CSIRO PUBLISHING
Sexual Health, 2014, 11, 495–504 http://dx.doi.org/10.1071/SH14102
Safer sex and condom use: findings from the Second Australian Study of Health and Relationships Richard O. de Visser A,I, Paul B. Badcock B,C, Chris Rissel D, Juliet Richters E, Anthony M. A. Smith B,H, Andrew E. GrulichF and Judy M. SimpsonG A
School of Psychology, Pevensey 1, University of Sussex, Falmer BN1 9QH, UK. Australian Research Centre in Sex, Health and Society, La Trobe University, 215 Franklin Street, Melbourne, Vic. 3000, Australia. C Centre for Youth Mental Health, University of Melbourne, Orygen Youth Health Research Centre, 35 Poplar Road, Parkville, Vic. 3052, Australia. D Sydney School of Public Health, Charles Perkins Centre (D17), University of Sydney, Sydney, NSW 2006, Australia. E School of Public Health and Community Medicine, University of New South Wales, Sydney, NSW 2052, Australia. F The Kirby Institute, Wallace Wurth Building, University of New South Wales, Sydney, NSW 2052, Australia. G Sydney School of Public Health, Edward Ford Building (A27), University of Sydney, Sydney, NSW 2006, Australia. H Deceased. I Corresponding author. Email: [email protected] B
Abstract. Background: It is important to have current and reliable estimates of the frequency and correlates of condom use among Australian adults. Methods: A representative sample of 20 094 men and women aged 16–69 years, from all states and territories, completed computer-assisted telephone interviews. The overall participation rate among eligible people was 66.2%. Results: Although most respondents had used a condom at some time in their lives, fewer than half of those who were sexually active in the year before being interviewed had used a condom in that year. Condom use in the last year was associated with youth, speaking a language other than English at home, bisexual identity, greater education, residence in major cities, lower income and having multiple sexual partners in the last year. One-quarter of respondents used a condom the last time they had vaginal intercourse and one-sixth of these were put on after genital contact. Condom use during most recent vaginal sex was associated with youth, lower income, having sex with a non-regular partner and not using another form of contraception. Condom use appears to have increased between 2001–02 and 2012–13. Conclusion: Consistent with other research, this study showed that condom use was strongly associated with partner type and use of other contraception. There may be a need to highlight among people with multiple sexual partners the fact that non-barrier methods of contraception do not offer protection against sexually transmissible infections. The finding that many condoms were applied after genital contact suggests a need to promote both use and correct use of condoms. Additional keywords: contraception, cross-sectional studies, health surveys, HIV prevention, national survey, sexual behaviour, STI prevention. Received 7 June 2014, accepted 23 August 2014, published online 7 November 2014
Introduction Condom use is an essential component of behavioural strategies to reduce the transmission of HIV and other sexually transmissible infections (STIs).1 Condoms are effective as contraceptives and provide protection against the transmission of many STIs because they provide a physical barrier to the passage of sperm and pathogens in semen or pre-ejaculatory fluid and because they limit possible Journal compilation CSIRO 2014
transmission of STIs via contact between infected mucosal surfaces.2,3 Research into patterns of condom use has taken three approaches: (1) to examine whether people have ever used condoms during a defined period (e.g. lifetime or in the past year); (2) to measure frequency of condom use during a defined period; and (3) to examine individual sexual encounters to identify factors that make condom use more or less likely. www.publish.csiro.au/journals/sh
496
Sexual Health
The First Australian Study of Health and Relationships (ASHR1) revealed that over 85% of people with experience of partnered sex had used condoms at some point in their lives, but that fewer than half of those who were sexually active in the last year reported condom use in the last year.4 These proportions were comparable to figures reported in studies of population-representative samples conducted around the same time, or subsequently, in the US, Britain and Europe.5–8 Surveys of population-representative samples have shown that condom use in the last year is most likely among people who are younger, better educated, have lower incomes, identify as bisexual or homosexual or have multiple sexual partners.4,7–12 Although many people report that they have used condoms at some point in their lives, fewer than half of heterosexually active men and women always use condoms.4,13 Consistent condom use appears to be more likely among younger people and is more likely with casual partners than with regular partners.4,13,14 Consistent condom use appears to be more likely among homosexually active men, especially for sex with casual partners.4 Surveys of representative samples in other countries reveal that fewer than 25% of respondents used a condom during their most recent heterosexual encounter.4,13,15 Studies of representative samples indicate that reporting condom use during people’s most recent instance of vaginal intercourse is significantly more likely among men, younger people, people who identify as bisexual, people with multiple partners in the last year, people who had sex with casual partners and people with higher levels of education.4,7,8,13–15 Condom use is significantly more likely if no other contraception is used, indicating that there tends to be a trade-off between condoms and other forms of contraception.4,10,12,16 Indeed, dual use of condoms and other forms of contraception is uncommon.10,17 However, increases in condom use among people using other forms of contraception could lead to significant reductions in the number of unplanned pregnancies and terminations or births.18 Homosexually active men are more likely than exclusively heterosexual men and women to report use of condoms during their most recent sexual encounter. Again, condom use is more likely among younger people and with casual partners than with regular partners.4,11 In addition to assessing whether condoms are used, it is important to determine whether they are used in ways that maximise their efficacy. The potential for transmission of STIs via pre-ejaculatory fluid or contact with infected mucosal surfaces means that condoms should be applied before penetration, not simply before ejaculation.19–21 However, research reveals that 10–15% of condoms are applied after initial genital contact.4,22 In addition to reducing the protection afforded to individuals, such behaviour has implications for the validity and reliability of self-reported condom use.22 Since ASHR1, ongoing efforts have been made to maintain and increase levels of condom use.1 One aim of this study was to provide up-to-date, reliable estimates of the frequency of condom use and correlates of condom use using data derived from a representative sample of Australian adults.
R. O. de Visser et al.
Methods The methodology used in the Second Australian Study of Health Relationships (ASHR2) is described elsewhere in this issue.23 Briefly, between October 2012 and November 2013, computer-assisted telephone interviews were completed by a representative sample of 20 094 Australian residents aged 16–69 years from all states and territories. Ethical approval was obtained from the researchers’ host universities. Respondents were selected using dual-frame modified random-digit dialling (RDD), combining directory-assisted landline-based RDD with RDD of mobile telephones. The overall participation rate among eligible people was 66.2%. To maximise the number of interviews with people who had engaged in less-common and/or more-risky behaviours, all respondents who had had no sexual partners in the previous year, who had had more than one partner in the previous year and/or who reported homosexual experience completed a long form of the survey instrument, which collected detailed data on their sexual attitudes, relationships and behaviours. Of the larger proportion of respondents who reported one partner in the previous year and no homosexual experience, 20% were randomly selected to complete the longform interview and the other 80% completed a short-form interview. As a consequence, 8577 completed the long-form interview and 11 517 completed the short-form interview. Answers to questions that occurred only in the long-form interview are reported after weighting to reflect the sample as a whole. All respondents who reported experience of vaginal or anal intercourse indicated whether they had ever used a condom. Respondents who were sexually active in the year before being interviewed indicated whether they had used a condom in the last year. Respondents who had been sexually active in the 6 months before being interviewed provided information about their frequency of condom use with casual partners during that period [Always/Often/Occasionally/Never]. This variable was recoded as dichotomous, distinguishing between respondents who always used condoms and those who did not. Casual partners were defined as someone other than a ‘live-in partner’ or a ‘regular non-live-in partner’. Respondents also provided information about condom use during their most recent sexual experience in the year before being interviewed. Questionnaire items pertaining to condom use are provided in Box 1. Respondents indicated their relationship to their partner and the age of their partner. They also indicated whether any contraception other than condoms was used. Respondents who used a condom indicated when it was put on. Correlates of the main outcomes examined in this paper included a range of demographic characteristics, which were recoded to facilitate analysis. Respondents’ ages were recoded into six groups (16–19, 20–29, 30–39, 40–49, 50–59 and 60–69 years). Languages spoken at home were recoded as English or a language other than English. Sexual identity (in answer to the question ‘Do you think of yourself as . . .’) was coded as heterosexual, homosexual or bisexual; too few respondents stated that they were ‘queer’, ‘other’, or ‘undecided’ to allow analysis of these groups. Respondents’ reported highest completed level of education was recoded to
Condom use in Australia
Sexual Health
497
Box 1. Ascertainment of men’s condom use
The questions were repeated with gender-appropriate wording for women and for men reporting male partners. Have you ever used condoms to have sex with a woman? Have you used condoms in the last 12 months to have sex with a woman? Have you used condoms in the last 6 months to have sex with a woman? During the past 6 months, how often was a condom used when having vaginal intercourse with your partner? Was that never, occasionally, often or always? The last time you had sex [with a woman], did you put your penis into her vagina? Was a condom used? Was the condom put on before your penis touched her vagina? Did you ejaculate inside her? [If respondent states ‘Yes’, interviewer clarifies whether ejaculation occurred in the condom.] The last time you had sex [with a woman], did you put your penis into her anus? Was a condom used when you did this? Was the condom put on before your penis touched her anus? Did you ejaculate inside her rectum? [If respondent states ‘Yes’, interviewer clarifies whether ejaculation occurred in the condom.]
