Tridimensionality of alcohol use in Canada: patterns of drinking, contexts and motivations to drink in the definition of Canadian drinking profiles according to gender.
Fortin, Marilyn ; Moulin, Stephane ; Picard, Elyse 等
Gender is one of the most important attributes that structure life
experience and behaviours. Men and women have different social roles,
are exposed to different opportunities and constraints, and have
different resources, responsibilities and privileges. (1) This appears
to shape the social norms regarding how a person may drink alcohol (2)
as well as his or her opportunities to drink. (3)
Numerous studies have shown that women drink smaller quantities,
drink less often, and get intoxicated on fewer occasions. (4-6) While
the hypothesis of a convergence between drinking patterns of men and
women has been debated over time, this hypothesis has found limited
support. (4,7,8)
In another way, gender differences in drinking go beyond drinking
patterns. Men and women differ with regards to drinking contexts (3) and
drinking motives. (9) These drinking practice dimensions are
intertwined. (10-12) Over and above gender, the individual's
position in life course and in social structure have been shown to be
key factors in the patterning of drinking. (8,13,14) Hence, a
simultaneous analysis among use, contexts and motivations in the
definition of drinking practices of men and women will permit us to
highlight the complexity of drinking behaviours, and will offer a guide
to better develop gender-sensitive prevention related to alcohol use.
The aim of this paper is to explore the variability of drinking
profiles, and its determinants, among Canadian men and women who drink
regularly. The following three questions are addressed: 1) What are the
underlying typologies of drinking practices among men and women that
adequately represent the variability in drinking practices in Canada? 2)
In which respects do those typologies differ and in which are they
similar? 3) Is variability in female and male drinking practice related
to certain socio-economic factors?
METHOD
Data
The data come from the GENACIS Canada survey as part of the Gender,
Alcohol, and Culture: An International Study (www.genacis.org)
(2004-2005). (15) This survey was specifically designed to observe
gender differences related to drinking. Respondents aged between 18 and
77 years were chosen randomly, using the Random Digit Dialing method
(RDD). Computer-Assisted Telephone Interview (CATI) was used to collect
data. A total of 8,055 women and 6,012 men aged 18-77 years participated
in the study for a response rate of 53% (N = 14,067), comparable to
other Canadian population-based telephone surveys. (16)
From the original sample, 22% of the respondents reported no
alcohol consumption over the previous year (N = 3,139) and 18.5%
reported drinking less than once a month (N = 2,615). Moreover,
questions on drinking reasons in GENACIS survey were asked only to a
randomly selected subsample (N = 2,053). After excluding cases with any
missing values on studied variables, the final subsample included 871
men and 843 women (N = 1,714).
Measures
Drinking patterns. Two indicators of the drinking patterns were
used: drinking frequency and usual quantity consumed per drinking day.
The frequency of drinking was measured by the average annual frequency
of drinking, ranging from 'less than once a month' to
'every day'. The variable was recoded into three categories:
'three or more times a week', 'once or twice a
week', and 'once to three times a month', with drinkers
drinking less than once a month being excluded from our subsample. Then,
the usual quantity consumed per drinking day was measured as a
continuous variable ranging from 0 to 30 ('In the past 12 months,
on those days when you had any kind of beverage containing alcohol, how
many drinks did you usually have?'). The variable was recoded into
three categories: 'one or two', 'three or four', and
'five or more' drinks per day.
Drinking contexts. The survey assessed frequency of drinking in
various contexts by asking: 'Thinking back over the last 12 months,
how often did you drink in the following circumstances/
situations'. The situations included: 1) the circumstances (meal
and party), 2) the location (home, friend's home, bar/disco/
nightclub and restaurant), and 3) whether or not alcohol was consumed
alone. The response categories ranged from 'every day or nearly
every day' through 'once or twice a year' to 'never
in the last 12 months'. Each variable was dichotomized into
drinking in a particular social context 'more than once a
month' vs. 'less than once a month'.
Drinking motivations. 19 items measured drinking reasons, 15 using
a 5-point scale ('never', 'rarely',
'sometimes', 'often', 'always') and 4
using a 3-point scale ('never', 'sometimes',
'usually'). Both scales were recoded into 'never',
'sometimes' (rarely and sometimes) and 'usually'
(often and always). An exploratory principal component analysis was
performed to reduce the data (data not shown). Four motivations
(metavariables) were derived by adding items and estimating the mean
score: social motives, enhancement motives, disinhibition motives and
compensatory motives (see Appendix 1). Questions related to
disinhibition motives were developed by GENACIS researchers to observe
gender differences in alcohol drinking, for instance related to sexual
practices. The other three motives are consistent with research in the
field of alcohol. (17-19) These variables were recoded in two
categories, 'rarely' (lower than 2) and ' usually'
(2 and over).
