Trends in attitudes toward people living with HIV, homophobia, and HIV transmission knowledge in Quebec, Canada (1996, 2002, and 2010)

Abstract

Abstract People living with HIV (PWHIV) face negative attitudes that isolate and discourage them from accessing services. Understanding negative attitudes and the social environment can lead to more effective health promotion strategies and programs. However, a scale to measure attitudes has been lacking. We developed and validated attitudes toward PWHIV Scale to examine trends in attitudes toward PWHIV in Quebec in 1996, 2002, and 2010. We also examined the relationship between negative attitudes toward PWHIV, homophobia, and knowledge about HIV transmission. The scale included 16 items and had a five-factor structure: F1 (fear of being infected), F2 (fear of contact with PWHIV), F3 (prejudicial beliefs toward groups at high risk of HIV), F4 (tolerance regarding sexual mores and behaviors), and F5 (social support for PWHIV). The validity and reliability of the scale were assessed and found to be high. Overall, Quebecers had positive attitudes toward PWHIV, with more negative attitudes observed in subgroups defined as male,≥50 years of age,<14 years of education, higher levels of homophobia, and below-average knowledge about HIV transmission. Scores were stable between 1996 and 2002, and increased in 2010. Negative attitudes were correlated with higher levels of homophobia and lesser knowledge about HIV transmission. The lowest scores for each factor were observed in the same subgroups that had low overall scores on the Attitudes Scale. The findings from this study can be used to intensify interventions that promote compassion for PWHIV, address attitudes toward homosexuality, and encourage greater knowledge about the transmission of HIV in these subgroups.

Full text

Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=caic20
AIDS Care
Psychological and Socio-medical Aspects of AIDS/HIV
ISSN: 0954-0121 (Print) 1360-0451 (Online) Journal homepage: www.tandfonline.com/journals/caic20
Trends in attitudes toward people living with HIV,
homophobia, and HIV transmission knowledge in
Quebec, Canada (1996, 2002, and 2010)
Alix Adrien, Marianne Beaulieu, Viviane Leaune, Michèle Perron & Clément
Dassa
To cite this article: Alix Adrien, Marianne Beaulieu, Viviane Leaune, Michèle Perron &
Clément Dassa (2013) Trends in attitudes toward people living with HIV, homophobia, and HIV
transmission knowledge in Quebec, Canada (1996, 2002, and 2010), AIDS Care, 25:1, 55-65,
DOI: 10.1080/09540121.2012.674195
To link to this article: https://doi.org/10.1080/09540121.2012.674195
Published online: 26 Apr 2012.
Submit your article to this journal
Article views: 559
View related articles
Citing articles: 1 View citing articles

Trends in attitudes toward people living with HIV, homophobia, and HIV transmission
knowledge in Quebec, Canada (1996, 2002, and 2010)
Alix Adrien
a,b
*, Marianne Beaulieu
c
, Viviane Leaune
a
, Miche
`le Perron
c
and Cle
´ment Dassa
c
a
Public Health Department Montreal Health and Social Services Agency, Montreal, QC, Canada;
b
Department of
Epidemiology, Biostatistics and Occupational Health, McGill University, Montreal, QC, Canada;
c
Department of Social and
Preventive Medicine, Universite´ de Montre´al, Montreal, QC, Canada
(Received 28 November 2011; final version received 6 March 2012)
People living with HIV (PWHIV) face negative attitudes that isolate and discourage them from accessing services.
Understanding negative attitudes and the social environment can lead to more effective health promotion
strategies and programs. However, a scale to measure attitudes has been lacking. We developed and validated
attitudes toward PWHIV Scale to examine trends in attitudes toward PWHIV in Quebec in 1996, 2002, and 2010.
We also examined the relationship between negative attitudes toward PWHIV, homophobia, and knowledge
about HIV transmission. The scale included 16 items and had a five-factor structure: F1 (fear of being infected),
F2 (fear of contact with PWHIV), F3 (prejudicial beliefs toward groups at high risk of HIV), F4 (tolerance
regarding sexual mores and behaviors), and F5 (social support for PWHIV). The validity and reliability of the
scale were assessed and found to be high. Overall, Quebecers had positive attitudes toward PWHIV, with more
negative attitudes observed in subgroups defined as male, ]50 years of age,B14 years of education, higher levels
of homophobia, and below-average knowledge about HIV transmission. Scores were stable between 1996 and
2002, and increased in 2010. Negative attitudes were correlated with higher levels of homophobia and lesser
knowledge about HIV transmission. The lowest scores for each factor were observed in the same subgroups that
had low overall scores on the Attitudes Scale. The findings from this study can be used to intensify interventions
that promote compassion for PWHIV, address attitudes toward homosexuality, and encourage greater
knowledge about the transmission of HIV in these subgroups.
Keywords: attitudes; HIV; scale; homophobia
Introduction
The HIV/AIDS epidemic in Quebec continues una-
bated among many members of the population, such
as injection drug users, prisoners, aboriginal people,
young gay men, and women (Public Health Agency of
Canada, 2010). People living with HIV (PWHIV) face
discrimination and moral judgment because HIV
infection is associated with behaviors portrayed in
the media as social taboo (Clarke, 2006; Li et al.,
2009; Mensah et al., 2008). The plight of PWHIV is
exacerbated by negative attitudes that stem from this
social construct of HIV/AIDS, which lead PWHIV to
isolate themselves, hide their HIV status, and deprive
themselves of available services, including HIV test-
ing and medical care (Aggleton, 2002; Chesney &
Smith, 1999; Fortenberry et al., 2002; Lichtenstein,
2003). Negative attitudes can also affect PWHIV if
they are present in health care professionals (Smith &
Mathews, 2007). One study found that health care
professionals with high homophobia were less willing
to care for PWHIV (Lohrmann et al., 2000).
There are no recent studies examining trends in
attitudes in developed countries since earlier work
conducted over 10 years ago in the USA and Europe
(Herek, Capitanio, & Widaman, 2002; Va
¨lima
¨ki,
Suominen, & Peate, 1998). Negative attitudes evolve
over many years and therefore, understanding the
social environment into which health promotion
strategies and programs are implemented is critical.
It is within this context that we began in 1996 to study
attitudes towards PWHIV in the general population
of Quebec. We also examined the relationship be-
tween negative attitudes toward PWHIV and homo-
phobic attitudes (Long & Millsap, 2008; Walch,
Orlosky, Sinkkanen, & Stevens, 2010) and lack of
knowledge about HIV (Brown, Macintyre, & Trujillo,
2003; Herek, 2002; Herek et al., 2002; National AIDS
Trust, 2011) in Quebec that have been described
elsewhere. We believed that public health initiatives
and programs aimed at promoting tolerance and
acceptance of PWHIV could potentially create a
supportive, rather than stigmatizing and confronta-
tional, environment. However, at that time little was
known about individuals who displayed negative
attitudes toward PWHIV and moreover, there was
no scale to measure such attitudes. Consequently, we
*Corresponding author. Email: aadrien@santepub-mtl.qc.ca
AIDS Care
Vol. 25, No. 1, January 2013, 5565
ISSN 0954-0121 print/ISSN 1360-0451 online
#2013 Taylor & Francis
http://dx.doi.org/10.1080/09540121.2012.674195
http://www.tandfonline.com