distinguish between those who had not (yet) completed secondary school, those who had completed secondary school and those who had completed post-secondary education. Respondents’ postcodes were used with the Accessibility/ Remoteness Index of Australia to determine whether respondents lived in a major city, a regional area or a remote area (i.e. areas with relatively unrestricted, restricted and very restricted access to goods, services and opportunities for social interaction respectively).24 To approximate the gross annual household income quintiles reported by the Australian Bureau of Statistics for 2009–2010,25 respondents’ reported annual household income was grouped into five categories: less than $28 000, $28 000–$52 000, $52 000–$83 000, $83 000– $125 000 and more than $125 000. Respondents’ reported occupation was coded into the nine major categories of the Australian Standard Classification of Occupations26 and then recoded to distinguish among managerial/professional occupations, white-collar occupations and blue-collar occupations. Respondents’ reports of their number of sexual partners were used to produce a dichotomous variable, identifying respondents who reported more than one sexual partner in the year before being interviewed. Respondents’ reports of alcohol consumption frequency and volume were used to determine whether respondents’ alcohol consumption was in excess of Australian National Health and Medical Research Council (NHMRC) guidelines at the time of ASHR1 (28 standard drinks a week for men and 14 for women).27 The number of men reporting homosexual activity in the past year (n = 201) was too small to allow detailed analyses of correlates of condom use for this group. The main outcome variables were compared with those of ASHR1 to identify significant changes over time. Data were weighted to adjust for the probability of each respondent being selected for a landline or mobile phone interview, a long-form interview and (for landline participants) the number of in-scope adults in the household. Data were then weighted to match the Australian population
on the basis of age, gender, area of residence and telephone ownership (i.e. mobile telephone only vs other), resulting in an adjusted sample of 10 056 men and 10 038 women (total 20 094). The data were thus weighted to account for the specifics of our sample design and the fact that particular types of people were slightly over- or under-represented. Therefore, the data presented describe the Australian population aged 16–69 years, subject to the biases noted elsewhere in this issue.23 Weighted data were analysed using the survey estimation commands in Stata Version 11.2.28 The data were analysed using univariate logistic regression for dichotomous outcomes, including comparisons between ASHR1 and ASHR2. Most percentages are presented in this article without standard errors or 95% confidence intervals (CI). This decision was made to maximise both readability and brevity and is in keeping with the style of other studies of a similar scope and intent.6,13,29 Further information about the precision of estimates is found elsewhere in this issue.23 Due to the number of participants in ASHR2, it is important to recognise that often there is the statistical power to detect even small changes as statistically significant, but these do not necessarily correspond to significant differences in a public health sense. Results Prevalence of condom use Although the majority (90%) of respondents had used a condom at some time in their lives, fewer than half of the people who were sexually active in the year before being interviewed had used a condom in the last year. Among respondents with heterosexual experience, women were significantly less likely than men to report that they had ever used condoms (88.4% vs 91.5%, OR = 0.72, CI: 0.65–0.79, P < 0.001). Among respondents who reported heterosexual activity in the last year, women were significantly less likely than men to report that they had used condoms in the year before being interviewed (33.8% vs 39.0%, OR = 0.80, CI: 0.74–0.86,
498
Sexual Health
R. O. de Visser et al.
areas (P < 0.001) or had lower incomes (P < 0.001). Condom use was most likely among men with white-collar occupations and least likely among managers/professionals (P < 0.001). Condom use in the last year was significantly more likely among men who had more than one sexual partner in the last year (P < 0.001) and significantly less likely among men who consumed alcohol in excess of NHMRC guidelines (P = 0.002). Condom use for heterosexual activity in the previous year was significantly more likely among women who were younger
P < 0.001). Among men who reported ever having sex with men, 56% had ever used a condom for such activity and among men who had sex with men in the year before being interviewed, 69% had used a condom for such activity. Table 1 displays correlates of condom use for heterosexual activity in the year before being interviewed. Condom use was significantly more likely among men who were younger (P < 0.001), spoke a language other than English at home (P < 0.001), were bisexually identified (P = 0.02), had higher levels of education (P < 0.001), lived in major cities or remote
Table 1. Sociodemographic correlates of having used a condom for heterosexual activity among men and women in the last year Unless indicated otherwise, data show the percentage of category using a condom. Unadjusted odds ratios (OR) and 95% confidence intervals (CI) are given for having used a condom, relative to reference category Correlate
Men (n = 8435)
OR (95% CI)
Women (n = 8056)
OR (95% CI)
Overall
39.0
Age (years) 16–19 20–29 30–39 40–49 50–59 60–69
86.5 73.4 47.6 26.7 16.8 7.3
2.32 (1.54–3.48) 1 0.33 (0.27–0.40) 0.13 (0.11–0.16) 0.07 (0.06–0.09) 0.03 (0.02–0.04)
85.0 62.4 37.7 22.7 8.1 3.6
3.40 (2.25–5.15) 1 0.36 (0.31–0.43) 0.18 (0.15–0.21) 0.05 (0.04–0.07) 0.02 (0.02–0.03)
Language at home English Other
37.3 57.8
1 2.30 (1.90–2.78)
32.9 48.1
1 1.89 (1.52–2.35)
Sexual identity Heterosexual Homosexual Bisexual
38.8 68.8 55.0
1 –A 1.92 (1.14–3.24)
33.2 41.4 61.3
1 –A 3.19 (2.10–4.83)
Education Less than secondary Secondary Post-secondary
30.5 40.6 40.1
1 1.56 (1.32–1.85) 1.53 (1.30–1.80)
25.7 38.7 33.9
1 1.83 (1.51–2.20) 1.48 (1.25–1.75)
Region of residence Major city Regional Remote
41.8 31.6 42.4
1 0.64 (0.57–0.72) 1.02 (0.70–1.50)
36.8 27.1 25.5
1 0.64 (0.57–0.72) 0.59 (0.34–1.02)
Household income $125 000
57.5 42.7 37.4 36.7 33.1
0.55 0.44 0.43 0.37
1 (0.44–0.69) (0.36–0.54) (0.35–0.53) (0.30–0.45)
53.3 35.8 34.7 27.5 28.5
0.49 0.47 0.33 0.34
Occupational category Blue collar White collar Manager/professional
38.5 45.8 34.9
1 1.35 (1.17–1.56) 0.86 (0.76–0.96)
35.6 34.6 31.2
1 0.96 (0.80–1.15) 0.82 (0.68–0.99)
Excessive alcohol consumption No Yes
39.7 26.0
1 0.53 (0.36–0.79)
34.8 25.4
1 0.64 (0.42–0.96)
Partners in last year 1 2 or more
32.0 80.2
1 8.58 (7.19–10.23)
29.7 79.2
1 9.00 (7.02–11.55)
A
33.8
Because there were 10 or fewer respondents, the OR was not calculated.
1 (0.39–0.60) (0.38–0.57) (0.27–0.40) (0.28–0.41)
Condom use in Australia
Sexual Health
(P < 0.001), spoke a language other than English at home (P < 0.001), were bisexually identified (P < 0.001), had higher levels of education (P < 0.001), lived in major cities (P < 0.001) or had lower incomes (P < 0.001). Women in managerial occupations were significantly less likely than women in blue-collar occupations to have used a condom (P = 0.02). Condom use was significantly more likely among women who had more than one partner in the last year (P < 0.001) and significantly less likely among women who consumed alcohol in excess of NHMRC guidelines (P = 0.03).
Frequency of condom use Just under half (49%) of the respondents who had heterosexual vaginal intercourse with casual partners in the 6 months before being interviewed reported that they always used condoms, with similar frequencies of consistent condom use reported by men and women (P = 0.65). The majority (58%) of the 116 men who had anal homosexual intercourse with casual partners in the 6 months before being interviewed reported that they always used condoms.
Table 2. Sociodemographic correlates of always using condoms for vaginal intercourse in the last 6 months with casual partners Unless indicated otherwise, data show the percentage of category always using a condom. Unadjusted odds ratios (OR) and 95% confidence intervals (CI) are given for always having used a condom, relative to the reference category, separately for men and women Correlate
Men (n = 931)
OR (95% CI)
Women (n = 587)
OR (95% CI)
Overall
49.1
Age (years) 16–19 20–29 30–39 40–49 50–59 60–69
60.1 48.4 47.7 41.8 47.0 30.5
1.63 (0.83–3.18) 1 0.91 (0.48–1.73) 0.89 (0.46–1.74) 0.86 (0.48–1.53) 0.46 (0.20–1.05)
55.3 63.8 25.4 32.0 3.5 0.0
0.79 (0.31–2.01) 1 0.23 (0.11–0.51) 0.62 (0.25–1.50) 0.08 (0.02–0.42) –A
Language at home English Other
49.0 38.5
1 0.89 (0.24–1.45)
47.2 79.8
1 6.18 (0.69–55.06)
48.1 100.0 33.7
1 –A 0.50 (0.17–1.49)
47.2 27.0 50.8
1 –A 1.06 (0.45–2.50)
Education Less than secondary Secondary Post-secondary
42.2 46.1 52.1
1 1.12 (0.57–2.20) 1.62 (0.83–3.16)
46.9 46.4 48.8
1 1.04 (0.38–2.81) 1.02 (0.37–2.84)
Region of residence Major city Regional Remote
50.3 41.8 56.1
1 0.67 (0.41–1.08) 1.50 (0.43–5.26)
43.4 49.9 80.5
1 1.28 (0.63–2.59) 4.63 (0.52–41.14)
Household income $125 000
41.7 43.5 53.3 53.5 50.0
1.11 1.50 1.45 1.25
1 (0.57–2.17) (0.80–2.80) (0.73–2.89) (0.62–2.51)
58.1 33.2 60.8 24.9 43.5
0.39 1.04 0.31 0.94
Occupational category Blue collar White collar Manager/professional
45.5 50.3 50.8
1 1.20 (0.71–2.01) 1.39 (0.77–1.99)
43.4 48.6 46.6
1 0.97 (0.40–2.36) 0.93 (0.29–2.94)
Excessive alcohol consumption No Yes
49.4 46.2
1 0.88 (0.45–1.72)
51.3 31.1
1 0.43 (0.19–0.97)
Partners in last year 1 2 or more
31.7 52.7
1 1.57 (1.02–4.41)
51.1 45.4
1 0.69 (0.31–1.12)
Sexual identity Heterosexual Homosexual Bisexual
A
499
49.4
Because there were 10 or fewer respondents, the OR was not calculated.