Demographics. Age was derived from the month/year of birth and
recoded into three categories corresponding to three generations (young
adult, adult and senior) (18-25, 26-55, 56+ years). Education was used
as a proxy variable of the economic status. The original 7-category
variable ('What is the highest level of education you have
completed?') was recoded into 4 categories: 1) less than high
school level, 2) completed high school, 3) completed technical or
community college, 4) completed bachelor's degree or higher.
Statistical analysis
Latent class analysis (LCA) was used to explore configurations in
drinking practice. LCA allows for analysis of the interdependence of
observed individual characteristics and inference of the homogeneous
non-observed grouping of individuals (latent classes). (20) The
analytical strategy seeks to simultaneously introduce drinking pattern,
drinking context and drinking motivation variables, as they are all
assumed to contribute to the defining of a drinking practice. In a first
step, we tested the hypothesis of gender invariance by comparing
non-constrained models (baseline models) with models with covariates
(gender-constrained models) for between 2 and 10 classes. The G2
difference test was used to assess significant differences between these
models. (21,22) Thus, we estimated the best fit model according to
gender. As recommended by Nylund et al. (2007), (23) we used the
Bayesian Information Criterion (BIC) to determine the number of relevant
latent classes and evaluate the fit of each model. The lowest BIC value
indicates the best fit model. In order to measure the best
classification based on individual posterior class membership
probabilities, we used the measure of entropy, where the nearest value
to 1 represents the highest certainty in classification.
Following the latent class analysis, we performed multivariate
multilogistic regressions in order to analyze the appurtenance of a
drinking profile according to age and education, adjusted by province of
residence. (24) The model is evaluated in function of the BIC fit test
obtained in the classification first model. (20) A common latent class
between men and women has been used as class reference. For each outcome
in the analyses, the groups with an Odd Ratio of 1.00 served as
reference categories. Estimation method for model parameters was the
Maximum Likelihood (ML). LCAs and multilogistic regressions were
performed with SAS version 4.11. Univariate descriptive statistics were
computed with SPSS 20.0.
RESULTS
Drinking practice: A gendered class-model
Table 1 displays the distribution of the drinking and demographic
characteristics by gender and includes a chi-square test. Overall,
demographic characteristics indicated gender differences not only in the
drinking patterns but also in the drinking contexts, the drinking
motivations and according to education.
To test the gender difference in drinking practice, latent class
models were derived. We tested the difference in the G2 statistic
between the baseline models and the gender-constrained models (data not
shown). The difference has been significant for all models (p <
0.001), providing evidence of variance across gender. Therefore,
stratified analyses by gender were performed.
Then, we evaluated the best fit model for men and women (data not
shown). According to the BIC, the best fit is found in a six-class model
for men (3757.20) and in a five-class model for women (3158.77). The
entropy gave a better fit for men for six- to ten-class models (0.75 and
0.76) and for the eight-class model for women (0.77). Based on these
test values, a six-class model for men and a five-class model for women
have been selected.
Women's drinking practice
Table 2 presents the drinking characteristics by class for women.
Latent classes can be described as follows:
Occasional--19%: members drink one to three times a month, in
moderation (usually one or two drinks) and display no context
preference. Members of this class are neither likely to drink to
accompany a meal nor to drink alone. They drink to socialize, and to a
lesser extent, to become disinhibited.
Nutritional--34%: members have one or two drinks once or twice a
week, mainly to accompany a meal, in private settings (at their home or
at friends' homes). Although their motivation to drink is to
socialize, they are also likely to drink alone. This class is mostly
represented by women.
Social--17%: members consume one or two drinks, once or twice a
week or less often, to socialize and to a lesser extent to enhance how
they are feeling or to become disinhibited. For them, drinking is
integrated in a large spectrum of contexts, from meals to festive
contexts, in private as well as in public settings. Women social
drinkers are not likely to drink alone.
Instrumental--16%: members show similarities with social drinkers
regarding the large variety of drinking contexts but differ from social
drinkers in their drinking patterns and their drinking motivations. For
them, drinking is likely to be integrated in their everyday life, even
when alone, and they are likely to consume more than women in the
previous classes. Besides socialization and enhancement functions, they
also use alcohol to become disinhibited or to compensate.
Festive--14%: members differ from other classes by an occasional
(less than weekly for most members of this group) but heavier alcohol
intake (39% reported a usual quantity of five drinks or more). They
drink in contexts of parties, at bars, discos or nightclubs or in
private homes, to socialize or to enhance but also to become
disinhibited.