developed and validated an attitude toward PWHIV
Scale (Attitudes Scale), which was administered to the
general population of Quebec in 1996, 2002, and
2010.
In this paper, we describe the development of
the attitudes toward PWHIV Scale and report on
the trends in attitudes toward PWHIV in Quebec in
1996, 2002, and 2010.We also examine the relation-
ship between attitudes toward PWHIV, homophobia,
and knowledge about HIV transmission.
Methods
Samples and procedures
The study design was cross-sectional. The Attitudes
Scale was tested and validated in telephone surveys in
1996, and used in 2002 and 2010 with minor changes.
The sampling frame for the three surveys was French-
speaking and English-speaking adults (aged 1564
years) residing in households with telephones in the
province of Quebec. The survey was conducted by
a polling firm under the supervision of the research-
ers. Samples were drawn with a list-assisted random-
digit-dialing procedure using the software program
‘‘ASDE Survey Sampler.’’ Samples were stratified
and proportional to the administrative regions of
Quebec.
The numbers of completed interviews (and re-
sponse rates) were: 3501 (72%) in 1996, 1300 (73%)
in 2002, and 1489 (74%) in 2010. English-speaking
residents were oversampled in 2002 (23.9% of the
sample) and 2010 (25.4% of the sample) to increase
the power and reliability of statistical tests. In
households with more than one member, the next
person to celebrate a birthday was selected. Post-
stratification weights were used to adjust the sample
data to conform to the population’s parameters.
Measures
Attitudes toward PWHIV Scale
The Attitudes Scale is a 16-item instrument with items
scored on a 4-point Likert scale: 1 (strongly disagree),
2 (disagree), 3 (agree), and 4 (strongly agree). Items
were coded, with a higher score indicating more
positive attitudes. The scale has a five-factor struc-
ture: F1 (Fear/concerns about being infected) assesses
the fear of being near a person with HIV infection;
F2 (Fear personal contact) assesses the fear of any
personal contact with someone believed to be infected
with HIV; F3 (Prejudicial views toward groups
at high risk) assesses negative beliefs about groups
believed to be at high risk of HIV infection; F4
(Liberalism) assesses tolerance or rigidity with respect
to sexual mores and behaviors; and F5 (Social
support) assesses the quality of social support avail-
able to persons with HIV infection.
The Attitudes Scale was developed and completed
before the 1996 survey. In order to define the domain
of reference for the scale items, we undertook a
literature review on attitudes scales that measured
tolerance, compassion, and solidarity in relation to
PWHIV. We also conducted focus groups with
individuals from the general population and inter-
views with support groups for PWHIV and family/
close friends of someone who died of AIDS.
Content validity was initially assessed in 1996
by 20 experts from community groups, as well as
academic and public health experts from Canada.
The survey instruments were slightly different in
the three surveys as some items were added or
dropped to take into account new elements that
could play a role in transforming attitudes. We
identified these elements through literature review
and interviews in 2002. In 2010, the validation was
done by administering a questionnaire to community
group representatives and to academic and public
health experts from Canada. To ensure invariance of
the constructs being measured, the present study was
based solely on the core original 16 items used in the
three surveys.
Concurrent criterion-related evidence of valid-
ity was obtained by relating attitudes with socio-
demographic variables, level of homophobia, and
knowledge about HIV transmission. As expected, the
typical profile of subjects presenting the highest
attitudes scores could be characterized as younger,
more educated, unmarried or in a common-law
relationship, knowing PWHIV, less religious, and
having above-average knowledge about HIV trans-
mission and positive attitudes toward male homo-
sexuality (multiple regression of attitudes on age,
education, marital status, religion, knowing PWHIV,
level of homophobia, and knowledge about HIV
transmission; R20.475, p0.000).
Construct validity was based on factor analyses,
which showed a five-factor structure that was then
used to define ad hoc factor scores and general
attitudes scores. The five factors were low to moder-
ately correlated (r0.0350.521) (Table 1). The
factorial structure invariance was assessed by testing
the hypothesis that the form of the factor model has
the same dimensions and the same patterns of fixed,
free, and constrained parameters. The five-factor
structure proved invariant over time and hence, was
used in the present study (chi-square845.788,
df282; RMSEA0.033; NFI 0.999; CFI 1;
SRMR0.028).
56 A. Adrien et al.