1 (0.19–0.79) (0.31–3.55) (0.12–0.80) (0.33–2.68)
500
Sexual Health
R. O. de Visser et al.
Table 2 displays correlates of consistent condom use for vaginal intercourse with casual partners in the 6 months before being interviewed. Among men, consistent condom use was not significantly related to age (P = 0.20), language spoken at home (P = 0.25), sexual identity (P = 0.21), education (P = 0.17), region of residence (P = 0.47), income (P = 0.70), occupation category (P = 0.64) or whether men consumed alcohol in excess of NHMRC guidelines (P = 0.71). However, men who had multiple sexual partners in the past year were significantly more likely to report consistent condom use (P = 0.04). Among women, consistent condom use was significantly less likely among women aged 30 years or over (P < 0.001). There was a significant, but variable, association between consistent condom use and income (P = 0.03). Women who drank in excess of NHMRC guidelines were significantly less likely to report consistent condom use (P = 0.042). There was no significant association between always using condoms and language spoken at home (P = 0.10), sexual identity (P = 0.76), education (P = 1.00), region of residence (P = 0.33), occupational category (P = 0.99) or having more than one sexual partner in the year before being interviewed (P = 0.35). Condom use during the most recent sexual encounter The data in Table 3 indicate the proportion of respondents’ most recent sexual encounters in which a condom was used. Condoms were used in 23.3% of instances of vaginal intercourse, with men significantly more likely than women to report condom use (P = 0.01). Of the condoms used for vaginal intercourse, 15.2% were put on after initial genital contact. Among the small number of respondents who had anal intercourse during their most recent heterosexual encounter, condoms were not commonly used, with no sex difference in reports of condom use (P = 0.18). All condoms used during heterosexual anal intercourse were put on before anal penetration. Men’s reports of their most recent homosexual encounters revealed that most used a condom for insertive or
receptive anal intercourse, or both. Late application of condoms was reported for very few of the condoms used for anal intercourse. Table 4 displays correlates of condom use for vaginal intercourse during respondents’ most recent sexual encounters. Among men, condom use was significantly associated with being younger (P < 0.001), speaking a language other than English at home (P = 0.02), having a lower income (P = 0.001), having a white-collar occupation (P < 0.001), having sex with a casual or occasional partner (P < 0.001) and not using another form of contraception (P < 0.001). Condom use was not significantly related to sexual identity (P = 0.54), education (P = 0.40), region of residence (P = 0.08) or drinking in excess of NHMRC guidelines (P = 0.074). Among women, condom use was significantly associated with being younger (P < 0.001), identifying as bisexual (P = 0.003), having a very low income (P < 0.001), having sex with a casual/occasional partner (P < 0.001) and not using another form of contraception (P < 0.001). Condom use was not significantly related to speaking a language other than English at home (P = 0.17), education (P = 0.18), region of residence (P = 0.13), occupation (P = 0.65), or drinking in excess of NHMRC guidelines (P = 0.10). Analyses of condom use during the most recent experience of vaginal intercourse were re-run, restricting the sample to those potentially at risk of STI transmission or unintended pregnancy. People in monogamous relationships of at least 1 years’ duration were considered not at risk of STI transmission, so were excluded from these analyses. In these analyses, condom use was reported by 55% of men and 46% of women (a significant sex difference, P = 0.031) and late application of condoms was reported by 6% of men and 9% of women (a non-significant sex difference, P = 0.26). When the analyses were further restricted to exclude those using other contraception or not exposed to the risk of pregnancy, condom use was reported by 72% of men and 60% of women (P = 0.13) and late application was reported by 7% of men and 11% of women (a non-significant sex difference, P = 0.43).
Table 3. Condom use during most recent sexual encounters among men and women Unadjusted odds ratio (OR) and 95% confidence intervals (CI) are given for women relative to men
Heterosexual behaviour No. engaged in vaginal sex No. using condom for vaginal sex (%) Condom put on before genital contact No. engaged in anal sex No. using condom for anal sex (%) Condom put on before genital contact Homosexual behaviour No. engaged in insertive anal sex No. using condom for insertive anal sex (%) Condom put on before genital contact No. engaged in receptive anal sex No. using condom for receptive anal sex (%) Condom put on before genital contact
Men
Women
7210 1836 (25.5%) 86.0% 90 16 (24.3%) 100.0%
6938 1460 (21.1%) 81.3% 29 3 (16.1%) 100.0%
80 46 (56.7%) 91.4% 65 38 (58.9%) 97.5%
OR (95% CI)
0.80 (0.64–0.94) 0.74 (0.47–1.15) 0.60 (0.12–2.99) 1
– –
– –
– –
– –
Condom use in Australia
Sexual Health
501
Table 4. Sociodemographic correlates of condom use among men and women during most recent event of vaginal intercourse Unless indicated otherwise, data show the percentage of respondents in a given category using a condom. Unadjusted odds ratios (OR) and 95% confidence intervals (CI) are given for having used a condom, relative to reference category, separately for men and women. Correlate
Men (n = 7210)
OR (95% CI)
Women (n = 6938)
OR (95% CI)
Overall
25.5
Age (years) 16–19 20–29 30–39 40–49 50–59 60–69
70.2 45.7 31.1 16.3 11.7 3.9
2.79 (1.51–5.15) 1 0.53 (0.36–0.78) 0.23 (0.16–0.34) 0.16 (0.10–0.25) 0.05 (0.03–0.09)
61.2 39.0 20.8 11.8 4.7 0.9
2.46 (1.37–4.43) 1 0.41 (0.28–0.61) 0.21 (0.14–0.32) 0.08 (0.04–0.14) 0.01 (0.00–0.05)
Language spoken at home English Other
24.6 36.9
1 1.79 (1.12–2.88)
20.7 29.9
1 1.64 (0.81–3.31)
25.4 100.0 30.0
1 –A 1.26 (0.61–2.61)
20.5 14.2 39.5
1 –A 2.52 (1.47–4.33)
Education Lower secondary Secondary Post-secondary
22.2 27.4 25.2
1 1.32 (0.87–2.00) 1.18 (0.78–1.76)
16.8 24.1 20.9
1 1.57 (0.97–2.54) 1.30 (0.84–2.03)
Region of residence Major city Regional Remote
26.7 22.1 36.6
1 0.78 (0.59–1.02) 1.58 (0.80–3.11)
22.7 16.3 27.6
1 0.66 (0.48–0.91) 1.29 (0.29–5.81)
Household income $125 000
38.8 30.7 25.2 23.9 20.5
0.69 0.53 0.49 0.41
1 (0.42–1.15) (0.33–0.86) (0.31–0.79) (0.26–0.64)
42.7 15.8 19.9 16.4 17.2
0.25 0.33 0.26 0.28
Socioeconomic status Blue collar White collar Manager/professional
26.2 31.9 19.7
1 1.31 (0.95–1.82) 0.69 (0.52–0.92)
20.0 21.9 19.5
1 1.12 (0.67–1.85) 0.96 (0.57–1.64)
Excessive alcohol consumption No Yes
25.8 18.8
1 0.66 (0.42–1.04)
21.4 15.2
1 0.66 (0.40–1.08)
Partner type Regular Other
21.4 64.8
1 6.77 (4.85–9.45)
19.0 55.8
1 5.38 (3.37–8.59)
Used other contraception Yes No No needB
23.3 50.4 7.3
1 3.36 (2.45–4.60) –
18.4 47.8 3.5
1 4.06 (2.91–5.69) –
Sexual identity Heterosexual Homosexual Bisexual
21.1
1 (0.15–0.41) (0.20–0.55) (0.17–0.42) (0.18–0.44)
A
Because there were 10 or fewer respondents, the OR was not calculated. Pregnant, past menopause, infertile (self or partner); odds ratio not calculated.
B
Selected comparisons between ASHR1 and ASHR2 Comparisons between selected outcomes of ASHR1 and ASHR2 are presented in Table 5. The data revealed nonsignificant increases in condom use for vaginal intercourse— an increase of 2% for use any time in the last year (P = 0.06 for
men and women) and a significant increase of at least 4% for both sexes for use during the most recent sexual encounter. Condom use for homosexual anal intercourse also appeared to increase, but numbers of respondents were small and the increases were not statistically significant.
502
Sexual Health
R. O. de Visser et al.
Table 5. Prevalence of condom use during intercourse among men and women aged 16–59 years: comparison of the ASHR1 and ASHR2 People aged 60–69 years have been removed from the ASHR2 sample for comparison. Unadjusted odds ratios (OR) and confidence intervals (CI) are for the Second Australian Study of Health and Relationships (ASHR2) versus the First Australian Study of Health and Relationships (ASHR1) Type of intercourse Vaginal intercourse in past year No. engaged in behaviour % Using a condom
Men ASHR1 ASHR2 8438 7239 42.5% 44.3%
P-value OR (95% CI)
Women ASHR1 ASHR2
P-value OR (95% CI)
0.06 1.08 (1.00–1.16)
8266 7238 34.1% 37.6%
0.06
Vaginal intercourse in most recent heterosexual encounter No. engaged in behaviour 7574 6249 % Using a condom 24.7% 28.8%
0.03 1.23 (1.03–1.49)
7231 6322 17.6% 23.0%
0.002 1.40 (1.13–1.75)
Homosexual anal intercourse in past year No. engaged in behaviour 177 185 % Using a condom 59.1% 69.5%
0.15 1.57 (0.85–2.93)
–
–
–
–
Anal intercourse in most recent homosexual encounter No. engaged in behaviour 69 75 % Using a condom 46.1% 58.0%
0.37 1.61 (0.56–4.67)
–
–
–
–
Discussion This study applied all three approaches used in previous studies of patterns of condom use by: (1) examining whether people had ever used condoms in their lifetime and in the past year; (2) measuring frequency of condom use during the past 6 months; and (3) examining individual sexual encounters to identify factors that make condom use more or less likely. The data presented here show that although experience of condom use is widespread, condoms were used in only one-quarter of respondents’ most recent heterosexual encounters. For all three key measures of condom use for vaginal intercourse—any use in the last year, consistency with casual partners in the last 6 months and use during the most recent vaginal intercourse—younger respondents were significantly more likely to report condom use. For the measures of any use in the last year and use during the most recent incident of vaginal intercourse, condom use was significantly more likely among people with lower incomes, those who have had casual and/or multiple sexual partners or those not using other contraception. These patterns of correlates reflect those reported in ASHR1 and in population-representative surveys conducted in other countries.4,7–12 Condom use was significantly less common among older people than among younger people. This may reflect the observation that older people are less likely to have patterns of partnering linked to increased risk of STIs,12,30,31 but it is still notable that older people were less likely to always use condoms during encounters with non-regular sexual partners (Table 3). Age-related variations in condom use may help to explain the counter-intuitive finding that condom use was more likely among people with higher levels of education, but less likely among those with higher incomes; although young adults tend to have greater education than people from older generations, they also tend to have lower incomes. As in studies of non-representative samples, dual use of condoms and other forms of contraception was not common,
1.16 (0.99–1.39)
occurring in 23% of people’s most recent experiences of vaginal intercourse.16,17 It has been argued that adding condoms to other methods should be considered seriously as the first line of defence against unplanned pregnancy and STIs, given calculations of the number of unplanned pregnancies and terminations of pregnancy that could be avoided if more women added condom use to current contraceptive practices.18 However, there is a need to identify the cognitive and motivational correlates of dual use in order to better promote this behaviour.17,31 Late application of condoms is still quite common for vaginal intercourse, occurring in approximately one-sixth of most recent sexual encounters, but appears to be less common for heterosexual and homosexual anal intercourse. There remains a need for promotion of the correct use of condoms, not just use of condoms.19–23 Moreover, there is a need to continue to promote and encourage condom use.1 To this end, use should be made of the findings of meta-analytic studies that have identified key correlates of intended and actual condom use.32,33 These studies suggest that it is important to provide normative support for condom use and to improve individuals’ attitudes towards condom use and confidence in being able to acquire and use condoms.32 Particular attention should be given to nurturing the skills and self-confidence required to successfully negotiate condom use with sexual partners.33 In the 11 years since ASHR1, condom use for heterosexual and homosexual activity has increased, with statistically significant increases for heterosexual condom use and with condoms now used in nearly half of men’s most recent experiences of insertive anal intercourse. This matches observed trends for temporal increases in condom use reported in some countries.34,35 The National STI Strategy recognises the importance of condom use,[1] but the data presented here indicate that condoms are not used in many situations when their use could prevent the transmission of STIs, unintended pregnancy or both.
