CanJPsychiatry 2013;58(6):344352

Original Research

Profiles and Mental Health Correlates of Alcohol and

Illicit Drug Use in the Canadian Population:
An Exploration of the J-Curve Hypothesis
Alexandre Dumais, MD, PhD, FRCPC1; Luigi De Benedictis, MD, MSc, FRCPC2;
Christian Joyal, PhD3; Jean-Franois Allaire, MSc4; Alain Lesage, MD, MPhil, FRCPC5;
Gilles Ct, PhD6
1

Psychiatrist and Researcher, Philippe-Pinel Institute of Montreal, Fernand-Seguin Research Centre, Department of Psychiatry, University of Montreal,
Montreal, Quebec.
Correspondence: Philippe-Pinel Institute of Montreal, Fernand-Seguin Research Centre, Department of Psychiatry, University of Montreal, 10905 HenriBourassa Boulevard East, Montreal, QC H1C 1H1; alexandre.dumais@umontreal.ca.

Psychiatrist, Louis-H Lafontaine Hospital, Fernand-Seguin Research Centre, Department of Psychiatry, University of Montreal, Montreal, Quebec.

Professor of Psychology, University of Quebec, Trois-Rivires, Quebec; Researcher, Philippe-Pinel Institute of Montreal, Montreal, Quebec.

Statistician, Philippe-Pinel Institute of Montreal, Montreal, Quebec.

Psychiatrist and Researcher, Louis-H Lafontaine Hospital, Fernand-Seguin Research Centre, Department of Psychiatry, University of Montreal, Montreal,
Quebec.

Director, Philippe-Pinel Research Centre, Philippe-Pinel Institute of Montreal, Montreal, Quebec; Professor of Psychology, University of Quebec,
Trois-Rivires, Quebec.

Key Words: lifetime

abstainers, alcohol, illicit drugs,
mental disorders, suicide,
multiple correspondence
analysis, psychological distress
Received June 2012, revised,
and accepted November 2012.

Objective: Alcohol and (or) illicit drug use (AIDU) problems are associated with mental
health difficulties, but low-to-moderate alcohol consumption may have mental health
benefits, compared with abstinence. Our study aimed to explore the hypothesis of a
nonlinear, or J-curve, relation between AIDU profiles and psychological distress, psychiatric
disorders, and mental health service use in the general Canadian population.
Methods: Data were collected from a representative sample of the Canadian population
(n = 36 984). Multiple correspondence analyses and cluster analyses were used to extract
AIDU profiles. Sociodemographics, psychological distress, psychiatric disorders, and
mental health service use were assessed and compared between profiles.
Results: Seven AIDU profiles emerged, including 3 involving risky or problematic AIDU that
correlate with major affective disorders, anxiety disorders, suicidal behaviours, and higher
levels of psychological distress. No J-curve relation was found for psychiatric disorders and
mental health service use. The lifetime-abstainer profile correlates with the lowest rates
of psychiatric disorders and mental health service use. Lifetime abstainers are also more
often female, immigrant, and unemployed. Compared with other profiles, spirituality is more
important in their life.
Conclusions: The hypothesis of a nonlinear relation between psychiatric disorders and
AIDU was not supported. Lifetime AIDU abstainers have specific sociodemographic and
cultural background characteristics in Canada.
WWW

Profils et corrlats de sant mentale de la consommation dalcool et

de drogues illicites dans la population canadienne : une exploration
de lhypothse de la courbe en J
Objectif : Les problmes de consommation dalcool et (ou) de drogues illicites (CADI)
sont associs des difficults de sant mentale, mais la consommation dalcool faible
modre peut avoir des avantages pour la sant mentale, compar labstinence. Notre
tude visait explorer lhypothse dune relation non linaire, ou de courbe en J, entre les
profils de la CADI et la dtresse psychologique, les troubles psychiatriques, et lutilisation
des services de sant mentale dans la population gnrale du Canada.
Mthodes : Les donnes ont t recueillies auprs dun chantillon reprsentatif de
la population canadienne (n = 36 984). Des analyses de correspondance multiple et
des analyses typologiques ont servi extraire les profils de la CADI. Les donnes
344 W La Revue canadienne de psychiatrie, vol 58, no 6, juin 2013

www.LaRCP.ca

Profiles and Mental Health Correlates of Alcohol and Illicit Drug Use in the Canadian Population: An Exploration of the J-Curve Hypothesis