The reliability of the scale was estimated with
Cronbach’s alpha and proved high: 0.8130.857.
Further estimation for single-factor scores showed
them to be adequate in 1996 (0.6000.679) evolving
to satisfactory in 2002 (0.6460.784) and 2010 (0.681
0.788).
Homophobia Scale
The Attitudes Scale was complemented by a 5-item
scale to measure attitudes toward male homosexu-
ality (Table 2). This scale was developed and vali-
dated in the US general population (Herek &
Capitanio, 1995, 1996). Factor analysis was per-
formed in 1996 and confirmed the unifactorial nature
of the scale, which was maintained in 2002 and 2010.
The scale was also found to have high reliability and
proved stable over time (Cronbach’s alpha: 0.813 in
1996, 0.805 in 2002 and 0.842 in 2010).
HIV Transmission Knowledge Scale
A 7-item scale to measure knowledge about HIV
transmission (Knowledge Scale) was administered in
1996 and 2010 (Table 3). Respondents assessed the
risk of transmission for each of seven situations on a
4-point scale (very low to very high); their answers
were then weighted to reflect experts’ criteria. The
final knowledge score was computed by averaging the
7 weighted scores and varied from 1 (very low
Table 2. Homophobia scale.
Item number Factor
1 Homosexual behavior between two men is just plain wrong.
2 Male homosexuality is merely a different kind of lifestyle that should not be condemned.
3 I think male homosexuality is disgusting.
4 Male homosexuality is a perversion.
5 Male homosexuality is a natural expression of sexuality in men.
Table 1. Attitudes toward PWHIV scale: five-factors structure and factors items.
Item
number Factors
F1: Fear/concerns about being infected
1 Being around a person with AIDS does not bother me.
2 I would not be worried about my health if a co-worker had AIDS.
3 It would not bother me if there was a rooming house for people with AIDS on my street.
F2: Fear personal contact
4 I could not be a friend of someone who has AIDS.
5 I would limit my contacts with a person who I know is infected with AIDS.
6 I would not hug someone with AIDS.
F3: Prejudicial views toward groups at high risk
7 People who are drug users deserve to get AIDS.
8 My support of a person living with AIDS depends on how the person was infected.
9 I am disgusted by persons who were infected during homosexual relations.
F4: Liberalism
10 To fight AIDS it is necessary that young people do not have sex.
11 Reinforcement of traditional sexual values will help to control AIDS.
12 The appearance of AIDS is linked to the fact that people have more sexual freedom.
13 The spread of AIDS is linked to the decline of moral values.
F5: Social support
14 Persons who have AIDS should have the right to work serving public as waiters, waitresses, cooks, hairdressers,
etc.
15 Children who are infected with the AIDS virus should be able to go to daycare.
16 Doctors with AIDS should be allowed to go on working with their patients.
AIDS Care 57

knowledge) to 4 (very high knowledge). The internal
consistency of the 1996 and 2010 7-item scales was
evaluated and presented Cronbach’s alpha values of
0.686 and 0.721, respectively, moderate values that
reflect the high homogeneity of the last two items
pertaining to transmission during sexual intercourse
with and without condom (the other five items form a
subscale with higher reliability: 0.756 in 1996 and
0.769 in 2010).
Data management and statistical analysis
Total scores and factor scores on the Attitudes Scale
were obtained by averaging the scores on the items
comprising them. One-factor ANOVA analyses fol-
lowed by post-hoc contrasts analyses were performed
on the scores using SPSS One-Way ANOVA and
Scheffe
´commands.
Results
Sociodemographic characteristics
Table 4 presents the sociodemographic data adjusted
for sex, age, region of residence, and language. Most
respondents were born in Canada. Sociodemographic
characteristics of respondents varied little between the
three surveys.
Attitudes toward PWHIV Scale
Total scores, and total scores by subgroup, are
presented by year in Table 5. Overall, Quebecers
had positive attitudes toward PWHIV, with stable
mean total scores between 1996 and 2002 (2.98 and
3.01, respectively) and a slight increase in 2010
(3.18).
Many of the differences observed between sub-
group categories in 1996 were also observed in 2002
and 2010 (Table 5). In all three years, scores were
significantly lower (more negative attitudes; pB0.01)
in respondents who were either male, aged ]50 years,
hadB14 years of education, or had higher levels of
homophobia. In 1996 and 2002, persons aged 3049
years had lower scores compared to persons B30
years old, whereas these age groups had the same
mean score in 2010. In both 1996 and 2002, persons
born outside Canada had lower scores than Cana-
dian-born respondents; no differences were observed
in 2010. In both 1996 and 2010, individuals with
below-average HIV transmission knowledge had low-
er total scores.
Among people born outside Canada, mean score
was stable between 1996 and 2002 with a slight
increase in 2010, whereas a steady, slight increase in
mean score was observed among Canadian-born
respondents. RespondentsB30 years of age showed
Table 3. HIV transmission knowledge scale.
Items
1. Kissing a person infected with the AIDS virus on the cheek?
2. Sharing a glass with a person infected with the AIDS virus?
3. When a person infected with the AIDS virus coughs or sneezes on you?
4. Shaking hands with a person infected with the AIDS virus?
5. Working in the same office as a person infected with the AIDS virus?
6. Having sex and using a condom with an infected person?
7. Having sex with an infected person without a condom?
Table 4. Sociodemographic characteristics of respondents adjusted for sex, age, region of residence, and language.
Year of survey
1996 2002 2010
n(%)
Sex: male 1716 (49.0) 631 (48.5) 751 (50.3)
Language spoken at home: French 3092 (88.3) 1053 (81.0) 1182 (79.3)
Place of birth: Canada 3219 (91.9) 1074 (82.6) 1313 (88.0)
Region of residence: Montreal 1971 (56.3) 598 (46.0) 702 (47.1)
Mean (SD)
Age (years) 37.0 (13.0) 40.8 (12.8) 41.8 (13.8)
Education (years) 12.4 (3.4) 14.1 (3.4) 14.3 (3.4)
SD, Standard deviation.
58 A. Adrien et al.

