Youth with substance use disorders (SUDs) often have problems with mental illnesses and youth with mental illnesses, including mood and anxiety disorders, often have problems with SUDs. In one study 50% of youth with SUDs also had non-psychotic mental health problems, and youth with SUDs plus either major depressive disorder or post-traumatic stress disorder (PTSD) had more total disorders, more substance use-related problems, and lower quality of life (1). Research has shown that the rates of alcohol dependence are significantly higher in people with combined anxiety and depression (2), and people (especially young males) with mood and anxiety disorders are at high risk of comorbid SUDs (3). The co-occurrence of SUDs and mental illnesses are important in youth because the combination of these conditions worsens prognosis and complicates treatment (4). In addition, SUDs predispose to completed suicide in youth and young adults (5), although in the case of cannabis use, this might be mediated through other risk factors (6).
A number of studies have demonstrated that cannabis use may precipitate the onset of bipolar disorder, depression, and anxiety (as well as psychosis) in genetically vulnerable youth (7). In fact, a growing body of research suggests an association between prior cannabis use and the onset of mood and anxiety disorders in youth (8–11). This is especially problematic because cannabis is the most commonly used recreational drug and rates of cannabis use remain high, at over 25%, for North American youth (12, 13).
For these reasons, we were interested in discovering the rates of SUD risk and cannabis use in a population of youth seeking specialized psychiatric evaluation and treatment for mood and/or anxiety disorders. We were also interested in whether or not cannabis use was associated with functional impairment in this population. This involved the clinical program known as the First Episode Mood and Anxiety Program (FEMAP) in London, Ontario, Canada. This is an outpatient psychiatric clinic affiliated with the London Health Sciences Centre that serves youth (age 16–26) with the recent onset of primary mood and/or anxiety concerns. This program screens out and refers youth presenting with primary substance-related concerns to a specialized SUD program in the community. However, it was noted by clinicians at FEMAP that cannabis use was common even in this population of youth with primary mood and anxiety disorders.
The current study sought to clarify the prevalence of youth at risk for SUDs within this psychiatric service aimed at youth clinically prescreened not to have primary SUDs. It also sought to understand some of the associated features of youth cannabis use, and whether or not cannabis use was predictive of degree of functional impairment. Based on our clinical experiences, we believed that higher cannabis use would be associated with certain demographic features as well as some traits, life experiences, and psychiatric symptoms. These included gender, age, socioeconomic status, inattentive and maladaptive coping style, age of first exposure to risky behaviors, rates of current substance use, and severity of psychiatric symptoms. Understanding the associations with cannabis use, the most common illicit substance used by youth today, in the population of youth seeking help for non-psychotic mental illnesses could help to clarify the need for addiction specialists within primary care as well as within specialized mood/anxiety clinics for youth.
All clients coming to FEMAP between October 2009 and July 2012 were invited to participate in the research as the inclusion and exclusion criteria for a clinical assessment at FEMAP were identical to those for the research study. Participants were provided with information about the study, had all their questions answered, and then signed written, informed consent if willing to participate. The study was approved by the Human Research Ethics Board affiliated with the London Health Sciences Centre.
Participants underwent a clinical evaluation by one of the clinical investigators (E.R., C.F., or C.S.) for assessment of their concerns, symptoms, function, substance use, living circumstances, and severity of illness. They completed self-report questionnaires as described below. Based on the clinical evaluation combined with the assessment instruments, a primary (single) final hypothesized diagnosis was made prior to the participant being scheduled with a psychiatrist for definitive diagnostic assessment. Trauma-related diagnoses (acute stress disorder and PTSD) were identified and reported separately from other anxiety disorders due to the differences in clinical needs for these conditions.
Youth, aged 16–26 years, were screened and excluded if they had a major medical problem, history of serious head injury, developmental delay, or the presence of an attention problem severe enough to have warranted psycho-stimulant medication since childhood. They were also excluded if they had previously been treated with medications for a total of 18 months in their lifetime, because this was exclusionary for entry into FEMAP. In addition, if the individual reported that his/her mood and/or anxiety symptoms only occurred after the onset of frequent use of a substance of abuse, then he/she was excluded and referred to an addiction service for youth in the community.
Participants were administered a demographic questionnaire that included age, gender, and information about their living situation and parents. Youth were asked about their family income as well as parental marital status and mother’s and father’s educational level. Unfortunately, many youth were not aware of their family’s income. As a proxy for family stability and socioeconomic status, parental marital status and mother’s educational level were used. Mother’s educational status was considered a better proxy than father’s because family income (and child academic achievement) better coincides with mother’s education after divorce (14).
