)⁎ 9.6 (5.9–15.3) 2.5 (1.4–4.5) 10.5 (5.4–19.5) 24.3 (12.2–42.7) 2.9 (1.1–7.1) 3.3 (2.5–4.4)⁎ 5.3 (3.7–7.4)⁎ 1.5 (1.0–2.4)⁎⁎ 7.0 (5.4–9.0)⁎ 11.6 (8.5–15.7)⁎ 3.7 (2.6–5.1)⁎ 6.3 (4.7–8.4)⁎ 8.2 (5.5–12.0)⁎ 4.5 (3.1–6.6)⁎ 9.9 (7.1–13.6)⁎ 14.8 (10.1–21.0)⁎ 5.1 (3.1–8.3)⁎⁎
12.0 (10.0–14.5)⁎ 13.1 (10.5–16.4)⁎ 9.1 (6.4–12.9)⁎ 5.6 (3.9–3.7)⁎ 10.1 (6.5–15.3)⁎⁎ 2.7 (1.6–4.7) 14.2 (8.5–22.8) 30.0 (16.9–47.4) 5.5 (2.7–10.6) 3.5 (2.7–4.5)⁎ 4.9 (3.9–6.8)⁎ 2.2 (1.5–3.2)⁎ 5.8 (4.5–7.5)⁎ 9.0 (6.5–12.4)⁎ 3.6 (2.6–5.0)⁎ 5.2 (3.8–7.0)⁎ 6.2 (3.9–9.7)⁎⁎ 4.2 (2.8–6.1)⁎ 9.8 (7.3–13.1)⁎ 13.9 (9.9–19.4)⁎ 5.8 (3.6–9.2)⁎⁎
Abbreviations: CUD = cannabis use disorders; PD = personality disorder. ⁎ Significantly different from respondents without any psychiatric disorder, p ≤ 0.0001. ⁎⁎ Significantly different from respondents without any psychiatric disorder, p b 0.001.
Individuals with alcohol and drug use disorders were at a particularly high risk for weekly cannabis use; more than 10% of these individuals used cannabis at least once per week, and rates of CUDs in this population were more than 10%. Multiple substance use may be influenced by common neurobiological factors, given that different substances act upon similar brain loci and involve similar neurotransmitter systems [29,30]. In addition, social factors and drug availability should be considered. Patterns of drug initiation have been found to vary across countries and cultures [3], suggesting that social settings that facilitate specific drug use may be particularly important. Though cannabis use and CUDs have been largely overlooked in treatment settings for substance use problems, the rates of admission to nationally funded treatment centers with cannabis stated as the primary problem-drug are gradually increasing. The rate of treatment admission with cannabis as the primary substance almost doubled between 1993 and 1999 [31]. In 2008, 16.6% of individuals discharged from treatment facilities reported cannabis as their primary substance of abuse [32]. Our findings emphasize the importance of assessing for frequent cannabis use and CUDs among individuals with any alcohol or drug use disorder. Among individuals with personality disorders, CUDs and at-least weekly cannabis use disorders were particularly prevalent among individuals with antisocial and histrionic personality disorders. Among these, antisocial personality disorder has been particularly shown to be associated with an increased risk for persistence of CUDs during a 3-year follow-up [33]. Though the increased association between
substance use disorders and cluster B personality disorders is well known [34], additional investigation of CUDs among individuals with personality disorders belonging to other clusters, which takes into account temporal relationships (such as that conducted by Hasin and colleagues [33]), can help in elucidating mechanisms of these co-morbidities. Almost without exception, the rates of cannabis use and CUDs were higher among men compared to women. This echoes findings on rates of cannabis use and CUDs in the general population in previous population-based surveys (for example, the National Longitudinal Alcohol Epidemiologic Survey (NLAES, 1991–1992)) as well as in the NESARC sample [5]. Though it is possible that these gender differences reflect differences in effects of cannabis, as well as susceptibility to frequent cannabis use and CUDs [35], it has also been suggested that these gender differences can be explained by differences in opportunities to use cannabis and other drugs, and that given the initial opportunity to use drugs, men and women are equally likely to move on to frequent cannabis use [36,37]. Among individuals with mental illness, both increased opportunities for use amongst men as well as potential increased susceptibility for frequent use and misuse should be taken into account even in those disorders which are more prevalent among women, such as depression and anxiety disorders. The association between mental illness and cannabis use was pervasive across most age groups. This implies that though cannabis use is generally more prevalent among younger people, the prevalence of cannabis use is higher among individuals with mental illness in older adults as well.
8
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx 25
20
15
10
5
0 0
1
2
3
4
>4
No. of 12-month Psychiatic Diagnoses Less than weekly cannabis use At least weekly cannabis use
Fig. 1. Patterns of cannabis use according to the number of 12-month DSMIV psychiatric diagnoses.
We found that the prevalence of cannabis use, both weekly and less than weekly, increases with the increase in number of DSM-IV psychiatric diagnoses. This is particularly important given reports that the vast majority of mental disorders are co-morbid disorders; results from the NCS show that 79% of individuals with a lifetime mental disorder had an additional co-morbid disorder. Frequent cannabis use among individuals with mental illness is associated with symptom exacerbation [38,39] and poorer treatment outcomes [39,40]. Individuals with co-morbid mental illness and substance use are at increased risk of higher rates of hospitalization [41], homelessness or housing instability [42], and increased treatment costs [43]. Nevertheless, there is still a paucity of clinical data on the impact of cannabis use on the prognosis of specific psychiatric disorders. Why do individuals with mental illness consume more cannabis? It has been suggested that such persons use cannabis as a means of self-medication of psychiatric symptoms [44]. Neurobiological research has implicated that the endocannabinoid (eCB) system is highly expressed in different brain regions and has regulatory functions and that it may be deeply involved in many mental disorders [45]. It is possible that cannabis use and mental illness share common factors; the same factors that predispose people to mental illness also increase their risk of cannabis use. These common factors may include biological, personality, social or environmental factors, or some combination of these factors [23]. Finally, it is possible that cannabis use increases the risk for developing mental illness. Evidence for an association between cannabis use and the development of psychotic disorders has accumulated [12]. There is a growing consensus that those who use cannabis, particularly heavy users and individuals who initiated cannabis use at a young age, are at increased risk for developing psychotic disorders.
