Content uploaded by George Mammen
Author content
All content in this area was uploaded by George Mammen on Jan 28, 2019
Content may be subject to copyright.
You are prohibited from making this PDF publicly available.
For reprints or permissions, contact permissions@psychiatrist.com. ♦ © 2018 Copyright Physicians Postgraduate Press, Inc.
It is illegal to post this copyrighted PDF on any website.
e1J Clin Psychiatry 79:4, July/August 2018
Review Article
Association of Cannabis With Long-Term Clinical
Symptoms in Anxiety and Mood Disorders:
A Systematic Review of Prospective Studies
George Mammen,PhDa,*; Sergio Rueda,PhDa; Michael Roerecke,PhDa;
Sarah Bonatob; Shaul Lev-Ran,MD,PhDc; and Jürgen Rehm,PhDa
ABSTRACT
Objective: To systematically review studies examining the longitudinal
associations between cannabis use and symptomatic outcomes among
individuals with an anxiety or mood disorder at baseline.
Data Sources: A search of the literature up to May 2017 was conducted
using several databases. Search terms related to the exposure (ie,
cannabis) and outcome (ie, symptoms) variables of interest. There were
no search restrictions.
Study Selection: In total, 10,191 citations were screened. Key inclusion
criteria related to (1) cohort-based longitudinal study design using
adults who met criteria for a mood or anxiety disorder at baseline, (2)
an independent variable focusing on at least baseline cannabis use, and
(3) a dependent variable focusing on the symptomatic course and/or
outcomes in anxiety and mood disorders (AMD).
Data Extraction: We followed the Preferred Reporting Items for
Systematic Reviews and Meta-Analyses guidelines. Methodological
characteristics and key findings were extracted from each study, and
quality assessments were conducted for each study.
Results: Twelve studies (with a total of 11,959 individuals) met inclusion
criteria related to posttraumatic stress disorder (n = 4), panic disorder
(n = 1), bipolar disorder (n = 5), and depressive disorder (n = 2). Across 11
studies, “recent” cannabis use (ie, any/greater frequency of use during
the last 6 months) was associated with higher symptomatic levels over
time relative to comparison groups (ie, no/lesser frequency of use). Ten
of these studies further suggested that cannabis use was associated
with less symptomatic improvement from treatment (eg, medication,
psychotherapy for AMD).
Conclusions: Recent cannabis use was associated with negative long-
term symptomatic and treatment outcomes across AMD. The findings
should be interpreted with caution, considering the observational
designs across studies and the biases associated with the samples (eg,
inpatients) and sources of cannabis consumed (ie, unregulated sources).
Nonetheless, clinicians can use the insight gained to inform their own
and their patients’ knowledge concerning potential risks of cannabis
with regard to symptoms of AMD.
J Clin Psychiatry 2018;79(4):17r11839
To c ite : Mammen G, Rueda S, Roerecke M, et al. Association of cannabis with
long-term clinical symptoms in anxiety and mood disorders: a systematic review
of prospec tive studies. J Clin Psychiatr y. 2018;79(4):17r11839.
To share: https://doi.org/10.4088/JCP.17r11839
© Copyright 2018 Physicians Postgraduate Press, Inc.
aInstitute for M ental Health Policy Research, Centre for Addiction and Mental
Health, Toronto, Ontario, Canada
bLibrary Services, Centre for Addic tion and Mental Health, Toronto, Ont ario,
Canada
cAddiction M edicine and Dual Disorders Clinic, Lev Hasharon Medic al Center,
Pardesya, Israel
*Corresponding author: George Mammen, PhD, Institute for Mental Health Policy
Research, Centre for Addiction and Mental Health, 33 Russell St, Toronto, ON,
Canada, M5S 2W6 (george.mammen@mail.utoronto.ca).
Cannabis is commonly used among individuals
living with anxiety and mood disorders (AMD),
which are the most prevalent mental health conditions
globally.1 For instance, those diagnosed with generalized
anxiety disorder, posttraumatic stress disorder (PTSD),
bipolar disorder, and depressive disorders have higher
rates of lifetime and recent use (eg, past year and month)
than individuals without such psychiatric conditions,2–4
with around 20%–30% of users consuming cannabis
dai l y. 5–9 This frequent use has been discussed in the
context of self-medication. In line with this, AMD
lead mental health conditions in which cannabis is
used for therapeutic purposes,10–12 as users report that
cannabis relieves acute symptoms in PTSD (eg, reduces
nightmares), bipolar disorder (eg, stabilizes mood), and
depressive disorders (eg, increases motivation).13–18
However, whether cannabis positively or negatively
influences long-term symptoms in AMD is highly
debated and is an understudied area of research.19–22
The majority of longitudinal studies examining
relationships between cannabis and psychiatric
disorders have focused on the general population and
the incidence of developing mental health conditions
as predicted by cannabis use. Systematic reviews show
that higher frequencies of use may increase risk in the
onset of anxiety, depressive, and bipolar disorders,23–26
in addition to schizophrenia and psychoses.27 On the
basis of this literature, mental health experts theorize
that cannabis use most likely does not benefit long-
term symptoms but rather exacerbates the course of
illness.19,28
To the authors’ knowledge, no systematic review
has focused on the longitudinal associations between
cannabis use and AMD in a clinical population.
Wilkinson and colleagues,29 in the Journal of Clinical
Psychiatry, published a systematic review to determine
the efficacy of cannabis in psychiatric indications.
However, their review did not focus on AMD, as they
examined PTSD, in addition to Alzheimer’s disease and
Tourette’s syndrome. Narrowing the focus to AMD is
warranted, particularly given the high prevalence of
cannabis consumption in PTSD and other anxiety
(eg, social anxiety disorder) and mood disorders (eg,
major depressive disorder).2–9 Pursuing this aim can
help address if cannabis is associated with negative or
positive symptomatic outcomes, which clinicians can
You are prohibited from making this PDF publicly available.
For reprints or permissions, contact permissions@psychiatrist.com. ♦ © 2018 Copyright Physicians Postgraduate Press, Inc.
It is illegal to post this copyrighted PDF on any website.
e2 J Clin Psychiatry 79:4, July/August 2018
Mammen et al
Table 1. Outline of Search Terms Used in the Study Search Strategy
Concept MeSH TermsaKeywords
Independent variable: cannabis cannabis
cannabis addiction
medical cannabis
marijuana smoking
marijuana abuse
medical marijuana
marijuana usage
cannabis*
marihuana*
marijuana*
hash
hashish
Dependent variable: anxiety and mood disorders anxiety
anxiety disorders
trauma/stressor related disorder
emotional trauma
mood disorders
adjustment disorder
depression
bipolar and related disorders
affective disorders
anxiety, anxious, phobia, phobic, panic disorder, panic attack
traumatic disorder*, post*, trauma*, stress, combat neuroses,
war neurosis, combat disorder*, combat neuroses, shellshock*,
psychological trauma, mood disorder*, mood disruptive
dysregulation, affective disorder*, seasonal affective disorder,
depress, dysthym*, premenstr* dysphoric disorder*, bipolar,
manic, mania
Association association
correlation
risk factors
relation*
causal*
aMeSH (Medical Subject Heading) search terms are the controlled vocabulary used by the US National Library of Medicine to index articles for MEDLINE/
PubMed. MeSH terminology provides a consistent way to retrieve information that may use different terminology for the same concepts; an asterisk (*)
denotes that the search term will look for any word with the given letter combination plus any combination of letters following the original combination.
use as evidence to inform patients regarding potential long-
term influence (eg, benefits, risks) of cannabis on AMD.
METHODS
To conduct this systematic review, we followed the
Preferred Reporting Items for Systematic Reviews and
Meta-Analyses (PRISMA) guidelines. The study protocol
was registered with PROSPERO (registration number:
CRD42017037733).
Databases and Search Terms
The following electronic databases were used for the
literature search: Embase, MEDLINE, MEDLINE in Process
and Other Non-Indexed Citations, MEDLINE Epub Ahead
of Print, and PsycInfo. Databases including gray literature (ie,
conference papers) were searched up until May 2017. There
were no year, language, or study type restrictions. Search
terms related to the exposure (ie, regulated and unregulated
cannabis) and outcome (ie, AMD) variables of interest
(Table 1). The search process was led by a professional health
science librarian (S.B.).