Men's drinking practice
Table 3 presents the drinking characteristics by class for men. The
five classes observed for women were echoed for men, but with a few
gender differences across all classes: men drink more and more often
than women, men are less likely than women to put forward disinhibition
motivations for drinking, and men are more likely to drink alone than
women.
Our analysis also reveals a sixth class for men that we labelled
private drinkers: Private drinkers (12%) mainly drink at home, mostly in
moderation and rarely more than once or twice a week. However, they are
not likely to drink to accompany a meal. For these drinkers, alcohol
serves multiple functions. Besides drinking to socialize and enhance,
they are also likely to drink to become disinhibited or to compensate.
Social status of women's and men's drinking practice
According to multivariate multilogistic regressions following the
LCA and performed on women and men separately, compared to adult women
between 26 and 55 years of age, young women are more likely to be
festive drinkers, and less likely to be nutritional drinkers than
occasional drinkers (see Table 4). Compared to women who did not finish
high school, women who obtained a university degree are more likely to
be nutritional drinkers, social drinkers or instrumental drinkers, and
less likely to be festive drinkers than occasional drinkers. Among men
(see Table 5), compared to the 26-55 year age group, young adults are
more likely to be instrumental or festive drinkers than occasional
drinkers, whereas those 56 years or older are more likely to be
nutritional drinkers and less likely to be instrumental or festive
drinkers. As for education, the likelihood of being nutritional drinkers
increases with the level of education and those with highest education
are less likely than those with lowest education to be festive drinkers.
DISCUSSION
The aim of this paper was to examine whether men and women present
distinctive drinking practices in Canada defined according to patterns,
contexts and motivations of drinking. Our multidimensional approach
defining types of drinking based on group definition allow us to observe
relationships between social dimensions of alcohol use and individual
drinking behaviours.
The results of the current study confirm the hypothesis of gender
variance in alcohol drinking practices. We confirmed that throughout all
typological structures that we could have retained in this study, men
and women were different enough in terms of drinking practice to
consider a stratified approach based on gender.
The gender difference was first observed in distinctive numbers of
types of drinking in each typology. While feminine typology presented
five types, the masculine typology expressed a sixth unique profile,
which is that of private drinkers. This type is particularly
distinguishable by consumption occurring solely at home. However, we did
not find an association between private consumption and social status,
preventing us from characterizing this subpopulation of drinkers. We
recommend that future studies explore other social and professional
statuses that could shed light on private consumption at home.
Conversely, it seems that the more alcohol is consumed daily, the
more that men and women differ in terms of the timing and motivation to
drink. This is especially true since men consume alcohol more frequently
than women and in larger quantities. These results confirm previous
findings in the literature. (24,25) Moreover, the more a man drinks, the
more he will express drinking in different social contexts, including in
a solitary context. However, we must also consider biological
differences between men and women which affect the ways of drinking.
(26)
On the other hand, an important association was observed between
feminine consumption and motivation to drink, in particular when
frequency and quantity are higher. Several motivators could account for
the increase in alcohol use. It seems that women used alcohol as a tool,
an outlet or a means much more frequently when consumption increased.
Social effects on drinking practice
In a second step, we examined how drinking practices are patterned
by age and education. Associations were clearly established between
young adults, a lower educational level and a festive consumption for
both men and women. Several studies point out that teenagers and
students consume larger quantities of alcohol, reflecting a practice
commonly referred to as binge drinking. (27,28) Nowadays, because of the
increase in the study period lasting beyond the age of 25 and the
pushing back of a graduation date and of the time when one earns a
decent revenue, young adults could have festive (and excessive) drinking
practices similar to those observed in teenagers and young students.
Drinking patterns of teenagers/students and young adults presenting
potentially similar risky drinking behaviours demonstrate the need to
adapt actions and prevention in alcohol use according to their similar
contexts of drinking, as well as their social differences. However, our
study also revealed two ways to express alcohol habits among young men.
Indeed, young men present two typical expressions of drinking: the
festive one discussed earlier, and the instrumental consumption. In that
sense, young age in men could be associated with two risky profiles:
binge drinking (festive profile) and a regular and solitary consumption
characterized by instrumental consumption.
With regard to the education level, the number of years at school
turned out to be an important factor in the establishment of three
female drinking practices: the nutritional, the social and the
instrumental. The association is even more pronounced between an
integrated consumption and a higher educational level. Previous findings
have highlighted a link between a riskier drinking practice in women
with a higher education level. (14) Instrumental consumption presents a
moderate to regular consumption and an association with compensatory
motivation not found in men. If we take into consideration alcohol use,
drinking occasions and several reasons to drink that could establish a
drinking habit, instrumental consumption by women could be viewed as a
risky practice.