The following well-validated questionnaires were administered to participants. The Beck Depression Inventory (BDI) (15) measures extent of depressive symptomatology; the Spielberger State Anxiety Inventory-State (STAI) (16) measures extent of state anxiety symptomatology; the Emotion Regulation Questionnaire (ERQ) (17) rates coping strategies related to the regulation of emotional states and is divided into two subscores – reappraisal and suppression; the Adult Attention Deficit Hyperactivity Disorder (ADHD) Self-Report Scale-version 1.1 (ASRS) (18) is a self-report measure of difficulties with attention; the NIDA Modified ASSIST (National Institute of Drug Abuse modified version of the WHO Alcohol, Smoking and Substance Involvement Screening Test, version 3.0) (19) measures substance use habits and predicts risk of SUDs; the Sheehan Disability Scale (SDS) (20) includes three subscales that measure school/work, family life, and social functional impairment; and the Youth Risk Behavior Survey (YRBS) (21) asks about a wide range of historical and recent risk behaviors common in youth populations.
Questionnaires representing trait coping styles were the ERQ and the ASRS. A total ERQ score was calculated as the sum of the Reappraisal subscale (considered to be an adaptive approach to emotion regulation) minus the sum of the suppression subscale (thought to be a maladaptive approach to emotion regulation) (22). The ASRS was being used here as a proxy for cognitive dysfunction in the spectrum of inattention, rather than as a predictor of the diagnosis of ADHD, a total score was derived by adding the total scores for part A and part B of the instrument.
Several items from the YRBS were analyzed as measures of early habits that we thought would be associated with subsequent substance use risk. These included the YRBS questions on age of first smoking of a cigarette, age of first alcoholic drink, age of first use of marijuana, and age of first sexual intercourse. The original items rank the answers from “never”, then “less than age 12” sequentially up to “17 or more years old”. For the linear regression analyses, these items were recoded such that the answer “never” was higher than the highest age of first use, thereby creating a logically consistent ordinal scale. The YRBS questions about alcohol, tobacco and cannabis use in the past 30 days were also hypothesized to be predictive of current cannabis use. The total scores for each substance from the ASSIST were used as risk scores for each of the substances of abuse, including cannabis, and the total ASSIST score provided a rating of overall risk for a SUD.
Descriptive, frequency, regression, and all other statistics were calculated using SPSS, version 20 (23). Data were explored for incorrect entries and these were corrected. Missing entries of data were treated by eliminating those cases from the analyses.
Backward selection multiple regression was chosen because this would provide the most parsimonious model to predict the dependent variable, given the identified independent variables which were all considered hypothetical predictors. A regression analysis was performed using the following variables to predict NIDA ASSIST total cannabis score: demographic variables of participants (age and gender); demographic variables of participants’ families (parental marital status, mother’s educational level); trait variables related to coping and cognition (ERQ and ASRS); participants’ early risk behaviors (age of first use of tobacco, alcohol, and cannabis and first sexual intercourse); recent substance use (alcohol, cannabis, and tobacco), and, lastly, measures of psychiatric symptoms and functional impairment (BDI, STAI, total SDS score).
In addition, to better understand the contributions to functional impairment in this population and the extent of the association of cannabis use (and other variables) to functional impairment, a multiple regression using level of functional impairment (SDS score) as the dependent variable was conducted with the other variables listed above as predictor variables.
Skewness was evaluated for all variables used in the regression analyses and was found to be <1.5 for all. Although BDI and STAI were strongly correlated (r=0.64, p<0.0005), collinearity statistics for the group of variables did not indicate that the combination was problematic in the regression analyses (collinearity tolerances for all variables=0.500).
Participants were 429 youth whose mean age was 19 (standard deviation, SD=2.7). These included 271 (63%) female and 158 (37%) male youth. The sample had well-educated mothers. The mean educational level was described as “some college or specialized education”, and the mode was “graduate from a college or university”. A total of 74% of respondents’ mothers had at least some college or university education. A slim majority (54%) of participants came from families with married parents (including common law married), whereas 46% came from families of separated, divorced, or widowed parents.
Of the participants, 325 (68%) received the primary hypothesized diagnosis of depression, anxiety, or the combination of both, whereas 22 (4.5%) were thought to have a primary SUD, notwithstanding the prescreen to exclude and refer such participants. Other diagnostic categories representing over 5% of the sample included: bipolar disorder (7.8%), trauma disorders (5.9%), and DSM-Axis IV disorders (9.4%). This latter category included participants who were evaluated to have problems related to life stressors (conflicts within the family, lack of stable housing or food, etc.), but did not have PTSD. For this category, the clinical evaluator had to believe that these participants would not have sought help were it not for these stressors, and that the individual would be symptom free if these were corrected. Another 4.4% had a mixture of other Axis I diagnoses.