There is evidence pointing to an association between heavy cannabis use and the development of depression [12]. Nevertheless, there is a scarcity of longitudinal studies examining the association between cannabis use and the development of various mental illnesses. Further longitudinal studies with multiple waves which account for temporal relationships between cannabis use and mental illness (such as that conducted by Wittchen and colleagues [10]) are important in elucidating the potential mechanisms involved in this co-morbidity. As is common in large-scale epidemiological surveys, limitations of this study should be recognized. First, information was based on self-reporting, allowing for recall and social desirability biases. Second, because the NESARC sample included only civilian households and quarters populations, information on individuals in prisons is missing. Since the prevalence of mental illness and substance use has been reported to be high in prison populations [46], this may affect calculations. Third, data do not include adolescents, a particularly vulnerable population for cannabis use [47]. Fourth, though the NESARC evaluated a large number of common mental disorders, additional disorders in which prevalence of cannabis use is known to be high have not been included in the NESARC survey or were not assessed using specific diagnostic criteria. Of particular interest may be rates of cannabis use and CUDs among individuals with psychotic disorders, which have been reported to be particularly high [45]. Psychotic disorders in the NESARC were assessed using a single question (“Did a doctor or other health professional diagnose schizophrenia or psychotic illness or episode in the last 12 months”). As this method of assessment is substantially different from the diagnostic assessment of all other DSM-IV disorders included in the NESARC, and may include an underestimation of psychotic disorders, we chose not to include these data in our study. It should be noted that the prevalence of cannabis use in this sample of individuals with anxiety disorders is lower than that reported in other samples [48]. There are several factors which may explain this. Differences in instruments used (to assess both anxiety disorders and substance use) have been reported to affect prevalence rates in population-based studies [27,28]. Difference in survey methodologies (e.g., coverage weight, data weighing) as well as question text may also affect estimations. It has been further suggested that respondents may be less forthcoming in surveys conducted by government agencies to reveal information pertaining to substance use [49]. Any or all of these considerations may have affected the rate of cannabis use in our sample. Finally, given the expected variability in potency of cannabis used and taking into account the variance of in estimating standard doses of cannabis (e.g., joints) [24], the limitation of applying the self-reported dose of cannabis used to the Lasser formulation should be acknowledged and the estimation of percentage of cannabis consumed by individuals with mental illness referred to with caution. Nevertheless, it seems that these challenges in
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
estimating doses of cannabis used will not resolve soon and even if not perfectly precise, this application of the Lasser formulation helps in elucidating the large impact of mental illness of cannabis use. Extrapolating our results to the general population, we estimate that persons with a diagnosable primary mental illness constitute almost 75% of all cannabis users, with a particularly high prevalence of more frequent cannabis use and CUDs in this population. Our findings emphasize the importance of proper screening for frequent cannabis use and CUDs particularly among individuals with mental illness, and focusing prevention and treatment efforts on the mentally ill. The largely disparate services for mental health and substance abuse pose serious challenges of service provision for this population. Treatment fragmentation between mental health and substance abuse services may mean that individuals with co-morbid cannabis use disorders and mental health problems do not receive adequate care. This results in suboptimal treatment outcomes in a population that already suffers some of the poorest outcomes in terms of illness. References [1] European Monitoring Centre for Drugs and Drug Addiction. Annual report. Luxembourg: Publications Office of the European Union; 2010. [2] United Nations Office on Drugs and Crime. World drug report. Vienna 2010. [3] Degenhardt L, Hall W. Extent of illicit drug use and dependence, and their contribution to the global burden of disease. Lancet 2012; 379(9810):55-70. [4] Substance Abuse and Mental Health Services Administration. National survey on drug use and health. Rockville MD: Substance Abuse and Mental Health Services Administration; 2008. [5] Compton WM, Grant BF, Colliver JD, Glantz MD, Stinson FS. Prevalence of marijuana use disorders in the United States: 1991–1992 and 2001–2002. JAMA 2004;291(17):2114-21. [6] Regier DA, Farmer ME, Rae DS, Locke BZ, Keith SJ, Judd LL, et al. Comorbidity of mental disorders with alcohol and other drug abuse. Results from the Epidemiologic Catchment Area (ECA) Study. JAMA 1990;264(19):2511-8. [7] Kessler RC, Nelson CB, McGonagle KA, Edlund MJ, Frank RG, Leaf PJ. The epidemiology of co-occurring addictive and mental disorders: implications for prevention and service utilization. Am J Orthopsychiatry 1996;66(1):17-31. [8] Buckner JD, Schmidt NB, Lang AR, Small JW, Schlauch RC, Lewinsohn PM. Specificity of social anxiety disorder as a risk factor for alcohol and cannabis dependence. J Psychiatr Res 2008;42(3): 230-9. [9] Hayatbakhsh MR, McGee TR, Bor W, Najman JM, Jamrozik K, Mamun AA. Child and adolescent externalizing behavior and cannabis use disorders in early adulthood: an Australian prospective birth cohort study. Addict Behav 2008;33(3):422-38. [10] Wittchen HU, Frohlich C, Behrendt S, Gunther A, Rehm J, Zimmermann P, et al. Cannabis use and cannabis use disorders and their relationship to mental disorders: a 10-year prospective-longitudinal community study in adolescents. Drug Alcohol Depend 2007;88(Suppl 1):S60-70. [11] Hall W, Degenhardt L. Adverse health effects of non-medical cannabis use. Lancet 2009;374(9698):1383-91. [12] Moore TH, Zammit S, Lingford-Hughes A, Barnes TR, Jones PB, Burke M, et al. Cannabis use and risk of psychotic or affective mental health outcomes: a systematic review. Lancet 2007;370(9584):319-28.
9
[13] Swift W, Coffey C, Degenhardt L, Carlin JB, Romaniuk H, Patton GC. Cannabis and progression to other substance use in young adults: findings from a 13-year prospective population-based study. J Epidemiol Community Health 2011. [14] Patton GC, Coffey C, Carlin JB, Degenhardt L, Lynskey M, Hall W. Cannabis use and mental health in young people: cohort study. BMJ 2002;325(7374):1195-8. [15] Grant BF, Moore TC, Kaplan K. Source and accuracy statement: wave 1 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). Bethesda, Md: National Institute on Alcohol Abuse and Alcoholism; 2003. [16] Huang B, Grant BF, Dawson DA, Stinson FS, Chou SP, Saha TD, et al. Race–ethnicity and the prevalence and co-occurrence of Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, alcohol and drug use disorders and Axis I and II disorders: United States, 2001 to 2002. Compr Psychiatry 2006;47(4):252-7. [17] Grant BF, Dawson DA, Stinson FS, Chou PS, Kay W, Pickering R. The Alcohol Use Disorder and Associated Disabilities Interview Schedule-IV (AUDADIS-IV): reliability of alcohol consumption, tobacco use, family history of depression and psychiatric diagnostic modules in a general population sample. Drug Alcohol Depend 2003;71(1):7-16. [18] Grant BF, Hasin D. The Alcohol Use Disorder and Associated Disabilites Schedule–DSM-IV Version. Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism; 2001. [19] Chatterji S, Saunders JB, Vrasti R, Grant BF, Hasin D, Mager D. Reliability of the alcohol and drug modules of the Alcohol Use Disorder and Associated Disabilities Interview Schedule–Alcohol/ Drug-Revised (AUDADIS-ADR): an international comparison. Drug Alcohol Depend 1997;47(3):171-85. [20] Grant BF, Stinson FS, Dawson DA, Chou SP, Dufour MC, Compton W, et al. Prevalence and co-occurrence of substance use disorders and independent mood and anxiety disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Arch Gen Psychiatry 2004;61(8):807-16. [21] Lev-Ran S, Imtiaz S, Taylor BJ, Shield KD, Rehm J, Le Foll B. Gender differences in health-related quality of life among cannabis users: results from the national epidemiologic survey on alcohol and related conditions. Drug Alcohol Depend 2011. [22] Lasser K, Boyd JW, Woolhandler S, Himmelstein DU, McCormick D, Bor DH. Smoking and mental illness: a population-based prevalence study. JAMA 2000;284(20):2606-10. [23] Degenhardt L, Hall W, Lynskey M. Exploring the association between cannabis use and depression. Addiction 2003;98(11):1493-504. [24] Mariani JJ, Brooks D, Haney M, Levin FR. Quantification and comparison of marijuana smoking practices: blunts, joints, and pipes. Drug Alcohol Depend 2011;113(2–3):249-51. [25] Research Triangle Institute. Software for survey data analysis (SUDAAN). 10 ed. Research Triangle Park, NC: Research Triangle Institute; 2004. [26] Cerullo MA, Strakowski SM. The prevalence and significance of substance use disorders in bipolar type I and II disorder. Subst Abuse Treat Prev Policy 2007;2:29. [27] Klerman GL. Diagnosis of psychiatric disorders in epidemiologic field studies. Arch Gen Psychiatry 1985;42(7):723-4. [28] Regier DA, Kaelber CT, Rae DS, Farmer ME, Knauper B, Kessler RC, et al. Limitations of diagnostic criteria and assessment instruments for mental disorders. Implications for research and policy. Arch Gen Psychiatry 1998;55(2):109-15. [29] Koob GF, Le Moal M. Drug addiction, dysregulation of reward, and allostasis. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 2001;24(2): 97-129. [30] Nutt DJ. The neurochemistry of addiction. Hum Psychopharmacol 1997;12:s53-8. [31] Substance Abuse and Mental Health Services Administration. National household survey on drug abuse: main findings, 1998. Rockville, MD:
10
[32]
[33]
[34]
[35] [36] [37]
[38]
[39]
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx Substance Abuse and Mental Health Services Administration. Office of Applied Studies; 2002. Substance Abuse and Mental Health Services Administration. Treatment Episode Data Set (TEDS): 2008. Discharges from Substance Abuse Treatment Services, DASIS Series: S-56, HHS Publication No. (SMA) 11–4628. Rockville, MD: Substance Abuse and Mental Health Services Administration; 2010. p. 55-6. Hasin D, Fenton MC, Skodol A, Krueger R, Keyes K, Geier T, et al. Personality disorders and the 3-year course of alcohol, drug, and nicotine use disorders. Arch Gen Psychiatry 2011;68(11):1158-67. Taylor J. Substance use disorders and Cluster B personality disorders: physiological, cognitive, and environmental correlates in a college sample. Am J Drug Alcohol Abuse 2005;31(3):515-35. Greenfield SF, O'Leary G. Sex differences in marijuana use in the United States. Harv Rev Psychiatry 1999;6(6):297-303. Van Etten ML, Neumark YD, Anthony JC. Male–female differences in the earliest stages of drug involvement. Addiction 1999;94(9):1413-9. Van Etten ML, Anthony JC. Comparative epidemiology of initial drug opportunities and transitions to first use: marijuana, cocaine, hallucinogens and heroin. Drug Alcohol Depend 1999;54(2):117-25. Arendt M, Rosenberg R, Fjordback L, Brandholdt J, Foldager L, Sher L, et al. Testing the self-medication hypothesis of depression and aggression in cannabis-dependent subjects. Psychol Med 2007;37(7): 935-45. Goldberg JF, Garno JL, Leon AC, Kocsis JH, Portera L. A history of substance abuse complicates remission from acute mania in bipolar disorder. J Clin Psychiatry 1999;60(11):733-40.