Finally, occasional and nutritional drinking practices are related
to an older age for both men and women, and nutritional drinkers are
significantly more educated than individuals in other profiles. As they
grow older, adults trade their past excessive drinking habits observed
in younger ages for a more regular and moderate consumption, which is in
synch with their lifestyle and their daily responsibilities. But what we
observed specifically in these profiles is the association of alcohol
and social dimensions that express non-risky types of drinking, in
particular by the intentionality to consume or to socialize around a
meal. (29)
Limitations
There are a number of limitations in the study. First, the small
spectrum of range of drinking context indicators does not allow us to
fully explore the drinking circumstances in the Canadian drinking
practices. Second, the relationships between alcohol intake, motivations
to drink and drinking contexts were analytically derived rather than
observed. To validate (or invalidate) the drinking profiles depicted in
this study, further research must examine how people drink in specific
contexts and according to specific motivations, as we did in other
studies. (10) Finally, we must acknowledge the limitation associated
with the under-reporting in data collection (e.g., quantity-frequency).
The validity of self-reported alcohol intake in surveys has often been
questioned. (30-35) Future research studies might solicit alcohol intake
information at several time points, asking the specific context and
specific reason to drink in each instance. This strategy would increase
the report of alcohol use in order to best evaluate the multidimensional
way of drinking.
CONCLUSION
While demonstrating both inter--and intra-variability in alcohol
consumption among a population with heterogeneous drinking profiles
according to gender, this study reinforces the idea of adapting
promotion strategies and interventions in public health based on social
environment and social status in order to make them more efficient and
better suited to the target populations. Several studies already point
out the efficiency of multidimensional and contextual framework in
alcohol research and public health. (10-12,36) A broad contextualization
of use, partly based on its underlying motivations, offers a track to
better understand drinking practices between genders in contrast to
traditional measures commonly used in alcohol study. Future research is
needed in order to explore health and harmful outcomes associated with
male and female drinking practices in order to evaluate the simultaneous
effects of contexts and motivations on a drinker's health.
Appendix 1. Reasons to drink according to each motive
Social motive To be sociable
Because that is what your
friends do when they get
together
Because it is customary on
special occasions
Because it makes a social
gathering more enjoyable
Enhancement motive To celebrate
I like the 'feeling'
Drinking is exciting
To get high
Because it is fun
Because drinking makes you
feel good
Disinhibition motive It is easier to be open with
people
Feeling less inhibited about
sex
Sexual activity is more
pleasurable for you
You feel more sexually
attractive
Compensatory motive To forget your worries
To feel more self-confident
Drinking helps when you feel
depressed or nervous
To cheer up when you're in a
bad mood
To relax
REFERENCES
(1.) Bird CE, Riecker PP. Gender and Health: The Effect of
Constrained Choices and Social Policies. New York, NY: Cambridge
University Press, 2008.
(2.) Keyes KM, Grant BF, Hasin DS. Evidence for a closing gender
gap in alcohol use, abuse, and dependence in the United States
population. Drug Alcohol Depend 2008;93:21-29. PMID: 17980512.
(3.) Paradis C, Demers A, Nadeau L, Picard E. Parenthood, alcohol
intake, and drinking contexts: Occasio Furem Facit. J Stud Alcohol Drugs
2011;72:259-69. PMID: 21388599.
(4.) Wilsnack RW Vogeltanz ND, Wilsnack SC, Harris TR, Ahlstrom S,
Bondy S, et al. Gender differences in alcohol consumption and adverse
drinking consequences: Cross-cultural patterns. Addiction
2000;95:251-65. PMID: 10723854.
(5.) Bloomfield K, Grittner U, Kramer S, Gmel G. Social
inequalities in alcohol consumption and alcohol-related problems in the
study countries of the EU concerted action 'Gender, Culture and
Alcohol Problems: A Multinational Study.'. Alcohol and Alcoholism
2006;41(Suppl. 1):i26-36. PMID: 17030500.
(6.) Graham K, Bernards S, Demers A. Gender differences by province
in alcohol consumption and consequences: Results of the GENACIS Canada
study. In: Poole N, Greaves L (Eds.), Highs & Lows: Canadian
Perspectives on Women and Substance Use. Toronto, ON: Centre for
Addiction and Mental Health, 2007.
(7.) McPherson M, Casswell S, Pledger M. Gender convergence in
alcohol consumption and related problems: Issues and outcomes from
comparisons of New Zealand survey data. Addiction 2004;99:738-48. PMID:
15139872.