sociodmographiques, la dtresse psychologique, les troubles psychiatriques, et

lutilisation des services de sant mentale ont t valus et compars entre les
profils.
Rsultats : Sept profils de la CADI sont ressortis, dont 3 comportaient une CADI
risque ou problmatique qui se corrle avec les principaux troubles affectifs, les
troubles anxieux, les comportements suicidaires, et des taux levs de dtresse
psychologique. Aucune relation de courbe en J na t observe pour les troubles
psychiatriques et lutilisation des services de sant mentale. Le profil de labstinent
vie se corrle avec les taux les plus faibles de troubles psychiatriques et dutilisation
des services de sant mentale. Les abstinents vie sont aussi plus souvent de sexe
fminin, immigrants, et sans emploi. Comparativement dautres profils, la spiritualit
est plus importante dans leur vie.
Conclusions : Lhypothse dune relation non linaire entre les troubles
psychiatriques et la CADI na pas t soutenue. Les abstinents de la CADI de dure
de vie ont des caractristiques sociodmographiques et des antcdents culturels
spcifiques au Canada.

lcohol and (or) illicit drug use constitutes a major health

problem. About 5% of all diseases and premature
deaths worldwide are associated with alcohol use.1 Alcohol
abuse is also linked with violent behaviours, unintentional
injuries, reduced job performance, absenteeism, and family
deprivation.1 The economic burden of alcohol use alone is
estimated to have been $650 billion worldwide in 2002.2
AIDU problems are also associated with mental disorders,
especially with mood and anxiety disorders,35 and to a
higher risk of hospitalization with psychiatric disorders.6
At the other end of the spectrum, though, the findings for
abstinence indicate that it too has negative consequences.7
Such conclusions have led to the J-curve or nonlinear
hypothesis regarding a cardioprotective effect for regular,
low-dose alcohol consumption.1 In addition, it has been
shown that abstainers and former heavy drinkers use health
services more than current alcohol drinkers.8,9 The nonlinear
hypothesis for mental health has also been evaluated, but
the studies were limited to alcohol use alone. Specifically,
studies using self-reporting quantitative questionnaires on
psychological distress, depression, or anxiety have found
evidence of a nonlinear relation with alcohol consumption
in different countries.1017 The results are not so clear for
psychiatric disorders. Some studies10,18 have found a
J-curve relation between personality and mental disorders
and alcohol consumption, while others14,19 have yielded no
evidence to support a nonlinear relation. To our knowledge,
no study has specifically investigated the relation between
the continuum of AIDU profiles in a general population
sample, including lifetime abstainers, and their correlates
Abbreviations
AIDU

alcohol and (or) illicit drug use

BA

bivariate analysis

MAD

major affective disorder

MCA

multiple correspondence analysis

MHP

mental health problem

VT

value test

www.TheCJP.ca

Clinical Implications

Problematic and excessive AIDU are associated with

mental disorders.

There is no clear J-curve relation between MHPs and

AIDU.

Lifetime abstainers are a specific subgroup of the

population not associated with a higher level of mental
disorders.

Limitations

This is a correlational, cross-sectional study.

The survey does not assess 2 key psychiatric disorders

that are often related to substance abuse and
dependence: personality and psychotic disorders.

No measures of lifetime AIDU problems have been

included.

regarding psychological distress, psychiatric disorders, and

mental health service use.
Our study aims, first, to describe AIDU profiles in a
representative sample of the general Canadian population
and, second, to explore the hypothesis of a nonlinear
relation between AIDU and mental health by evaluating
psychological distress, psychiatric disorders, and mental
health service use correlates.

Method

The sample was drawn from the Canadian Community

Health Survey: Mental Health and Well-Being, a crosssectional Statistics Canada survey that collected information
on Canadians mental health status, mental health care use,
and health determinants.20 The data were gathered between
May and December 2002 (n = 36 984) and are representative
of about 98% of the population aged 15 or older in the
10 provinces. Residents of the 3 territories, First Nations
Reserves, Crown lands, and certain remote regions, as
well as institutional clientele and full-time members of the
Canadian Armed Forces, were excluded from the survey.
The Canadian Journal of Psychiatry, Vol 58, No 6, June 2013 W 345

Original Research

Alcohol and (or) Illicit Drug Use

Past-12-month and lifetime AIDU was assessed in

face-to-face interviews. The illicit substances include
cannabis, cocaine, amphetamines, 3,4-methylendioxymethamphetamine (commonly referred to as MDMA or
ecstasy), hallucinogens, solvents, heroin, and steroids.
Because a disproportionate number of drug users reported
cannabis use, compared with other drugs, illicit substances
were dichotomized into cannabis and other illicit drugs.
Past-12-month excessive alcohol use was measured by
the consumption of 5 or more drinks on any 1 occasion.
Problematic use was judged by interference by AIDU with
daily activities and responsibilities during the previous 12
months; that is, whether consumption interfered significantly
with a persons normal routine, occupational (or academic)
functioning, social activities, or relationships. There were
no measures of lifetime AIDU problems.