relatively stable mean scores from 1996 to 2002
and 2010, whereas mean scores increased for respon-
dents aged 3049 years and ]50 years from 2002 to
2010.
Changes in attitudes by scale factor
Table 6 shows total scores for each factor by year
and scores from subgroup analyses if the changes
in subgroup scores differed from changes observed in
the total scores or the difference in subgroup scores
was significant in 1996 but not in 2010.
Similar to overall scores on the Attitudes Scale, the
lowest total scores for each factor were observed
in respondents displaying the following characteristics
(scores not shown): greater homophobia,B14 years of
education, aged]50 years, born outside Canada, and
below-average HIV transmission knowledge.
In 1996, scores were significantly different by age
(pB0.01) for each of the five factors. Within each
factor, the younger age groups showed progressively
higher scores. In 1996 and 2002, more positive
attitudes were observed among Canadian-born
Table 5. Attitudes toward PWHIV scale (1996, 2002, and 2010): total scores and total scores by selected sociodemographic
characteristics, HIV transmission knowledge, and level of homophobia.
Year of survey
1996 2002 2010
NMean (SD) NMean (SD) NMean (SD) For tdf pConclusion
Total Score 3501 2.98 (0.49) 1300 3.01 (0.54) 1491 3.18 (0.50) 88.17 2 0.000 19962002 B2010
Region
Outside Montreal 1529 3.01 (0.48) 702 3.03 (0.51) 789 3.15 (0.46) 23.87 2 0.000 19962002B2010
Montreal 1971 2.96 (0.49) 598 2.99 (0.58) 702 3.22 (0.53) 67.41 2 0.000 1996 2002B2010
p0.002 p0.214 p0.009
Language
English 409 2.91 (0.54) 247 2.87 (0.63) 309 3.17 (0.57) 24.75 2 0.000 1996 2002B2010
French 3092 2.99 (0.48) 1053 3.05 (0.52) 1182 3.19 (0.48) 70.94 2 0.000 1996B2002B2010
p0.009 p0.000 p0.742
Place of birth
Outside Canada 282 2.80 (0.55) 226 2.75 (0.62) 179 3.09 (0.57) 19.25 2 0.000 1996 2002B2010
Canada 3219 3.00 (0.48) 1074 3.07 (0.51) 1313 3.20 (0.48) 81.42 2 0.000 1996B2002 B2010
p0.000 p0.000 p0.013
Sex
Male 1716 2.93 (0.48) 631 2.96 (0.52) 751 3.13 (0.49) 44.97 2 0.000 19962002B2010
Female 1785 3.03 (0.49) 669 3.07 (0.56) 741 3.24 (0.49) 45.62 2 0.000 1996 2002B2010
p0.000 p0.000 p0.000
Age (years)
B30 1098 3.11 (0.43) 266 3.18 (0.50) 337 3.23 (0.46) 9.64 2 0.000 19962002 2010
3049 1690 2.98 (0.48) 668 3.03 (0.52) 637 3.23 (0.48) 64.37 2 0.000 19962002B2010
]50 690 2.78 (0.51) 357 2.87 (0.57) 500 3.09 (0.53) 52.23 2 0.000 1996 B2002B2010
p0.000 p0.000 p0.000
Education (years)
B14 2324 2.91 (0.48) 575 2.93 (0.55) 631 3.06 (0.49) 26.00 2 0.000 19962002B2010
]14 1165 3.13 (0.46) 717 3.08 (0.53) 854 3.27 (0.48) 34.47 2 0.000 19962002B2010
p0.000 p0.000 p0.000
HIV transmission knowledge
a
BMean Score 1531 2.77 (0.49) - - 579 2.95 (0.50) -7.19 2108 0.000 1996 B2010
]Mean Score 1969 3.14 (0.42) - - 912 3.34 (0.43) -11.51 2880 0.000 1996B2010
p0.000 p0.000
Level of Homophobia
b
BMean Score 1818 3.21 (0.40) 733 3.26 (0.43) 852 3.41 (0.40) 67.85 2 0.000 1996B2002B2010
]Mean Score 1681 2.73 (0.45) 566 2.69 (0.51) 639 2.88 (0.45) 32.76 2 0.000 19962002B2010
p0.000 p0.000 p0.000
SD, Standard deviation.
a
Higher scores indicate greater knowledge about HIV transmission. The scale was not administered in 2002.
b
Higher scores indicate greater homophobia and more negative attitudes toward male homosexuality.
AIDS Care 59