(8.) Wilsnack RW, Wilsnack SC, Kristjanson AF, Vogeltanz-Holm ND,
Gmel G. Gender and alcohol consumption: Patterns from the multinational
GENACIS project. Addiction 2009;104:1487-500. PMID: 19686518. doi:
10.1111/j.1360-0443.2009.02696.x.
(9.) Bailly RC, Carman RS, Forslund MA. Gender difference in
drinking motivations and outcomes. J Psychol 1991;125(6):649-56. PMID:
1806665.
(10.) Demers A, Kairouz S, Adlaf EM, Gliksman L, Newton-Taylor B,
Marchand A. Multilevel analysis of situational drinking among Canadian
undergraduates. Soc Sci Med 2002;55(3):415-24. PMID: 12144149. doi:
10.1016/S0277-9536(01)00258-1.
(11.) Kairouz S, Gliksman L, Demers A, Adlaf E. For all these
reasons, I do drink: A multilevel analysis of contextual reasons for
drinking among Canadian undergraduates. J Stud Alcohol 2002;63:600-8.
PMID: 12380857.
(12.) Kairouz S, Greenfield T. A comparative multi-level analysis
of contextual drinking in American and Canadian adults. Addiction
2007;102:71-80. PMID: 17207125.
(13.) Adlaf E, Blackburn J, Demers A, Kellner F, Single E, Webster
I. Social Determinants, Alcohol Consumption and Health: A Secondary
Analysis of Canada's Alcohol and Other Drugs Survey 1994. Ottawa,
ON: Canadian Centre on Substance Abuse, 1997.
(14.) Grittner U, Kuntsche S, Graham K, Bloomfield K. Social
inequalities and gender differences in the experience of alcohol-related
problems. Alcohol and Alcoholism 2013;47(5):597-605. PMID: 22542707.
doi: 10.1093/alcalc/ags040.
(15.) GENACIS. The European Commission (contract
QLG4-CT-2001-0196): "Gender & Alcohol--A Multinational
Study", the U.S. National Institute on Alcohol Abuse and Alcoholism
(grants R01AA04610 and R21AA12941), the Swiss Federal Office of
Education and Science (contract 01.0366) the German Ministry of Health
and Social Security, and the World Health Organization (WHO), 2005.
Available at: http://www.genacis.org/ (Accessed February 1, 2015).
(16.) CCSA. Canadian Addiction Survey, 2004: Microdata eGuide.
Ottawa: CCSA, 2007.
(17.) Cooper ML, Russell M, Skinner JB, Windle M. Development and
validation of a three-dimensional measure of drinking motives. Psychol
Assess 1992;4:123-32. doi: 10.1037/1040-3590.4.2.123.
(18.) Cox WM, Klinger E. A motivational model of alcohol use. J
Abnorm Psychol 1998;97:168-80. PMID: 3290306.
(19.) Kuntsche E, Knibbe R, Gmel G, Engels R. Why do young people
drink? A review of drinking motives. Clin Psychol Rev 2005;25:841-61.
PMID: 16095785.
(20.) Collins LM, Lanza ST. Latent Class and Latent Transition
Analysis. With Applications in the Social, Behavioral, and Health
Sciences. Wiley series in probability and statistics. Hoboken, NJ: John
Wiley & Sons, Inc., 2010.
(21.) Lanza ST, Collins LM, Lemmon DR, Schafer JL. PROC LCA: A SAS
procedure for latent class analysis. Struct Equation Modeling
2007;14(4):671-94. PMID: 19953201.
(22.) Agresti A, Yang M. An empirical investigation of some effects
of sparseness in contingency tables. Computational Statistics Data Anal
1986;5:9-21. doi:10.1016/0167-9473(87)90003-X.
(23.) Nylund KL, Asparouhov T, Muthen BO. Deciding on the number of
classes in latent class analysis and growth mixture modeling: A Monte
Carlo simulation study. Structural Equat Model: Multidisciplinary J
2007;14(4): 535-69.
(24.) Paradis C, Demers A, Picard E. Alcohol consumption: A
different kind of Canadian mosaic. Can J Public Health
2010;101(4):275-80. PMID: 21033531.
(25.) Health Canada. Enquete de surveillance canadienne de la
consommation d'alcool et de drogues. Faits saillants, 2011.
Available at: http://www.hc-sc.
gc.ca/hc-ps/drugs-drogues/stat/_2011/summary-sommaire-fra.php (Accessed
June 28, 2012).
(26.) Cole-Harding S, Wilson JR. Ethanol metabolism in men and
women. J Stud Alcohol 1987;48:380-87. PMID: 3613588.