[40] Bricker JB, Russo J, Stein MB, Sherbourne C, Craske M, Schraufnagel TJ, et al. Does occasional cannabis use impact anxiety and depression treatment outcomes? Results from a randomized effectiveness trial. Depress Anxiety 2007;24(6):392-8. [41] Grace R, Shenfield G, Tennant C. Cannabis and psychosis in acute psychiatric admissions. Drug Alcohol Rev 2000;19(3):287-90. [42] Drake RE, Osher FC, Wallach MA. Homelessness and dual diagnosis. Am Psychol 1991;46(11):1149-58. [43] Bartels SJ, Teague GB, Drake RE, Clark RE, Bush PW, Noordsy DL. Substance abuse in schizophrenia: service utilization and costs. J Nerv Ment Dis 1993;181(4):227-32. [44] Khantzian EJ. The self-medication hypothesis of substance use disorders: a reconsideration and recent applications. Harv Rev Psychiatry 1997;4(5):231-44. [45] Leweke FM, Koethe D. Cannabis and psychiatric disorders: it is not only addiction. Addict Biol 2008;13(2):264-75. [46] Fazel S, Baillargeon J. The health of prisoners. Lancet 2011; 377(9769):956-65. [47] Lynskey MT, Heath AC, Bucholz KK, Slutske WS, Madden PA, Nelson EC, et al. Escalation of drug use in early-onset cannabis users vs co-twin controls. JAMA 2003;289(4):427-33. [48] Agosti V, Nunes E, Levin F. Rates of psychiatric comorbidity among US residents with lifetime cannabis dependence. Am J Drug Alcohol Abuse 2002;28(4):645-54. [49] Grucza RA, Abbacchi AM, Przybeck TR, Gfroerer JC. Discrepancies in estimates of prevalence and correlates of substance use and disorders between two national surveys. Addiction 2007;102(4):623-9.
Comprehensive Psychiatry xx (2013) xxx – xxx www.elsevier.com/locate/comppsych
Cannabis use and cannabis use disorders among individuals with mental illness Shaul Lev-Ran a, b, c, d,⁎, Bernard Le Foll c, e, f, g , Kwame McKenzie a, b, g, h , Tony P. George g, i, j , Jürgen Rehm b, k a
Social Aetiology of Mental Illness (SAMI) CIHR Training Program, Centre for Addiction and Mental Health, Toronto, Ontario, Canada b Social and Epidemiological Research Department, Centre for Addiction and Mental Health, Toronto, Ontario, Canada c Addictions Program, Centre for Addiction and Mental Health, Toronto, Ontario, Canada d Addiction Medicine Services, Department of Psychiatry, Sheba Medical Center, Tel Hashomer, Israel e Translational Addiction Research Laboratory, Centre for Addiction and Mental Health, Toronto, Ontario, Canada f Departments of Family and Community Medicine, Pharmacology and Toxicology, University of Toronto, Toronto, Ontario, Canada g Department of Psychiatry, University of Toronto, Toronto, Ontario, Canada h Social Equity and Health Research Program, Centre for Addiction and Mental Health, Toronto, Ontario, Canada i Schizophrenia Program, Centre for Addiction and Mental Health, Toronto, Ontario, Canada j Division of Brain and Therapeutics, Department of Psychiatry, University of Toronto, Toronto, Ontario, Canada k Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada
Abstract Background: National epidemiological surveys have reported increased rates of cannabis use and cannabis use disorders (CUDs) among individuals with mental illness. However, this subject has not been sufficiently investigated, particularly given limitations in diagnostic tools used and lack of data pertaining to frequency of cannabis used. Objectives: To examine the prevalence of cannabis use and CUDs among individuals with a wide range of mental illness. Method: We analyzed data on 43,070 respondents age 18 and above from the National Epidemiologic Survey on Alcohol and Related Conditions, a nationally representative survey conducted from 2001 to 2002. Main outcome measures included rates of cannabis use by frequency (at least weekly and less than weekly use) and DSM-IV CUDs according to the number and type of axis I and axis II psychiatric diagnoses, assessed by the Alcohol Use Disorders and Associated Disabilities Interview Schedule-IV. We estimated the proportion of cannabis used by individuals with mental illness using reported daily dose and frequency of cannabis used by individuals with and without mental illness. Results: Rates of weekly cannabis use, less than weekly cannabis use and CUDs among individuals with 12-month mental illness were 4.4%, 5.4% and 4.0%, respectively, compared to 0.6%, 1.1% and 0.4%, respectively, among individuals without any 12-month mental illness (P b 0.0001 for all comparisons). The odds ratio for cannabis use among individuals with 12-month mental illness vs. respondents without any mental illness was 2.5, and the odds of having a CUD among individuals with 12-month mental illness were 3.2, after adjusting for sociodemographic variables and additional substance use disorders. Cannabis use and CUDs were particularly associated with bipolar disorder, substance use disorders and specific (anti-social, dependant and histrionic) personality disorders. Persons with a mental illness in the past 12 months represented 72% of all cannabis users and we estimated they consumed 83% of all cannabis consumed by this nationally representative sample. Conclusions: The current study provides further evidence of the strong association between cannabis use and a broad range of primary mental illness. This emphasizes the importance of proper screening for frequent cannabis use and CUDs among individuals with primary mental illness and focusing prevention and treatment efforts on this population. © 2013 Elsevier Inc. All rights reserved.
1. Introduction ⁎ Corresponding author. Centre for Addiction and Mental Health, Toronto, ON, Canada M5N2N5. Tel.: +1 416 535 8501; fax: +1 416 260 4156. E-mail address: [email protected] (S. Lev-Ran). 0010-440X/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.comppsych.2012.12.021
Cannabis is the most widely used illicit substance worldwide [1–3]. Globally, the number of people who have used cannabis at least once is estimated to be between
2
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
125 and 203 million [3], and lifetime prevalence of cannabis use among young adults in the United States (US) has been reported to be around 50% [4]. Among past-year cannabis users, prevalence of cannabis use disorders (CUDs) has been estimated to be more than 35% [5]. Previous large national epidemiological surveys have reported increased rates of cannabis use and CUDs among individuals with mental illness. Population-based data collected in the early 1980s by the Epidemiologic Catchment Area (ECA) study showed that roughly 50% of individuals with DSM-III psychiatric disorders meet criteria for lifetime diagnosis of CUDs [6]. The National Comorbidity Survey (NCS) conducted in the early 1990s reported that 90% of individuals with cannabis dependence have lifetime DSM-III psychiatric disorders and that cannabis dependence significantly increased the odds for a mood (OR = 2.2) or anxiety (OR = 2.6) disorder [7]. Additional longitudinal reports have explored the association between cannabis use and specific psychiatric disorders. Buckner and colleagues [8] reported that the odds of cannabis dependence among individuals with social anxiety disorder were almost 5 times more than among individuals without social anxiety disorder. Hayatbakhsh and colleagues [9] reported that individuals with “externalizing behaviors” during childhood and adolescence had a significantly increased risk (OR = 2.5) of having a CUD in young adulthood. Wittchen and colleagues [10] reported that in longitudinal analyses, non-cannabis substance use disorders (SUDs) and mood and anxiety disorders were associated with increased risk of cannabis use and CUDs. Nevertheless, many of these reports have not used assessment tools which clearly differentiate primary psychiatric disorders and substance-induced disorders. This is particularly important when exploring the association between mental illness and cannabis use, as it is the only way to conclude whether specific psychiatric disorders (such as mood and anxiety disorders) are associated with higher rates of cannabis use even when cannabis-induced psychiatric disorders are ruled out. In addition, though previous research implies that the association between mental health problems and cannabis use is particularly affected by the frequency of cannabis use [3,11], most reports on rates of cannabis use among individuals with mental illness have clustered all cannabis users into one category, without reporting separately on different frequencies of cannabis use and CUDs. Finally, previous reports have not reported differentially on rates of cannabis use and CUDs across a wide range of axis I and axis II psychiatric disorders. The aim of this study was to analyze rates of cannabis use and CUDs among individuals with a wide range of concurrent primary mental illness, ruling out substance-induced psychiatric disorders. Though there is currently no standard for “high” and “low” frequencies of cannabis use, we focused differentially on individuals using cannabis less than once per week and those using cannabis at least weekly. This was based on several studies using these categories of frequency of cannabis use when examining the association between
cannabis use and mental illness [11–14]. We hypothesized that individuals with concurrent primary mental illness have significantly higher rates of frequent cannabis use and CUDs compared to individuals without mental illness and that these rates further increase as the number of primary mental disorders increases. We used data from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) to examine the relationship between types of mental illness and patterns of cannabis use. The NESARC is the largest epidemiological survey to-date on mental illness and substance use, and the first and only national survey to include specific diagnoses of primary and substance-induced psychiatric disorders.