Mental Health

Participants reported various current and lifetime MHP.

Lifetime and past-12-month episodes of mania, major
depressive disorder, panic disorder, and social phobia were
appraised using an adapted version of the World Mental
Health Composite International Diagnostic Interview21;
data on suicide were derived from the depression module.
In addition, the Kessler Psychological Distress Scale
(commonly referred to as K10) was employed to determine
the level of psychological distress.22 This scale uses 10
questions to establish the respondents level of distress.
Higher scores indicate more distress.

Use of Mental Health Services

Contacts with professionals about emotions, mental health,

or AIDU were determined by asking whether the participant
had ever discussed these matters with a psychiatrist,
general practitioner, psychologist, nurse, or social worker.
The number of hospitalizations for MHP or AIDU-related
problems was also elicited.

Statistical Analyses

Using SAS for Windows version 9.1 (SAS Institute Inc,

Cary, NC), we employed 2 complementary statistical
techniques to extract AIDU profiles: MCA and cluster
analysis. Sampling weights provided by Statistics Canada
to account for the complex sampling procedure were used
for all analyses.
An MCA23 was performed to assess the dimensions of
AIDU (see Table 1 for specific variable categories). MCA
is an extension of simple correspondence analysis designed
to analyze relations between variables represented in a
2-way frequency cross-tabulation table. It is an exploratory
graphical technique that allows one to identify personprofiles using the variables included in the analysis. The
rows and columns of the table are assumed to be points
in a high-dimensional Euclidean space. Associations are
ascertained by calculating distances between points in the
space, that is, the chi-square distances between people in
346 W La Revue canadienne de psychiatrie, vol 58, no 6, juin 2013

different categories of the variables under study. The aim is

to redefine the principal dimensions or axes of the space to
capture most of the inertia (which may be interpreted as the
explained variance or R2). If a dimension can be interpreted
in a meaningful way, it is deemed to be justified. If it cannot,
it is considered a random fluctuation or noise in the data.
In accordance with Cattells24 suggestion in the context of
factor analysis, a scree test was applied to determine how
many dimensions to retain. The eigenvalues obtained from
MCA are plotted by increasing dimensionality and resulting
in a falling curve. The number of dimensions is determined
by the point at which the curve bends and flattens out (the
elbow) after sloping relatively steeply downward.25
Although the MCA provided a good picture of the data,
cluster analysis was needed to group people of similar
AIDU into their respective classes. A hierarchical clustering
technique applied to a limited number of the dimensions
obtained from the MCA has been shown to be more
efficient for classifying people.26 Wards criterion, which
results in minimum loss of inertia, was used.27 To determine
the number of clusters to retain, we used the dendrogram
(aggregation tree) that emerged from the hierarchical
clustering to establish the number of classes. To describe
the specific characteristics of each profile, we used the VT.28
A cut-off of 4 (P < 0.001) was chosen; this cut-off is for
descriptive purposes only and should not be interpreted as
a hypothesis test as all the variables were used to create the
clusters.
Further, BAs were carried out using the VT, which is
equivalent to the standardized adjusted residuals for a
2 2 cross-tabulation. The VT compared the percentage of
people within a class who share a particular trait with the
percentage in the total sample who have the trait. The VT
approximately follows a standard normal distribution. A VT
of 4 or more (P < 0.001) was chosen to identify variables
that are significantly associated with each profile. This cutoff was chosen because of the large number of comparisons
made. Finally, psychological distress was compared
between groups by the Kruskal-Wallis test as the variable
was not normally distributed.

Results

Lifetime abstainers were not included in the MCA as they

already represent a specific profile. For the remaining
people, 3 dimensions were found to explain 52.2% of the
Greenacre adjusted inertia. The firstand principal
dimension, accounting for 28.9%, pertains to the level
of alcohol, cannabis, and other illicit drug use (high and
regular use, compared with nonuse without lifetime
abstention). The second dimension explains 13.3% of the
inertia and distinguishes between the extreme types of
alcohol users (heavy drinkers and past-year abstainers)
and low-to-moderate alcohol users. The third dimension
accounts for 9.9% of the inertia and differentiates between
regular alcohol users without excessive drinking and heavy
drinkers and past-year abstainers. We found that a flag
for an AIDU problem was not a significant factor in the
www.LaRCP.ca

Profiles and Mental Health Correlates of Alcohol and Illicit Drug Use in the Canadian Population: An Exploration of the J-Curve Hypothesis

dimensions retained. This variable was then evaluated in

the BA.
Clustering yielded 6 different profiles (Table 1). Lifetime
abstainers comprise a seventh profile. All are outlined
below.

