Table 1:

Cannabis and the onset of mood and anxiety disorders organized by symptom/disorder type

Author, year	Participants	N	Follow up (yr)	Cannabis Use	Outcome	Adjustment Variables	Relevant Findings	NOS
MANIA SYMPTOMS AND BIPOLAR DISORDERS
Marwaha et al. 2018	Avon UK 1991 birth cohort	3,370	5	CU frequency (age 17)	Hypomania symptoms (HCL-32) (age 22)	Family adversity index, history of abuse, alcohol use, drug use, psychotic symptoms, depression history	Adolescent CU associated with f hypomania symptoms at age 22; dose dependent: >2x/wk (OR = 2.87 (1.68 – 4.91); any CU (OR=1.82 (1.45 – 2.28))	7
Ratheesh et al. 2015	Youth ↑risk BD (BAR)	52	1	CUD	BD-1, BD −2 (DSM IV-TR)	N/A	25% with CUD and 17% without CUD developed BD (OR=1.7 (0.2–18.1). Outcome rate (4/52) too small to achieve significance	5
Tjissen et al. 2010	Munich 1994 (age 14–17) represent.	543	8.3	Lifetime CU > 5 uses	Mania symptoms Munich-CIDI (DSM IV)	Age, sex, SES, family history of mood episodes, exposure to trauma, loss of a parent, alcohol use, personality style	Any CU associated with f mania symptoms (OR= 4.26 (1.42–12.76) p=0.010)	9
Henquet et al. 2006	Dutch adult represent.	4,815	3	CU frequency	Mania, psychosis symptoms (CIDI)	Age, sex, ethnicity, education, marital status, neuroticism, lifetime drug use, last year alcohol use, baseline depression/ mania	Any CU associated with f mania symptoms after adjusting for covariates: daily CU (AOR 3.43 (1.42–8.26), monthly CU (AOR 2.23 (0.82–6.07)	8
DEPRESSION SYMPTOMS AND DEPRESSIVE DISORDERS
Schoeler et al. 2018	London males born in 1953	411	40	CU frequency at age 14, 18, 32, 48	Lifetime diagnosis of MDD (DSM-IV)	Alcohol, cigarette and other illicit drug use; psychiatric illness; childhood anxiety, conduct problems and behaviour and emotional problems	CU onset <18 yrs associated with f lifetime MDD (AOR= 2.41 (1.22–4.76) p= 001) and ↑ time to MDD for low risk users (HR=2.09 (1.16–3.74)) and high risk users (HR = 8.69 (2.07–36.5))	8
Wilkinson et al. 2016	US (1994) adolescents age 12–18	11,995	12	30-day CU frequency	Depression symptoms (CES- D)	Race, ethnicity, educational attainment of parents and respondent, age, sex	CU in earlier waves was not significantly associated with depressive symptoms in later waves	7
Womack et al. 2016	Pittsburgh males from low SES families	264	11	Age 17, 20, 22: past year CU >2– 3x/week	Age 17; 20, 22: Depressive symptoms (BDI)	Caregiver BDI score, parent income, ethnicity, highest level of education age 22, youth antisocial behaviour, tobacco and alcohol use age 17, youth IQ, adult court records	Significant positive association between adolescent CU and mild depression at age 22 (B=0.493 SE=0.207, p<0.05).	6
Baggio et al. 2014	Switzerland male conscripts	5,223	1.25	CU frequency; trajectory; CUD	Depression symptoms (WHO-MDI)	Age at first CU, language (German, English)	Only CUD was associated with increase in depression symptoms (B=0.087, P<0.001)	4
Otten & Engels 2013	Netherlands adolescents +/− 5-HTTLP R allele	306	4	Lifetime CU; CU frequency in past month	Depression symptoms (DML)	Personality scores, alcohol use, tobacco use, parental education, parenting practices	CU associated with later depression symptoms in the presence of the short allele (B =0.34 (b=0.10), P<0.001), but not in its absence (B=−1.379 (b=−0.14), P=0.51)	7
Rasic et al. 2013	Nova Scotia Grade 10 students	1,582	2	Past 30 day CU	Depression (CES- D >24 females; >22 males)	Alcohol use, other illicit drug use, living situation, school marks, age, gender	Adolescents with CU have f odds of depression (AC)R=1.10 (1.01–1.19), p<0.05); higher odds in those with heavier CU (AOR = 1.16(1.04–1.29), P<0.05)	5
Manrique-Garcia et al. 2012	Sweden male conscripts (1969–1970)	45,087	35	Total number lifetime CU	Depression (ICD-8. 9. 10)	Personality disorder, IQ, disturbed behaviour in childhood, social adjustment, popularity, relationships, alcohol use, smoking, early adulthood SES, urbanicity	Heavy CU (> 50 uses) does not f risk of depression after adjusting for confounders (AHR= 0.9 (0.5–1.6))	9