2. Methods 2.1. Participants and procedure We analyzed cross-sectional data from a populationbased national representative sample, the National Epidemiological Survey of Alcohol and Related Conditions (NESARC) study (Wave 1, 2001–2002) [15] conducted by the National Institute on Alcohol Abuse and Alcoholism (NIAAA). The research protocol, including informed consent procedures, received full ethical review and approval from the U.S. Census Bureau and U.S. Office of Management and Budget. The interview was developed to advance measurement of substance use and mental disorders in largescale surveys. Face-to-face interviews were conducted with 43,093 adults (response rate, 81%), aged 18 years and older from the civilian non-institutionalized population residing in the US, including the District of Columbia, Hawaii and Alaska. The NESARC sample was weighted to adjust for probabilities of selection of a sample housing unit or housing unit equivalent, the non-response at the household and person levels, the selection of one person per household and the oversampling of African–Americans, Hispanics and young adults (ages 18–24). The weighted data were poststratified and adjusted to match the target population based on the 2000 decennial census in terms of region, age, sex, race and ethnicity [16]. Details regarding sampling, purpose and weighting have been previously published [17]. Characteristics of interviewers, training and field quality control have been described elsewhere [5]. 2.2. Assessments The Alcohol Use Disorder and Associated Disabilities Interview Schedule–DSM-IV Version (AUDADIS-IV) was used to assess substance use and psychiatric disorders [18]. The AUDADIS-IV has been reported to have excellent reliability and validity in the US [17] and internationally [19]. It includes an extensive list of symptom questions that separately operationalizes DSM-IV criteria for SUDs and additional axis I and axis II diagnoses.
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
2.3. Cannabis use and cannabis use disorders (CUDs) Cannabis use in this study referred to cannabis use in the last 12 months. Respondents were asked about frequency of cannabis use. We categorized cannabis users according to frequency of use, dividing them into “at least weekly” and “less than weekly” cannabis users. Individuals who did not provide this information (n = 23) were excluded from analyses pertaining to frequency of cannabis use. Doses of cannabis used were measured as number of joints consumed on days that cannabis was used in the last 12 months. CUDs referred to cannabis abuse or dependence in the last 12 months as defined by DSM-IV. 2.4. Mental Illness Any psychiatric disorder in this study referred to any axis I, axis II or SUD in the last 12 months. Axis I disorders included mood (major depressive disorder, bipolar I, bipolar II and dysthymia) and anxiety (panic disorder, social anxiety disorder, specific phobia, generalized anxiety disorder) disorders. In the case of all mood and anxiety disorders, we included only primary mental illness and excluded all cases of substance-induced mental disorders as diagnosed in the AUDADIS-IV. Primary (non substance-induced) mood and anxiety disorders were defined if: (1) the respondent did not use alcohol or drugs in the previous 12 months; (2) the episode(s) did not occur in the context of drug or alcohol intoxication or withdrawal; (3) the episode(s) started before initiation of drug or alcohol or (4) the episode(s) began after drug or alcohol consumption began, but persisted for more than 1 month after cessation of intoxication or withdrawal. Respondents were classified as having independent (primary) mood and anxiety disorders if none or only some of their episodes were substance-induced [20]. SUDs referred to any alcohol or drug (excluding cannabis) abuse or dependence. The substances included in this study were cocaine (including crack cocaine), heroin, hallucinogens, inhalants/solvents, sedatives, tranquilizers, opioids and amphetamines. Axis II disorders included any of the following personality disorders: paranoid, schizoid, histrionic, antisocial, obsessive–compulsive, dependent, and avoidant personality disorders. 2.5. Statistical analysis Respondents with any 12-month mental illness were examined with respect to demographic characteristics (sex, age, educational level, household income, marital status, urbanicity, race/ethnicity and region). Prevalence of weekly cannabis use, less than weekly cannabis use and CUDs among individuals with and without any primary psychiatric diagnosis in the last 12-months were calculated with crosstabulations providing 95% confidence intervals. We used the χ 2-statistic to compare differences between groups in the proportion of cannabis users and individuals with CUDs. As it has been repeatedly shown that patterns of cannabis use, as
3
well as prevalence of CUDs vary greatly between men and women [5,21], these analyses were also conducted separately by gender. We used logistic regression to analyze the odds for any 12-month mental illness among different demographic groups, and to analyze the odds for cannabis use and CUDs among individuals with 12-month mental illnesses, while controlling for sociodemographic variables (age, sex, household income, region, educational level, marital status and urbanicity) and any non-cannabis SUDs. In order to examine the effect of age on the relationship between cannabis use and mental illness, we calculated the odds for 12-month cannabis use among individuals with and without mental illness by age group and controlling for other sociodemographic variables. We used the t-test to compare daily dose of cannabis used (measured as number of joints used on days cannabis is consumed) among individuals with and without 12-month mental illness. We used the Mantel– Haenszel χ 2 test for trend to compare rates of cannabis use according to the number of 12-month mental diagnoses. In order to correct for multiple testing, we used an adjusted p-value according to the maximum number of categories included in an analysis (ie, 8 categories of axis I diagnoses). Based on the Bonferroni correction, we calculated that in order to maintain 95% level of confidence we use an adjusted p value of p b 0.00625. Based on the formula presented by Lasser et al. [22], we estimated the proportion of all cannabis used in the US that was consumed by persons with mental illness using the following calculation: ðM ÞðD1 ÞðF1Þ=½ðN ÞðD 2 ÞðF2Þ þ ðM ÞðD1 ÞðF1Þ where M = the percentage of current cannabis users with mental illness; D1 = the mean dose of joints per day consumed by current cannabis users with mental illness; F1 = the mean number of days in the last year in which cannabis was consumed by current cannabis users with mental illness; N = the percentage of current cannabis users without mental illness; D2 = the mean dose of joints per day consumed by current cannabis users without mental illness and F2 = the mean number of days in the last year in which cannabis was consumed by current cannabis users without mental illness. Though this formulation was developed for cigarette smoking, in which standardization is more readily possible, it has merit when dealing with cannabis as well. One of the major challenges in cannabis research is assessing intensity of cannabis used and both dose and frequency have been proposed as proxy terms for the intensity of use [23]. Moreover, substance use clinical trials frequently use daily dose of substances as the primary outcome measure [24]. Accordingly, despite potential errors in estimation, this formulation is useful as it combines daily dose and frequency. We used the same formulation to calculate the portion of cannabis used among occasional and regular cannabis users with mental illness and among individuals with SUDs.
4
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
All results were based on population weighted data to allow conclusions representative for the US. To accurately estimate variances taking the NESARC sample design components into account, analyses were conducted with Software for Survey Data Analysis (SUDAAN) Version 10 [25], a software program that uses Taylor series linearization to make adjustments for the NESARC's sample design characteristics.
3. Results 3.1. Sociodemographic characteristics and association between 12-month mental illness and concurrent cannabis use and CUDs The prevalence of cannabis use and of CUDs among individuals with 12-month mental illness according to demographics is presented in Table 1. The population prevalence of 12-month cannabis use was 4.1%. In the general population, 2.4% reported less than weekly cannabis
use, 1.7% reported at least weekly cannabis use and 1.5% had a CUD. The majority of cannabis users (56.4%) were in the young (18–29) age group: among individuals with 12month mental illness, 59.2% of cannabis users were in the young age group vs 49.2% of cannabis users among individuals without 12-month mental illness. The prevalence of cannabis use and CUDs among individuals with 12-month mental illness was 9.9% and 4.0%, respectively, compared to 1.6% and 0.4% among individuals without any mental illness in the last 12 months. Individuals with 12-month mental illness represented 72.2% of all individuals who used cannabis in the last 12 months and 81.8% of individuals with 12-month CUDs. Prevalence of mental illness was significantly higher among individuals using cannabis at least weekly (77.2%) compared to individuals using cannabis less than weekly (68.7%; p b 0.01). Respondents with any axis I, axis II or SUD had elevated rates of both weekly cannabis use, less than weekly cannabis use and CUDs in the last 12 months (Table 2). Rates of cannabis use and CUDs were particularly elevated among individuals with Bipolar I
Table 1 Prevalence and logistic regression analyses of 12-month mental illness by selected sociodemographic characteristics. Variable Sex Men (ref) Women Age 18–29 (ref) 30–44 45–64 65+ Education less than high school (ref) high school some college or higher Household income $0–$19,999 (ref) $20,000–$39,999 $40,000–$59,000 $60,000 or above Marital status Married (ref) separated/divorced/widowed never married Urbanicity Urban (ref) Rural Race/ethnicity White (ref) Black American Indian Asian Hispanic Region Northeast (ref) Midwest South West Abbreviations: OR = Odds ratio, ref = reference.