Condom use in Australia
Conflicts of interest None declared. Acknowledgements This study was funded by the National Health and Medical Research Council (grant no. 1002174). The authors are indebted to David Shellard and the staff of the Hunter Valley Research Foundation for managing data collection and undertaking the interviews for this study; and to the Social Research Centre for producing weights for the data. The authors also thank the 21 139 Australians who took part in the two phases of the project and so freely shared the sometimes intimate aspects of their lives. Professor Anthony Smith died during the course of this project and we intend this work to be a tribute to, and further example of, the extraordinary contribution his work has made to the sexual health and wellbeing of Australians.
References 1 Department of Health and Ageing (DoHA). The Second National Sexually Transmissible Infections Strategy: 2010–2013. Canberra: DoHA; 2010. 2 Roper WL, Peterson HB, Curran JW. Commentary: condoms and HIV/STD prevention: clarifying the message. Am J Public Health 1993; 83: 501–3. doi:10.2105/AJPH.83.4.501 3 Sparling P. Biology of Neisseria gonorrhoeae. In Holmes K, Mårdh P-A, Sparling P, Wiesner P, editors. Sexually transmitted diseases, 2nd edn. New York: McGraw Hill; 1990. pp. 131–47. 4 de Visser RO, Smith AM, Rissel CE, Richters J, Grulich AE. Sex in Australia: safer sex and condom use among a representative sample of adults. Aust N Z J Public Health 2003; 27: 223–9. doi:10.1111/ j.1467-842X.2003.tb00812.x 5 Dubois-Arber F, Spencer B. Condom use. In Hubert M, Bajos N, Sandfort T, editors. Sexual Behaviour and HIV/AIDS in Europe. London: UCL Press; 1998. pp. 287–302. 6 Laumann E, Gagnon J, Michael R, Michaels S. The social organisation of sexuality: sexual practices in the United States. Chicago: University of Chicago Press; 1994. 7 Jeffries WL, Dodge B. Male bisexuality and condom use at last sexual encounter: results from a national survey. J Sex Res 2007; 44: 278–89. doi:10.1080/00224490701443973 8 Mercer CH, Copas AJ, Sonnenberg P, Johnson AM, McManus S, Erens B, et al. Who has sex with whom? Characteristics of heterosexual partnerships reported in the second National Survey of Sexual Attitudes and Lifestyles. Int J Epidemiol 2009; 38: 206–14. doi:10.1093/ije/dyn216 9 Carrasco-Garrido P, López de Andrés A, Hernández Barrera V, Jiménez-Trujillo I, Santos-Sancho J, Jiménez-García R. Predictors of contraceptive methods among adolescents and young women residing in Spain. J Sex Med 2011; 8: 2431–8. doi:10.1111/j.17436109.2011.02346.x 10 Martinez G, Copen CE, Abma JC. Teenagers in the United States: sexual activity, contraceptive use, and childbearing, 2006–2010 national survey of family growth. Vital Health Stat 23 2011; Oct: 1–35. 11 Rosenberger JG, Reece M, Schick V, Herbenick D, Novak DS, Van Der Pol B, et al. Condom use during most recent anal intercourse event among a U.S. sample of men who have sex with men. J Sex Med 2012; 9: 1037–47. doi:10.1111/j.1743-6109.2012.02650.x 12 Schick V, Herbenick D, Reece M, Sanders SA, Dodge B, Middlestadt S, et al. Sexual behavior, condom use, and sexual health of Americans over 50: implications for sexual health promotion for aging adults. J Sex Med 2010; 7(Suppl. 5): 315–29. doi:10.1111/j.1743-6109. 2010.02013.x
Sexual Health
503
13 Layte RD, McGee HP, Quail A, Rundle K, Cousins G, Donnelly CD, et al. Irish study of sexual health and relationships. Dublin: Crisis Pregnancy Agency, and Department of Health and Children (DOHC); 2006. 14 Reece M, Herbenick D, Schick V, Sanders SA, Dodge B, Fortenberry JD. Condom use rates in a national probability sample of males and females ages 14 to 94 in the United States. J Sex Med 2010; 7266–76. (Suppl. 5): doi:10.1111/j.1743-6109.2010.02017.x 15 Sanders SA, Reece M, Herbenick D, Schick V, Dodge B, Fortenberry JD. Condom use during most recent vaginal intercourse event among a probability sample of adults in the United States. J Sex Med 2010; 7(Suppl. 5): 362–73. doi:10.1111/j.1743-6109.2010.0 2011.x 16 Ott MA, Adler NE, Millstein SG, Tschann JM, Ellen JM. The trade-off between hormonal contraceptives and condoms among adolescents. Perspect Sex Reprod Health 2002; 34: 6–14. doi:10.2307/3030227 17 de Visser R. Why do heterosexual young adults who use reliable contraception also use condoms? Results from a diary-based prospective longitudinal study. Br J Health Psychol 2007; 12: 305–13. doi:10.1348/135910706X111294 18 Pazol K, Kramer MR, Hogue CJ. Condoms for dual protection: patterns of use with highly effective contraceptive methods Public Health Rep 2010; 125: 208–17. 19 Ilaria G, Jacobs J, Polsky B, Koll B, Baron P, MacLow C, et al. Detection of HIV-1 DNA sequences in pre-ejaculatory fluid. Lancet 1992; 340: 1469. doi:10.1016/0140-6736(92)92658-3 20 Pudney J, Oneta M, Mayer K, Seage G, Anderson D. Pre-ejaculatory fluid as potential vector for sexual transmission of HIV-1. Lancet 1992; 340: 1470. doi:10.1016/0140-6736(92)92659-4 21 Killick SR, Leary C, Trussell J, Guthrie KA. Sperm content of preejaculatory fluid. Hum Fertil 2011; 14: 48–52. doi:10.3109/14647273. 2010.520798 22 de Visser RO, Smith AMA. When always isn’t enough: implications of the late application of condoms for the validity and reliability of self-reported condom use. AIDS Care 2000; 12: 221–4. doi:10.1080/ 09540120050001904 23 Richters J, Badcock PB, Simpson JM, Shellard D, Rissel C, de Visser RO, et al. Design and methods of the Second Australian Study of Health and Relationships. Sex Health 2014; 11: 383–96. doi:10. 1071/SH14115 24 Department of Health and Aged Care (DHAC). Measuring remoteness: accessibility/remoteness index of Australia (ARIA). Canberra: DHAC; 2001. 25 Australian Bureau of Statistics (ABS). Household wealth and wealth distribution, Australia, 2009–2010. Canberra: ABS; 2011. 26 Australian Bureau of Statistics (ABS). Australian standard classification of occupations, 2nd edn. Catalogue no. 1220.0. Canberra: ABS; 1997. 27 National Health and Medical Research Council (NHMRC). Australian alcohol guidelines: health risks and benefits. Canberra: NHMRC; 2001. 28 StataCorp. Stata statistical software: release 11.2. College Station, TX: StataCorp LP; 2009. 29 Smith AMA, Rissel CE, Richters J, Grulich AE, de Visser RO. Sex in Australia: the rationale and methods of the Australian study of health and relationships. Aust N Z J Public Health 2003; 27: 106–17. doi:10.1111/j.1467-842X.2003.tb00797.x 30 Badcock PB, Smith AMA, Richters J, Rissel C, de Visser RO, Simpson JM, et al. Characteristics of heterosexual regular relationships among a representative sample of adults: the Second Australian Study of Health and Relationships. Sex Health 2014; 11: 427–38. doi:10.1071/SH14114 31 Pack RP, Li X, Stanton BF, Cottrell LA. Psychosocial correlates of dual methods for contraception and STI protection in urban