The mean difference score for the ERQ (Reappraisal minus Suppression) was 5.3 (SD 8.7, range –19 to 36). The ASRS mean total score was 8.5 (SD 4.4, range 0–18).
Participant risk behaviors
Related to age of onset, the most frequently reported age of first smoking a cigarette in this sample was “never” (43% of sample). Only 38% of participants endorsed smoking a cigarette before the age of 17 (the remainder endorsed smoking for the first time at the age of 17 or older). In contrast to this, only 8% of subjects endorsed never having had a drink of alcohol and 78% reported drinking for the first time before the age of 17. Furthermore, 30% endorsed never having tried cannabis and just over half, or 53%, reported using cannabis for the first time before the age of 17. Lastly, 33% of these youth reported never having had sexual intercourse, whereas 43% had sex for the first time before the age of 17.
The YRBS includes 30-day frequency of use of multiple substances, including cigarettes, alcohol, and cannabis. A total of 64% of our sample endorsed using no cigarettes in the past 30 days. In contrast to this, just under one-third (31.5%) of participants reported no alcoholic drinks in the past 30 days, whereas 37.5% endorsed alcohol use on more than 3 days in the past 30 days (the remainder used alcohol between 1 and 3 days in the past 30 days). Lastly, 62% of our sample reported no cannabis use in the past 30 days, and 21% reported using cannabis 10 or more times in the past 30 days (the remainder used cannabis 1–9 times). As expected, there was a high correlation between 30-day cannabis use frequency and ASSIST total cannabis score (r=0.86, p<0.0005). Therefore, 30-day cannabis use frequency was not used in the regression analysis but 30-day tobacco and alcohol use frequencies were.
Psychiatric and SUD symptoms
The symptoms score means, SDs, and ranges were as follows for the various symptom questionnaires. BDI mean=31 (SD=11.6, range=0–59), which was in the severe range of depressive symptomatology; STAI mean=56 (SD=12.0, range=20–80), which was at the high end of the moderate range of anxiety; SDS mean=20 (SD=6.3, range=1–30); ASSIST total cannabis mean=8 (SD=10.9, range=0–39). In addition, the ASSIST defines score ranges to identify low, moderate, and high substance use risk based on questionnaire total score. In this sample, 50.3% were at low, 32.6% were at moderate, and 17.0% were at high risk for SUDs.
Use of other substances was also evaluated by the ASSIST, including cocaine, stimulants, methamphetamine, inhalants, sedatives, hallucinogens, street opioids, prescription opioids, and “other” drugs. In this sample, rates of use of these drugs were low. The most commonly used illicit drugs after cannabis were sedatives (16.3% of the sample had non-zero total scores) and hallucinogens (16.1% had non-zero total scores). For all other substances, <10% of participants admitted to having used each of them.
Using the backward selection method, a significant model emerged that associated other variables with ASSIST total cannabis score (F7, 421=37.943, p<0.0005. adjusted R2=0.377). Significant variables are shown in Table 1.
|Predictor variable||Adjusted β||Significance p-Value||95% Confidence interval for B|
|Lower bound||Upper bound|
|Gender (male=1, female=2)||–0.195||<0.0005||–6.184||–2.664|
|Age at first use of marijuana||–0.263||<0.0005||–2.961||–1.404|
|30-Day cigarette use per day||0.278||<0.0005||1.241||2.399|
|Sheehan Disability Scale Score||0.137||<0.001||0.104||0.370|
Additionally, a significant model emerged that associated total level of functional impairment per SDS score with other variables (F5, 423=42.442, p<0.0005; adjusted R2=0.326). Significant predictor variables are shown in Table 2.
|Predictor variable||Adjusted β||Significance p-Value||95% Confidence interval for B|
|Lower bound||Upper bound|
|ADHD Self-Rating Scale Score||0.097||0.029||0.014||0.264|
|Beck Depression Inventory Score||0.494||<0.0005||0.223||0.316|
|NIDA Modified ASSIST Total Cannabis Score||0.167||<0.0005||0.045||0.148|
In this study, we investigated the risk levels of SUDs, cannabis use, and functional impairment in a sample population of 429 youth aged 16–26 who presented for evaluation and treatment of mood and/or anxiety problems, and who were screened to exclude youth with primary SUDs. In spite of this screening, a small percentage (4.5%) was thought to have had a primary SUD after clinical assessment. A much larger percentage of the youth, approximately half, were found to be at either moderate or high risk for a SUD.