12-month mental illness (N = 12,659) prevalence, % (95% CI)
OR
30.0 (28.8–31.3) 29.6 (28.3–30.9)
1.0 0.98 (0.93–1.04)
38.4 (36.5–40.4) 32.2 (30.6–33.8) 28.1 (26.8–29.4) 16.8 (15.7–17.8)
1.0 0.76 (0.71–0.81) 0.63 (0.58–0.67) 0.32 (0.29–0.35)
22.6 (20.5–24.9) 30.4 (29.1–31.8) 30.1 (28.9–31.4)
1.0 1.50 (1.32–1.70) 1.48 (1.30–1.68)
31.4 (29.9–32.9) 31.0 (29.3–32.7) 29.7 (28.3–31.1) 28.0 (26.7–29.4)
1.0 0.98 (0.90–1.07) 0.92 (0.86–0.99) 0.85 (0.78–0.92)
26.4 (25.2–27.6) 32.21 (30.6–33.7) 37.7 (36.0–39.4)
1.0 1.32 (1.34–1.41) 1.69 (1.59–1.79)
30.8 (28.6–33.1) 29.3 (28.3–30.4)
1.0 0.93 (0.84–1.04)
30.5 (29.4–31.6) 29.5 (27.9–31.2) 41.6 (37.1–46.3) 19.9 (16.9–23.2) 27.1 (25.3–28.9)
1.0 0.95 (0.89–1.03) 1.63 (1.35–1.95) 0.56 (0.47–0.68) 0.85 (0.77–0.93)
27.9 (25.6–30.3) 32.6 (29.8–35.6) 27.8 (26.5–29.2) 31.6 (28.7–34.6)
1.0 1.25 (1.05–1.49) 1.0 (0.87–1.14) 1.19 (1.00–1.43)
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
5
Table 2 Prevalence of cannabis use and cannabis use disorders according to 12-month mental illness status (n = 43,093).
Total No mental illness Men Women Any mental illness Men Women Any axis I disorder a Men Women Any SUD b Men Women Any axis II disorder c Men Women
Less than weekly, % (95% CI)
At least weekly, % (95% CI)
CUD, % (95% CI)
2.4 (2.2–2.6) 1.1 (0.9–1.2) 1.4 (1.1–1.6) 0.8 (0.6–1.0) 5.4 (4.9–6.0)⁎ 7.2 (6.3–8.2)⁎ 3.8 (3.3–4.4)⁎ 4.4 (3.8–5.1)⁎ 6.5 (5.2–7.9)⁎ 3.3 (2.7–3.9)⁎ 12.3 (11.0–13.7)⁎ 12.7 (11.0–14.5)⁎ 11.5 (9.4–13.8)⁎ 5.0 (4.3–5.8)⁎ 6.2 (5.0–7.5)⁎ 3.8 (3.1–4.7)⁎
1.7 (1.5–1.9) 0.6 (0.5–0.7) 0.8 (0.6–1.0) 0.3 (0.2–0.5) 4.4 (3.9–5.0)⁎ 6.6 (5.7–7.5)⁎ 2.4 (2.0–3.0)⁎ 3.8 (3.2–4.5)⁎ 7.1 (5.7–8.8)⁎ 2.0 (1.6–2.6)⁎ 10.4 (9.2–11.8)⁎ 11.6 (10.0–13.5)⁎ 7.7 (6.0–9.8)⁎ 5.3 (4.5–6.2)⁎ 7.4 (6.2–9.0)⁎ 3.1 (2.5–4.0)⁎
1.5 (1.3–1.6) 0.4 (0.3–0.5) 0.6 (0.4–0.7) 0.2 (0.1–0.3) 4.0 (3.6–4.5)⁎ 6.0 (5.3–6.9)⁎ 2.1 (1.7–2.5)⁎ 3.5 (3.0–4.2)⁎ 6.4 (5.0–8.1)⁎ 2.0 (1.6–2.5)⁎ 10.2 (9.1–11.4)⁎ 11.3 (9.8–13.0)⁎ 7.1 (6.1–9.8)⁎ 4.8 (4.1–5.5)⁎ 6.8 (5.7–8.2)⁎ 2.6 (2.1–3.4)⁎
Abbreviations: CUD = cannabis use disorders. a Any mood (Bipolar I, Bipolar II, Major Depressive Disorder, dysthymia) or anxiety (panic, social phobia, specific phobia, Generalized Anxiety Disorder) disorder. b Any alcohol or drug use disorder (excluding cannabis). c Any antisocial, avoidant, dependent, paranoid, schizoid, histrionic personality disorder. ⁎ Significantly different from respondents without any psychiatric disorder, p ≤ 0.0001.
disorder (Table 3) and specific (anti-social, and histrionic) personality disorders (Table 4). The relationship between cannabis use and mental illness persisted when we controlled for sociodemographic variables. Compared with respondents without 12-month mental illness, those with any 12-month mental illness were significantly more likely to use cannabis in the last 12 months (OR = 5.2, 95% CI = 4.5–5.9). When additionally controlling for any non-cannabis SUD, the increased odds for cannabis use among individuals with mental illness were retained (OR = 2.5, 95% CI = 2.1–3.0). Adjusting for sociodemographic variables, the odds of having a CUD were 8.0 (95% CI = 6.2–10.3) among individuals with 12-month mental illness compared to those without 12-month mental illness. After additional adjustment for non-cannabis SUDs, the odds ratio for having a CUD among individuals with mental illness compared to those without mental illness was 3.2 (95% CI = 2.2–4.6). We calculated the odds for 12-month cannabis use among individuals with and without mental illness by age group and controlling for other sociodemographic variables. In the 18– 29 year age group, individuals with any 12-month mental illness were significantly more likely to use cannabis in the last 12 months (OR = 5.8, 95% CI = 4.8–7.1) compared to those without mental illness. The association was further retained in other age groups: in the 30–44 and the 45– 64 year age group, individuals with any 12-month mental illness were significantly more likely to use cannabis in the last 12 months (OR = 8.2 (95% CI = 5.6–12.0) and 7.4 (95% CI = 4.6–11.9), respectively) compared to those without any mental illness. In the 65+ age group, this association was not
retained (OR = 5.0, 95% CI = 0.5–48.3), though it should be noted that the prevalence of cannabis use in this age group was generally low (0.04%). 3.2. Daily dose of cannabis used The mean number of joints consumed (on days that cannabis was used) by individuals with 12-month mental illness was 2.2, compared to 1.6 among individuals without any mental illness in the last 12 months (p b 0.0001). Among individuals with 12-month prevalence of mental illness, the mean number of joints among individuals using cannabis less than weekly was 1.4 joints per day, compared to 3.1 joints per day among individuals using cannabis at least weekly (p b 0.0001). 3.3. Cannabis use and multiple psychiatric diagnoses Persons with multiple 12-month psychiatric diagnoses had higher rates of weekly and less than weekly cannabis use than persons with only 1 DSM-IV diagnosis (p b 0.0001, Fig. 1). 3.4. Proportion of cannabis used by individuals with mental illness We estimated that persons with a 12-month mental illness consumed 83% of all cannabis consumed in the US. Among individuals with mental illness, 2.4% of the cannabis was used by individuals using cannabis less than weekly, and 97.6% was used by individuals using cannabis at least weekly. Sixty percent of all cannabis was consumed by individuals with SUDs. Of these, 46% used cannabis at least
6
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
Table 3 Prevalence of cannabis use and cannabis use disorders according to specific 12-month DSM-IV Axis I diagnoses (n = 43,093).