504
Sexual Health
R. O. de Visser et al.
adolescents. ISRN Obstet Gynecol 2011; 2011: 469610. doi:10.5402/ 2011/469610 32 Albarracín D, Johnson BT, Fishbein M, Muellerleile PA. Theories of reasoned action and planned behavior as models of condom use: a meta-analysis. Psychol Bull 2001; 127: 142–61. doi:10.1037/00332909.127.1.142 33 Noar SM, Carlyle K, Cole C. Why communication is crucial: metaanalysis of the relationship between safer sexual communication and condom use. J Health Commun 2006; 11: 365–90. doi:10.1080/ 10810730600671862
34 Bajos N, Bozon M, Beltzer N, Laborde C, Andro A, Ferrand M, et al. Changes in sexual behaviours: from secular trends to public health policies. AIDS 2010; 24: 1185–91. doi:10.1097/QAD.0b013e328336 ad52 35 Herlitz CA, Forsberg M. Sexual behaviour and risk assessment in different age cohorts in the general population of Sweden (1989–2007) Scand J Public Health 2010; 38: 32–9. doi:10.1177/14 03494809355072
www.publish.csiro.au/journals/sh
Sexual Health, 2014, 11, 495–504 http://dx.doi.org/10.1071/SH14102
Safer sex and condom use: findings from the Second Australian Study of Health and Relationships Richard O. de Visser A,I, Paul B. Badcock B,C, Chris Rissel D, Juliet Richters E, Anthony M. A. Smith B,H, Andrew E. GrulichF and Judy M. SimpsonG A
School of Psychology, Pevensey 1, University of Sussex, Falmer BN1 9QH, UK. Australian Research Centre in Sex, Health and Society, La Trobe University, 215 Franklin Street, Melbourne, Vic. 3000, Australia. C Centre for Youth Mental Health, University of Melbourne, Orygen Youth Health Research Centre, 35 Poplar Road, Parkville, Vic. 3052, Australia. D Sydney School of Public Health, Charles Perkins Centre (D17), University of Sydney, Sydney, NSW 2006, Australia. E School of Public Health and Community Medicine, University of New South Wales, Sydney, NSW 2052, Australia. F The Kirby Institute, Wallace Wurth Building, University of New South Wales, Sydney, NSW 2052, Australia. G Sydney School of Public Health, Edward Ford Building (A27), University of Sydney, Sydney, NSW 2006, Australia. H Deceased. I Corresponding author. Email: [email protected] B
Abstract. Background: It is important to have current and reliable estimates of the frequency and correlates of condom use among Australian adults. Methods: A representative sample of 20 094 men and women aged 16–69 years, from all states and territories, completed computer-assisted telephone interviews. The overall participation rate among eligible people was 66.2%. Results: Although most respondents had used a condom at some time in their lives, fewer than half of those who were sexually active in the year before being interviewed had used a condom in that year. Condom use in the last year was associated with youth, speaking a language other than English at home, bisexual identity, greater education, residence in major cities, lower income and having multiple sexual partners in the last year. One-quarter of respondents used a condom the last time they had vaginal intercourse and one-sixth of these were put on after genital contact. Condom use during most recent vaginal sex was associated with youth, lower income, having sex with a non-regular partner and not using another form of contraception. Condom use appears to have increased between 2001–02 and 2012–13. Conclusion: Consistent with other research, this study showed that condom use was strongly associated with partner type and use of other contraception. There may be a need to highlight among people with multiple sexual partners the fact that non-barrier methods of contraception do not offer protection against sexually transmissible infections. The finding that many condoms were applied after genital contact suggests a need to promote both use and correct use of condoms. Additional keywords: contraception, cross-sectional studies, health surveys, HIV prevention, national survey, sexual behaviour, STI prevention. Received 7 June 2014, accepted 23 August 2014, published online 7 November 2014
Introduction Condom use is an essential component of behavioural strategies to reduce the transmission of HIV and other sexually transmissible infections (STIs).1 Condoms are effective as contraceptives and provide protection against the transmission of many STIs because they provide a physical barrier to the passage of sperm and pathogens in semen or pre-ejaculatory fluid and because they limit possible Journal compilation CSIRO 2014
transmission of STIs via contact between infected mucosal surfaces.2,3 Research into patterns of condom use has taken three approaches: (1) to examine whether people have ever used condoms during a defined period (e.g. lifetime or in the past year); (2) to measure frequency of condom use during a defined period; and (3) to examine individual sexual encounters to identify factors that make condom use more or less likely. www.publish.csiro.au/journals/sh
496
Sexual Health
The First Australian Study of Health and Relationships (ASHR1) revealed that over 85% of people with experience of partnered sex had used condoms at some point in their lives, but that fewer than half of those who were sexually active in the last year reported condom use in the last year.4 These proportions were comparable to figures reported in studies of population-representative samples conducted around the same time, or subsequently, in the US, Britain and Europe.5–8 Surveys of population-representative samples have shown that condom use in the last year is most likely among people who are younger, better educated, have lower incomes, identify as bisexual or homosexual or have multiple sexual partners.4,7–12 Although many people report that they have used condoms at some point in their lives, fewer than half of heterosexually active men and women always use condoms.4,13 Consistent condom use appears to be more likely among younger people and is more likely with casual partners than with regular partners.4,13,14 Consistent condom use appears to be more likely among homosexually active men, especially for sex with casual partners.4 Surveys of representative samples in other countries reveal that fewer than 25% of respondents used a condom during their most recent heterosexual encounter.4,13,15 Studies of representative samples indicate that reporting condom use during people’s most recent instance of vaginal intercourse is significantly more likely among men, younger people, people who identify as bisexual, people with multiple partners in the last year, people who had sex with casual partners and people with higher levels of education.4,7,8,13–15 Condom use is significantly more likely if no other contraception is used, indicating that there tends to be a trade-off between condoms and other forms of contraception.4,10,12,16 Indeed, dual use of condoms and other forms of contraception is uncommon.10,17 However, increases in condom use among people using other forms of contraception could lead to significant reductions in the number of unplanned pregnancies and terminations or births.18 Homosexually active men are more likely than exclusively heterosexual men and women to report use of condoms during their most recent sexual encounter. Again, condom use is more likely among younger people and with casual partners than with regular partners.4,11 In addition to assessing whether condoms are used, it is important to determine whether they are used in ways that maximise their efficacy. The potential for transmission of STIs via pre-ejaculatory fluid or contact with infected mucosal surfaces means that condoms should be applied before penetration, not simply before ejaculation.19–21 However, research reveals that 10–15% of condoms are applied after initial genital contact.4,22 In addition to reducing the protection afforded to individuals, such behaviour has implications for the validity and reliability of self-reported condom use.22 Since ASHR1, ongoing efforts have been made to maintain and increase levels of condom use.1 One aim of this study was to provide up-to-date, reliable estimates of the frequency of condom use and correlates of condom use using data derived from a representative sample of Australian adults.
R. O. de Visser et al.
Methods The methodology used in the Second Australian Study of Health Relationships (ASHR2) is described elsewhere in this issue.23 Briefly, between October 2012 and November 2013, computer-assisted telephone interviews were completed by a representative sample of 20 094 Australian residents aged 16–69 years from all states and territories. Ethical approval was obtained from the researchers’ host universities. Respondents were selected using dual-frame modified random-digit dialling (RDD), combining directory-assisted landline-based RDD with RDD of mobile telephones. The overall participation rate among eligible people was 66.2%. To maximise the number of interviews with people who had engaged in less-common and/or more-risky behaviours, all respondents who had had no sexual partners in the previous year, who had had more than one partner in the previous year and/or who reported homosexual experience completed a long form of the survey instrument, which collected detailed data on their sexual attitudes, relationships and behaviours. Of the larger proportion of respondents who reported one partner in the previous year and no homosexual experience, 20% were randomly selected to complete the longform interview and the other 80% completed a short-form interview. As a consequence, 8577 completed the long-form interview and 11 517 completed the short-form interview. Answers to questions that occurred only in the long-form interview are reported after weighting to reflect the sample as a whole. All respondents who reported experience of vaginal or anal intercourse indicated whether they had ever used a condom. Respondents who were sexually active in the year before being interviewed indicated whether they had used a condom in the last year. Respondents who had been sexually active in the 6 months before being interviewed provided information about their frequency of condom use with casual partners during that period [Always/Often/Occasionally/Never]. This variable was recoded as dichotomous, distinguishing between respondents who always used condoms and those who did not. Casual partners were defined as someone other than a ‘live-in partner’ or a ‘regular non-live-in partner’. Respondents also provided information about condom use during their most recent sexual experience in the year before being interviewed. Questionnaire items pertaining to condom use are provided in Box 1. Respondents indicated their relationship to their partner and the age of their partner. They also indicated whether any contraception other than condoms was used. Respondents who used a condom indicated when it was put on. Correlates of the main outcomes examined in this paper included a range of demographic characteristics, which were recoded to facilitate analysis. Respondents’ ages were recoded into six groups (16–19, 20–29, 30–39, 40–49, 50–59 and 60–69 years). Languages spoken at home were recoded as English or a language other than English. Sexual identity (in answer to the question ‘Do you think of yourself as . . .’) was coded as heterosexual, homosexual or bisexual; too few respondents stated that they were ‘queer’, ‘other’, or ‘undecided’ to allow analysis of these groups. Respondents’ reported highest completed level of education was recoded to
Condom use in Australia
Sexual Health
497
Box 1. Ascertainment of men’s condom use
The questions were repeated with gender-appropriate wording for women and for men reporting male partners. Have you ever used condoms to have sex with a woman? Have you used condoms in the last 12 months to have sex with a woman? Have you used condoms in the last 6 months to have sex with a woman? During the past 6 months, how often was a condom used when having vaginal intercourse with your partner? Was that never, occasionally, often or always? The last time you had sex [with a woman], did you put your penis into her vagina? Was a condom used? Was the condom put on before your penis touched her vagina? Did you ejaculate inside her? [If respondent states ‘Yes’, interviewer clarifies whether ejaculation occurred in the condom.] The last time you had sex [with a woman], did you put your penis into her anus? Was a condom used when you did this? Was the condom put on before your penis touched her anus? Did you ejaculate inside her rectum? [If respondent states ‘Yes’, interviewer clarifies whether ejaculation occurred in the condom.]