Screening Process and Study Eligibility Criteria
The study screening and retrieval process was conducted
in duplicate by 4 trained reviewers and documented by
DistillerSR.30 Reference lists of relevant studies and literature
reviews, in addition to “related articles” in electronic
databases, were further examined. Meeting the inclusion
criteria meant that the study (1) employed a cohort-based
longitudinal design; (2) focused on adults (ie, 18+ years
of age) meeting criteria for a mood or anxiety disorder
at baseline (without comorbidities related to physical
illness, schizophrenia, or psychoses), as determined by
either clinician interviews or screening instruments with
established cutoff thresholds; (3) assessed symptomatic
course (operationalized as using multiple follow-up
assessments in analysis) and/or symptomatic outcome
(operationalized as using only 1 follow-up measure) as the
dependent variable; (4) assessed at least baseline cannabis
use as the independent variable (isolated cannabis without
polysubstance use); and (5) included at least 1 comparison/
control group.
Data Extraction and Quality Assessments
Methodological characteristics and key findings were
extracted from each study. Authors of eligible studies were
contacted to provide missing or additional data. To assess
study quality, we used the Newcastle-Ottawa Quality
Assessment Scale for Cohort Studies,31 which examines
quality via indicators related to sample recruitment, group
comparability, and ascertainment of the exposure and
outcome variables of interest. Two appraisers performed
these assessments independently. Interrater agreement was
assessed using the κ statistic (κ = 0.79), and any discrepancies
were resolved with the coauthors of this review.
RESULTS
Retrieved Results
The search process yielded 16,251 citations. After
deduplication, 10,191 titles/abstracts were screened, resulting
Clinical Points
■ Whether cannabis is associated with long-term positive
or negative symptomatic outcomes in anxiety and mood
disorders is an understudied and controversial area of
research.
■ Clinicians can use the evidence presented in this review
to help inform their own and their patients’ knowledge
concerning potential risks of cannabis on symptom and
treatment outcomes in anxiety and mood disorders.
You are prohibited from making this PDF publicly available.
For reprints or permissions, contact permissions@psychiatrist.com. ♦ © 2018 Copyright Physicians Postgraduate Press, Inc.
It is illegal to post this copyrighted PDF on any website.
e3J Clin Psychiatry 79:4, July/August 2018
Cannabis and Anxiety and Mood Disorders
Figure 1. PRISMA Flow Diagram
Abbreviation: PTSD = posttraumatic stress disorder.
10,191
Records screened
10,048
Records excluded as
study not pertinent
to this topic or not
longitudinal by
design
Additional records
identied through
other sources
8
16,251
Records identied
through database
searching
10,191
Records after
duplicates removed
PTSD
Panic disorder
Bipolar disorder
Depressive disorder
4
1
5
2
Studies included in
systematic review
12
Full-text articles
excluded:
Not clinical sample at baseline
Not adult population at baseline
Exposure variable not
cannabis specic
Outcome variable not specic
to anxiety/mood symptoms
131
23
26
27
36
No control group
Comorbid sample
with physical illness
11
8
Full-text articles
assessed for
eligibility
143
Included
Eligibility
Screening
Identication
in 12 included studies. Figure 1 displays the flow diagram
and reasoning for study exclusions. Five anxiety disorder–
based studies (PTSD, n = 4, panic disorder, n = 1) and 7 mood
disorder–based studies (bipolar disorder, n = 5; depressive
disorder, n = 2) were used in the synthesis, using data from
a total sample of 11,959 individuals (2,588 individuals “more
exposed” to cannabis, operationalized as any recent use or
greater use; 9,371 individuals “less exposed” to cannabis,
operationalized as no recent use or lesser use).
Study Characteristics and Quality
Tables 2 and 3 highlight methodological characteristics
and findings, respectively, of each study. The majority of
studies were based in the United States (n = 8), followed
by Europe (n = 3) and Australia (n = 1). Study samples
ranged from 62 to 22,948 individuals who qualified for an
anxiety or mood disorder at baseline and were psychiatric-
based patients (majority inpatients) receiving symptomatic
treatment. The baseline age ranged from 18 to 65+ years,
with the baseline mean age across the cohorts being close
to 43 years. There was fairly even representation of genders
across the bipolar, panic, and depressive disorder studies.
Among the 4 PTSD studies, 3 were heavily focused on males
(93%–100%). In terms of study design, 9 were observational
cohort studies, and 3 were based from secondary analyses of
interventions aimed at treating the anxiety or mood disorder.
Follow-up periods ranged from 2.5 months to 5 years.
Cannabis use was assessed subjectively across studies,
primarily using self-report measures via clinical interviews.
All studies focused on frequency of recent use during the
past 7 days, 1 month, 3 months, or 6 months. Six studies used
only baseline assessments of cannabis use, and 6 studies used
multiple assessments to help determine how changes in use
(eg, reducing it) affect symptomatic outcomes. In terms of
source of cannabis, all studies examined use of illicit “street
cannabis” (ie, unregulated cannabis) in relation to outcomes
of AMD.
The majority of studies (n = 10) used a version of the
DSM (eg, -III, -IV) for diagnosing AMD. In terms of the
dependent variable, various measures of symptoms were
used across the AMD (see Table 2), though all but 1 study
(ie, number of symptoms)32 used a scale of symptom severity.
Two studies used only 1 follow-up outcome of symptoms, 7
studies used multiple follow-up measures of symptoms to
You are prohibited from making this PDF publicly available.
For reprints or permissions, contact permissions@psychiatrist.com. ♦ © 2018 Copyright Physicians Postgraduate Press, Inc.
It is illegal to post this copyrighted PDF on any website.
e4 J Clin Psychiatry 79:4, July/August 2018
Mammen et al
Table 2. Study Characteristics
First Author,
Country Sample (% Male)
Cannabis Use
Measure
Diagnosis and Symptom
Measure(s)
Design, Follow-Up,
Longitudinal Analysis Accounted Confounder Variables
Treatment Status for Anxiety
and Mood Disorder
PTSD
Bonn-Miller 2013,33
United States
N: 260 (100%)
Baseline age:
mean = 52.6 y
Inpatients
Interview assessing
current cannabis
abuse and
dependence
(using DSM-IV)
Diagnosis: DSM-IV
PTSD symptoms: PTSD
Checklist-Military Version
(17-item)
Design: observational
Follow-up: 2.5 months
Analysis: hierarchical linear
regression
Baseline symptoms/age/trauma/
psychological distress/polysubstance use
Patients enrolled in Veterans
Affairs residential rehabilitation
program. Medication status
unreported
Manhapra 2015,34
United States
N: 22,948 (93.2%)
Baseline age:
mean = 51.2 y
Inpatients/
outpatients
Interview assessing
days of use in
the past 1 month
(ASI)
Diagnosis: DSM-IV
PTSD symptoms: Short Form
Mississippi Scale for Combat-
Related PTSD (11-item)
Design: observational
Follow-up: 4 months after program
discharge
Analysis: analysis of covariance
Baseline symptoms/age/marital status/race/
other psychiatric issues/employment/car
ownership/sexual trauma in military
Patients enrolled in specialized
intensive PTSD program.
Medication status unreported
Ruglass 2017,35
United States
N: 136 (47.7%)
Baseline age:
mean = 42.8 y
Outpatients
Interview assessing
days of use in
the past 7 days
(Substance Use
Inventory)
Diagnosis: DSM-IV
PTSD symptoms: Clinician-
Administered PTSD scale
Design: intervention (secondary
analysis)
Follow-up: 3 months
Analysis: multivariate regression
analyses with bootstrapping
Baseline symptoms/age/sex/days of
substance use
Patients enrolled in an
intervention to reduce
PTSD (via medications
including sertraline and
riboflavin + psychotherapy)
Wilkinson 2015,36
United States
N: 2,276 (96.7%)
Baseline age:
mean = 51.7 y
Inpatients
Interview assessing
days of use in
the past 1 month
(using ASI)
Diagnosis: DSM-III/IV
PTSD symptoms: Short Form
Mississippi Scale for Combat-
Related PTSD (11-item)
Design: observational
Follow-up: 4–6 months
Analysis: analysis of covariance and
linear multiple regression
Baseline symptoms/age/marital status/race/
incarceration history/psychosis/chronic
medical problems/war zone service/length
of stay/expulsion from treatment/drug and
alcohol abuse/employment
Patients enrolled in specialized
Veterans Affairs treatment
programs. Medication status
unreported
Panic Disorder
Bricker 2007,37
United States
N: 232 (39.4%)
Baseline age:
mean = 40.4 y
Outpatients
Interview assessing
frequency of
use in the past 1
month
Diagnosis: DSM-IV
Core panic: Anxiety Sensitivity
Index
Social phobia: Social Phobia
subscale of the Fear
Questionnaire
Depression: CES-D
Design: Intervention (secondary
analysis)
Follow-up: 1 year
Analysis: random coefficient
hierarchical models and mixed-
effects linear regressions
Education/social phobia Participants enrolled in an
intervention to reduce
anxiety/panic (via
unspecified antianxiety
medications + psychotherapy)
Bipolar Disorder
Kim 2015,38
Australia
N: 213 (40.6%)
Baseline age:
mean = 41.8 y
Inpatient/
outpatient
Interview assessing
frequency of
use in the past 3
months
Diagnosis: Mini-International
Neuropsychiatric Interview
version 5
Mania: YMRS
Depression: HDRS-21
Design: observational
Follow-up: 2 years
Analysis: Kruskal-Wallis test, Mann-
Whitney U test
None reported On medication (lithium
carbonate, sodium valproate,
carbamazepine, olanzapine)
Kvitland 2015,39
Norway
N: 62 (40%)
Baseline age:
mean = 30.9 y
Inpatient/
outpatient
Interview assessing
frequency of
use in the past 6
months
Diagnosis: DSM-IV
Mania: YMRS
Depression: Inventory of
Depressive Symptoms—
Clinician Rated
Design: observational
Follow-up: 1 year
Analysis: hierarchical blockwise
multiple linear regression
Baseline symptoms/age/sex/premorbid
functioning
On medication (unspecified)
Strakowski 2007,40
United States
N: 144 (56.4%)
Baseline age:
mean = 21.6 y
Inpatients
Interview assessing
frequency of
use in the past
1 month (using
ASI)
Diagnosis: DSM-IV and
YMRS > 20
Mania: YMRS
Depression: HDRS-17
Rapid cycling: > 4 episodes
within 52 weeks
Design: observational
Follow-up: mean = 2.6 years (5 years
max)
Analysis: survival analysis and Cox
proportional hazards regression
model
Baseline symptoms/age/sex/education/
presence of psychosis/alcohol and drug
use history
On medication and hospitalized
for mania (putative mood
stabilizers including lithium,
divalproex, carbamazepine,
lamotrigine, topiramate,
atypical antipsychotic drugs)
(continued)
You are prohibited from making this PDF publicly available.