Any mood disorder Men Women Bipolar I Disorder Men Women Bipolar 2 Disorder Men Women Major depression Men Women Dysthymia Men Women Any anxiety disorder Men Women Panic disorder Men Women Social phobia Men Women Specific phobia Men Women GAD Men Women
Less than weekly, % (95% CI)
At least weekly, % (95% CI)
CUD, % (95% CI)
5.8 (4.8–6.9)⁎ 7.9 (6.0–10.2)⁎ 4.6 (3.7–5.7)⁎ 8.4 (6.3–11.0)⁎ 9.8 (6.6–14.1)⁎ 7.3 (5.0–10.5)⁎ 7.6 (4.8–12.0)⁎ 5.6 (2.5–12.1) 9.0 (5.0–15.7) 4.8 (3.7–6.2)⁎ 7.8 (5.4–11.3)⁎ 3.3 (2.4–4.5)⁎ 3.8 (2.5–5.8) 4.9 (2..6–9.0) 3.3 (1.9–5.9) 4.0 (3.4–4.8)⁎ 6.2 (4.8–8.0)⁎ 2.9 (2.3–3.7)⁎ 4.4 (3.1–6.3)⁎ 7.3 (4.4–11.8) 3.3 (1.9–5.4) 5.7 (4.2–7.7)⁎ 9.5 (6.3–14.0)⁎⁎ 3.5 (2.3–5.3)⁎⁎ 3.4 (2.7–4.4)⁎ 5.0 (3.5–7.2)⁎⁎ 2.7 (2.0–3.7)⁎ 4.5 (3.1–6.4)⁎⁎ 5.9 (3.3–10.5) 3.9 (2.4–6.2)
4.9 (4.1–5.9)⁎ 8.9 (7.0–11.3)⁎ 2.7 (2.1–3.6)⁎ 9.6 (7.2–12.7)⁎ 15.5 (10.7–22.1)⁎ 5.2 (3.3–8.0)⁎ 4.8 (2.8–8.3)⁎ 7.6 (3.6–15.1) 3.0 (1.3–6.7) 3.4 (2.4–4.2)⁎ 6.2 (4.2–9.0)⁎ 1.8 (1.2–2.8)⁎⁎ 6.0 (4.3–8.4)⁎ 10.8 (6.9–16.3)⁎⁎ 3.8 (2.3–6.1)⁎⁎ 3.6 (2.9–4.5)⁎ 7.0 (5.2–9.3)⁎ 2.0 (1.5–2.7)⁎ 5.9 (4.1–8.4)⁎ 11.9 (7.3–18.8) 3.5 (2.2–5.4)⁎⁎ 3.4 (2.4–4.9)⁎ 4.8 (2.9–7.8) 2.6 (1.5–4.5) 3.8 (2.9–4.9)⁎ 7.9 (5.5–11.0)⁎ 2.0 (1.3–2.9)⁎ 5.8 (4.0–8.3)⁎ 12.3 (7.1–19.5) 3.2 (2.0–5.1)⁎⁎
4.9 (4.1–5.9)⁎ 6.3 (6.5–10.6)⁎ 3.0 (2.3–3.9)⁎ 9.4 (7.0–12.5)⁎ 14.6 (10.2–20.3)⁎ 5.5 (3.5–8.6)⁎ 6.1 (3.6–10.0)⁎ 8.8 (4.3–16.8) 4.3 (2.1–8.4) 3.1 (2.3–4.0)⁎ 5.2 (3.6–7.5)⁎ 2.0 (1.3–3.1)⁎⁎ 5.7 (3.9–8.4)⁎ 8.7 (5.0–14.6)⁎⁎ 4.4 (2.7–7.1) 3.2 (2.5–3.9)⁎ 5.7 (4.2–7.7)⁎ 2.0 (1.5–2.6)⁎ 5.0 (3.6–7.1)⁎ 8.8 (5.4–14.2)⁎⁎ 3.5 (2.2–5.6)⁎ 3.9 (2.8–5.4)⁎ 5.0 (3.0–8.4) 3.2 (2.0–5.1)⁎⁎ 2.9 (2.2–3.9)⁎ 5.6 (3.6–8.5)⁎⁎ 1.8 (1.2–2.6)⁎ 5.5 (3.7–8.3)⁎⁎ 10.1 (5.5–17.9) 3.6 (2.3–5.8)⁎⁎
Abbreviations: CUD = cannabis use disorders. ⁎ Significantly different from respondents without any psychiatric disorder, p ≤ 0.0001. ⁎⁎ Significantly different from respondents without any psychiatric disorder, p b 0.001.
weekly; we estimated they consumed 97.5% of all cannabis consumed by individuals with SUDs.
4. Discussion We found that persons with mental illness are almost 10 times as likely to use cannabis weekly or suffer from a CUD, and more than 5 times as likely to use cannabis less than weekly, compared to individuals without mental illness. Though increased rates of cannabis use among individuals with mental illness are well-known, our findings add to previous findings in the following: (1) we included only individuals with primary (e.g., not substance-induced) mood and anxiety disorders based on DSM-IV diagnostic criteria; (2) we differentiated cannabis users by frequency (3) we report on a wide range of primary axis I and axis II disorders and (4) we estimated percentage of cannabis consumed by individuals with mental illness based on frequency of use and reported daily doses, as an additional measure of the association between mental illness and cannabis use. The unique data collected in the NESARC pertaining to cannabis use and mental illness allowed for these analyses.
Among non-SUD axis I mental disorders, CUDs and atleast weekly cannabis use were particularly prevalent among individuals with bipolar I disorder, with approximately 10% of individuals fulfilling criteria of a CUD or reporting using cannabis at least once per week. This is lower than rates of cannabis related problems cited in various studies among individuals with bipolar disorder, which are as high as 50% [26]. These differences in prevalence rates may be accounted for by both criteria and information variance [27]. Differences in instruments used (for example AUDADIS-IV in the NESARC vs. the National Institute of Mental Health Diagnostic Interview Schedule (DIS) and Composite International Diagnostic Interview (CIDI) in the ECA and NCS, respectively) have been reported to affect prevalence rates in population-based studies [28]. It has been further suggested that respondents may be less forthcoming in surveys conducted by government agencies to reveal information pertaining to substance use. Nevertheless, this would not seem to specifically affect individuals with bipolar disorder; hence, the increased rate of frequent cannabis use and CUDs among individuals with bipolar disorder relative to individuals with other psychiatric disorders should be noted.
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
7
Table 4 Prevalence of cannabis use and cannabis use disorders according to specific 12-month DSM-IV Axis II diagnoses (n = 43,093).
Antisocial PD Men Women Avoidant PD Men Women Dependent PD Men Women Obsessive–Compulsive PD Men Women Paranoid PD Men Women Schizoid PD Men Women Histrionic PD Men Women
Less than weekly, % (95% CI)
At least weekly, % (95% CI)
CUD, % (95% CI)
10.1 (8.1–12.5)⁎ 9.8 (7.3–13.0)⁎ 10.8 (7.6–15.2)⁎ 5.7 (4.1–8.0)⁎ 9.7 (6.2–14.7)⁎⁎ 3.2 (1.9–5.4) 7.1 (3.5–13.9) 11.4 (3.9–28.9) 4.7 (2.0–10.8) 3.7 (2.9–4.7)⁎ 4.8 (3.4–6.6)⁎⁎ 2.6 (1.9–2.7)⁎⁎ 5.7 (4.5–7.2)⁎ 6.9 (4.6–10.3)⁎⁎ 4.9 (3.6–6.7)⁎ 3.9 (2.8–5.5)⁎ 5.0 (3.2–7.8) 2.9 (1.7–5.0) 9.0 (7.1–11.5)⁎ 10.2 (7.2–14.3)⁎ 7.9 (5.4–11.5)⁎
13.4 (11.2–15.9)⁎ 13.6 (11.0–16.7)⁎ 12.7 (9.4–17.0)⁎ 5.3 (3.6–7.7)⁎ 9.6 (5.9–15.3) 2.5 (1.4–4.5) 10.5 (5.4–19.5) 24.3 (12.2–42.7) 2.9 (1.1–7.1) 3.3 (2.5–4.4)⁎ 5.3 (3.7–7.4)⁎ 1.5 (1.0–2.4)⁎⁎ 7.0 (5.4–9.0)⁎ 11.6 (8.5–15.7)⁎ 3.7 (2.6–5.1)⁎ 6.3 (4.7–8.4)⁎ 8.2 (5.5–12.0)⁎ 4.5 (3.1–6.6)⁎ 9.9 (7.1–13.6)⁎ 14.8 (10.1–21.0)⁎ 5.1 (3.1–8.3)⁎⁎
12.0 (10.0–14.5)⁎ 13.1 (10.5–16.4)⁎ 9.1 (6.4–12.9)⁎ 5.6 (3.9–3.7)⁎ 10.1 (6.5–15.3)⁎⁎ 2.7 (1.6–4.7) 14.2 (8.5–22.8) 30.0 (16.9–47.4) 5.5 (2.7–10.6) 3.5 (2.7–4.5)⁎ 4.9 (3.9–6.8)⁎ 2.2 (1.5–3.2)⁎ 5.8 (4.5–7.5)⁎ 9.0 (6.5–12.4)⁎ 3.6 (2.6–5.0)⁎ 5.2 (3.8–7.0)⁎ 6.2 (3.9–9.7)⁎⁎ 4.2 (2.8–6.1)⁎ 9.8 (7.3–13.1)⁎ 13.9 (9.9–19.4)⁎ 5.8 (3.6–9.2)⁎⁎
Abbreviations: CUD = cannabis use disorders; PD = personality disorder. ⁎ Significantly different from respondents without any psychiatric disorder, p ≤ 0.0001. ⁎⁎ Significantly different from respondents without any psychiatric disorder, p b 0.001.