distinguish between those who had not (yet) completed secondary school, those who had completed secondary school and those who had completed post-secondary education. Respondents’ postcodes were used with the Accessibility/ Remoteness Index of Australia to determine whether respondents lived in a major city, a regional area or a remote area (i.e. areas with relatively unrestricted, restricted and very restricted access to goods, services and opportunities for social interaction respectively).24 To approximate the gross annual household income quintiles reported by the Australian Bureau of Statistics for 2009–2010,25 respondents’ reported annual household income was grouped into five categories: less than $28 000, $28 000–$52 000, $52 000–$83 000, $83 000– $125 000 and more than $125 000. Respondents’ reported occupation was coded into the nine major categories of the Australian Standard Classification of Occupations26 and then recoded to distinguish among managerial/professional occupations, white-collar occupations and blue-collar occupations. Respondents’ reports of their number of sexual partners were used to produce a dichotomous variable, identifying respondents who reported more than one sexual partner in the year before being interviewed. Respondents’ reports of alcohol consumption frequency and volume were used to determine whether respondents’ alcohol consumption was in excess of Australian National Health and Medical Research Council (NHMRC) guidelines at the time of ASHR1 (28 standard drinks a week for men and 14 for women).27 The number of men reporting homosexual activity in the past year (n = 201) was too small to allow detailed analyses of correlates of condom use for this group. The main outcome variables were compared with those of ASHR1 to identify significant changes over time. Data were weighted to adjust for the probability of each respondent being selected for a landline or mobile phone interview, a long-form interview and (for landline participants) the number of in-scope adults in the household. Data were then weighted to match the Australian population
on the basis of age, gender, area of residence and telephone ownership (i.e. mobile telephone only vs other), resulting in an adjusted sample of 10 056 men and 10 038 women (total 20 094). The data were thus weighted to account for the specifics of our sample design and the fact that particular types of people were slightly over- or under-represented. Therefore, the data presented describe the Australian population aged 16–69 years, subject to the biases noted elsewhere in this issue.23 Weighted data were analysed using the survey estimation commands in Stata Version 11.2.28 The data were analysed using univariate logistic regression for dichotomous outcomes, including comparisons between ASHR1 and ASHR2. Most percentages are presented in this article without standard errors or 95% confidence intervals (CI). This decision was made to maximise both readability and brevity and is in keeping with the style of other studies of a similar scope and intent.6,13,29 Further information about the precision of estimates is found elsewhere in this issue.23 Due to the number of participants in ASHR2, it is important to recognise that often there is the statistical power to detect even small changes as statistically significant, but these do not necessarily correspond to significant differences in a public health sense. Results Prevalence of condom use Although the majority (90%) of respondents had used a condom at some time in their lives, fewer than half of the people who were sexually active in the year before being interviewed had used a condom in the last year. Among respondents with heterosexual experience, women were significantly less likely than men to report that they had ever used condoms (88.4% vs 91.5%, OR = 0.72, CI: 0.65–0.79, P < 0.001). Among respondents who reported heterosexual activity in the last year, women were significantly less likely than men to report that they had used condoms in the year before being interviewed (33.8% vs 39.0%, OR = 0.80, CI: 0.74–0.86,
498
Sexual Health
R. O. de Visser et al.
areas (P < 0.001) or had lower incomes (P < 0.001). Condom use was most likely among men with white-collar occupations and least likely among managers/professionals (P < 0.001). Condom use in the last year was significantly more likely among men who had more than one sexual partner in the last year (P < 0.001) and significantly less likely among men who consumed alcohol in excess of NHMRC guidelines (P = 0.002). Condom use for heterosexual activity in the previous year was significantly more likely among women who were younger
P < 0.001). Among men who reported ever having sex with men, 56% had ever used a condom for such activity and among men who had sex with men in the year before being interviewed, 69% had used a condom for such activity. Table 1 displays correlates of condom use for heterosexual activity in the year before being interviewed. Condom use was significantly more likely among men who were younger (P < 0.001), spoke a language other than English at home (P < 0.001), were bisexually identified (P = 0.02), had higher levels of education (P < 0.001), lived in major cities or remote
Table 1. Sociodemographic correlates of having used a condom for heterosexual activity among men and women in the last year Unless indicated otherwise, data show the percentage of category using a condom. Unadjusted odds ratios (OR) and 95% confidence intervals (CI) are given for having used a condom, relative to reference category Correlate
Men (n = 8435)
OR (95% CI)
Women (n = 8056)
OR (95% CI)
Overall
39.0
Age (years) 16–19 20–29 30–39 40–49 50–59 60–69
86.5 73.4 47.6 26.7 16.8 7.3
2.32 (1.54–3.48) 1 0.33 (0.27–0.40) 0.13 (0.11–0.16) 0.07 (0.06–0.09) 0.03 (0.02–0.04)
85.0 62.4 37.7 22.7 8.1 3.6
3.40 (2.25–5.15) 1 0.36 (0.31–0.43) 0.18 (0.15–0.21) 0.05 (0.04–0.07) 0.02 (0.02–0.03)
Language at home English Other
37.3 57.8
1 2.30 (1.90–2.78)
32.9 48.1
1 1.89 (1.52–2.35)
Sexual identity Heterosexual Homosexual Bisexual
38.8 68.8 55.0
1 –A 1.92 (1.14–3.24)
33.2 41.4 61.3
1 –A 3.19 (2.10–4.83)
Education Less than secondary Secondary Post-secondary
30.5 40.6 40.1
1 1.56 (1.32–1.85) 1.53 (1.30–1.80)
25.7 38.7 33.9
1 1.83 (1.51–2.20) 1.48 (1.25–1.75)
Region of residence Major city Regional Remote
41.8 31.6 42.4
1 0.64 (0.57–0.72) 1.02 (0.70–1.50)
36.8 27.1 25.5
1 0.64 (0.57–0.72) 0.59 (0.34–1.02)
Household income $125 000
57.5 42.7 37.4 36.7 33.1
0.55 0.44 0.43 0.37
1 (0.44–0.69) (0.36–0.54) (0.35–0.53) (0.30–0.45)
53.3 35.8 34.7 27.5 28.5
0.49 0.47 0.33 0.34
Occupational category Blue collar White collar Manager/professional
38.5 45.8 34.9
1 1.35 (1.17–1.56) 0.86 (0.76–0.96)
35.6 34.6 31.2
1 0.96 (0.80–1.15) 0.82 (0.68–0.99)
Excessive alcohol consumption No Yes
39.7 26.0
1 0.53 (0.36–0.79)
34.8 25.4
1 0.64 (0.42–0.96)
Partners in last year 1 2 or more
32.0 80.2
1 8.58 (7.19–10.23)
29.7 79.2
1 9.00 (7.02–11.55)
A
33.8
Because there were 10 or fewer respondents, the OR was not calculated.
1 (0.39–0.60) (0.38–0.57) (0.27–0.40) (0.28–0.41)
Condom use in Australia
Sexual Health
(P < 0.001), spoke a language other than English at home (P < 0.001), were bisexually identified (P < 0.001), had higher levels of education (P < 0.001), lived in major cities (P < 0.001) or had lower incomes (P < 0.001). Women in managerial occupations were significantly less likely than women in blue-collar occupations to have used a condom (P = 0.02). Condom use was significantly more likely among women who had more than one partner in the last year (P < 0.001) and significantly less likely among women who consumed alcohol in excess of NHMRC guidelines (P = 0.03).
Frequency of condom use Just under half (49%) of the respondents who had heterosexual vaginal intercourse with casual partners in the 6 months before being interviewed reported that they always used condoms, with similar frequencies of consistent condom use reported by men and women (P = 0.65). The majority (58%) of the 116 men who had anal homosexual intercourse with casual partners in the 6 months before being interviewed reported that they always used condoms.
Table 2. Sociodemographic correlates of always using condoms for vaginal intercourse in the last 6 months with casual partners Unless indicated otherwise, data show the percentage of category always using a condom. Unadjusted odds ratios (OR) and 95% confidence intervals (CI) are given for always having used a condom, relative to the reference category, separately for men and women Correlate
Men (n = 931)
OR (95% CI)
Women (n = 587)
OR (95% CI)
Overall
49.1
Age (years) 16–19 20–29 30–39 40–49 50–59 60–69
60.1 48.4 47.7 41.8 47.0 30.5
1.63 (0.83–3.18) 1 0.91 (0.48–1.73) 0.89 (0.46–1.74) 0.86 (0.48–1.53) 0.46 (0.20–1.05)
55.3 63.8 25.4 32.0 3.5 0.0
0.79 (0.31–2.01) 1 0.23 (0.11–0.51) 0.62 (0.25–1.50) 0.08 (0.02–0.42) –A
Language at home English Other
49.0 38.5
1 0.89 (0.24–1.45)
47.2 79.8
1 6.18 (0.69–55.06)
48.1 100.0 33.7
1 –A 0.50 (0.17–1.49)
47.2 27.0 50.8
1 –A 1.06 (0.45–2.50)
Education Less than secondary Secondary Post-secondary
42.2 46.1 52.1
1 1.12 (0.57–2.20) 1.62 (0.83–3.16)
46.9 46.4 48.8
1 1.04 (0.38–2.81) 1.02 (0.37–2.84)
Region of residence Major city Regional Remote
50.3 41.8 56.1
1 0.67 (0.41–1.08) 1.50 (0.43–5.26)
43.4 49.9 80.5
1 1.28 (0.63–2.59) 4.63 (0.52–41.14)
Household income $125 000
41.7 43.5 53.3 53.5 50.0
1.11 1.50 1.45 1.25
1 (0.57–2.17) (0.80–2.80) (0.73–2.89) (0.62–2.51)
58.1 33.2 60.8 24.9 43.5
0.39 1.04 0.31 0.94
Occupational category Blue collar White collar Manager/professional
45.5 50.3 50.8
1 1.20 (0.71–2.01) 1.39 (0.77–1.99)
43.4 48.6 46.6
1 0.97 (0.40–2.36) 0.93 (0.29–2.94)
Excessive alcohol consumption No Yes
49.4 46.2
1 0.88 (0.45–1.72)
51.3 31.1
1 0.43 (0.19–0.97)
Partners in last year 1 2 or more
31.7 52.7
1 1.57 (1.02–4.41)
51.1 45.4
1 0.69 (0.31–1.12)
Sexual identity Heterosexual Homosexual Bisexual
A
499
49.4
Because there were 10 or fewer respondents, the OR was not calculated.