Profile 1 (Problematic AIDU; 4.6%)

This group is characterized by regular consumption

of cannabis, use of other drugs, and regular excessive
consumption of alcohol. A large proportion (20.6%) of the
group may have an AIDU problem.

Profile 2 (Regular, Excessive AIDU; 7.9%)

This group is comprised of people who report regular

consumption of cannabis and regular and excessive alcohol
intake, though to a lesser degree than in profile 1. A smaller
proportion (9.3%) than in profile 1 may have an AIDU
problem.

Profile 3 (Excessive Alcohol Users; 14.3%)

This group is defined by excessive use of alcohol on

a regular basis, but not as often as in profiles 1 and 2.
Although people report occasional use of cannabis in the
past year, most have not used it during that period. A small
proportion (2.2%) may have an AIDU problem.

Profile 4 (Regular, Low Alcohol Users and Lifetime

Cannabis Abstainers; 15.5%)

This group is characterized by regular low use of alcohol.

Most have never used cannabis.

Profile 5 (Low Alcohol Users and Illicit Drugs

Abstainers; 25.5%)

This profile is associated with occasional use of alcohol,

mostly in small quantities, and is also characterized by
cannabis and other illicit drug abstention.

Profile 6 (Occasional or Former Alcohol Users and

Lifetime Illicit Drug Abstainers; 23.1%)

This group is characterized by occasional alcohol use

or past-year alcohol abstention and lifetime illicit drug
abstention.

Profile 7 (Lifetime AIDU Abstainers; 9.2%)

This profile was created prior to performance of the MCA

and the clustering analysis as the people involved represent a
particular group: lifetime alcohol and illicit drug abstainers.

Sociodemographic and Cultural Background

Correlates

Profiles 1 to 3 are significantly associated with younger

people (Table 2). Profiles 1 to 4 are significantly linked
with male sex and profiles 5 to 7 with female sex. A larger
proportion of the people in profiles 1 to 3 are single than
in the other profiles. People with profiles 1 to 5 more often
report having a regular job. Among the people who declared
www.TheCJP.ca

they had not worked during the preceding year, those with
profiles 4 and 6 are mainly retired; those with profiles 1
to 3 have more often been laid off; and those with profile
5 report having family responsibilities. In addition, people
with profiles 4, 6, and 7 attribute greater importance to
spirituality, and those with profile 7 are more commonly
immigrants.

Mental Health Correlates

Comparison of the profiles revealed that higher proportions

of people in profiles 1 and 2 report MADs, anxiety disorders,
and suicidal behaviours (Table 3). Profiles 1 to 3 report a
higher median level of psychological distress. The lowest
proportion of psychiatric disorders was found among AIDU
lifetime abstainers (profile 7), but the lowest median level
of psychological distress was found for profiles 4 and 6.

Mental Health Service Correlates

Comparison of mental health service use revealed similar

levels for people with profiles 1 and 2 and to a lesser extent
for those in profile 3 (Table 4). People with profile 1 further
exhibit the highest rate of hospitalization for MHPs. AIDU
lifetime abstainers exhibit the lowest level of mental health
service use.

Discussion

The goal of our study was to explore the hypothesis of a

J-curve relation between AIDU and psychological distress,
psychiatric disorders, and mental health service use.
AIDU profiles were drawn from a representative sample
of the general Canadian population. Six different AIDU
profiles emerged from a descriptive multivariate statistical
analysis; the seventh profile comprised lifetime abstainers.
The hypothesis of a nonlinear or J-curve relation was not
supported for psychiatric disorders or mental health service
use.
Specifically, profiles 1 and 2 represent the most problematic
pattern of AIDU. These groups are largely comprised
of younger men reporting MADs, anxiety disorders,
and suicidal behaviour. Mental health care services are
consequently used more often and at an earlier age. These
findings support earlier ones showing a strong association
between AIDU problems and mental disorders35 and higher
levels of psychological distress.10
Profiles 4 to 6 represent former to moderate alcohol drinkers,
mostly with lifetime illicit drug use. The 3 groups exhibit
comparable levels of MHP and use of services, falling
between the level for profiles 1 to 3 and lifetime abstainers.
Sareen et al19 similarly found that moderate drinkers fall
between abstainers and problem drinkers. Profiles 4 and 6
show the lowest median levels of psychological distress.
This finding, especially for profile 4, which displays highfrequency but not excessive alcohol use, supports the
nonlinear hypothesis presented in previous studies using
quantitative scales.14 This pattern of drinking may have
nonspecific psychological benefits, but it is not associated
The Canadian Journal of Psychiatry, Vol 58, No 6, June 2013 W 347