Marmorstein & Iacono 2011	Minnesota Twin Family Study	1,252	5	CU frequency (age 17);’ CUD	Age 17. 20. 24: MDD Diagnosis (DSM III-R)	Gender; baseline MDD, AUD, nicotine dependence; psychosocial risks (not graduating high school by age 20, period of unemployment >6 months), crime	Adolescent CUD associated with f odds of later MDD (AOR=2.62 (1.22–5.65). Relationship partially mediated by psychosocial failure (AOR = 2.54 (1.40–4.60), p<0.05).	9
Harder et al. 2008	Mid-Atlantic cohort (1985– 2001)	1,494	7	CUD before age 17	MDE (DSM-IV) between ages 19– 24	Demographics, SES, other drug use, childhood disturbances of psychological well being, parental monitoring, behavioural intervention status variables, pre exposure depression/anxiety	Early CUD not associated with MDE (OR=1.33 (0.76–2.23), p=0.32), when propensity scores used to adjust for confounders	9
Pederson 2008	Norway 14 year olds (1992)	2,902	13	Past 12 month CU quantity at age 14. 16. 21. 27	Depressed mood (Kandel and Davies score >9)	Parent SES, parental monitoring and support, parental substance use, pubertal development, student academics, school completion, conduct problems, alcohol intoxication, alcohol problems unemployment, daily smoking	No significant association found between early or late CU and later depression symptoms when adjusting for confounders (AOR = 0.9 (0.4–2.5))	7
Georgiades & Boyle 2007	Ontario birth cohorts (1966– 1979)	3,294	18	Past year CU frequency	12 month prevalence MDD (CIDI-SF)	Family SES, family functioning, sex, grade failure, other medical condition, general health status, externalizing and internalizing symptom scales, tobacco use	CU in adulthood alone (AOR=2.58 (1.67– 3.99), p<0.001) or adolescence + adulthood (AOR=4.45 (2.05–9.66), p<0.001), associated with ↑ MDD diagnosis, but not adolescence alone (AOR=1.48 (0.65–3.40), p > 0.5)	7
Harder et al. 2006	US 1979 birth cohort	12,686	4	Past year CU at age 19	Depression (CES- Dscore>16) age 23	Age, sex, aptitude, survey weight, general health limitations, region of residence, criminal activity, residence age 14, cigarette use, excessive alcohol use, hard drugs use	CU was not associated with depression when compared to non users weighted for other depression risk factors (AOR= 1.51 (0.64– 3.54))	7
Patton et al. 2002	Australia Victoria adolescents Age 15–21	1,601	7	Highest CU frequency over a 6 month period	Depression and anxiety symptoms (CIS-R >12);	Teenage depression, anxiety, alcohol use, tobacco use, other illicit drugs, antisocial behaviour, parental separation, parental education, sex, age, rural vs urban residence, parental education	Weekly CU associated with j depression/anxiety symptoms in females (AOR=1.9 (1.1–3.3) p=0.01); but not males (AOR = 0.47 (0.17–1.3))	8
Bovasso et al. 2001	Baltimore sstudents (1980)	1,920	15	CUD(DSM III-R)	Depression symptoms (DIS DSMIII-R)	Demographic variables, stressful life events and chronic illnesses, baseline depression symptoms, mental health treatment, psychiatric disorder, substance abuse/dependence	CUD associated with j depression symptoms (OR=4.49 (1.51–13.26)) p<0.01)	9
ANXIETY SYMPTOMS AND DISORDERS
Duperrouzel et al. 2018	Miami adolescents	250	1	Past month CU frequency at baseline	Depression and anxiety symptoms (DASS-21)	Gender, alcohol use, nicotine use, and history of mood disorder	People with CU had a more gradual decline in anxiety symptoms over time (b=0.28, p=0.024)	6
Feingold et al. 2016	US adult represent.	34,653	3	Past year CU frequency	AD (DSM IV- TR)	AUD, SUD, sex, race, education level, household income, marital status, age, region, other DSM diagnosis	Daily CU associated with later SAD after controlling for all confounders (AOR=1.98 (0.99–3.94))	8
Bechtold et al. 2015	Pittsburgh adolescent males	506	22	CU onset (early, late); CU chronicity	AD (DSM IV)	Past year substance use age 36, SES age 36, health insurance, health status, mental and physical health age 14	CU groups did not differ in lifetime diagnoses of anxiety disorders	7
Zvolensky et al. 2008	Oregon adolescents	1,790	10	Lifetime CU or CUD	DSM IV diagnosis panic attack or panic disorder	Life time history drug dependence, daily cigarette smoking status	CU not associated with developing panic attack (AOR 1.3 (0.55–3.2)); PD (AOR = 1.0 (0.34–3.2)) after adjusting for cigarette smoking	7
PTSD