1 (0.19–0.79) (0.31–3.55) (0.12–0.80) (0.33–2.68)
500
Sexual Health
R. O. de Visser et al.
Table 2 displays correlates of consistent condom use for vaginal intercourse with casual partners in the 6 months before being interviewed. Among men, consistent condom use was not significantly related to age (P = 0.20), language spoken at home (P = 0.25), sexual identity (P = 0.21), education (P = 0.17), region of residence (P = 0.47), income (P = 0.70), occupation category (P = 0.64) or whether men consumed alcohol in excess of NHMRC guidelines (P = 0.71). However, men who had multiple sexual partners in the past year were significantly more likely to report consistent condom use (P = 0.04). Among women, consistent condom use was significantly less likely among women aged 30 years or over (P < 0.001). There was a significant, but variable, association between consistent condom use and income (P = 0.03). Women who drank in excess of NHMRC guidelines were significantly less likely to report consistent condom use (P = 0.042). There was no significant association between always using condoms and language spoken at home (P = 0.10), sexual identity (P = 0.76), education (P = 1.00), region of residence (P = 0.33), occupational category (P = 0.99) or having more than one sexual partner in the year before being interviewed (P = 0.35). Condom use during the most recent sexual encounter The data in Table 3 indicate the proportion of respondents’ most recent sexual encounters in which a condom was used. Condoms were used in 23.3% of instances of vaginal intercourse, with men significantly more likely than women to report condom use (P = 0.01). Of the condoms used for vaginal intercourse, 15.2% were put on after initial genital contact. Among the small number of respondents who had anal intercourse during their most recent heterosexual encounter, condoms were not commonly used, with no sex difference in reports of condom use (P = 0.18). All condoms used during heterosexual anal intercourse were put on before anal penetration. Men’s reports of their most recent homosexual encounters revealed that most used a condom for insertive or
receptive anal intercourse, or both. Late application of condoms was reported for very few of the condoms used for anal intercourse. Table 4 displays correlates of condom use for vaginal intercourse during respondents’ most recent sexual encounters. Among men, condom use was significantly associated with being younger (P < 0.001), speaking a language other than English at home (P = 0.02), having a lower income (P = 0.001), having a white-collar occupation (P < 0.001), having sex with a casual or occasional partner (P < 0.001) and not using another form of contraception (P < 0.001). Condom use was not significantly related to sexual identity (P = 0.54), education (P = 0.40), region of residence (P = 0.08) or drinking in excess of NHMRC guidelines (P = 0.074). Among women, condom use was significantly associated with being younger (P < 0.001), identifying as bisexual (P = 0.003), having a very low income (P < 0.001), having sex with a casual/occasional partner (P < 0.001) and not using another form of contraception (P < 0.001). Condom use was not significantly related to speaking a language other than English at home (P = 0.17), education (P = 0.18), region of residence (P = 0.13), occupation (P = 0.65), or drinking in excess of NHMRC guidelines (P = 0.10). Analyses of condom use during the most recent experience of vaginal intercourse were re-run, restricting the sample to those potentially at risk of STI transmission or unintended pregnancy. People in monogamous relationships of at least 1 years’ duration were considered not at risk of STI transmission, so were excluded from these analyses. In these analyses, condom use was reported by 55% of men and 46% of women (a significant sex difference, P = 0.031) and late application of condoms was reported by 6% of men and 9% of women (a non-significant sex difference, P = 0.26). When the analyses were further restricted to exclude those using other contraception or not exposed to the risk of pregnancy, condom use was reported by 72% of men and 60% of women (P = 0.13) and late application was reported by 7% of men and 11% of women (a non-significant sex difference, P = 0.43).
Table 3. Condom use during most recent sexual encounters among men and women Unadjusted odds ratio (OR) and 95% confidence intervals (CI) are given for women relative to men
Heterosexual behaviour No. engaged in vaginal sex No. using condom for vaginal sex (%) Condom put on before genital contact No. engaged in anal sex No. using condom for anal sex (%) Condom put on before genital contact Homosexual behaviour No. engaged in insertive anal sex No. using condom for insertive anal sex (%) Condom put on before genital contact No. engaged in receptive anal sex No. using condom for receptive anal sex (%) Condom put on before genital contact
Men
Women
7210 1836 (25.5%) 86.0% 90 16 (24.3%) 100.0%
6938 1460 (21.1%) 81.3% 29 3 (16.1%) 100.0%
80 46 (56.7%) 91.4% 65 38 (58.9%) 97.5%
OR (95% CI)
0.80 (0.64–0.94) 0.74 (0.47–1.15) 0.60 (0.12–2.99) 1
– –
– –
– –
– –
Condom use in Australia
Sexual Health
501
Table 4. Sociodemographic correlates of condom use among men and women during most recent event of vaginal intercourse Unless indicated otherwise, data show the percentage of respondents in a given category using a condom. Unadjusted odds ratios (OR) and 95% confidence intervals (CI) are given for having used a condom, relative to reference category, separately for men and women. Correlate
Men (n = 7210)
OR (95% CI)
Women (n = 6938)
OR (95% CI)
Overall
25.5
Age (years) 16–19 20–29 30–39 40–49 50–59 60–69
70.2 45.7 31.1 16.3 11.7 3.9
2.79 (1.51–5.15) 1 0.53 (0.36–0.78) 0.23 (0.16–0.34) 0.16 (0.10–0.25) 0.05 (0.03–0.09)
61.2 39.0 20.8 11.8 4.7 0.9
2.46 (1.37–4.43) 1 0.41 (0.28–0.61) 0.21 (0.14–0.32) 0.08 (0.04–0.14) 0.01 (0.00–0.05)
Language spoken at home English Other
24.6 36.9
1 1.79 (1.12–2.88)
20.7 29.9
1 1.64 (0.81–3.31)
25.4 100.0 30.0
1 –A 1.26 (0.61–2.61)
20.5 14.2 39.5
1 –A 2.52 (1.47–4.33)
Education Lower secondary Secondary Post-secondary
22.2 27.4 25.2
1 1.32 (0.87–2.00) 1.18 (0.78–1.76)
16.8 24.1 20.9
1 1.57 (0.97–2.54) 1.30 (0.84–2.03)
Region of residence Major city Regional Remote
26.7 22.1 36.6
1 0.78 (0.59–1.02) 1.58 (0.80–3.11)
22.7 16.3 27.6
1 0.66 (0.48–0.91) 1.29 (0.29–5.81)
Household income $125 000
38.8 30.7 25.2 23.9 20.5
0.69 0.53 0.49 0.41
1 (0.42–1.15) (0.33–0.86) (0.31–0.79) (0.26–0.64)
42.7 15.8 19.9 16.4 17.2
0.25 0.33 0.26 0.28
Socioeconomic status Blue collar White collar Manager/professional
26.2 31.9 19.7
1 1.31 (0.95–1.82) 0.69 (0.52–0.92)
20.0 21.9 19.5
1 1.12 (0.67–1.85) 0.96 (0.57–1.64)
Excessive alcohol consumption No Yes
25.8 18.8
1 0.66 (0.42–1.04)
21.4 15.2
1 0.66 (0.40–1.08)
Partner type Regular Other
21.4 64.8
1 6.77 (4.85–9.45)
19.0 55.8
1 5.38 (3.37–8.59)
Used other contraception Yes No No needB
23.3 50.4 7.3
1 3.36 (2.45–4.60) –
18.4 47.8 3.5
1 4.06 (2.91–5.69) –
Sexual identity Heterosexual Homosexual Bisexual
21.1
1 (0.15–0.41) (0.20–0.55) (0.17–0.42) (0.18–0.44)
A
Because there were 10 or fewer respondents, the OR was not calculated. Pregnant, past menopause, infertile (self or partner); odds ratio not calculated.
B
Selected comparisons between ASHR1 and ASHR2 Comparisons between selected outcomes of ASHR1 and ASHR2 are presented in Table 5. The data revealed nonsignificant increases in condom use for vaginal intercourse— an increase of 2% for use any time in the last year (P = 0.06 for
men and women) and a significant increase of at least 4% for both sexes for use during the most recent sexual encounter. Condom use for homosexual anal intercourse also appeared to increase, but numbers of respondents were small and the increases were not statistically significant.
502
Sexual Health
R. O. de Visser et al.
Table 5. Prevalence of condom use during intercourse among men and women aged 16–59 years: comparison of the ASHR1 and ASHR2 People aged 60–69 years have been removed from the ASHR2 sample for comparison. Unadjusted odds ratios (OR) and confidence intervals (CI) are for the Second Australian Study of Health and Relationships (ASHR2) versus the First Australian Study of Health and Relationships (ASHR1) Type of intercourse Vaginal intercourse in past year No. engaged in behaviour % Using a condom
Men ASHR1 ASHR2 8438 7239 42.5% 44.3%
P-value OR (95% CI)
Women ASHR1 ASHR2
P-value OR (95% CI)
0.06 1.08 (1.00–1.16)
8266 7238 34.1% 37.6%
0.06
Vaginal intercourse in most recent heterosexual encounter No. engaged in behaviour 7574 6249 % Using a condom 24.7% 28.8%
0.03 1.23 (1.03–1.49)
7231 6322 17.6% 23.0%
0.002 1.40 (1.13–1.75)
Homosexual anal intercourse in past year No. engaged in behaviour 177 185 % Using a condom 59.1% 69.5%
0.15 1.57 (0.85–2.93)
–
–
–
–
Anal intercourse in most recent homosexual encounter No. engaged in behaviour 69 75 % Using a condom 46.1% 58.0%
0.37 1.61 (0.56–4.67)
–
–
–
–
Discussion This study applied all three approaches used in previous studies of patterns of condom use by: (1) examining whether people had ever used condoms in their lifetime and in the past year; (2) measuring frequency of condom use during the past 6 months; and (3) examining individual sexual encounters to identify factors that make condom use more or less likely. The data presented here show that although experience of condom use is widespread, condoms were used in only one-quarter of respondents’ most recent heterosexual encounters. For all three key measures of condom use for vaginal intercourse—any use in the last year, consistency with casual partners in the last 6 months and use during the most recent vaginal intercourse—younger respondents were significantly more likely to report condom use. For the measures of any use in the last year and use during the most recent incident of vaginal intercourse, condom use was significantly more likely among people with lower incomes, those who have had casual and/or multiple sexual partners or those not using other contraception. These patterns of correlates reflect those reported in ASHR1 and in population-representative surveys conducted in other countries.4,7–12 Condom use was significantly less common among older people than among younger people. This may reflect the observation that older people are less likely to have patterns of partnering linked to increased risk of STIs,12,30,31 but it is still notable that older people were less likely to always use condoms during encounters with non-regular sexual partners (Table 3). Age-related variations in condom use may help to explain the counter-intuitive finding that condom use was more likely among people with higher levels of education, but less likely among those with higher incomes; although young adults tend to have greater education than people from older generations, they also tend to have lower incomes. As in studies of non-representative samples, dual use of condoms and other forms of contraception was not common,
1.16 (0.99–1.39)
occurring in 23% of people’s most recent experiences of vaginal intercourse.16,17 It has been argued that adding condoms to other methods should be considered seriously as the first line of defence against unplanned pregnancy and STIs, given calculations of the number of unplanned pregnancies and terminations of pregnancy that could be avoided if more women added condom use to current contraceptive practices.18 However, there is a need to identify the cognitive and motivational correlates of dual use in order to better promote this behaviour.17,31 Late application of condoms is still quite common for vaginal intercourse, occurring in approximately one-sixth of most recent sexual encounters, but appears to be less common for heterosexual and homosexual anal intercourse. There remains a need for promotion of the correct use of condoms, not just use of condoms.19–23 Moreover, there is a need to continue to promote and encourage condom use.1 To this end, use should be made of the findings of meta-analytic studies that have identified key correlates of intended and actual condom use.32,33 These studies suggest that it is important to provide normative support for condom use and to improve individuals’ attitudes towards condom use and confidence in being able to acquire and use condoms.32 Particular attention should be given to nurturing the skills and self-confidence required to successfully negotiate condom use with sexual partners.33 In the 11 years since ASHR1, condom use for heterosexual and homosexual activity has increased, with statistically significant increases for heterosexual condom use and with condoms now used in nearly half of men’s most recent experiences of insertive anal intercourse. This matches observed trends for temporal increases in condom use reported in some countries.34,35 The National STI Strategy recognises the importance of condom use,[1] but the data presented here indicate that condoms are not used in many situations when their use could prevent the transmission of STIs, unintended pregnancy or both.