For reprints or permissions, contact permissions@psychiatrist.com. ♦ © 2018 Copyright Physicians Postgraduate Press, Inc.
It is illegal to post this copyrighted PDF on any website.
e5J Clin Psychiatry 79:4, July/August 2018
Cannabis and Anxiety and Mood Disorders
Table 2 (continued).
First Author,
Country Sample (% Male)
Cannabis Use
Measure
Diagnosis and Symptom
Measure(s)
Design, Follow-Up,
Longitudinal Analysis Accounted Confounder Variables
Treatment Status for Anxiety
and Mood Disorder
van Rossum 2009,41
14 European
countries
N: 3,426 (44.6%)
Baseline age:
mean = 44.6 y
Inpatient/
outpatient
Interview assessing
frequency of
use in the past 3
months
Diagnosis: DSM-IV, ICD-10,
clinical judgment for manic/
mixed episode; CGI-BP
for overall illness, mania,
depression
Design: observational
Follow-up: 1 year
Analysis: multilevel random
regression
Baseline symptoms/age/sex/medication
compliance/age of mood onset/alcohol
use/drug use
On medication (antipsychotics,
anticonvulsants, lithium)
Zorrilla 2015,42
14 European
countries
N: 1,922 (58.8%)
Baseline age:
mean = 39.1 y
Inpatient/
outpatient
Interview assessing
frequency of
use in the past 3
months
Diagnosis: DSM-IV, ICD-10,
clinical judgment for manic/
mixed episode
Mania: CGI-BP
Design: observational
Follow-up: 2 years
Analysis: logistic regression
Alcohol/polydrug use On medication (olanzapine
antipsychotics, mood
stabilizers)
Depressive Disorder
Bahorik 2017,43
United States
N: 307 (100%)
Baseline age:
mean = 52.6 y
Outpatients
Interview assessing
days of use in the
past 1 month
Diagnosis: PHQ-9 (> 5 mild
depression)
Depression: PHQ-9
Anxiety: GAD-7
Design: intervention (secondary
analysis)
Follow-up: 6 months
Analysis: hierarchical linear
modeling (mixed-effects growth
models)
Age/sex/marital status/race/polysubstance
use/time-varying psychiatry visits
Unreported
Feingold 2017,32
United States
N: 2,348 (41.3%)
Baseline age:
18–65+ y
Unspecified
patient status
Interview assessing
any use and
cannabis use
disorder using
AUDADIS-IV
Diagnoses: DSM-IV (5 of
9 symptoms for major
depressive disorder)
Depression: DSM-IV
Design: observational
Follow-up: 3 years
Analysis: linear regression analysis
and multivariate logistic
regression analyses
Baseline symptoms/age/sex/race/education/
income/marital status/age/region
Unreported
Abbreviations: ASI = Addiction Severity Index, AUDADIS = Alcohol Use Disorder and Associated Disabilities Interview Schedule, CES-D = Center for Epidemiologic Studies Depression scale, CGI-BP = Clinical Global
Impression Bipolar Disorder Scale, DSM = Diagnostic and Statistical Manual of Mental Disorders, GAD-7 = Generalized Anxiety Disorder 7 scale, HDRS = Hamilton Depression Rating Scale, ICD = International Classification of
Diseases, PHQ = Patient Health Questionnaire, PTSD = posttraumatic stress disorder, YMRS = Young Mania Rating Scale.
help determine how cannabis affects the course
of symptoms, and 3 studies measured both course
and outcomes of symptoms.
In terms of study quality (Table 4), out
of a possible 8 stars, 3 studies were rated
7 stars, 6 studies were rated 6 stars, and 3
studies were rated 5 stars, suggesting that the
majority of studies can be considered of good
overall methodological rigor. Specifically, all
studies received maximum stars for cohort
representativeness and assessment of exposure,
and most studies received maximum stars for
the following: controlling for baseline symptoms
(n = 9) and multiple additional confounding
variables in final statistical models (n = 10) (eg,
age, sex, comorbidities, polysubstance use);
conducting follow-up assessments 1 year after
baseline (n = 7); and having high retention rates
(> 80%; n = 10). Detailed scoring of each study is
available upon request.
Cannabis and PTSD or Panic Disorder
Four studies examined the association of
cannabis with long-term symptoms in PTSD.33–36
Three studies collectively showed that recent
cannabis use (eg, past month) was associated with
a negative course33,34 and negative outcomes36 of
PTSD symptom severity, while 1 study found
no significant relationships (note that this study
included only 32 individuals in the “exposure”
group).35 Among the significant findings, 2 studies
found that “any” level of baseline cannabis use
(F = 81.83, P < .0001)34 or sustained use over time
(F = 21.47, P < .01)36 was associated with greater
PTSD symptom severity 4 months following
baseline assessments, compared to abstinence.
These studies also supported that stopping use
is associated with less severe symptoms (Cohen
d = −0.18)36 and greater improvements in
symptoms from treatment (Cohen d = −0.61)34
compared to continuing or starting cannabis
use. Bonn-Miller and colleagues33 revealed that
a baseline current cannabis use disorder (CUD)
diagnosis was associated with less improvement
from treatment regarding PTSD symptoms
(β = –0.14, P < .05) over the course of 2.5 months,
relative to the comparison group (ie, no CUD),
specifically concerning avoidance numbing
(β = –0.13, P < .05) and hyperarousal (β = –0.13,
P < .05) symptoms. In these 3 studies finding
significance, it is worth noting that all participants
were predominantly men who were enrolled
in a Veterans Affairs residential rehabilitation
program to help treat their combat-related PTSD.
Hence, the findings may not be generalizable to
females experiencing non–combat-related PTSD,
for instance.
You are prohibited from making this PDF publicly available.
For reprints or permissions, contact permissions@psychiatrist.com. ♦ © 2018 Copyright Physicians Postgraduate Press, Inc.
It is illegal to post this copyrighted PDF on any website.
e6 J Clin Psychiatry 79:4, July/August 2018
Mammen et al
Table 3. Study Findings
First Author
Exposure Variables—
Cannabis Use
Dependent Variables—
Symptom Change/Outcome Key Findings
Association of Cannabis
With Symptoms
PTSD
Bonn-Miller 201333 CUD (n = 81): current CUD diagnosis
No CUD diagnosis (n = 179): no current CUD diagnosis
Used 1 assessment (at baseline)
Course of symptom severity for
PTSD
Used multiple assessments
CUD associated with higher symptom severity/less improvement in
PTSD symptoms (β = –0.14, P < .05) including avoidance numbing
(β = –0.13, P < .05) and hyperarousal (β = –0.13, P < .05) symptoms
relative to comparison group. No CUD associated with greater
improvements in PTSD
Cannabis use associated
with higher symptom
severity
Manhapra 201534 Not abstaining (n = 353): any level of cannabis use in past
1 month
Abstaining (n = 270): stopping use in past 1 month
Used 1 assessment (at baseline)
Course of symptom severity for
PTSD
Used multiple assessments
Abstaining from cannabis associated with lower PTSD symptom
severity and greater improvements in symptoms (Cohen d = −0.61;
F = 81.83, P < .0001) relative to comparison group
Cannabis use associated
with higher symptom
severity
Ruglass 201735 Cannabis use (n = 32): any level of cannabis use in past 7
days
No use (n = 104): no use in past 7 days
Used multiple assessments
Course and outcome of
symptom severity for PTSD
Used multiple assessments
Cannabis use not associated with PTSD symptom severity No association
Wilkinson 201536 Continued use (n = 268): continuing any level of cannabis
use in past 1 month
Starting use (n = 738): starting any level of cannabis use in
past 1 month
No use (n = 767): no cannabis use in past 1 month
Stopping use (n = 263): stopping any level of cannabis use
in past 1 month
Used multiple assessments
Outcome of symptom severity
for PTSD
Used 1 assessment (at follow-
up)
Starting and continuing use associated with higher PTSD symptom
severity (F = 21.47, P < .01) relative to no use and stopping use.