Table 6. Attitudes toward PHIV scale (1996, 2002, and 2010): total scores by scale factor and selected subgroups.
Year of Survey
1996 2002 2010
NMean (SD) NMean (SD) NMean (SD) For tdf pConclusion
F1: Fear/concerns about being infected
Total score 3501 3.43 (0.59) 1300 3.32 (0.68) 1491 3.49 (0.57) 27.06 2 0.000 19962002 B2010
Place of birth
Outside Canada 282 3.26 (0.69) 226 3.10 (0.81) 179 3.41 (0.65) 9.23 2 0.000 1996 2002B2010
Canada 3219 3.45 (0.58) 1074 3.37 (0.64) 1313 3.50 (0.55) 14.6 2 0.000 19962002B2010
p0.000 p0.000 p0.066
Age (years)
B30 1098 3.54 (0.55) 266 3.50 (0.59) 337 3.56 (0.50) 0.91 2 0.402 1996 20022010
3049 1690 3.40 (0.59) 668 3.31 (0.67) 637 3.50 (0.57) 15.78 2 0.000 19962002B2010
]50 690 3.35 (0.65) 357 3.21 (0.74) 500 3.44 (0.61) 12.41 2 0.000 1996 2002B2010
p0.000 p0.000 p0.011
HIV transmission knowledge
a
BMean Score 1531 3.25 (0.66)  579 3.28 (0.61) 0.89 1112 0.371 1996 2010
]Mean Score 1969 3.57 (0.49)  912 3.63 (0.49) 2.63 2880 0.009 1996 B2010
p0.000 p0.000
F2: Fear personal contact
Total score 3498 3.30 (0.65) 1299 3.25 (0.79) 1491 3.46 (0.66) 41.37 2 0.000 19962002 B2010
Place of birth
Outside Canada 282 3.12 (0.73) 226 2.96 (0.93) 179 3.40 (0.71) 15.92 2 0.000 1996 2002B2010
Canada 3217 3.31 (0.64) 1073 3.31 (0.74) 1313 3.47 (0.65) 28.64 2 0.000 19962002B2010
p0.000 p0.000 p0.194
Age (years)
B30 1098 3.47 (0.56) 266 3.48 (0.66) 337 3.56 (0.57) 3.30 2 0.037 1996 20022010
3049 1688 3.28 (0.63) 668 3.24 (0.79) 637 3.50 (0.62) 31.53 2 0.000 19962002B2010
]50 690 3.07 (0.75) 356 3.09 (0.82) 500 3.35 (0.74) 22.06 2 0.000 1996 2002B2010
p0.000 p0.000 p0.000
Factor 3: Prejudicial views toward groups at high risk
Total score 3498 3.05 (0.80) 1300 3.19 (0.82) 1491 3.38 (0.70) 93.62 2 0.000 1996B2002 B2010
Age (years)
B30 1096 3.10 (0.77) 266 3.21 (0.80) 337 3.38 (0.64) 17.52 2 0.000 19962002 B2010
3049 1690 3.09 (0.79) 668 3.26 (0.80) 637 3.45 (0.67) 53.15 2 0.000 1996B2002B2010
]50 688 2.87 (0.85) 357 3.05 (0.84) 500 3.30 (0.74) 40.61 2 0.000 1996 B2002B2010
p0.000 p0.000 p0.002
Factor 4: Liberalism
Total score 3501 2.45 (0.69) 1299 2.64 (0.76) 1491 2.75 (0.74) 104.10 2 0.000 1996B2002B2010
Age (years)
B30 1098 2.66 (0.61) 266 2.92 (0.65) 337 2.82 (0.67) 21.03 2 0.000 1996B2002 2010
3049 1690 2.45 (0.68) 667 2.64 (0.73) 637 2.84 (0.74) 73.63 2 0.000 1996B2002B2010
]50 690 2.12 (0.69) 357 2.44 (0.82) 500 2.60 (0.76) 65.29 2 0.000 1996 B2002B2010
p0.000 p0.000 p0.000
Factor 5: Social support
Total score 3501 2.83 (0.75) 1299 2.78 (0.76) 1491 2.97 (0.75) 27.03 2 0.000 19962002 B2010
Place of birth
Outside Canada 282 2.68 (0.77) 226 2.57 (0.79) 179 2.94 (0.80) 11.75 2 0.000 1996 2002B2010
Canada 3219 2.84 (0.74) 1074 2.82 (0.74) 1313 2.98 (0.74) 17.79 2 0.000 19962002B2010
p0.000 p0.000 p0.565
60 A. Adrien et al.

respondents for F1, F2, and F5. However, scores
were not different by place of birth in 2010.
Some notable trends in factor-based scores were
observed for subgroups defined by age and HIV
transmission knowledge (Table 6). For each factor,
trends in subgroup scores for the 3049 and ]50
years age groups were similar to trends in total scores.
However, for each factor, trends in subgroup scores
for theB30 years age group were different from
trends in total scores. For F1, subgroup (B30 years
old) scores were stable during 1996, 2002, and 2010,
whereas total scores decreased from 1996 to 2002 and
increased from 2002 to 2010. Scores for this subgroup
were also stable over the three survey years for F2
and F5, which showed stable total scores from 1996
to 2002 only and increased scores from 2002 to 2010.
For F3 and F4, total scores increased each year,
whereas subgroup scores increased from 2002 to 2010
only for F3 and from 1996 to 2002 only for F4.
Association between attitudes toward PWHIV,
homophobia, and HIV transmission knowledge
Scores on the Homophobia Scale were significantly
correlated to scores on the Attitudes Scale in each
year (Table 7). In 1996 and 2010, scores on the
Knowledge Scale were moderately correlated to
scores on the Attitudes Scale and, to a lesser degree,
to scores on the Homophobia Scale.
Homophobia Scale
As shown in Table 8, total scores on the Homophobia
Scale decreased significantly (less homophobic atti-
tudes) during 1996, 2002, and 2010 (2.38, 2.26, and
1.96, respectively). In each year, greater homophobic
attitudes were reported in: persons with less positive
attitudes toward PWHIV, individuals born outside
Canada, men, and respondents with B14 years of
education. Persons with below-average HIV transmis-
sion knowledge also showed greater homophobic
attitudes in 1996 and 2010.
Discussion
Attitudes toward PWHIV Scale
Our study suggests that the Attitudes Scale is a viable
tool for measuring attitudes toward PWHIV. The
results of the telephone surveys showed that the
psychometric properties of the scale were satisfactory
Table 6 (Continued )
Year of Survey
1996 2002 2010
NMean (SD) NMean (SD) NMean (SD) For tdf pConclusion
Age (years)
B30 1098 2.92 (0.73) 266 2.87 (0.73) 337 2.97 (0.76) 1.22 2 0.295 199620022010
3049 1690 2.83 (0.73) 668 2.78 (0.75) 637 3.01 (0.74) 18.58 2 0.000 19962002B2010
]50 690 2.69 (0.79) 357 2.70 (0.79) 500 2.93 (0.75) 16.03 2 0.000 1996 2002B2010
p0.000 p0.017 p0.207
SD, Standard deviation.
a
Higher scores indicate greater knowledge about HIV transmission. The scale was not administered in 2002.
Table 7. Correlations between attitudes toward PWHIV, HIV transmission knowledge, and homophobia scales (1996, 2002,
and 2010).
Year of survey
1996 2002 2010
Scales NrNrNr
Attitudes-Homophobia 3499 0.61** 1299 0.63** 1491 0.59**
Attitudes-Knowledge
a
3501 0.49** - - 1491 0.44**
Homophobia-Knowledge
a
3499 0.29** - - 1491 0.24**
a
The HIV Transmission Knowledge Scale was not administered in 2002.
**p0.000.
AIDS Care 61