4.7
0
0.5
43.1a
51.7a

13.8
18.3
20.9
27.8
9.3

Past 12-month abstention

<1/month

13/month

13/week

4/week

348 W La Revue canadienne de psychiatrie, vol 58, no 6, juin 2013

0
2.3
91.9a

18.3
10.4
6.5

<1/month

13/month

1/week

43.5a
63.0
20.6a

5.9
14.2
2.1

1/month

Flag for 12-month interference with daily activities and responsibilities by AIDU; BA

0.2

0.3

Has tried one of the following drugs at least once: cocaine, amphetamines, 3,4-methylendioxymethamphetamine, hallucinogens, solvents, heroin, steroids

0.6

13.2

9.6

90.4a

100a

2.4

3.2

36.0a

60.8a

9.5

25.5a

65.0a

25.6

100a

100a

0
0

0
48.2a

100a

51.8a

7
Lifetime AIDU
abstainers
9.2%

Positively associated major or significant characteristic of profile (VT 4 or P < 0.001)

9.3a

2.2

28.7

64.0

2.8

40.3a

1.9

16.0

79.8a

52.6a

21.8

5
Low alcohol
users and
illicit-drugs
abstainers
25.5%

6
Occasional
or former
alcohol
users and
lifetime
illicit-drugs
abstainers
23.1%

22.7

16.5
a

73.7a

0.9

5.2

3.2

27.5
a

5.0

86.7a

31.9a

51.2a

0.9

5.6

10.4

67.1a

5.4

4.3

45.7a

41.9a

8.2

3
Excessive
alcohol
users
14.3%

42.3a

1.0

19.3a

43.2
a

18.2

19.4

18.0a

60.7a

16.3

5.0

2
Regular
excessive
AIDU
7.9%

4
Regular
low alcohol
users and
lifetime
cannabis
abstainers
15.5%

Flag for alcohol or drug problemc

Other drugs

12.0

6.2

<1/month

29.8

29.1

Past 12-month abstention

a

14.7

58.8

Never

Cannabis use

5.8

64.8

Never

Frequency of excessive alcohol use

(5 drinks on any single occasion)

Never

1
Problematic
AIDU
4.6%

Frequency of alcohol use

Profile

Percentage
of total
sample

Table 1 Profiles of AIDU in the Canadian population, %

Original Research

www.LaRCP.ca

www.TheCJP.ca

81.1a
18.9

49.0
51.0

Profile

Male

Female

23.6
3.4

30.5
15.0

4564

65

11.4
42.6a
57.1a

12.8
25.4
50.6

Separated, divorced, or widowed

Single (never married or common

law)

43.0

30.6
2.3
19.7
91.3
39.6

14.7
10.1
15.6
77.5
63.2

Looking for job or laid off

Family responsibility

Other

Born in Canada

Spirituality important, yes

25.4

29.5

Significant correlate positively associated with profile (VT 4 or P < 0.001)