Starting use associated with more severe symptoms (Cohen
d = 0.34) relative to never use. Stopping use associated with less
severe symptoms (Cohen d = −0.18) relative to never use. Increase
in days of use associated with increase in symptom severity
(β = 0.17, t = 4.08, P < .0001)
Cannabis use associated
with higher symptom
severity
Panic Disorder
Bricker 200737 Monthly use (n = 29): use of cannabis at least once in the
past 1 month (but no more than once/wk)
Less than monthly use (n = 203): no use in past 1 month
Used 1 assessment (at baseline)
Course of symptom severity
for anxiety and depression
symptoms, outcome of
symptom severity for social
phobia
Used multiple assessments
for symptom severity and 1
assessment (at follow-up) for
social phobia outcome
Monthly use associated with higher severity of depression (mean
CES-D adjusted score = 28.54; 95% CI, 24.20–32.87) relative to
comparison group (mean = 21.73; 95% CI, 19.92–23.55). No
associations found for anxiety or social phobia
Cannabis use associated
with higher symptom
severity for depression
(but not anxiety)
Bipolar Disorder
Kim 201538 Regular use (n = 25): use of cannabis > 3 days/wk
Nonregular use (n = 209): no use or use < 3 days/wk
1 assessment used (at baseline)
Course of symptomatic
remission for mania (< 12 on
YMRS) and depression (< 8 on
HDRS-21)
Used multiple assessments
Regular use associated with less likelihood of total remission (P = .025)
relative to comparison group. Female users had lower remission
rates for depression only (P = .002); male users had lower remission
rates for mania only (P = .019)
Cannabis use associated
with less symptomatic
remission rates
Kvitland 201539 Continued use (n = 6): use of cannabis 2–3 times/wk
No continued use (n = 56): no use or stopping/starting use
Multiple assessments used
Outcome of symptom severity
for mania and depression
Used 1 assessment (at
follow-up)
Continued use associated with higher mania symptom severity
(β = 0.360, t = 2.985, P = .004) relative to comparison group. No
associations found for depression
Cannabis use associated
with higher symptom
severity
(continued)
You are prohibited from making this PDF publicly available.
For reprints or permissions, contact permissions@psychiatrist.com. ♦ © 2018 Copyright Physicians Postgraduate Press, Inc.
It is illegal to post this copyrighted PDF on any website.
e7J Clin Psychiatry 79:4, July/August 2018
Cannabis and Anxiety and Mood Disorders
Table 3 (continued).
First Author
Exposure Variables—
Cannabis Use
Dependent Variables—
Symptom Change/Outcome Key Findings
Association of Cannabis
With Symptoms
Strakowski 200740 CUD i: bipolar disorder preceded CUD (n = 36)
CUD ii: CUD preceded bipolar disorder (n = 33)
No CUD (n = 75): no history of CUD
Used 1 assessment (at baseline)
Course of symptomatic recovery
(> 1 and < 5 on YMRS; < 7 on
HDRS-17), recurrence (> 5 on
YMRS; > 7 on HDRS-17), and
cycling (> 4 affective episodes
within 52 weeks) for mania
and depression
Used multiple assessments
CUD i associated with more weeks in affective episode (F2,134 = 5.9;
P = .004), manic episodes (F2,134 = 2.8; P = .06), and mixed episodes
(F2,134 = 3.8; P = .03) relative to no CUD. CUD i and ii associated with
more rapid cycling (χ21 = 4.3; P = .04) relative to no CUD
Cannabis use associated
with more symptomatic
recurrence and cycling
van Rossum 200941 Any use (n = 436): any level of current cannabis use
No use (n = 2,990): no use
Used multiple assessments
Course of symptom severity for
mania, depression
Used multiple assessments
Any use associated with higher severity of overall symptoms (β = 0.13;
CI, 0.04–0.22; P = .004) and mania symptoms (β = 0.15; CI, 0.06–0.24;
P = .001) relative to comparison group; no associations found for
depression
Cannabis use associated
with higher symptom
severity
Zorrilla 201542 Current use (n = 132): any level of cannabis use
Previous use (n = 89): stopping use
Never use (n = 1,701)
Used multiple assessments
Course of symptomatic
remission (< 3 on CGI-BP
Mania) and recurrence (> 4 on
CGI-BP Mania) for mania
Used multiple assessments
Current use associated with higher recurrence rates of mania
(OR = 1.59, P = .048) and quicker time to recurrence (OR = 1.47,
P = .034) relative to never use. Previous use not associated with any
negative symptom outcomes
Cannabis use associated
with less symptomatic
remission and more
symptomatic recurrence
Depressive Disorder
Bahorik 201743 Cannabis use (n = 125): any level of cannabis use during
past 1 month
No use (n = 182): no cannabis use during past 1 month
Used multiple assessments
Course of symptom severity for
depression and anxiety
Used multiple assessments
Cannabis use associated with higher severity/less improvement of
depressive (β = 1.24; CI, 0.466–2.015; P < .001) and anxiety (β = 0.80;
CI, 0.101–1.509; P < .001) symptoms relative to comparison group
Cannabis use associated
with higher symptom
severity
Feingold 201732 Any use, no CUD (n = 173): any level of cannabis use
(without CUD diagnoses)
CUD (n = 121): diagnosed with CUD
No use (n = 2,283): not using cannabis
Used 1 assessment (for between Waves 1 and 2)
Outcome of number of
depressive symptoms and
course of symptomatic
recurrence (depression at
Waves 1 and 2 using CES-D)
and remission (depression at
Wave 1, but not Wave 2) for
depression
Used 1 assessment (at
follow-up) for symptom
severity proxy and multiple
assessments for recurrence/
remission
CUD associated with a greater number of depressive symptoms
(proxy for severity; β = 0.62, P = .0019) relative to no use. CUD
associated with anhedonia (OR = 2.62; CI, 1.36–5.08; P = .0048) and
insomnia/hypersomnia (OR = 2.30; CI, 1.29–4.12; P = .0055) relative
to no use. No association found between any use, no CUD, and
rates of remission
Cannabis use associated
with higher symptom
severity
Abbreviations: CES-D = Center for Epidemiologic Studies Depression scale, CGI-BP = Clinical Global Impression Bipolar Disorder Scale, CUD = cannabis use disorder, HDRS = Hamilton Depression Rating Scale,
PTSD = posttraumatic stress disorder, YMRS = Young Mania Rating Scale.
You are prohibited from making this PDF publicly available.
For reprints or permissions, contact permissions@psychiatrist.com. ♦ © 2018 Copyright Physicians Postgraduate Press, Inc.
It is illegal to post this copyrighted PDF on any website.
e8 J Clin Psychiatry 79:4, July/August 2018
Mammen et al
Table 4. Study Quality Assessment Scoresa
Study Selection (3) Comparability (2) Outcome (3) Total Score (8)
PTSD
Bonn-Miller 201333
★★★ ★★
★6
Manhapra 201534
★★★ ★★
★6
Ruglass 201735
★★★ ★★
5
Wilkinson 201536
★★★ ★★
★6
Panic Disorder
Bricker 200737
★★★
★ ★★ 6
Bipolar Disorder
Kim 201538
★★★
★★ 5
Kvitland 201539
★★★ ★★
★6
Strakowski 200740
★★★ ★★
★★ 7
van Rossum 200941
★★★ ★★
★★ 7
Zorrilla 201442
★★★
★ ★★ 6
Depression
Bahorik 201743
★★★
★ ★ 5
Feingold 201732
★★★
★★ ★★ 7
aEach study can be awarded a maximum of 8 stars: 3 stars for its “selection” of the
exposed cohort (1 star if cohort is representative) and nonexposed cohorts (1 star
if cohort is drawn from same community as exposed cohort), in addition to its
selection of the exposure assessment (1 star if exposure assessed via secure record or
structured interview); 2 stars for “comparability” of cohorts based on accounting for
the most important factor (1 star if study controls for baseline symptomatic scores)
and additional factors (1 star if study accounts for age, gender, and comorbidities,
polysubstance use); and 3 stars for “outcome” criteria concerning assessment of
outcome (1 star for independent blind assessment or record linkage), length of
follow-up (1 star for > 6 months follow-up), and adequacy of follow-up (1 star for
retention rate > 80%).