and stable over time. The invariance of the 16-item
scale allowed us to assess changes in attitudes from
1996 to 2010 and to evaluate its potential use in the
future.
Overall changes
Although the Attitudes Scale is composed of many
distinct factors, it is the scale taken as a whole that
provides the most complete representation of atti-
tudes toward PWHIV. We found positive trends in
attitudes toward PWHIV among Quebec residents.
This has also been observed in the Canadian popula-
tion (Calzavara et al., 2011).
We observed lower scores in persons with greater
homophobia and more negative attitudes toward
male homosexuality, havingB14 years of education,
born outside Canada, having below-average knowl-
edge about HIV transmission, and aged]50 years.
This is consistent with results from recent studies
conducted in Canada and the UK (Calzavara et al.,
2011; National AIDS Trust, 2011).
Our trends analysis showed that the score in 2010
for personsB30 years old was no longer different
from that of persons aged 3049 years old. This may
be explained in large part by a cohort effect, such that
the score in 2010 was increased by persons aged
3049 years who in 1996 and 2002 were part of
theB30 years age group. These changes in attitudes
may be attributed to changes in the perception of
HIV/AIDS. In the early years of the epidemic, HIV/
AIDS was often portrayed discriminatingly by the
Table 8. Homophobia scale (1996, 2002, and 2010): total scores and total scores by sociodemographic characteristics, HIV
transmission knowledge, and attitudes toward PWHIV.
Year of survey
1996 2002 2010
NMean (SD) NMean (SD) NMean (SD) For tdf pConclusion
Total score 3499 2.38 (1.03) 1299 2.26 (1.03) 1491 1.96 (0.96) 89.73 2 0.000 199620022010
Language
English 409 2.43 (1.20) 247 2.64 (1.26) 309 2.14 (1.17) 11.89 2 0.000 1996 20022010
French 3090 2.37 (1.00) 1052 2.18 (0.95) 1182 1.91 (0.90) 98.88 2 0.000 1996 20022010
p0.369 p0.000 p0.001
Place of birth
Outside Canada 282 2.89 (1.14) 225 2.98 (1.19) 179 2.43 (1.22) 12.38 2 0.000 199620022010
Canada 3217 2.33 (1.01) 1074 2.11 (0.93) 1313 1.89 (0.91) 99.77 2 0.000 199620022010
p0.000 p0.000 p0.000
Sex
Male 1715 2.64 (1.05) 631 2.47 (1.05) 751 2.17 (1.03) 53.12 2 0.000 199620022010
Female 1784 2.13 (0.95) 668 2.07 (0.98) 741 1.75 (0.84) 45.00 2 0.000 1996 20022010
p0.000 p0.000 p0.000
Age (years)
B30 1098 2.23 (1.01) 266 2.10 (1.07) 337 1.80 (0.95) 22.92 2 0.000 199620022010
3049 1689 2.37 (1.00) 668 2.27 (1.05) 637 1.96 (0.98) 36.92 2 0.000 199620022010
]50 690 2.63 (1.05) 357 2.37 (0.97) 500 2.05 (0.94) 48.89 2 0.000 199620022010
p0.000 p0.006 p0.001
Education (years)
B14 2323 2.48 (1.03) 575 2.37 (1.00) 631 2.13 (1.00) 29.38 2 0.000 199620022010
]14 1165 2.17 (0.99) 717 2.17 (1.05) 854 1.82 (0.91) 37.39 2 0.000 199620022010
p0.000 p0.001 p0.000
HIV transmission knowledge
a
BMean Score 1530 2.64 (1.05)  579 2.20 (1.04) 8.71 1 0.000 1996 2010
]Mean Score 1969 2.17 (0.96)  912 1.81 (0.88) 10.09 1 0.000 19962010
p0.000 p0.000
Attitudes toward PWHIV
b
BMean score 1664 2.91 (0.99) 632 2.82 (1.02) 670 2.50 (1.01) 38.94 2 0.000 1996 20022010
]Mean score 1835 1.90 (0.79) 667 1.74 (0.72) 822 1.51 (0.64) 76.29 2 0.000 1996 20022010
p0.000 p0.000 p0.000
SD, Standard deviation.
a
Higher scores indicate greater knowledge about HIV transmission. The scale was not administered in 2002.
b
Higher scores indicate a more positive attitude toward PWHIV.
62 A. Adrien et al.