(27.) Hugues SO, Power TG, Frances DJ. Defining patterns of
drinking in adolescence: A cluster analytic approach. J Stud Alcohol
1992;52:40-47.
(28.) Cooper ML. Motivations for alcohol use among adolescents:
Development and validation of a four-factor model. Psychol Assess
1994;6(2):117-28. doi: 10.1037/1040-3590.6.2.117.
(29.) Jayne M, Valentine G, Holloway SL. Alcohol, Drinking,
Drunkenness: (Dis) Orderly Spaces. Surrey, England: Ashgate Publishing
Limited, 2011.
(30.) Midanik LT. Perspectives on the validity of self-reported
alcohol use. Br J Addict 1989;84:1419-23. PMID: 2692741.
(31.) Gronbaek M, Heitmann BL. Validity of self-reported intakes of
beer, wine and spirits in population studies. Eur J Clin Nutr
1996;50:487-90. PMID: 8862487.
(32.) Sommers MS, Dyehouse JM, Howe SR, Lemmink J, Volz T, Manharth
M. Validity of self-reported alcohol consumption in nondependent drinker
with unintentional injuries. Alcohol Clin Exp Res 2000;24:1406-13. PMID:
11003207.
(33.) Gmel G, Graham K, Kuendig H, Kuntsche S. Measuring alcohol
consumption Should the 'graduated frequency approach become the
norm in survey research? Addiction 2006;101(1):16-30. PMID: 16393189.
doi: 10.1111/j.1360-0443.2005.01224.x.
(34.) Greenfield TK, Kerr WC. Alcohol measurement methodology in
epidemiology: Recent advances and opportunities. Addiction
2008;103(7):1082-99. doi: 10.1111/j.1360-0443.2008.02197.x.
(35.) Rehm J. Measuring quantity, frequency and volume of drinking.
Alcohol Clin Exp Res 1998;22(Suppl. 2):4S-14S. doi:
10.1111/j.1530-0277.1998.tb04368.x.
(36.) Frohlich KL, Corin E, Potvin L. La relation entre contexte et
maladie: une proposition theorique. In: Frohlich KL, De Koninck M,
Demers A, Bernard P (Eds.), Les inegalites sociales de sante au Quebec.
Montreal, Quebec: Les Presses de l'Universite de Montreal, 2008.
Received: July 4, 2014
Accepted: December 26, 2014
Marilyn Fortin, PhD, [1,2] Stephane Moulin, PhD, [3] Elyse Picard,
MSc, [4] Richard E. Belanger, MD, [2,5] Andree Demers, PhD [3,6]
Author Affiliations
[1.] School of Psychology, Laval University, Quebec, QC
[2.] Population Health and Optimal Health Practices Research Unit,
CHU de Quebec Research Centre, Quebec, QC
[3.] Department of Sociology, University of Montreal, Quebec, QC
[4.] Lifestyle and Addiction Research Lab, Concordia University,
Quebec, QC
[5.] Department of Paediatrics, Centre mere-enfant Soleil du CHU de
Quebec, Laval University, Quebec, QC
[6.] Public Health Research Institute, University of Montreal,
Quebec, QC
Correspondence: Marilyn Fortin, Population Health and Optimal
Health Practices Research Unit, CHU de Quebec Research Centre, 2875,
boulevard Laurier, Edifice Delta II, Bureau 600, 6e etage, Quebec, QC
G1V 2M2, Tel: 418-525-4444, ext. 46581, E-mail:
marilyn.fortin.2@ulaval.ca
Sources of Funding: Funding for this research was provided by the
Canadian Institutes of Health Research (CIHR) to Katryn Graham (PI) and
Andree Demers (Co-PI) (Application no. 108626). Marilyn Fortin received
a Doctoral research award (no 182174) from the CIHR. English editing was
funded by the Public Health Research Institute of the University of
Montreal.
Disclaimer: The content of this study is solely the responsibility
of the authors.
Conflict of Interest: None to declare.
Table 1. Distribution of drinking characteristics and
demographic variables by gender
Men Women
(N = 871) (N = 843)
Frequency of drinking
1 to 3 times per month 29.9 45.0
1 or 2 times per week 40.4 36.4
3+ times per week 29.7 18.6
Number of drinks per
drinking day
1 or 2 49.7 70.1
3 or 4 29.2 21.4
5+ 21.1 8.5
Contexts of drinkingf
Meal 64.4 64.3
Party 56.5 50.4
Home 75.9 68.1
Friend's home 59.4 55.9
Bar, disco or nightclub 42.4 28.0
Restaurant 45.9 40.3
Alone 33.4 18.6
Motivations to drink!