Abbreviation: PTSD = posttraumatic stress disorder.
One study examined the association of cannabis with symptoms
in panic disorder, using secondary data from a trial aimed to reduce
symptoms via medication and psychotherapy. Bricker et al37 showed that,
in the intervention group, there were no differences between monthly
users and less than monthly users in terms of symptomatic outcomes
over 1 year, suggesting that cannabis use did not impede recovery.
Within the control group not receiving the intervention, however,
monthly cannabis use (Center for Epidemiologic Studies Depression
Scale [CES-D] adjusted mean = 28.54, P < .01) was associated with
higher levels of depressive symptoms over 1 year compared to less than
monthly use (CES-D adjusted mean = 21.73, P < .01). No associations
were found for core panic symptoms or social phobia symptoms. The
limited findings in this study could be due to the small sample of
monthly users (ie, 29 individuals) and the focus on lower frequencies
of use (ie, using cannabis at least once in the past month but no more
than once per week was operationalized as monthly use).
Cannabis and Bipolar Disorder
Five studies examined the association of cannabis with long-term
symptoms in bipolar disorder.38–42 Each study provided indication
that recent cannabis use (eg, past month) was associated with negative
symptomatic outcomes. For instance, 2 studies revealed that “any recent
use” of cannabis (eg, 2–3 times per week) over 1 year was associated
with greater symptom severity for mania throughout the course of 1
year (β = 0.15, P = .001)41 and at 1-year follow-up39 compared to no use
or stopping use. Both studies found no significant associations related
to depressive symptoms, likely attributed, as the authors note, to the
focus on patients with more severe baseline mania than depression.
The remaining 3 studies38,40,42 collectively showed that cannabis use
was associated with greater symptom severity, as measured through
thresholds concerning symptomatic remission, recovery, and recurrence
for mania and depressive symptoms. Specifically,
Kim and colleagues38 showed that baseline
“regular” use (ie, > 3 days/week) was associated
with less occurrence of remission for mania and
depressive symptoms throughout the course of 2
years compared to nonregular use (ie, < 3 days/
week). Aligned with this finding, Zorrilla et
al42 showed that continued use over 1 year (ie,
any use) was associated with higher recurrence
rates of mania (OR = 1.59, P = .048) and quicker
time to recurrence (OR = 1.47, P = .034) over
the course of 2 years, relative to never use.
Zorrilla and colleagues’ study further suggested
that stopping cannabis use was associated with
more favorable symptomatic outcomes over
time. Lastly, Strakowski and colleagues40 found
that a baseline CUD diagnosis was associated
with more weeks spent in an affective episode
(F2,134 = 5.9; P = .004), more time in mixed
episodes (F2,134 = 3.8; P = .03), and more rapid
cycling between episodes over the course of up to
5 years compared to having no CUD diagnosis.
All bipolar disorder studies were conducted in
the context of symptomatic treatment including
medication and inpatient care, which revealed
that cannabis use was associated with less
improvement from treatment.
Cannabis and Depressive Disorder
Two studies examined the association of
cannabis with long-term symptoms in depressive
disorders. Both studies revealed that cannabis use
was associated with higher symptom severity43
and the number of symptoms.32 Specifically,
Feingold and colleagues32 showed that a CUD
diagnosis between baseline and follow-up was
associated with a greater number of depressive
symptoms (β = 0.62, P = .0019), specifically
related to anhedonia (OR = 2.62; P = .0048) and
insomnia/hypersomnia (OR = 2.30 P = .0055) at
3-year follow-up relative to no use. Bahorik and
colleagues43 showed that continued cannabis
use was associated with greater symptom
severity and less symptomatic improvement
for depressive (β = 1.24, P < .001) and anxiety
symptoms (β = 0.80, P < .001) over the course of
6 months relative to no use. This study was in the
context of cognitive behavior therapy, suggesting
that cannabis use may potentially interfere with
psychotherapy effectiveness.
DISCUSSION
This review provides consistent evidence
that—among individuals living with a baseline
PTSD, panic disorder, bipolar disorder, or
depressive disorder—recent cannabis use was
You are prohibited from making this PDF publicly available.
For reprints or permissions, contact permissions@psychiatrist.com. ♦ © 2018 Copyright Physicians Postgraduate Press, Inc.
It is illegal to post this copyrighted PDF on any website.
e9J Clin Psychiatry 79:4, July/August 2018
Cannabis and Anxiety and Mood Disorders
associated with negative symptomatic outcomes (including
course of symptoms) over time. Specifically, the collective
findings suggest that individuals using cannabis (ie, any/
greater frequency of use in the last 6 months) experienced
greater symptom severity and number of symptoms and less
occurrence of symptomatic remission and recovery up to 5
years following baseline assessment relative to the comparison
groups (ie, no/lesser frequency of use). All but 1 study32 across
the review was in the context of treatment for an anxiety or
mood disorder (eg, medication, psychotherapy), implying
that cannabis use may potentially interfere with recovery
efforts and contribute to long-term persistent symptoms.
These results are supported by the broader substance use
literature inferring a detrimental effect of various substances
(eg, alcohol, tobacco) on the clinical course and treatment
outcomes in anxiety, depressive, and bipolar disorders.44–46
Although cannabis is considered less harmful than most
psychoactive substances,47,48 the results nonetheless support
its link to negative symptomatic outcomes in AMD. Specific to
the cannabis literature, the results support studies (ineligible
for inclusion in the current review; eg, no comparison
groups, not longitudinal) showing that cannabis users with
AMD experience “negative” symptoms,49,50 psychological
distress,51 and a low quality of life52 and that reducing use
may benefit symptoms.53 The evidence from the general
population, which shows cannabis use to increase the risk
in developing AMD over time23–26 and other adverse mental
health effects,54 further supports association of cannabis with
negative symptomatic outcomes in a clinical population.
Our review provides no indication that cannabis benefits
AMD over time. This finding opposes other studies
(ineligible for inclusion) suggesting that cannabis can
play a role in alleviating symptoms in PTSD (eg, reduces
nightmares), bipolar disorder (eg, stabilizes mood), and
depressive disorders (eg, increases motivation).13–18,55,56
However, these studies are primarily based on acute
therapeutic effects of cannabis. Mechanistically, acute effects
of cannabis are mediated by the human endocannabinoid
system.57 This homeostatic system serves to regulate mood,
cognition, appetite, and sleep, among other functions,
by the interactions between endogenous cannabinoids
(anandamide and 2-arachidonoylglycerol) and G-protein
cannabinoid receptors (CB1, CB2).58 When cannabis is
consumed, its constituents such as tetrahydrocannabinol
(THC) and cannabidiol (CBD), which structurally resemble
the mentioned endogenous cannabinoids, activate the
endocannabinoid system by reacting with the brain’s CB1
receptors. This mechanism is the basis by which cannabis
facilitates a “high” effect, emulating antianxiety and
antidepressive states in some individuals.
Particularly among individuals living with AMD, research
shows there are deficiencies in cannabinoid production and
signaling dysfunctions within the endocannabinoid system
that may contribute to the disorder.59–63 Targeting this system
has therefore been recommended to help treat AMD62,63 by
helping to synthesize endocannabinoids, regulate signaling,
and overall facilitate the endocannabinoid system. However,
whether exogenous cannabinoids, such as THC, can intervene
in the endocannabinoid system to sustainably improve
related symptoms is controversial and understudied. Based
on the literature and mechanisms explaining acute effects
of cannabis, in conjunction with the review’s longitudinal-
associative findings, it can be speculated that cannabis may
serve as a “Band-Aid” strategy to relieve acute symptoms, but
over time the drug may contribute to persistent symptoms
and the prevention of symptomatic recovery.
Study Limitations and Future Research
Despite the consistent results of cannabis’s association with
negative long-term symptomatic and treatment outcomes,
the review’s findings need to be interpreted with caution
when considering the individual studies’ methodological
limitations in conjunction with the broader limitations
of the systematic review. First, the observational designs
across studies disallow causal inferences to be made between
cannabis use and persisting symptoms of AMD. Though the
review found a consistent longitudinal association between
cannabis use and symptoms in AMD, the conclusion that
cannabis can negatively influence the course and outcomes
of symptoms and treatment efforts over time can only be
speculative. Second, the review’s findings may be biased
toward a sample with a higher severity of symptoms, as
the large majority of the total sample (11,959 individuals)
included psychiatric inpatients and outpatients. Including
individuals with varying degrees of symptom severity can
provide further knowledge on the influence of cannabis in
AMD, particularly among those not needing psychiatric
intervention (ie, lower severity).