Our results showed that heavy cannabis use (a high ASSIST total cannabis score) was associated with being male, early first use of marijuana, high 30-day use of cigarettes, and high levels of functional impairment. Contrary to what might be expected, lower socioeconomic level, using the proxies of maternal educational level and parental marital status, was not predictive in this model. However, it should be noted that the socioeconomic status of this population appeared to be higher than average compared with the general population. The coping styles identified by ASRS and ERQ scores were also not predictive. This sample excluded subjects with ADHD diagnosed and treated in childhood, so participants were not representative of the full range of ASRS scores in a general clinical population. This may account for an absence of predictive significance for this variable, although the ASRS score was associated with functional impairment in the regression analysis with the same participants.
Anxiety or depression scores were not associated with total cannabis score, which was also counter to our predictions, although total depression score (but not anxiety) was associated with overall functional impairment.
Interestingly, age of first cigarette and alcohol use were not predictors of cannabis total score, but age of first cannabis use was, as would be expected. Age of onset of smoking tobacco did not predict total cannabis score, but 30-day tobacco use did. Although the rate of tobacco use found here was rather low compared with that of cannabis, our cohort nevertheless reported almost twice the rate of cigarette use of high school students in North America (12, 13). Previous reports have found higher rates of tobacco use in individuals with mental health symptoms, including non-psychotic symptoms (24–26). Our findings indicate that recent tobacco use is associated with higher scores of total cannabis use.
Our results lead to the conclusion that total cannabis use was most associated with being male, younger age of onset of cannabis use (but not cigarette smoking or alcohol use or age of first sexual intercourse), high frequency of recent cigarette (but not alcohol) use, and total degree of functional impairment.
This last association, with functional impairment, was perhaps our most clinically relevant variable. Investigated as the dependent variable as well, total level of functional impairment in our sample was associated with ASRS (measuring cognitive coping style), BDI score (measuring current depressive symptom severity), and the total cannabis score, but not associated with demographic variables, age of onset of risk behaviors, emotion regulation style, state anxiety, or recent substance use. These results confirmed our clinical impression that greater cannabis use was associated with a higher degree of dysfunction in our youth seeking help for mood and/or anxiety symptoms. We believe this to be the first report that has investigated cannabis use and functional impairment in youth with mood and/or anxiety disorders. Because cannabis is the most commonly used illicit substance of youth in North America (12, 13), this association is clinically important.
There has been a dramatic decrease in the use of tobacco products by youth since 1997, whereas rates of cannabis use have fluctuated around a largely unchanged rate of use (12, 13). The cigarette smoking decrease may be related to public health campaigns educating about the negative health effects of tobacco use (27). No similar public health campaign exists yet aimed at reducing cannabis use in youth.
The limitation of this study is the unknown representativeness of this clinical sample to youth with mood and/or anxiety disorders in general. Nonetheless, this research has suggested that approximately half of youth seeking help for mood and/or anxiety disorders screened to exclude those with a primary SUD are at risk for a SUD. It also showed that there is an association between cannabis use and functional impairment. Those at risk for cannabis use were more likely to be male, to have engaged in marijuana use at an earlier age, to have higher recent cigarette use, and to endorse greater functional impairment. This indicates a strong need for substance use specialists in the youth population of individuals with mood and/or anxiety complaints even when they do not appear to have a primary SUD. Including such services in primary care and specialized clinics treating youth with mental health concerns appears to be important for ensuring the best care to reduce symptom severity and negative outcomes associated with co-occurring mental illness and SUD, as well as improve treatment response.
Lubman DI, Allen NB, Rogers N, Cementon E, Bonomo Y. The impact of co-occurring mood and anxiety disorders among substance-abusing youth. J Affect Disord 2007;103: 105–12.
Boschloo L, Vogelzangs N, Smit JH, van den Brink W, Veltman DJ, et al. Comorbidity and risk indicators for alcohol use disorders among persons with anxiety and/or depressive disorders: findings from the Netherlands Study of Depression and Anxiety (NESDA). J Affect Disord 2011;131:233–42.
Liang W, Chikritzhs T, Lenton S. Affective disorders and anxiety disorders predict the risk of drug harmful use and dependence. Addiction 2011;106:1126–34. [Web of Science]
Baker KD, Lubman DI, Cosgrave EM, Killackey EJ, Yuen HP, et al. Impact of co-occurring substance use on 6 month outcomes for young people seeking mental health treatment. Aust NZ J Psychiatry 2007;41:896–902.