Lee et al. 2018	African American, Puerto Rican East Harlem students	674	22	No CU, chronic CU, moderate CU, early vs late quitters	PTSD symptoms at age 36 (PCL-S)	Gender, race/ethnicity, alcohol use, cigarette use, other illicit drug use, delinquency, low self control, depressive symptoms age 14, victimization, sexual assault age 19	People with CU and exposed to trauma more likely to have PTSD symptoms: Chronic (AOR= 4.27 (1.28–14.20), p<0.05); Late quitters (AOR= 6.67 (1.62–27.44); p<0.01); Moderate users (AOR= 3.32 (1.0710.34); p< 0.05); but not early quitters (AOR = 1.75 (0.36–8.44))	7
MULTIPLE SYMPTOMS OR DISORDERS
Guttmann-ova et al. 2017	Seattle youth	808	20	Age of CU, Regular CU (weekly), Duration of CU	Generalized and social anxiety; Depression symptom count (DIS-IV)	Adolescent tobacco and alcohol use, gender, ethnicity, childhood poverty, early environmental risk, baseline psychopathology	All CU groups, except adolescent limited regular users, had ↑ symptoms of GAD than non-users after controlling for all confounders. No significant association was found between CU and depression symptoms.	8
Danielsson et al. 2016	Sweden adult represent.	8,598	3	Lifetime CU	Anxiety symptoms (SPRAS) Depression symptoms (MDI)	Substance use, sex, age, education, childhood adverse circumstances, ethnicity, place of upbringing	Baseline CU was not associated with later depression (RR=0.99 (0.82–1.17)) or anxiety (RR = 1.09 (0.98 – 1.20)) symptom scores after adjusting for all confounders	7
Scholes-Balog et al. 2016	Australia grade 5 youth represent.	927	12	No CU; CU by age 12 or 19	Depression and anxiety symptoms (K-10)	Alcohol, cigarette and other substance use, gender, parent education, school grades and antisocial behaviour (age 12)	CU (mean 1–2x/year) not significantly associated with depression and anxiety symptom scores	6
Gage et al. 2015	Avon UK 1991 birth cohort	1,791	2	CU frequency (age 16)	AD or MDD (ICD 10)(age 18)	Alcohol and illicit drug use, family history depression, maternal education, urban living, childhood IQ, personality traits, victimization, conduct disorder, depression/ anxiety age 16	CU frequency was not significantly associated with ↑ odds of AD (AOR = 0.96 (0.75–1.24) or MDD (AOR = 1.3 (0.98 – 1.72), p=0.065) by age 18 after adjusting for confounders	7
Feingold et al. 2015	US adult represent.	43,093	3	CU frequency over last 12 months	BD-1, BD-2 MDD (DSM IV-TR)	Sex, age, education, income, marital status, urbanity, alcohol use, other substance use, other psychiatric disorder	Past year CU not associated with increased incidence of BD (AOR 1.17 (0.65–2.11) or MDD (AOR = 0.58 (0.22–1.51)) after adjusting for confounders	8
Degenhardt et al. 2013	Victoria, Australia adolescents	1,388	15	CU frequency (age 16)	AD (ICD 10) (age 29)	Alcohol, nicotine, and illicit drug use, age, education level, nationality, adolescent anxiety/depression	Weekly CU in adolescence did not affect odds of MDD (OR = 1.2 (0.73–2.0), p = 0.6) or AD (OR = 1.4 (0.84–2.5)) by age 29	8
Van Laar et al. 2007	Holland adult represent (1996)	5618 total	3	CU frequency	AD, MDD, dysthymia, BD (DSM III-R)	AUD/SUD, age, gender, education, urbanicity, employment, partner status, neuroticism, parental psych history, childhood trauma, life time psychiatric disorder	Any baseline CU associated with ↑ MDD (AOR = 1.68 (1.11–2.55), p < 0.05), but not dysthymia (AOR = 1.55 (0.67–3.58), p >0.05), BD (AOR = 5.38 (1.93–14.9) p< 0.01), or any AD (AOR = 1.27 (0.77–2.12)	8
Hayatbakhsh et al. 2007	Australia 1981 birth cohort	2,854	7	CU frequency (age 21); Age first CU	Depression and anxiety symptoms (YASR)	Age 14 smoking/alcohol use, gender, maternal factors, family income	Frequent CU ↑ mood and anxiety symptoms at 21 years with early onset (OR=3.0 (1.8–5.2)) and late onset (OR = 2.3 (1.5–3.6) users	8
Fergusson et al. 1996	New Zealand birth cohort (1977)	1,265	2	Any CU (age 14)	AD, MDD, or dysthymia age 16 (DSM III-R)	Substance use age 12; family social position, functioning, substance abuse; childhood behaviour problems, cognitive ability, truancy, plan to enter university, peer affiliations, conduct problems	CU by age 15 not associated with ↑ depressive disorder (AOR=1.4 (0.7–2.4)) or AD (AOR = 1.2 (0.5–2.8)) after adjusting for confounding variables	8