Third, unaccounted confounding variables around
cannabis consumption in each study may have further
biased the results. For example, the earlier age at onset for
cannabis use significantly increases risk in developing AMD
in addition to cannabis dependence.64 Hence, the age at onset
could have mediated the association between cannabis use
and negative symptomatic outcomes.
Further, none of the studies captured the dose of cannabis
consumed, but rather the broader, subjective frequency of
recent use, which is subject to recall and social desirability
biases. Within frequency of use, there was heterogeneity in the
definitions and operationalizations of individuals “exposed”
and “nonexposed” to cannabis. For example, the exposed
groups in multiple studies may have included those who use
cannabis daily, in addition to those who used cannabis “at
least once in the past month,” whereas the nonexposed groups
in multiple studies may have included those who have never
used cannabis with those who may use cannabis at lower
frequencies. Distinguishing these users in analysis, among
low, moderate, and heavy/daily consumption, for example, is
important since the adverse psychological effects of cannabis
are considered to be frequency and dose dependent.57,65
Regarding frequency and dose, it is imperative to also
collect information on the concentrations of cannabinoids
consumed (eg, THC). A proxy for this in the review was based
on the source of cannabis used, which were unregulated-illicit
You are prohibited from making this PDF publicly available.
For reprints or permissions, contact permissions@psychiatrist.com. ♦ © 2018 Copyright Physicians Postgraduate Press, Inc.
It is illegal to post this copyrighted PDF on any website.
e10 J Clin Psychiatry 79:4, July/August 2018
Mammen et al
sources. Over the last several decades, “street cannabis” has
contained increasingly potent levels of THC and decreases
in CBD production.66 Researchers have noted that the
drug’s effects and risks can depend upon the potency and
dose of THC and its ratio with CBD.57,65 At higher levels,
THC—the primary and psychoactive constituent of cannabis
responsible for the “high”—may overstimulate CB1 receptors
and contribute to adverse effects including increased acute
anxiety, paranoia, memory impairment, and sedation and
subsequent addiction issues related to withdrawal and
tolerance.58,67 Thus, the frequency of cannabis use among a
clinical population (eg, daily)68,69 and the increasingly potent
THC levels in street cannabis may help explain the review’s
findings of cannabis’s association with negative symptomatic
and treatment outcomes.
Interestingly, emerging evidence shows that THC’s adverse
effects can be limited by CBD, which is cannabis’s secondary
and nonpsychoactive constituent that has been bred out of
street cannabis. Unlike THC, CBD does not yield a “high”
experience, is perceived as having limited side effects, and
is generally well tolerated across doses.70 One role of CBD
is to mitigate THC’s effects, mechanistically explained by
its indirect antagonist actions71,72 and low affinity for CB1
receptors.73 This action helps prevent THC from acting at full
strength.74,75 Independent of THC, CBD further contains its
own antianxiety properties involving mechanisms with other
mood regulatory receptors (eg, GPR55, 5-HT1A).76 Hence,
due to its nonpsychoactive property, safety and tolerability,
and encouraging evidence as an antianxiety drug, “CBD is
possibly the cannabinoid more likely to have initial findings
translated into clinical practice.”70(p1224)
Fourth, there was further heterogeneity regarding the
outcome assessments. For instance, the symptoms across
AMD were measured and operationalized by severity,
remission, recurrence, recovery, relapse, cycling, or
number of symptoms. Compounding the heterogeneity
was the review’s inclusion of individual AMD in PTSD,
panic disorder, bipolar disorder, and depressive disorder.
This decision was based on the limited number of studies
available for each disorder, and it limited the ability to
conduct a meaningful meta-analysis. Of notice was the lack
of cohort studies on generalized anxiety disorder (n = 0),
social anxiety disorder (n = 0), and depressive disorders
(n = 2). Considering that AMD are the most common mental
health conditions,1 with strong links to frequent cannabis
use,2–9 more monitoring of consumption and symptoms
within each disorder is warranted.
Overall, future research in this area needs to address
the limitations of the current literature by increasing
methodological rigor to better ascertain the influence of
cannabis on the course and outcomes of symptoms in AMD.
In addition to accounting for the potential confounding
variables noted above (eg, age at onset of use, dose of
cannabinoid concentration), greater focus is needed in
examining changes in severity scale scores between baseline
and follow-up assessments. These data would help determine
if cannabis “worsens” symptoms over time within subjects,
as opposed to the current review, which moreover indicates
that cannabis users are more likely to experience greater
symptom severity over time compared to those abstaining
from use. Additionally, clinical trials are needed to rigorously
examine cannabis’s short- and long-term medical application
for AMD, with specifics on cannabinoid concentrations
(eg, THC, CBD), route of administration (eg, pill form,
vaporization), dosage (eg, low vs moderate), different forms
of cannabis (eg, oils, dried cannabis), interactions with
medications (eg, antianxiety drugs), and mechanisms (eg,
endocannabinoid system functional magnetic resonance
imaging studies).
Clinical Implications
Though the findings of the review have clear limitations,
this systematic review is the first to provide unique insight
into the longitudinal associations between cannabis use and
symptomatic outcomes among those living with a baseline
anxiety or mood disorder. Clinicians can use this “best
available” evidence to inform their own and their patients’
knowledge concerning potential long-term risks of cannabis
on symptoms and recovery. The results can be useful for
health care professionals (eg, psychiatrists, family doctors,
nurse practitioners) who are asked to prescribe medical
cannabis from patients living with AMD. With increasing
legalization of recreational cannabis in North America (eg,
Canada in 2018), communicating this evidence to patients
requesting medical cannabis is timely and important when
one considers they will arguably have easier and quicker
access to regulated recreational cannabis than regulated
medical cannabis.
CONCLUSION
Across AMD, recent cannabis use was associated with
negative symptomatic and treatment outcomes over time.
The findings should be interpreted with caution when
considering the observational designs across studies, biases
linked with the samples (eg, inpatients) and sources of
cannabis consumed (ie, unregulated sources), and limitations
surrounding the heterogeneity in exposure and outcome
measurements. This review can inform future research to
provide more rigorous data to better ascertain cannabis’s
influence on the course and outcomes of symptoms in AMD.
Clinicians can use the insight gained from the review to help
inform their own and their patients’ knowledge concerning
potential risks of cannabis on long-term symptoms and
recovery.
Submitted: August 1, 2017; accepted November 3, 2017.
Published online: June 5, 2018.
Potentia l conicts of in terest: None.
Funding/support: This review was conducted when Dr Mammen was a
postdoctoral fellow funded by the Centre for Addiction and Mental Health
(CAMH) Fellowship Program.
Role of the sponsor: CAMH had no role in the conduct or publication of the
stu dy.
Acknowledgments: The authors acknowledge and thank Eva Pila, PhD
(University of Toronto, Toronto, Ontario, Canada); Rose Wang, MSc (Lakehead
You are prohibited from making this PDF publicly available.
For reprints or permissions, contact permissions@psychiatrist.com. ♦ © 2018 Copyright Physicians Postgraduate Press, Inc.
It is illegal to post this copyrighted PDF on any website.
e11J Clin Psychiatry 79:4, July/August 2018
Cannabis and Anxiety and Mood Disorders
University, Thunder Bay, Ontario, Canada); and
Tereza Florica, BSc (McMaster University, Hamilton,
Ontario, Cana da), all of whom repor ted no conict
of interests, for their valuable contributions toward
the study screening and retrieval processes.
REFERENCES
1. Whiteford HA, Ferrari AJ, De genhardt L, et al.
The global burden of mental, neurological and
substance use disorders: an analysis from the
Global Burden of Disease Study 2010. PLoS One.
2015;10(2):e0116820.
PubMed
CrossRef
2. Blanco C, Hasin DS, Wall MM, et al. Cannabis
use and risk of psychiatric disorders:
prospective evidence from a US national
longitudinal study. JAMA Psychiatry.
2016;73(4):388–395.
PubMed
CrossRef
3. Lev-Ran S, Le Fo ll B, McKenzie K, et al. Cannabis
use and cannabis use disorders among
individuals w ith mental illness. Compr
Psychiatry. 2013;54(6):589–598.
PubMed
CrossRef
4. Martins SS, Gorelick DA. Conditional substance
abuse and dependence by diagnosis of mood
or anxiet y disorder or schizophrenia in the US
population. Drug Alcohol Depend. 20 11;119 (1-
2):28–36.