Condom use in Australia
Conflicts of interest None declared. Acknowledgements This study was funded by the National Health and Medical Research Council (grant no. 1002174). The authors are indebted to David Shellard and the staff of the Hunter Valley Research Foundation for managing data collection and undertaking the interviews for this study; and to the Social Research Centre for producing weights for the data. The authors also thank the 21 139 Australians who took part in the two phases of the project and so freely shared the sometimes intimate aspects of their lives. Professor Anthony Smith died during the course of this project and we intend this work to be a tribute to, and further example of, the extraordinary contribution his work has made to the sexual health and wellbeing of Australians.
References 1 Department of Health and Ageing (DoHA). The Second National Sexually Transmissible Infections Strategy: 2010–2013. Canberra: DoHA; 2010. 2 Roper WL, Peterson HB, Curran JW. Commentary: condoms and HIV/STD prevention: clarifying the message. Am J Public Health 1993; 83: 501–3. doi:10.2105/AJPH.83.4.501 3 Sparling P. Biology of Neisseria gonorrhoeae. In Holmes K, Mårdh P-A, Sparling P, Wiesner P, editors. Sexually transmitted diseases, 2nd edn. New York: McGraw Hill; 1990. pp. 131–47. 4 de Visser RO, Smith AM, Rissel CE, Richters J, Grulich AE. Sex in Australia: safer sex and condom use among a representative sample of adults. Aust N Z J Public Health 2003; 27: 223–9. doi:10.1111/ j.1467-842X.2003.tb00812.x 5 Dubois-Arber F, Spencer B. Condom use. In Hubert M, Bajos N, Sandfort T, editors. Sexual Behaviour and HIV/AIDS in Europe. London: UCL Press; 1998. pp. 287–302. 6 Laumann E, Gagnon J, Michael R, Michaels S. The social organisation of sexuality: sexual practices in the United States. Chicago: University of Chicago Press; 1994. 7 Jeffries WL, Dodge B. Male bisexuality and condom use at last sexual encounter: results from a national survey. J Sex Res 2007; 44: 278–89. doi:10.1080/00224490701443973 8 Mercer CH, Copas AJ, Sonnenberg P, Johnson AM, McManus S, Erens B, et al. Who has sex with whom? Characteristics of heterosexual partnerships reported in the second National Survey of Sexual Attitudes and Lifestyles. Int J Epidemiol 2009; 38: 206–14. doi:10.1093/ije/dyn216 9 Carrasco-Garrido P, López de Andrés A, Hernández Barrera V, Jiménez-Trujillo I, Santos-Sancho J, Jiménez-García R. Predictors of contraceptive methods among adolescents and young women residing in Spain. J Sex Med 2011; 8: 2431–8. doi:10.1111/j.17436109.2011.02346.x 10 Martinez G, Copen CE, Abma JC. Teenagers in the United States: sexual activity, contraceptive use, and childbearing, 2006–2010 national survey of family growth. Vital Health Stat 23 2011; Oct: 1–35. 11 Rosenberger JG, Reece M, Schick V, Herbenick D, Novak DS, Van Der Pol B, et al. Condom use during most recent anal intercourse event among a U.S. sample of men who have sex with men. J Sex Med 2012; 9: 1037–47. doi:10.1111/j.1743-6109.2012.02650.x 12 Schick V, Herbenick D, Reece M, Sanders SA, Dodge B, Middlestadt S, et al. Sexual behavior, condom use, and sexual health of Americans over 50: implications for sexual health promotion for aging adults. J Sex Med 2010; 7(Suppl. 5): 315–29. doi:10.1111/j.1743-6109. 2010.02013.x
Sexual Health
503
13 Layte RD, McGee HP, Quail A, Rundle K, Cousins G, Donnelly CD, et al. Irish study of sexual health and relationships. Dublin: Crisis Pregnancy Agency, and Department of Health and Children (DOHC); 2006. 14 Reece M, Herbenick D, Schick V, Sanders SA, Dodge B, Fortenberry JD. Condom use rates in a national probability sample of males and females ages 14 to 94 in the United States. J Sex Med 2010; 7266–76. (Suppl. 5): doi:10.1111/j.1743-6109.2010.02017.x 15 Sanders SA, Reece M, Herbenick D, Schick V, Dodge B, Fortenberry JD. Condom use during most recent vaginal intercourse event among a probability sample of adults in the United States. J Sex Med 2010; 7(Suppl. 5): 362–73. doi:10.1111/j.1743-6109.2010.0 2011.x 16 Ott MA, Adler NE, Millstein SG, Tschann JM, Ellen JM. The trade-off between hormonal contraceptives and condoms among adolescents. Perspect Sex Reprod Health 2002; 34: 6–14. doi:10.2307/3030227 17 de Visser R. Why do heterosexual young adults who use reliable contraception also use condoms? Results from a diary-based prospective longitudinal study. Br J Health Psychol 2007; 12: 305–13. doi:10.1348/135910706X111294 18 Pazol K, Kramer MR, Hogue CJ. Condoms for dual protection: patterns of use with highly effective contraceptive methods Public Health Rep 2010; 125: 208–17. 19 Ilaria G, Jacobs J, Polsky B, Koll B, Baron P, MacLow C, et al. Detection of HIV-1 DNA sequences in pre-ejaculatory fluid. Lancet 1992; 340: 1469. doi:10.1016/0140-6736(92)92658-3 20 Pudney J, Oneta M, Mayer K, Seage G, Anderson D. Pre-ejaculatory fluid as potential vector for sexual transmission of HIV-1. Lancet 1992; 340: 1470. doi:10.1016/0140-6736(92)92659-4 21 Killick SR, Leary C, Trussell J, Guthrie KA. Sperm content of preejaculatory fluid. Hum Fertil 2011; 14: 48–52. doi:10.3109/14647273. 2010.520798 22 de Visser RO, Smith AMA. When always isn’t enough: implications of the late application of condoms for the validity and reliability of self-reported condom use. AIDS Care 2000; 12: 221–4. doi:10.1080/ 09540120050001904 23 Richters J, Badcock PB, Simpson JM, Shellard D, Rissel C, de Visser RO, et al. Design and methods of the Second Australian Study of Health and Relationships. Sex Health 2014; 11: 383–96. doi:10. 1071/SH14115 24 Department of Health and Aged Care (DHAC). Measuring remoteness: accessibility/remoteness index of Australia (ARIA). Canberra: DHAC; 2001. 25 Australian Bureau of Statistics (ABS). Household wealth and wealth distribution, Australia, 2009–2010. Canberra: ABS; 2011. 26 Australian Bureau of Statistics (ABS). Australian standard classification of occupations, 2nd edn. Catalogue no. 1220.0. Canberra: ABS; 1997. 27 National Health and Medical Research Council (NHMRC). Australian alcohol guidelines: health risks and benefits. Canberra: NHMRC; 2001. 28 StataCorp. Stata statistical software: release 11.2. College Station, TX: StataCorp LP; 2009. 29 Smith AMA, Rissel CE, Richters J, Grulich AE, de Visser RO. Sex in Australia: the rationale and methods of the Australian study of health and relationships. Aust N Z J Public Health 2003; 27: 106–17. doi:10.1111/j.1467-842X.2003.tb00797.x 30 Badcock PB, Smith AMA, Richters J, Rissel C, de Visser RO, Simpson JM, et al. Characteristics of heterosexual regular relationships among a representative sample of adults: the Second Australian Study of Health and Relationships. Sex Health 2014; 11: 427–38. doi:10.1071/SH14114 31 Pack RP, Li X, Stanton BF, Cottrell LA. Psychosocial correlates of dual methods for contraception and STI protection in urban
504
Sexual Health
R. O. de Visser et al.
adolescents. ISRN Obstet Gynecol 2011; 2011: 469610. doi:10.5402/ 2011/469610 32 Albarracín D, Johnson BT, Fishbein M, Muellerleile PA. Theories of reasoned action and planned behavior as models of condom use: a meta-analysis. Psychol Bull 2001; 127: 142–61. doi:10.1037/00332909.127.1.142 33 Noar SM, Carlyle K, Cole C. Why communication is crucial: metaanalysis of the relationship between safer sexual communication and condom use. J Health Commun 2006; 11: 365–90. doi:10.1080/ 10810730600671862
34 Bajos N, Bozon M, Beltzer N, Laborde C, Andro A, Ferrand M, et al. Changes in sexual behaviours: from secular trends to public health policies. AIDS 2010; 24: 1185–91. doi:10.1097/QAD.0b013e328336 ad52 35 Herlitz CA, Forsberg M. Sexual behaviour and risk assessment in different age cohorts in the general population of Sweden (1989–2007) Scand J Public Health 2010; 38: 32–9. doi:10.1177/14 03494809355072
www.publish.csiro.au/journals/sh