Seguin M, Renaud J, Lesage A, Robert M, Turecki G. Youth and young adult suicide: a study of life trajectory. J Psychiatr Res 2011;45:863–70. [Web of Science]
Price C, Hemmingsson T, Lewis G, Zammit S, Allebeck P. Cannabis and suicide: longitudinal study. Br J Psychiatry 2009;195:492–7.
Lagerberg TV, Sundet K, Aminoff SR, Berg AO, Ringen PA, et al. Excessive cannabis use is associated with earlier age at onset in bipolar disorder. Eur Arch Psychiatry Clin Neurosci 2011;261:397–405. [Web of Science]
Wittchen HU, Frohlich C, Behrendt S, Gunther A, Rehm J, 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. [Web of Science]
Henquet C, Krabbendam L, de Graaf R, ten Have M, van Os J. Cannabis use and expression of mania in the general population. J Affect Disord 2006;95:103–10.
Patton GC, Coffey C, Carlin JB, Degenhardt L, Lynskey M, et al. Cannabis use and mental health in young people: cohort study. Br Med J 2002;325:1195–8.
van Laar M, van Dorsselaer S, Monshouwer K, de Graaf R. Does cannabis use predict the first incidence of mood and anxiety disorders in the adult population? Addiction 2007;102:1251–60. [Web of Science]
Paglia-Boak A, Adlaf EM, Mann RE. Drug use among Ontario students, 1977–2011: OSDUHS highlights. 2011; CAMH research document series no. 33. Available from: http://www.camh.ca/en/research/news_and_publications/ontario-student-drug-use-and-health-survey/Documents/2011%20OSDUHS%20Docs/2011OSDUHS_Highlights_DrugUseReport.pdf. Accessed on 22 August, 2012.
US Department of Health and Human Services. Drug facts. Atlanta, GA: US Department of Health and Human Services, 2012. Available from: http://www.drugabuse.gov/publications/drugfacts/high-school-youth-trends. Accessed on 22 August, 2012.
DeGarmo DS, Forgatch MS, Martinez CR Jr. Parenting of divorced mothers as a link between social status and boys’ academic outcomes: unpacking the effects of socioeconomic status. Child Dev 1999;70:1231–45.
Beck AT, Steer RA, Brown GK. Manual for Beck Depression Inventory-II. San Antonio, TX: Psychological Corporation, 1996.
Spielberger CD, Gorsuch RL, Lushene RE. Manual for the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press, 1970.
Gross JJ, John OP. Individual differences in two emotion regulation preocesses: implications for affect, relationships, and well-being. J Pers Soc Psychol 2003;85:348–62.
Kessler RC, Adler L, Ames M, Demler O, Faraone S, et al. The World Health Organization Adult ADHD Self-Report Scale (ASRS): a short screening scale for use in the general population. Psychol Med 2005;35:245–56.
National Institute of Drug Abuse. Screening for drug use in general medical settings: resource guide. National Institute of Drug Abuse, 2009. Available at: http://www.drugabuse.gov/publications/resource-guide. Accessed on 15 April, 2012.
Sheehan DV. The anxiety disease. New York, NY: Charles Scribner, 1983.
Brener ND, Collins JL, Kann L, Warren CW, Williams BI. Reliability of the youth risk behavior survey questionnaire. Am J Epidemiol 1995;141:575–80.
Gross JJ. Emotion regulation: affective, cognitive, and social consequences. Psychophysiology 2002;39:281–91.
IBM. IBM SPSS statistics v. 20.0.0. Somers, NY: An IBM Company, 2011.
Lawrence D, Considine J, Mitrou F, Zubrick SR. Anxiety disorders and cigarette smoking: results from the Australian Survey of Mental Health and Wellbeing. Aust NZ J Psychiatry 2010;44:520–7.
Lawrence D, Mitrou F, Zubrick SR. Smoking and mental illness: results from population surveys in Australia and the United States. BMC Public Health 2009;9:285. [Web of Science]
Lawrence D, Mitrou F, Zubrick SR. Non-specific psychological distress, smoking status and smoking cessation: United States National Health Interview Survey 2005. BMC Public Health 2011;11:256. [Web of Science]
US Department of Health and Human Services. Preventing tobacco use among youth and young adults: a report of the Surgeon General. Atlanta, GA: US Department of Health and Human Services, 2012.