13.7

16.4

16.5

14.6

47.0

49.2

91.9a

90.1

22.6a

11.2

20.2a

8.3

18.5

Disability

15.7

13.2

11.8

Studying

17.7

8.6

64.8a

33.3a

8.3

58.4

1.0

21.2

33.2

9.9

62.0a

42.5a

7.2

50.3

1.0

22.1

36.4
a

34.8a

42.6

57.4a

40.4a

32.5

67.5a

3
Excessive
alcohol
users
14.3%

Retired

Not working in the last week, reason

Has had a job throughout past year

46.1

61.8

Married or common law

Marital status

35.0

30.6

3044
a

38.0a

24.0

1529

Age, years

1
Problematic
AIDU
4.6%

Percentage
of total
sample

2
Regular
excessive
AIDU
7.9%

Table 2 Sociodemographic and cultural background correlates, %

68.0a

75.4

14.3

5.7

9.3

63.9

80.5a

16.8

13.2a

15.6

12.4

10.9

3.1
13.0

31.2

57.6a

23.9

10.7

65.4a

26.5

72.8a

72.1

12.8

10.1

8.9

18.1a

10.6

39.4a

38.9a

20.7

18.3a

61.1

9.6

23.3

17.3
32.9a

34.4

37.5
62.5a

32.1a

23.9

53.3a

46.7

5
Low alcohol
users and
illicit-drugs
abstainers
25.5%

54.7a

54.3a

11.8

14.3

73.9a

26.3
a

42.4a

22.4

9.0

47.2

52.9a

4
Regular low
alcohol users
and lifetime
cannabis
abstainers
15.5%

6
Occasional
or former
alcohol
users and
lifetime
illicit-drugs
abstainers
23.1%

76.6a

45.6

13.2

12.9

15.8

12.0

21.1a

25.1

33.1

28.8a

15.3a

55.4

21.2a

25.0

25.9

28.0a

67.5a

32.5

7
Lifetime AIDU
abstainers
9.2%

Profiles and Mental Health Correlates of Alcohol and Illicit Drug Use in the Canadian Population: An Exploration of the J-Curve Hypothesis

The Canadian Journal of Psychiatry, Vol 58, No 6, June 2013 W 349

350 W La Revue canadienne de psychiatrie, vol 58, no 6, juin 2013

8.9
8.3a
1.8a
5.9a
2.3
5.5a
2.9a
11.4a
4.6
5

3.6
3.0
0.5
2.4
1.0
3.7
1.5
8.1
3.0
4

Lifetime suicidal thoughts

Past 12-month suicidal thoughts

Lifetime suicide attempt

Past 12-month suicide attempt

Lifetime mania

Past 12-month mania

Lifetime panic disorder

Past 12-month panic disorder

Lifetime social phobia

Past 12-month social phobia

Psychological distress,b median

Significant correlate positively associated with profile (VT 4 or P < 0.001)

Kessler Psychological Distress Scale (Kruskal-Wallis test; P < 0.001)

1.3

4.6

4.6
a

10.9a

1.8

11.7a

3.7a

5.3a

4.7a

2.2
a

2.9

0.6

4.5a

0.9

3.7

4.7

5.5a

6.8

5.8
17.7a

7.4
22.1a

13.8

3
Excessive
alcohol users
14.3%

16.7a

2
Regular
excessive
AIDU
7.9%

7.9
26.1a

4.8
13.2

Past 12-month major depression

a

17.6a

12.1

Lifetime major depression

Profile

1
Problematic
AIDU
4.6%

Percentage
of total
sample

Table 3 Mental health correlates, %

2.8

8.5

1.0

3.8

0.9

2.7

0.5

3.4

2.7

11.9

4.1

11.7

4
Regular low
alcohol users
and lifetime
cannabis
abstainers
15.5%

2.3

7.2

1.2

3.2

0.6

1.6

0.3

1.9

2.6

10.6

4.3

12.0

5
Low alcohol
users and
illicit-drugs
abstainers
25.5%

2.4

6.8

1.6

3.3

0.6

1.7

0.4

2.6

3.3

11.8

4.1

10.3

6
Occasional
or former
alcohol
users and
lifetime
illicit-drugs
abstainers
23.1%

2.4

4.4

0.8

1.9

0.5

1.1

0.2

1.0

1.9

6.0

3.4

8.5

7
Lifetime AIDU
abstainers
9.2%

Original Research

www.LaRCP.ca

Profiles and Mental Health Correlates of Alcohol and Illicit Drug Use in the Canadian Population: An Exploration of the J-Curve Hypothesis

3.3
5.0
3.1
6.0
4.7

9.8

4.1
6.4
b

Psychiatric hospitalization

Age at first consultation

During lifetime, has ever seen or talked on the telephone to a [professional] about emotions, mental health, or AIDU
b
Significant correlate positively associated with profile (VT 4 or P < 0.001)
c
If person has seen any of these professionals, age at first counselling or therapy (<20 years old)
d
At least one hospitalization for mental or substance use disorder

32.5
37.2
42.7
24.8

71.3

67.1
69.1
76.7

Never consulted

Social workera

Nurse

Psychologista

General practitioner

Profile

Psychiatrista

17.5
19.2
15.3

30.9

87.2b
79.5b
77.9
76.2

2.9

0.5
1.7

5.8

1.3

6.3

1.6

7.0

1.6

10.3b
13.5b

2.8
1.6

3.2

12.7b

10.5b

17.0
20.0

12.2b

17.2

12.6b
8.1

14.7

7.4

2.3

8.1

8.2
9.2

5.8

14.0
15.4

13.9

4.4
7.4
6.6
9.1

11.2b
7.9

9.0
11.9b

1
Problematic
AIDU
4.6%
www.TheCJP.ca

Percentage
of total
sample

Table 4 Mental-health services correlates, %

2
Regular
excessive
AIDU
7.9%

3
Excessive
alcohol users
14.3%

5
Low alcohol
users and
illicit-drugs
abstainers
25.5%
4
Regular low
alcohol users
and lifetime
cannabis
abstainers
15.5%