Individuals with alcohol and drug use disorders were at a particularly high risk for weekly cannabis use; more than 10% of these individuals used cannabis at least once per week, and rates of CUDs in this population were more than 10%. Multiple substance use may be influenced by common neurobiological factors, given that different substances act upon similar brain loci and involve similar neurotransmitter systems [29,30]. In addition, social factors and drug availability should be considered. Patterns of drug initiation have been found to vary across countries and cultures [3], suggesting that social settings that facilitate specific drug use may be particularly important. Though cannabis use and CUDs have been largely overlooked in treatment settings for substance use problems, the rates of admission to nationally funded treatment centers with cannabis stated as the primary problem-drug are gradually increasing. The rate of treatment admission with cannabis as the primary substance almost doubled between 1993 and 1999 [31]. In 2008, 16.6% of individuals discharged from treatment facilities reported cannabis as their primary substance of abuse [32]. Our findings emphasize the importance of assessing for frequent cannabis use and CUDs among individuals with any alcohol or drug use disorder. Among individuals with personality disorders, CUDs and at-least weekly cannabis use disorders were particularly prevalent among individuals with antisocial and histrionic personality disorders. Among these, antisocial personality disorder has been particularly shown to be associated with an increased risk for persistence of CUDs during a 3-year follow-up [33]. Though the increased association between
substance use disorders and cluster B personality disorders is well known [34], additional investigation of CUDs among individuals with personality disorders belonging to other clusters, which takes into account temporal relationships (such as that conducted by Hasin and colleagues [33]), can help in elucidating mechanisms of these co-morbidities. Almost without exception, the rates of cannabis use and CUDs were higher among men compared to women. This echoes findings on rates of cannabis use and CUDs in the general population in previous population-based surveys (for example, the National Longitudinal Alcohol Epidemiologic Survey (NLAES, 1991–1992)) as well as in the NESARC sample [5]. Though it is possible that these gender differences reflect differences in effects of cannabis, as well as susceptibility to frequent cannabis use and CUDs [35], it has also been suggested that these gender differences can be explained by differences in opportunities to use cannabis and other drugs, and that given the initial opportunity to use drugs, men and women are equally likely to move on to frequent cannabis use [36,37]. Among individuals with mental illness, both increased opportunities for use amongst men as well as potential increased susceptibility for frequent use and misuse should be taken into account even in those disorders which are more prevalent among women, such as depression and anxiety disorders. The association between mental illness and cannabis use was pervasive across most age groups. This implies that though cannabis use is generally more prevalent among younger people, the prevalence of cannabis use is higher among individuals with mental illness in older adults as well.
8
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx 25
20
15
10
5
0 0
1
2
3
4
>4
No. of 12-month Psychiatic Diagnoses Less than weekly cannabis use At least weekly cannabis use
Fig. 1. Patterns of cannabis use according to the number of 12-month DSMIV psychiatric diagnoses.
We found that the prevalence of cannabis use, both weekly and less than weekly, increases with the increase in number of DSM-IV psychiatric diagnoses. This is particularly important given reports that the vast majority of mental disorders are co-morbid disorders; results from the NCS show that 79% of individuals with a lifetime mental disorder had an additional co-morbid disorder. Frequent cannabis use among individuals with mental illness is associated with symptom exacerbation [38,39] and poorer treatment outcomes [39,40]. Individuals with co-morbid mental illness and substance use are at increased risk of higher rates of hospitalization [41], homelessness or housing instability [42], and increased treatment costs [43]. Nevertheless, there is still a paucity of clinical data on the impact of cannabis use on the prognosis of specific psychiatric disorders. Why do individuals with mental illness consume more cannabis? It has been suggested that such persons use cannabis as a means of self-medication of psychiatric symptoms [44]. Neurobiological research has implicated that the endocannabinoid (eCB) system is highly expressed in different brain regions and has regulatory functions and that it may be deeply involved in many mental disorders [45]. It is possible that cannabis use and mental illness share common factors; the same factors that predispose people to mental illness also increase their risk of cannabis use. These common factors may include biological, personality, social or environmental factors, or some combination of these factors [23]. Finally, it is possible that cannabis use increases the risk for developing mental illness. Evidence for an association between cannabis use and the development of psychotic disorders has accumulated [12]. There is a growing consensus that those who use cannabis, particularly heavy users and individuals who initiated cannabis use at a young age, are at increased risk for developing psychotic disorders.
There is evidence pointing to an association between heavy cannabis use and the development of depression [12]. Nevertheless, there is a scarcity of longitudinal studies examining the association between cannabis use and the development of various mental illnesses. Further longitudinal studies with multiple waves which account for temporal relationships between cannabis use and mental illness (such as that conducted by Wittchen and colleagues [10]) are important in elucidating the potential mechanisms involved in this co-morbidity. As is common in large-scale epidemiological surveys, limitations of this study should be recognized. First, information was based on self-reporting, allowing for recall and social desirability biases. Second, because the NESARC sample included only civilian households and quarters populations, information on individuals in prisons is missing. Since the prevalence of mental illness and substance use has been reported to be high in prison populations [46], this may affect calculations. Third, data do not include adolescents, a particularly vulnerable population for cannabis use [47]. Fourth, though the NESARC evaluated a large number of common mental disorders, additional disorders in which prevalence of cannabis use is known to be high have not been included in the NESARC survey or were not assessed using specific diagnostic criteria. Of particular interest may be rates of cannabis use and CUDs among individuals with psychotic disorders, which have been reported to be particularly high [45]. Psychotic disorders in the NESARC were assessed using a single question (“Did a doctor or other health professional diagnose schizophrenia or psychotic illness or episode in the last 12 months”). As this method of assessment is substantially different from the diagnostic assessment of all other DSM-IV disorders included in the NESARC, and may include an underestimation of psychotic disorders, we chose not to include these data in our study. It should be noted that the prevalence of cannabis use in this sample of individuals with anxiety disorders is lower than that reported in other samples [48]. There are several factors which may explain this. Differences in instruments used (to assess both anxiety disorders and substance use) have been reported to affect prevalence rates in population-based studies [27,28]. Difference in survey methodologies (e.g., coverage weight, data weighing) as well as question text may also affect estimations. It has been further suggested that respondents may be less forthcoming in surveys conducted by government agencies to reveal information pertaining to substance use [49]. Any or all of these considerations may have affected the rate of cannabis use in our sample. Finally, given the expected variability in potency of cannabis used and taking into account the variance of in estimating standard doses of cannabis (e.g., joints) [24], the limitation of applying the self-reported dose of cannabis used to the Lasser formulation should be acknowledged and the estimation of percentage of cannabis consumed by individuals with mental illness referred to with caution. Nevertheless, it seems that these challenges in
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx
estimating doses of cannabis used will not resolve soon and even if not perfectly precise, this application of the Lasser formulation helps in elucidating the large impact of mental illness of cannabis use. Extrapolating our results to the general population, we estimate that persons with a diagnosable primary mental illness constitute almost 75% of all cannabis users, with a particularly high prevalence of more frequent cannabis use and CUDs in this population. Our findings emphasize the importance of proper screening for frequent cannabis use and CUDs particularly among individuals with mental illness, and focusing prevention and treatment efforts on the mentally ill. The largely disparate services for mental health and substance abuse pose serious challenges of service provision for this population. Treatment fragmentation between mental health and substance abuse services may mean that individuals with co-morbid cannabis use disorders and mental health problems do not receive adequate care. This results in suboptimal treatment outcomes in a population that already suffers some of the poorest outcomes in terms of illness. References [1] European Monitoring Centre for Drugs and Drug Addiction. Annual report. Luxembourg: Publications Office of the European Union; 2010. [2] United Nations Office on Drugs and Crime. World drug report. Vienna 2010. [3] Degenhardt L, Hall W. Extent of illicit drug use and dependence, and their contribution to the global burden of disease. Lancet 2012; 379(9810):55-70. [4] Substance Abuse and Mental Health Services Administration. National survey on drug use and health. Rockville MD: Substance Abuse and Mental Health Services Administration; 2008. [5] Compton WM, Grant BF, Colliver JD, Glantz MD, Stinson FS. Prevalence of marijuana use disorders in the United States: 1991–1992 and 2001–2002. JAMA 2004;291(17):2114-21. [6] Regier DA, Farmer ME, Rae DS, Locke BZ, Keith SJ, Judd LL, et al. Comorbidity of mental disorders with alcohol and other drug abuse. Results from the Epidemiologic Catchment Area (ECA) Study. JAMA 1990;264(19):2511-8. [7] Kessler RC, Nelson CB, McGonagle KA, Edlund MJ, Frank RG, Leaf PJ. The epidemiology of co-occurring addictive and mental disorders: implications for prevention and service utilization. Am J Orthopsychiatry 1996;66(1):17-31. [8] Buckner JD, Schmidt NB, Lang AR, Small JW, Schlauch RC, Lewinsohn PM. Specificity of social anxiety disorder as a risk factor for alcohol and cannabis dependence. J Psychiatr Res 2008;42(3): 230-9. [9] Hayatbakhsh MR, McGee TR, Bor W, Najman JM, Jamrozik K, Mamun AA. Child and adolescent externalizing behavior and cannabis use disorders in early adulthood: an Australian prospective birth cohort study. Addict Behav 2008;33(3):422-38. [10] Wittchen HU, Frohlich C, Behrendt S, Gunther A, Rehm J, Zimmermann P, et al. Cannabis use and cannabis use disorders and their relationship to mental disorders: a 10-year prospective-longitudinal community study in adolescents. Drug Alcohol Depend 2007;88(Suppl 1):S60-70. [11] Hall W, Degenhardt L. Adverse health effects of non-medical cannabis use. Lancet 2009;374(9698):1383-91. [12] Moore TH, Zammit S, Lingford-Hughes A, Barnes TR, Jones PB, Burke M, et al. Cannabis use and risk of psychotic or affective mental health outcomes: a systematic review. Lancet 2007;370(9584):319-28.