media (Clarke, 2006; Li et al., 2009; Mensah et al.,
2008), whereas subsequently people aged30 years
were more likely to be exposed to efforts to counter
the stigma and discrimination targeted at groups
believed to be at high risk.
Changes by scale factor
Stronger intervention programs can be built by
examining attitudes by factor, which allows for a
better understanding of the underlying subcatego-
ries of attitudes: avoidance (particularly groups
considered at high risk), extreme precautions to
avoid contamination, lack of regard for victims, and
verbal expressions of fear of catching the disease
(Meisenhelder & LaCharite, 1989). An irrational fear
of HIV infection also significantly affects the atti-
tudes of health care workers toward PWHIV (Ding et
al., 2005; Hossain & Kippax, 2011). Of particular
concern, our findings show no improvements since
1996 in the score for social support (F5) in theB30
years age group. Social support is necessary if we are
to have programs (e.g., needle exchange programs
and secure injection sites) that will help maintain
contact with groups who are particularly vulnerable
to HIV.
In 1996, scores for all factors and overall total
score by HIV transmission knowledge were different
by birthplace. By 2010, respondents born outside
Canada were different from Canadian-born respon-
dents only with respect to two factors: F3 and F4,
which measure attitudes related to social and cultural
norms. This is consistent with results from 2010
which showed that persons born outside Canada no
longer presented scores on the HIV Transmission
Knowledge Scale that were different from Canadian-
born individuals. One result of their increased knowl-
edge may be the improvement in the social support
factor (F5) observed from 2002 to 2010. Indeed, it
would appear that persons born outside Canada,
despite showing more rigid social and cultural norms
regarding sexual and social behavior, may be sensitive
to public discussions and messages to promote great-
er compassion for PWHIV.
Homophobia Scale
The relationship between negative attitudes toward
PWHIV and homophobic attitudes has long been
established (Long & Millsap, 2008; Walch et al.,
2010). Some of the subgroups we studied (e.g.,
individuals born outside Canada) had significantly
higher scores on the Homophobia Scale and conse-
quently, more negative attitudes toward male homo-
sexuality. Although respondents born outside
Canada had scores on the Attitudes Scale that
increased steadily, their scores in 2010 were never-
theless different from those of Canadian-born re-
spondents for factors measuring attitudes related
to social and cultural norms (Factors F3 and F4).
Given that these factors contribute to the shaping of
homophobic attitudes in these communities (Adrien
et al., 1996), it is not surprising that persons born
outside Canada still had relatively high scores on the
Homophobia Scale in 2010. Identifying a champion
for change among health professionals and margin-
alized groups (Hackett, 2005; Satterlund, Cassady,
Treiber, & Lemp, 2001) and facilitating the path to
strong leadership and voice for PWHIV (Friedman &
Mottiar, 2004) are some ways to combat these
negative attitudes (Brown et al., 2003).
Survey limitations
Households and individuals without telephones, such
as the homeless and street youth, are underrepre-
sented. Also, given that the Knowledge Scale pre-
sented moderate internal consistency, these results
should be interpreted cautiously.
Conclusion
The value of this study lies not only in its findings,
but also in the development and validation of a vi-
able Attitudes Scale that can be used for measur-
ing attitudes toward PWHIV in other settings. The
results of this study are encouraging in that Quebe-
cers generally show positive attitudes toward
PWHIV. However, our findings also point to the
importance of intensifying public campaigns that
promote compassion for PWHIV, address attitudes
toward homosexuality, and encourage greater knowl-
edge about the transmission of HIV, particularly
in certain subgroups. If we are to see any progress
in preventing HIV and sexually transmitted infec-
tions, it is important to continue research aimed at
identifying the psychological dimensions that shape
attitudes toward PWHIV, as well as developing
public health policies that promote tolerance and
acceptance of PWHIV.
Acknowledgements
The authors gratefully acknowledge the contribution of
everyone who took part in the study interviews, focus
groups and consultations during the development of the
questionnaire. In particular, the authors thank the members
of the telephone survey consulting committees for their
invaluable insight and advice, as well as members of the
expert panel from community groups, academia and public
AIDS Care 63