6
Occasional
or former
alcohol users
and lifetime
illicit-drugs
abstainers
23.1%

7
Lifetime AIDU
abstainers
9.2%

with a lower risk of more severe forms of

psychological disturbance, such as MADs.
People in profile 7 are lifetime abstainers. Female
sex and birth outside Canada are associated factors.
This finding is consistent with the results of an earlier
study in the United States that revealed an association
between lifetime abstinence from alcohol and female
sex and birth outside the country.29 Spirituality is
another important attribute of profile 7, as well as
of profiles 4 and 6. These findings support previous
ones to the effect that spirituality is associated with
a lower level of psychiatric disorders and AIDU
problems.30 Spirituality may also be associated
with a decreased level of distress. People in profile
7 exhibit the lowest level of psychiatric disorders,
a result in line with previous studies.14,19 The same
pattern emerges for mental health service use and
is consistent with previous findings regarding
immigrants to Canada.31

Limitations

Our study has several limitations that should be

noted. First, despite a large representative sample
and the originality of the statistical analysis, it
remains a correlational, cross-sectional study.
Our ability to infer a causal relation between the
variables was thus limited. Second, the survey did
not assess 2 key psychiatric disorders that are often
related to substance abuse and dependence: namely,
personality and psychotic disorders. Third, no
measures of lifetime AIDU problems were included.
Fourth, there was a possible recall bias, especially
for lifetime MHP. Finally, as the sample was large,
it is possible that the small median difference for the
psychological distress variable, between abstainers
and low and regular alcohol user, does not have any
significant practical or clinical impact.

Conclusion

To our knowledge, this is the first study to

simultaneously investigate AIDU that used a
clustering method in a general population sample
to evaluate the level of psychological distress,
psychiatric disorders, and mental health service
use correlates. Seven AIDU profiles were found
in the Canadian population. A linear relation
between psychiatric disorders and AIDU profiles
was found. Lifetime AIDU abstainers represent
a specific group with particular characteristics
regarding sociodemographics, cultural background,
and spiritual beliefs; they display the lowest level
of MHP and use of services for MHPs. Evidence
for a J-curve relation was found for the level of
psychological distress. This suggests that regular,
low alcohol consumption without illicit drug use may
have mental health benefits for the level of distress.
However, profiles have different sociodemographic
The Canadian Journal of Psychiatry, Vol 58, No 6, June 2013 W 351

Original Research

and cultural backgrounds that may account for the

differences between the groups, which also differ in the
importance attributed to spirituality. Longitudinal profiling
to better investigate the possible mental health benefits
of regular, low alcohol drinking should be considered for
future research.

Acknowledgements

Dr Dumais is supported by a postdoctoral grant from the

Quebec Health Research Fund.
The authors have no conflicts of interest to declare.

References

1. Anderson P, Chisholm D, Fuhr DC. Effectiveness and costeffectiveness of policies and programmes to reduce the harm caused
by alcohol. Lancet. 2009;373(9682):22342246.
2. Baumberg B. The global economic burden of alcohol: a review and
some suggestions. Drug Alcohol Rev. 2006;25(6):537551.
3. Compton WM, Thomas YF, Stinson FS, et al. Prevalence, correlates,
disability, and comorbidity of DSM-IV drug abuse and dependence
in the United States: results from the National Epidemiologic
Survey on Alcohol and Related Conditions. Arch Gen Psychiatry.
2007;64(5):566576.
4. Currie SR, Patten SB, Williams JV, et al. Comorbidity of major
depression with substance use disorders. Can J Psychiatry.
2005;50(10):660666.
5. Grant BF, Stinson FS, Dawson DA, et al. Prevalence and cooccurrence of substance use disorders and independent mood
and anxiety disorders: results from the National Epidemiologic
Survey on Alcohol and Related Conditions. Arch Gen Psychiatry.
2004;61(8):807816.
6. Flensborg-Madsen T, Becker U, Gronbaek M, et al. Alcohol
consumption and later risk of hospitalization with psychiatric
disorders: prospective cohort study. Psychiatry Res.
2011;187(12):214219.
7. Tolstrup J, Jensen MK, Tjonneland A, et al. Prospective study of
alcohol drinking patterns and coronary heart disease in women and
men. BMJ. 2006;332(7552):12441248.
8. Baumeister SE, Schumann A, Nakazono TT, et al. Alcohol
consumption and out-patient services utilization by abstainers and
drinkers. Addiction. 2006;101(9):12851291.
9. Zarkin GA, Bray JW, Babor TF, et al. Alcohol drinking patterns and
health care utilization in a managed care organization. Health Serv
Res. 2004;39(3):553570.
10. Rodgers B, Parslow R, Degenhardt L. Affective disorders, anxiety
disorders and psychological distress in non-drinkers. J Affect
Disord. 2007;99(13):165172.
11. ODonnell K, Wardle J, Dantzer C, et al. Alcohol consumption and
symptoms of depression in young adults from 20 countries. J Stud
Alcohol. 2006;67(6):837840.
12. Manninen L, Poikolainen K, Vartiainen E, et al. Heavy drinking
occasions and depression. Alcohol Alcohol. 2006;41(3):293299.
13. Alati R, Kinner S, Najman JM, et al. Gender differences in the
relationships between alcohol, tobacco and mental health in
patients attending an emergency department. Alcohol Alcohol.
2004;39(5):463469.