9
[13] Swift W, Coffey C, Degenhardt L, Carlin JB, Romaniuk H, Patton GC. Cannabis and progression to other substance use in young adults: findings from a 13-year prospective population-based study. J Epidemiol Community Health 2011. [14] Patton GC, Coffey C, Carlin JB, Degenhardt L, Lynskey M, Hall W. Cannabis use and mental health in young people: cohort study. BMJ 2002;325(7374):1195-8. [15] Grant BF, Moore TC, Kaplan K. Source and accuracy statement: wave 1 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). Bethesda, Md: National Institute on Alcohol Abuse and Alcoholism; 2003. [16] Huang B, Grant BF, Dawson DA, Stinson FS, Chou SP, Saha TD, et al. Race–ethnicity and the prevalence and co-occurrence of Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, alcohol and drug use disorders and Axis I and II disorders: United States, 2001 to 2002. Compr Psychiatry 2006;47(4):252-7. [17] Grant BF, Dawson DA, Stinson FS, Chou PS, Kay W, Pickering R. The Alcohol Use Disorder and Associated Disabilities Interview Schedule-IV (AUDADIS-IV): reliability of alcohol consumption, tobacco use, family history of depression and psychiatric diagnostic modules in a general population sample. Drug Alcohol Depend 2003;71(1):7-16. [18] Grant BF, Hasin D. The Alcohol Use Disorder and Associated Disabilites Schedule–DSM-IV Version. Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism; 2001. [19] Chatterji S, Saunders JB, Vrasti R, Grant BF, Hasin D, Mager D. Reliability of the alcohol and drug modules of the Alcohol Use Disorder and Associated Disabilities Interview Schedule–Alcohol/ Drug-Revised (AUDADIS-ADR): an international comparison. Drug Alcohol Depend 1997;47(3):171-85. [20] Grant BF, Stinson FS, Dawson DA, Chou SP, Dufour MC, Compton W, et al. Prevalence and co-occurrence of substance use disorders and independent mood and anxiety disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Arch Gen Psychiatry 2004;61(8):807-16. [21] Lev-Ran S, Imtiaz S, Taylor BJ, Shield KD, Rehm J, Le Foll B. Gender differences in health-related quality of life among cannabis users: results from the national epidemiologic survey on alcohol and related conditions. Drug Alcohol Depend 2011. [22] Lasser K, Boyd JW, Woolhandler S, Himmelstein DU, McCormick D, Bor DH. Smoking and mental illness: a population-based prevalence study. JAMA 2000;284(20):2606-10. [23] Degenhardt L, Hall W, Lynskey M. Exploring the association between cannabis use and depression. Addiction 2003;98(11):1493-504. [24] Mariani JJ, Brooks D, Haney M, Levin FR. Quantification and comparison of marijuana smoking practices: blunts, joints, and pipes. Drug Alcohol Depend 2011;113(2–3):249-51. [25] Research Triangle Institute. Software for survey data analysis (SUDAAN). 10 ed. Research Triangle Park, NC: Research Triangle Institute; 2004. [26] Cerullo MA, Strakowski SM. The prevalence and significance of substance use disorders in bipolar type I and II disorder. Subst Abuse Treat Prev Policy 2007;2:29. [27] Klerman GL. Diagnosis of psychiatric disorders in epidemiologic field studies. Arch Gen Psychiatry 1985;42(7):723-4. [28] Regier DA, Kaelber CT, Rae DS, Farmer ME, Knauper B, Kessler RC, et al. Limitations of diagnostic criteria and assessment instruments for mental disorders. Implications for research and policy. Arch Gen Psychiatry 1998;55(2):109-15. [29] Koob GF, Le Moal M. Drug addiction, dysregulation of reward, and allostasis. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 2001;24(2): 97-129. [30] Nutt DJ. The neurochemistry of addiction. Hum Psychopharmacol 1997;12:s53-8. [31] Substance Abuse and Mental Health Services Administration. National household survey on drug abuse: main findings, 1998. Rockville, MD:
10
[32]
[33]
[34]
[35] [36] [37]
[38]
[39]
S. Lev-Ran et al. / Comprehensive Psychiatry xx (2013) xxx–xxx Substance Abuse and Mental Health Services Administration. Office of Applied Studies; 2002. Substance Abuse and Mental Health Services Administration. Treatment Episode Data Set (TEDS): 2008. Discharges from Substance Abuse Treatment Services, DASIS Series: S-56, HHS Publication No. (SMA) 11–4628. Rockville, MD: Substance Abuse and Mental Health Services Administration; 2010. p. 55-6. Hasin D, Fenton MC, Skodol A, Krueger R, Keyes K, Geier T, et al. Personality disorders and the 3-year course of alcohol, drug, and nicotine use disorders. Arch Gen Psychiatry 2011;68(11):1158-67. Taylor J. Substance use disorders and Cluster B personality disorders: physiological, cognitive, and environmental correlates in a college sample. Am J Drug Alcohol Abuse 2005;31(3):515-35. Greenfield SF, O'Leary G. Sex differences in marijuana use in the United States. Harv Rev Psychiatry 1999;6(6):297-303. Van Etten ML, Neumark YD, Anthony JC. Male–female differences in the earliest stages of drug involvement. Addiction 1999;94(9):1413-9. Van Etten ML, Anthony JC. Comparative epidemiology of initial drug opportunities and transitions to first use: marijuana, cocaine, hallucinogens and heroin. Drug Alcohol Depend 1999;54(2):117-25. Arendt M, Rosenberg R, Fjordback L, Brandholdt J, Foldager L, Sher L, et al. Testing the self-medication hypothesis of depression and aggression in cannabis-dependent subjects. Psychol Med 2007;37(7): 935-45. Goldberg JF, Garno JL, Leon AC, Kocsis JH, Portera L. A history of substance abuse complicates remission from acute mania in bipolar disorder. J Clin Psychiatry 1999;60(11):733-40.
[40] Bricker JB, Russo J, Stein MB, Sherbourne C, Craske M, Schraufnagel TJ, et al. Does occasional cannabis use impact anxiety and depression treatment outcomes? Results from a randomized effectiveness trial. Depress Anxiety 2007;24(6):392-8. [41] Grace R, Shenfield G, Tennant C. Cannabis and psychosis in acute psychiatric admissions. Drug Alcohol Rev 2000;19(3):287-90. [42] Drake RE, Osher FC, Wallach MA. Homelessness and dual diagnosis. Am Psychol 1991;46(11):1149-58. [43] Bartels SJ, Teague GB, Drake RE, Clark RE, Bush PW, Noordsy DL. Substance abuse in schizophrenia: service utilization and costs. J Nerv Ment Dis 1993;181(4):227-32. [44] Khantzian EJ. The self-medication hypothesis of substance use disorders: a reconsideration and recent applications. Harv Rev Psychiatry 1997;4(5):231-44. [45] Leweke FM, Koethe D. Cannabis and psychiatric disorders: it is not only addiction. Addict Biol 2008;13(2):264-75. [46] Fazel S, Baillargeon J. The health of prisoners. Lancet 2011; 377(9769):956-65. [47] Lynskey MT, Heath AC, Bucholz KK, Slutske WS, Madden PA, Nelson EC, et al. Escalation of drug use in early-onset cannabis users vs co-twin controls. JAMA 2003;289(4):427-33. [48] Agosti V, Nunes E, Levin F. Rates of psychiatric comorbidity among US residents with lifetime cannabis dependence. Am J Drug Alcohol Abuse 2002;28(4):645-54. [49] Grucza RA, Abbacchi AM, Przybeck TR, Gfroerer JC. Discrepancies in estimates of prevalence and correlates of substance use and disorders between two national surveys. Addiction 2007;102(4):623-9.