health for their counsel during the development of the 1996
questionnaire and for their critique of the subsequent 2002
and 2010 questionnaires. The authors would especially like
to posthumously acknowledge the dedication and talent
of their colleague and co-author, Miche
`le Perron, who
contributed significantly to the data management and
analyses of the first two surveys. Funding for this study
was provided by the Service de lutte aux infections
transmissibles sexuellement et par le sang (SLITSS) of the
Ministe
`re de la Sante
´et des Services Sociaux du Que
´bec.
References
Adrien, A., Godin, G., Cappon, P., Singer, S.M., Maticka-
Tyndale, E., & Willms, D. (1996). Overview of the
Canadian study on the determinants of ethnoculturally
specific behaviours related to HIV/AIDS. Revue cana-
dienne de sante´ publique,87(Suppl. 1), S4S10.
Aggleton, P. (2002). Barcelona 2002: Law, ethics, and
human rights. HIV/AIDS-related stigma and discrimi-
nation: A conceptual framework. Canadian HIV-AIDS
Policy Law Review,7, 115116.
Brown, L., Macintyre, K., & Trujillo, L. (2003). Interven-
tions to reduce HIV/AIDS stigma: What have we
learned? AIDS Education and Prevention,15,4969.
Calzavara, L., Bunting, C., Allman, D., Worthington, C.,
Tyndall, M., & Adrien, A. (2011). Public attitudes,
knowledge and perception of HIV and AIDS in Canada:
A population study (Preliminary Report). Toronto,
ON: CIHR Social Research Centre in HIV Prevention,
Dalla Lana School of Public Health Social and
Behavioural Health Sciences.
Chesney, M.A., & Smith, A.W. (1999). Critical delays in
HIV testing and care: The potential role of stigma.
The American Behavioral Scientist,42, 11621174.
Clarke, J.N. (2006). Homophobia out of the closet in the
media portrayal of HIV/AIDS 1991, 1996 and 2001:
Celebrity, heterosexism and the silent victims. Critical
Public Health,16, 317330.
Ding, L., Landon, B.E., Wilson, I.B., Wong, M.D.,
Shapiro, M.F., & Cleary, P.D. (2005). Predictors and
consequences of negative physician attitudes toward
HIV-infected injection drug users. Archives of Internal
Medicine,165, 618623.
Fortenberry, J.D., McFarlane, M., Bleakley, A., Bull, S.,
Fishbein, M., Grimley, D.M., ... Stoner, B.P. (2002).
Relationships of stigma and shame to gonorrhea and
HIV screening. American Journal of Public Health,92,
378381.
Friedman, S., & Mottiar, S. (2004). A moral to the tale:
The treatment action campaign and the politics of
HIV/AIDS. Retrieved from http://ccs.ukzn.ac.za/files/
friedman%20mottier%20tac%20research%20report%
20short.pdf
Hackett, V. (Ed.). (2005). Champions for change. Reducing
HIV/AIDS stigma and discrimination in the Caribbean.
Greater Georgetown, Guyana: Caribbean Community
Secretariat / Pan Caribbean Partnership Against HIV/
AIDS.
Herek, G.M. (2002). Heterosexuals’ attitudes toward bisex-
ual men and women in the United States. Journal
of Sex Research,39, 264274.
Herek, G.M., & Capitanio, J.M. (1995). Black heterosex-
uals’ attitudes toward lesbians and gay men in the
United States. Journal of Sex Research,32,95105.
Herek, G.M., & Capitanio, J.M. (1996). Some of my best
friends: Intergroup contact, concealable stigma,
and heterosexuals’ attitudes toward gay men and
lesbians. Personality & Social Psychology Bulletin,22,
412424.
Herek, G.M., Capitanio, J.P., & Widaman, K.F. (2002).
HIV-related stigma and knowledge in the United
States: Prevalence and Trends, 1991/1999. American
Journal of Public Health,92, 371377.
Hossain, M.B., & Kippax, S. (2011). Stigmatized attitudes
toward people living with HIV in Bangladesh: Health
care workers’ perspectives. Asia-Pacific Journal of
Public Health,23, 171182.
Li, L., Wu, Z., Lin, C., Guan, J., Rotheram-Borus, M.J.,
Lu, Y. & NIMH Collaborative HIV/STD Prevention
Trial Group. (2009). Mass media and HIV/AIDS in
China. Journal of Health Communication,, 14, 424438.
Lichtenstein, B. (2003). Stigma as a barrier to treatment of
sexually transmitted infection in the American deep
south: Issues of race, gender and poverty. Social
Science & Medicine,57, 24352445.
Lohrmann, C., Va
¨lima
¨ki, M., Suominen, T., Muinonen, U.,
Dassen, T., & Peate, I. (2000). German nursing
students’ knowledge of and attitudes to HIV and
AIDS: Two decades after the first AIDS cases.
Journal of Advanced Nursing,31, 696703.
Long, W., & Millsap, C.A. (2008). Fear of AIDS and
homophobia scales in an ethnic population of uni-
versity students. The Journal of Social Psychology,148,
637640.
Meisenhelder, J.B., & LaCharite, C.L. (1989). Fear and
contagion: A stress response to AIDS. Advances in
Nursing Science,11,2938.
Mensah, M.N., Waugh, T., Lavoie, R., Dumas, J., Bernier,
M., Garneau, M.J., ... Otis, J. (2008). ‘‘The VIHsibi-
lite Project’’: HIV-positive people in the Quebec press
and community responses. AIDS Care,20, 596600.
National AIDS Trust. (2011). HIV: Public knowledge and
attitudes, 2010. Retrieved from http://www.nat.org.
uk/Media%20library/Files/Communications%20and%
20Media/HIV_awareness_report_2011DOWNLOAD.pdf
Public Health Agency of Canada. (2010). HIV and AIDS in
Canada. Surveillance Report to December 31, 2009.
Retrieved from http://www.phac-aspc.gc.ca/aids-sida/
publication/survreport/2009/dec/index-eng.php
Satterlund, T.D., Cassady, D., Treiber, J., & Lemp, C.
(2001). Strategies implemented by 20 local tobacco
control agencies to promote smoke-free recreation
areas, California, 20042007. Preventing Chronic
Disease,8, A111.
64 A. Adrien et al.

Smith, D.M., & Mathews, W.C. (2007). Physicians’ atti-
tudes toward homosexuality and HIV: Survey of a
California medical society-revisited (PATHH-II).
Journal of Homosexuality,52,19.
Va
¨lima
¨ki, M., Suominen, T., & Peate, I. (1998). Attitudes
of professionals, students and the general public to
HIV/AIDS and people with HIV/AIDS: A review
of the research. Journal of Advanced Nursing,27,
752759.
Walch, S.E., Orlosky, P.M., Sinkkanen, K.A., & Stevens,
H.R. (2010). Demographic and social factors asso-
ciated with homophobia and fear of AIDS in a
community sample. Journal of Homosexuality,57,
310324.
AIDS Care 65