352 W La Revue canadienne de psychiatrie, vol 58, no 6, juin 2013

14. Graham K, Massak A, Demers A, et al. Does the association

between alcohol consumption and depression depend on how they
are measured? Alcohol Clin Exp Res. 2007;31(1):7888.
15. Alati R, Lawlor DA, Najman JM, et al. Is there really a J-shaped
curve in the association between alcohol consumption and
symptoms of depression and anxiety? Findings from the MaterUniversity Study of Pregnancy and its outcomes. Addiction.
2005;100(5):643651.
16. Caldwell TM, Rodgers B, Jorm AF, et al. Patterns of association
between alcohol consumption and symptoms of depression and
anxiety in young adults. Addiction. 2002;97(5):583594.
17. Skogen JC, Harvey SB, Henderson M, et al. Anxiety and depression
among abstainers and low-level alcohol consumers. The NordTrondelag Health Study. Addiction. 2009;104(9):15191529.
18. Skogen JC, Mykletun A, Ferri CP, et al. Mental and personality
disorders and abstinence from alcohol: results from a national
household survey. Psychol Med. 2011;41(4):809818.
19. Sareen J, McWilliams L, Cox B, et al. Does a U-shaped relationship
exist between alcohol use and DSM-III-R mood and anxiety
disorders? J Affect Disorders. 2004;82(1):113118.
20. Gravel R, Beland Y. The Canadian Community Health
Survey: mental health and well-being. Can J Psychiatry.
2005;50(10):573579.
21. Kessler RC, Ustn TB. The World Mental Health (WMH) Survey
Initiative Version of the World Health Organization (WHO)
Composite International Diagnostic Interview (CIDI). Int J Methods
Psychiatr Res. 2004;13(2):93121.
22. Kessler RC, Andrews G, Colpe LJ, et al. Short screening scales
to monitor population prevalences and trends in non-specific
psychological distress. Psychol Med. 2002;32(6):959976.
23. Greenacre M, Blasius J. Multiple correspondence analysis and
related methods. Boca-Raton (FL): Chapman-Hall; 2006.
24. Cattell RB. The scree test for the number of factors. Multivariate
Behav Res. 1966;1:245276.
25. Clausen SE. Applied correspondence analysis: an introduction. Sage
university paper series on quantitative applications in the social
sciences. Thousand Oaks (CA): Sage; 1998.
26. Lebart L. Complementary use of correspondence analysis and
cluster analysis. In: Greenacre M, Blasius J, editors. Correspondence
analysis in the social sciences. San Diego (CA): Academic Press;
1994. p 163178.
27. Greenacre M. Clustering the rows and columns of a contingency
table. J Classification. 1988;5:3951.
28. Lebart L, Morineau A, Piron M. Statistique exploratoire
multidimensionnelle. Paris (FR): Dunod; 2000.
29. Kerr WC, Greenfield TK, Bond J, et al. Racial and ethnic differences
in all-cause mortality risk according to alcohol consumption
patterns in the national alcohol surveys. Am J Epidemiol.
2011;174(7):769778.
30. Koenig HG. Research on religion, spirituality, and mental health:
a review. Can J Psychiatry. 2009;54(5):283291.
31. Kirmayer LJ, Weinfeld M, Burgos G, et al. Use of health care
services for psychological distress by immigrants in an urban
multicultural milieu. Can J Psychiatry. 2007;52(5):295304.

www.LaRCP.ca