PubMed
CrossRef
5. Bonn-Miller MO, Harris AH, Trafton JA.
Prevalence of cannabis use disorder diagnoses
among veterans in 2002, 2008, and 2009.
Psy chol S erv. 2012; 9(4) :40 4– 416.
PubMed
CrossRef
6. Cheung JT, Mann RE, Ialomiteanu A, et al.
Anxiety and mood disorders and cannabis use.
Am J Drug Alcohol Ab use. 2010;36(2):118–122.
PubMed
CrossRef
7. Cougle JR, Bonn-Miller MO, Vujanovic AA, et al.
Posttraumatic stress disorder and cannabis use
in a nationally representative sample. Psych ol
Addict Behav. 2011;25(3):554–558.
PubMed
CrossRef
8. Gentes EL, Schry AR, Hicks TA, et al. Prevalence
and correlates of cannabis use in an outpatient
VA posttraumatic stress disorder clinic. Ps ychol
Addict Behav. 2016;30(3):415–421.
PubMed
CrossRef
9. Hunt GE, Malhi GS, Clear y M, et al. Prevalence
of comorbid bipolar and subs tance use
disorders in clinical set tings, 1990–2015:
systematic review and meta-analysis. JAffect
Disord. 2016;20 6:331–349.
PubMed
CrossRef
10. Hazek amp A, Ware MA, Muller-Vahl KR, et al.
The medicinal use of cannabis and
cannabinoids: an international cross-sectional
survey on administration forms. JPsycho active
Drugs. 2013;45(3):199–210.
PubMed
CrossRef
11. Ogborne AC, Smar t RG, Weber T, et al. Who is
using cannabis as a medicine and why: an
exploratory study. JPsychoactive Drugs.
2000;32(4):435–443.
PubMed
CrossRef
12. Walsh Z, Callaway R, Belle-Isle L, et al. Cannabis
for therapeutic purposes: patient
characteristics, access, and reasons for use. Int J
Drug Policy. 2013;24(6): 511–516 .
PubMed
CrossRef
13. Bonn-Miller MO, Babson K A, Vandrey R. Using
cannabis to help you sleep: heightened
frequenc y of medical cannabis use among
those with PTSD. Drug Alcohol Depend.
2014;136:16 2–165.
PubMed
CrossRef
14. Denson TF, Earleywine M . Decreased
depression in marijuana user s. Addict Behav.
20 06 ;31(4) :7 38 –742.
PubMed
CrossRef
15. Elliott L, Golub A, Bennett A, et al. P TSD and
cannabis-related coping among recent
veterans in New York Cit y. Contemp D rug Probl.
2015;42(1):6 0 –76.
PubMed
CrossRef
16. Gruber AJ, Pope HG Jr, Brown ME. Do patients
use marijuana as an antidepressant?
Depression. 1996; 4(2):77–80.
PubMed
CrossRef
17. Healey C, Peters S, Kinderman P, et al. Reasons
for substance use in dual diagnosis bipolar
disorder and substance use disorders: a
qualitative study. JAffect Disord. 2 00 9;11 3(1-
2) :11 8 –12 6 .
PubMed
CrossRef
18. Osbor n LA, Lauritsen K J, Cross N, et al. Self-
medication of somatic and ps ychiatric
conditions using botanical marijuana.
JPsychoactive D rugs. 2015;47(5):345–350.
PubMed
CrossRef
19. Haney M, Evins AE. Does cannabis cause,
exacerbate or ameliorate ps ychiatric
disorders? an oversimplified debate discussed.
Neuropsychopharmacology.
2016;41(2):393– 401.
PubMed
CrossRef
20. Khoury JM, Neves MCLD, Roque MAV, et al. Is
there a role for cannabidiol in psychiatr y?
World J Biol Psychiatr y. 2017;1 –16.
PubMed
21. Turna J, Patterson B, Van Ameringen M. Is
cannabis treatment for anxiety, mood, and
related disorders ready for prime time?
Depress Anxiet y. 2 017;3 4(11):10 06– 1017.
PubMed
CrossRef
22. Walsh Z, G onzalez R, Crosby K, et al. Medical
cannabis and mental health: A guided
systematic review. Clin Psychol Rev.
2017 ;5 1:15 –2 9.
PubMed
CrossRef
23. Gibbs M, Winsper C , Marwaha S, et al.
Cannabis use and mania symptoms: a
systematic review and meta-analysis. JAffect
Disord. 201 5;171: 39 –47.
PubMed
CrossRef
24. Kedzior KK , Laeber LT. A positive association
between anxiety disorders and cannabis use
or cannabis use disorders in the general
population—a meta-analysis of 31 studies.
BMC Psychiatry. 2014;14(1):13 6.
PubMed
CrossRef
25. Lev- Ran S, Roerecke M, Le Foll B, et al. The
association b etween cannabis use and
depression: a s ystematic review and meta -
analysis of longitudinal studies. Psychol Med.
2014;44(4):797–810.
PubMed
CrossRef
26. Twomey CD. Association of cannabis use with
the development of elevated anxiety
symptoms in the general population: a meta-
analysis. JEpidemiol Commu nity Health.
2017;71(8):811–816.
PubMed
CrossRef
27. Moore TH, Z ammit S, Lingford-Hughes A, et al.
Cannabis use and risk of psychotic or affective
mental health outcomes: a sys tematic review.
Lancet. 2007;370(9584):319–328.
PubMed
CrossRef
28. Johns A. Psychiatric effe cts of cannabis. Br J
Psychiatry. 20 01; 178(2 ):116 –122 .
PubMed
CrossRef
29. Wilkinson ST, Radhak rishnan R, D’Souza DC. A
systematic review of t he evidence for medical
marijuana in psychiatric indications. JClin
Psychiatry. 2016;7 7(8):1050 –1064.
PubMed
CrossRef
30. DistillerSR [computer program]. Ot tawa,
Ontario, Cana da: Evidence Partners; 2011.
31. Wells G, Shea B, O’Connell D, et al. Newcastle-
Ottawa Quality Ass essment Scale, Cohort
Studies. http://www.ohri.ca/programs/
clinical_epidemiology/oxford.asp. 2014 .
32. Feingold D, Rehm J, Lev-Ran S. Cannabis use
and the course and outcome of major
depressive disorder: a population based
longitudinal study. Psychiatry Res.
2017;251:225–234.
PubMed
CrossRef
33. Bonn-Miller MO, Boden MT, Vujanovic AA, et
al. Prospec tive investigation of the impact of
cannabis use disorders on posttraumatic stress
disorder sy mptoms among veterans in
residential treatment. Psychol Trauma.
2013;5(2):193–200.
CrossRef
34. Manhapra A, Stefanovics E, Rosenheck R.
Treatment outcomes for veterans with PTSD
and substance use: impact of specific
substances and achievement of abstinence.
Drug Alcohol Depend. 2015;156 :70–77.
PubMed
CrossRef
35. Ruglass LM, Shevorykin A, Radoncic V, et al.
Impact of cannabis use on treatment
outcomes among adults recei ving cognitive-
behavioral treatment for PTSD and substance
use disorder s. JClin Med. 2017;6 (2) :14.
PubMed
CrossRef
36. Wilkinson ST, Stefanovics E, Rosenh eck RA.
Marijuana use is associated with worse
outcomes in symptom severit y and violent
behavior in patients with posttraumatic stress
disorder. JClin Psychiatry. 2 015; 76( 9): 1174– 1180 .
PubMed
CrossRef
37. Bricker JB, Russo J, Stein MB, et al. Does
occasional cannabis use impact anxiety and
depression treatment outcomes? results from a
randomized effectiveness trial. De press Anxiety.
2007;24(6):392–398.
PubMed
CrossRef
38. Kim S-W, Dodd S, Berk L, et al. Impact of
cannabis use on long-term remission in bip olar
I and schizoaffective disorder. Psychiatry
Investig. 2015;12(3):349 –355.
PubMed
CrossRef
39. Kvitland LR, Melle I, Aminof f SR, et al.
Continued cannabis use at one year follow up
is associated with elevated mood and lower
global func tioning in bipolar I disorder. BMC
Psychiatry. 2015;15(1) :11.
PubMed
CrossRef
40. Strakowski SM, DelBello MP, Fleck DE, et al.
Effec ts of co-occurring cannabis use disorders
on the course of bipolar disorder after a first
hospitaliz ation for mania. Arch Gen Psychiatry.
2007;64(1):57–6 4.
PubMed
CrossRef
41. van Rossum I, Boomsma M, Tenback D, et al;
EMBLEM Advisory Board . Does cannabis use
affect treatment outcome in bipolar disorder?
A longitudinal analysis. JNerv Ment Dis .
2009;197(1):35–40.