To socialize 75.7 67.1
To enhance 45.6 38.7
To become disinhibited 25.7 31.7
To compensate 20.7 19.7
Social characteristics
Age, years
18-25 13.5 12.7
26-55 67.2 68.9
56+ 19.3 18.4
Education
Less than secondary school 13.2 7.5
High school 25.1 22.4
Technical, community college 35.4 39.0
or some university
Bachelor, post graduate or 26.3 31.1
professional degree
P Total
(N = 1714)
Frequency of drinking 0.001
1 to 3 times per month 37.3
1 or 2 times per week 38.4
3+ times per week 24.3
Number of drinks per 0.001
drinking day
1 or 2 59.7
3 or 4 25.3
5+ 14.9
Contexts of drinkingf
Meal ns 64.4
Party 0.05 53.5
Home 0.001 72.1
Friend's home ns 57.6
Bar, disco or nightclub 0.001 35.3
Restaurant 0.01 43.2
Alone 0.001 26.1
Motivations to drink!
To socialize 0.001 71.5
To enhance 0.05 42.2
To become disinhibited 0.05 28.6
To compensate ns 20.2
Social characteristics
Age, years
18-25 13.1
26-55 68.0
56+ 18.8
Education 0.001
Less than secondary school 10.4
High school 23.8
Technical, community college 37.2
or some university
Bachelor, post graduate or 28.6
professional degree
Note: ([dagger]) = More than once a month; ([double dagger]) =
Usually. ns = not significant.
Table 2. Latent classes marginal and conditional probabilities for
drinking patterns, drinking contexts and drinking motivations,
women (N = 843)
Latent class
Occasional Nutritional Social
Marginal probability 0.19 0.34 0.17
Conditional probability
Frequency of drinking
1 to 3 times per month 0.85 0.38 0.40
1 to 2 times per week 0.14 0.40 0.47
3+ times per week 0.02 0.22 0.13
Number of drinks per
drinking day
1 or 2 0.82 0.85 0.81
3 or 4 0.15 0.14 0.19
5+ 0.02 0.01 0.00
Contexts of drinkingf
Meal 0.12 0.88 0.83
Party 0.09 0.37 0.88
Home 0.21 0.83 0.80
Friend's home 0.14 0.46 0.92
Bar, disco or nightclub 0.11 0.04 0.55
Restaurant 0.06 0.32 0.79
Alone 0.03 0.21 0.07
Motivations ([double
dagger])
To socialize 0.55 0.46 0.75
To enhance 0.13 0.18 0.21
To become disinhibited 0.21 0.16 0.23
To compensate 0.07 0.05 0.01
N 162 283 140
Latent class
Instrumental Festive
Marginal probability 0.16 0.14
Conditional probability
Frequency of drinking
1 to 3 times per month 0.03 0.61
1 to 2 times per week 0.47 0.34
3+ times per week 0.50 0.05
Number of drinks per
drinking day
1 or 2 0.58 0.20
3 or 4 0.28 0.41
5+ 0.14 0.39
Contexts of drinkingf
Meal 0.95 0.22
Party 0.75 0.65
Home 0.97 0.51
Friend's home 0.86 0.59
Bar, disco or nightclub 0.46 0.56
Restaurant 0.78 0.19
Alone 0.54 0.07
Motivations ([double
dagger])
To socialize 0.92 0.95
To enhance 0.92 0.82
To become disinhibited 0.62 0.57
To compensate 0.61 0.45
N 139 119
([dagger]) = More than once a month; $ = Usually.