PubMed
CrossRef
42. Zorrilla I, Aguado J, Haro JM, e t al. Cannabis
and bipolar disorder: does quitting cannabis
use during manic/mixed episode improve
clinical/functional outcomes? Acta Psychiatr
Scand. 2015;131(2):100 –110.
PubMed
CrossRef
43. Bahorik AL, Leibowitz A , Sterling SA, et al.
Patterns of marijuana use among psychiatr y
patients with depression and its impac t on
rec overy. JAffect D isord. 2017; 213:16 8–171 .
PubMed
CrossRef
44. Bahorik AL, Leibowitz A, Sterling SA , et al. The
role of h azardous drinking re ductions in
predicting depression and anxiety symptom
improvement among psychiatr y patients: a
longitudinal study. JAffect Disord.
2016; 206:169 –173.
PubMed
CrossRef
45. Fontana A, Rosenheck R, D esai R. Comparison
of treatment outcomes for veterans with
posttraumatic stress disorder with and without
comorbid substance use/dependence.
JPsychiatr Res. 2 012;46(8) :1008–1014.
PubMed
CrossRef
46. Strakowski SM , DelBello MP, Fleck DE, et al. The
impact of substance abuse on the course of
bipolar disorder. Biol Psychiatry.
2000;48(6):477–485.
PubMed
CrossRef
47. Lachenmeier DW, Rehm J. Comparative risk
assessment of alcohol, tobacco, cannabis and
other illicit drugs using the margin of exposure
approach. Sci Rep. 2015;5 (1):812 6.
PubMed
CrossRef
48. Nutt DJ, King LA, Phillips LD; Indepen dent
Scientific Committee on Drugs. Drug harms in
the UK: a multicriteria decision analysis. Lancet.
2010 ;376 (97 52) :155 8–156 5.
PubMed
CrossRef
49. Bersani G, Bersani FS, Caroti E, et al. Negative
symptoms as key features of depression
among cannabis users: a preliminary report.
Eur Rev Med Pharmacol S ci. 2016;20(3):5 47–552.
PubMed
50. Galynker II, Coh en LJ, Cai J. Negative symptoms
in patients with major depressive disorder : a
preliminary report. Neu ropsychiatry
Neuropsychol Behav Neurol. 2000 ;13(3 ):171–176 .
PubMed
51. Aspis I, Feingold D, Weiser M , et al. Cannabis
use and mental health-related quality of life
among individuals with depressive disorders.
Psychiatry Res. 2015;230(2):341–349.
PubMed
CrossRef
52. Lev-Ran S, Le Fo ll B, McKenzie K, et al. Cannabis
use and mental health-related quality of life
among individuals with anxiety disorders.
JAnxiety D isord. 2012;26(8):799–810.
PubMed
CrossRef
53. Moitra E, Anderson BJ, Stein MD. Reductions in
cannabis use are associated with mood
improvement in female emerging adults.
Depress Anxiet y. 2016;33(4):332–338.
PubMed
CrossRef
54. Volkow ND, Baler RD, Compton WM, et al.
Adverse health effec ts of marijuana use. N Engl
J Med. 2014;370(23):2219 –2227.
PubMed
CrossRef
You are prohibited from making this PDF publicly available.
For reprints or permissions, contact permissions@psychiatrist.com. ♦ © 2018 Copyright Physicians Postgraduate Press, Inc.
It is illegal to post this copyrighted PDF on any website.
e12 J Clin Psychiatry 79:4, July/August 2018
Mammen et al
55. Cameron C, Watson D, Robinson J. Use of a
synthetic cannabinoid in a correctional
population for posttraumatic stress disorder-
related insomnia and nightmares, chronic pain,
harm reduction, and other indications: a
retrospective evaluation. JClin
Psychopharmacol. 2014;34(5):559–564.
PubMed
CrossRef
56. Fraser GA. The use of a synthetic c annabinoid
in the management of treatment-resistant
nightmares in posttraumatic stress disorder
(PTSD). CNS Neurosci Ther. 2009;15(1):84–88.
PubMed
CrossRef
57. Volkow ND, Hampson AJ, Baler RD. Don’t worry,
be happy: endocannabinoids and cannabis at
the intersec tion of stress and reward. An nu Rev
Pharmacol Toxicol. 2 017;57 (1):2 85– 308 .
PubMed
CrossRef
58. Mechoulam R, Parker LA. The
endocannabinoid system and the brain. Annu
Rev Psychol. 2013; 64 (1) :21 –47.
PubMed
CrossRef
59. Ashton CH, Mo ore PB. Endocannabinoid
system dysfunction in mood and related
disorders. Acta Psychiatr Scand.
20 11;1 24 (4 ) :2 5 0 –2 61 .
PubMed
CrossRef
60.Micale V, Tabiova K, Kucerova J, et al. Role of the
endocannabinoid system in depression: from
preclinical to clinical evidence. In: Campolongo
T, Fattore L, eds. Cannabinoid Modulation of
Emotion, Memory, and Motivation. New York,
NY: Springer; 2015:97–129.
61.Patel S, Hillard CJ. Role of endocannabinoid
signaling in anxiety and depression. In: Kendall
D, Alexander S, e ds. Behavioral Neurobiology of
the Endocannabinoid System. Berlin, Germany:
Srpinger-Verlag; 2009:347–371.
62. Saito VM, Wotjak CT, Moreira FA.
Pharmacological exploitation of the
endocannabinoid system: new perspectives
for the treatment of depression and anxiet y
disorders?. Rev Bras Psiquiatr. 2010;32(suppl
1): S7–S14 .
PubMed
63. Serra G, Fratt a W. A possible role for the
endocannabinoid system in the neurobiology
of depression. Clin Pract Epidemol Ment Health.
2007;3(1):25.
PubMed
CrossRef
64. Paruk S, Burns JK. Cannabis and mental illness
in adolescents: a review. S Afr Fam Pract.
2015;58(suppl 1):S18–S21.
65. Hall W, Degenhardt L . Adverse health effe cts of
non-medical cannabis use. Lancet.
20 09; 374(969 8) :138 3–1391.
PubMed
CrossRef
66. Cascini F, Aiello C, D i Tanna G. Increasing delta-
9-tetrahydrocannabinol (Δ-9-THC) content in
herbal cannabis over time: systematic review
and meta-analysis. Curr Drug Abuse Re v.
2012 ;5(1) :32– 4 0 .
PubMed
CrossRef
67. Diana M, Melis M, Gessa GL. Increase in meso-
prefrontal dopaminergic activity after
stimulation of CB1 receptors by cannabinoids.
Eur J Neurosci. 1998;10(9):2825 –2830.
PubMed
CrossRef
68. Sex ton M, Cuttler C, Finnell JS, et al. A cross-
sectional survey of medical cannabis users:
patterns of use and perceive d efficacy.
Cannabis Cannabinoid Res. 2016;1(1):131 –138.
PubMed
CrossRef
69. Ware MA, Adams H, Guy GW. The medicinal use
of cannabis in the UK: results of a nationwide
sur vey. Int J Clin Pract. 20 05;59(3):291–295.
PubMed
CrossRef
70. Bergamaschi MM, Queiroz RH, Zuardi AW, et al.
Safety and side effects of cannabidiol, a
Cannabis sativa constituent. Cur r Drug Saf.
2011;6(4):237–249.
PubMed
CrossRef
71. Hayakawa K , Mishima K, Hazekawa M, et al.
Cannabidiol potentiates pharmacological
effects of Δ(9)-tetrahydrocannabinol via CB(1)
receptor-dependent mechanism. Brain Res.
20 0 8 ; 118 8 :157 –1 6 4 .
PubMed
CrossRef
72. Witkin JM, Tzavara ET, Nomikos GG. A role for
cannabinoid CB1 receptors in moo d and
anxiety disorders. Behav Pharmacol. 20 05;16(5 -
6): 315–331.
PubMed
CrossRef
73. Per twee RG. The pharmacology of cannabinoid
receptors and their ligands: an overview. Int J
Obes. 20 06;30(s uppl 1) :S13–S18.
PubMed
CrossRef
74. McPartland JM, Duncan M, Di Marzo V, et al.
Are cannabidiol and Δ(9)
-tetrahydrocannabivarin negative modulators
of the endocannabinoid system? a systematic
revi ew. Br J Pharmacol. 2015;172(3):737–753.
PubMed
CrossRef
75. Niesink RJ, van Laar MW. Does cannabidiol
protect against adverse psychological effects
of THC? Front Psychiat ry. 2013; 4:130.
PubMed
CrossRef
76. Blessing EM, Steenkamp MM, Manzanares J, et
al. Cannabidiol as a potential treatment for
anxiety disorders. Neurotherapeutics.
2015;12(4):825–836.
PubMed
CrossRef