Table 3. Latent classes marginal and conditional probabilities
for drinking patterns, drinking contexts and drinking
motivations, men (N = 871)
Latent class
Occasional Private Nutritional
Marginal probability 0.17 0.12 0.18
Conditional probability
Frequency of drinking
1 to 3 times per month 0.66 0.46 0.16
1 to 2 times per week 0.30 0.41 0.38
3+ times per week 0.03 0.14 0.46
Number of drinks per
drinking day
1 or 2 0.65 0.45 0.88
3 or 4 0.26 0.36 0.11
5+ 0.09 0.17 0.01
Contexts of
drinking ([dagger])
Meal 0.27 0.28 0.88
Party 0.17 0.11 0.33
Home 0.34 0.58 0.96
Friend's home 0.18 0.15 0.42
Bar, disco or nightclub 0.15 0.17 0.04
Restaurant 0.15 0.13 0.37
Alone 0.09 0.30 0.43
Motivations ([double
dagger])
To socialize 0.52 0.98 0.66
To enhance 0.05 0.73 0.17
To become disinhibited 0.10 0.38 0.06
To compensate 0.00 0.52 0.03
N 145 101 160
Latent class
Social Instrumental Festive
Marginal probability 0.18 0.20 0.16
Conditional probability
Frequency of drinking
1 to 3 times per month 0.19 0.01 0.44
1 to 2 times per week 0.53 0.38 0.41
3+ times per week 0.28 0.61 0.16
Number of drinks per
drinking day
1 or 2 0.57 0.28 0.13
3 or 4 0.33 0.38 0.32
5+ 0.10 0.34 0.55
Contexts of drinking
([dagger])
Meal 0.90 0.96 0.36
Party 0.76 0.93 0.91
Home 0.91 0.98 0.66
Friend's home 0.92 0.91 0.78
Bar, disco or nightclub 0.61 0.75 0.72
Restaurant 0.81 0.77 0.33
Alone 0.34 0.63 0.15
Motivations ([double
dagger])
To socialize 0.63 0.98 0.83
To enhance 0.19 0.89 0.81
To become disinhibited 0.09 0.45 0.51
To compensate 0.00 0.51 0.24
N 156 171 138
Note: ([dagger]) = More than once a month; ([double dagger]) =
Usually.
Table 4. Logistic regression model of women's drinking profiles by
age and education
Nutritional Social
OR 95% CI OR 95% CI
Age, years
18-25 0.20 * [0.07-0.57] 1.64 [0.68-3.95]
26-55 1 -- 1 --
56+ 1.41 [0.82-2.45] 0.83 [0.40-1.75]
Education
Less than 1 -- 1 --
secondary
school
([dagger])
High school 1.05 [0.45-2.46] 0.69 [0.21-2.21]
diploma
Technical, 1.65 [0.72-3.78] 1.40 [0.47-4.16]
community,
college or
some university
Bachelor, post 3.42 * [1.37-8.51] 4.83 * [1.55-15.04]
graduate or
professional
degree
Instrumental Festive
OR 95% CI OR 95% CI
Age, years
18-25 1.29 [0.58-2.87] 4.99 * [2.37-10.48]
26-55 1 -- 1 --
56+ 0.92 [0.48-1.76] 0.08 * [0.02-0.33]
Education
Less than 1 -- 1 --
secondary
school
([dagger])
High school 1.70 [0.51-5.72] 0.25 * [0.09-0.68]
diploma
Technical, 2.59 [0.78-8.57] 0.28 * [0.11 -0.74]
community,
college or
some university
Bachelor, post 8.97 * [2.58-31.20] 0.32 * [0.10-0.96]
graduate or
professional
degree
Note: ([dagger]) Reference categories.
'Occasional' = reference profile.
*p <0.05.
Table 5. Logistic regression model of men's drinking profiles by
age and education
Private Nutritional
OR 95% CI OR 95% CI
Age, years
18-25 0.89 [0.22-3.58] 0.29 [0.04-2.05]
26-55 ([dagger]) 1 --
56+ 0.82 [0.36-1.85] 4.59 * [2.31-9.15]
Education
Less than secondary 1 --
school ([dagger])
High school diploma 1.44 [0.52-4.04] 2.16 [0.78-6.00]
Technical, community, 1.56 [0.57-4.26] 3.64 * [1.31-10.13]
college or some
university
Bachelor, post 1.17 [0.42-3.31] 4.32 * [1.58-11.78]
graduate or
professional
degree
Social Instrumental
OR 95% CI OR 95% CI
Age, years
18-25 0.82 [0.27-2.52] 4.14 * [1.74-9.85]
26-55 ([dagger]) 1 -- 1 --
56+ 0.71 [0.32-1.58] 0.51 * [0.24-1.06]
Education
Less than secondary 1 -- 1 --
school ([dagger])
High school diploma 2.32 [0.93-5.76] 1.20 [0.52-277]
Technical, community, 1.94 [0.77-4.85] 0.91 [0.40-2.10]
college or some
university
Bachelor, post 2.24 [0.88-5.75] 1.51 [0.66-3.47]
graduate or
professional
degree
Festive
OR 95% CI
Age, years
18-25 8.58 * [3.58-20.55]
26-55 ([dagger]) 1 --
56+ 0.10 * [0.02-0.46]
Education
Less than secondary 1 --
school ([dagger])
High school diploma 0.55 [0.23-1.32]
Technical, community, 0.72 [0.32-1.66]
college or some
university
Bachelor, post 0.16 * [0.04-0.53]
graduate or
professional
degree
Note: ([dagger]) Reference categories.
Occasional' = reference profile
* p < 0.05.
