Sexual Health, 2011, 8,

Transcription

1 CSIRO PUBLISHING Review Sexual Health, 2011, 8, Behavioural surveillance among gay men in Australia: methods, findings and policy implications for the prevention of HIV and other sexually transmissible infections Iryna B. Zablotska A,B,C, Susan Kippax B, Andrew Grulich A, Martin Holt B and Garrett Prestage A A National Centre in HIV Epidemiology and Clinical Research, University of New South Wales, Sydney, NSW 2052, Australia. B National Centre in HIV Social Research, University of New South Wales, Sydney, NSW 2052, Australia. C Corresponding author. izablotska@nchecr.unsw.edu.au Abstract. Background: The Australian HIV and sexually transmissible infection (STI) behavioural surveillance system (the repeated cross-sectional Gay Community Periodic Surveys, GCPS) has been conducted since 1998 and covers six main Australian jurisdictions. In this paper, we review its history and methodology, and the available indicators, their trends and their use. Methods: We describe the design and history of GCPS. For analyses of indicators, we use Pearson s c 2 -test and test for trend where appropriate. Results: About 90% of gay men in Australia have been tested for HIV (60% to 70% of men who were not HIV-positive) have been tested as recommended in the preceding 12 months. STI testing levels (~70% in the preceding 12 months) are high, but remain insufficient for STI prevention. In general, unprotected anal intercourse with regular (UAIR) and casual (UAIC) sex partners has increased over time. The prevalence and increasing trends in UAIR were similar across jurisdictions (P-trend <0.01), while trends in UAIC differed across the states: during , UAIC declined in NSW (P-trend <0.01) and increased elsewhere (P-trend <0.01). Trends in UAIC were associated with HIV diagnoses. Conclusion: This review of the design, implementation and findings of the Australian HIV/STI behavioural surveillance highlights important lessons for HIV/STI behavioural surveillance among homosexual men, particularly the need for consistent data collection over time and across jurisdictions. Investment in systematic behavioural surveillance appears to result in a better understanding of the HIV epidemic, the availability of a warning system and a better targeted HIV prevention strategy. Additional keywords: homosexual, men who have sex with men, sexual behaviour, testing. Introduction In Australia, as in Northern America and many European countries, homosexual men are the population group at highest risk of acquiring HIV. 1,2 Since the late 1990s, behavioural surveillance has been recommended by the UNAIDS and World Health Organization (WHO) Working Group of Global HIV/AIDS and sexually transmissible infection (STI) Surveillance as a part of a tailored surveillance system to track the state of the HIV epidemic in this high-priority population group. 3 Although many countries followed these recommendations, few have established systematic surveillance of sexual practices among homosexual men. Reasons include small population size and high stigma against HIV and homosexual identity, which make this population group hard to reach and quantify. 4 This lack of behavioural trend data limits the understanding of the changes in the HIV epidemic, particularly in the context of recent increases in HIV diagnoses among homosexual men. 5,6 In Australia, HIV was first diagnosed in homosexual men in 1985 and the national reporting of HIV diagnoses started in HIV-related behavioural data have been collected since the early days of the epidemic, 8,9 and systematic behavioural data collection for the purposes of surveillance started in the late 1990s. This paper reviews the history and methodology of the Australian HIV/STI behavioural surveillance, available indicators and their trends, and the use of these indicators for HIV/STI policy and prevention. Methods Surveillance system set-up A standardised ongoing behavioural surveillance system the Australian Gay Community Periodic Surveys (GCPS) was launched in In New South Wales (NSW), GCPSs have been conducted biannually since February 1996; annually since 1998 in Victoria (Vic.) and Queensland (Qld); in alternating years since 1998 in South Australia (SA) and Western Australia (WA) and every 3 years since 2000 in the Australian Capital Territory (ACT) This surveillance system covers all the large urban gay communities in Australia. 10 As recommended by WHO, 11 the Australian GCPS target the segments of the population where most new HIV infections CSIRO /SH /11/030272

2 Behavioural surveillance in Australia Sexual Health 273 have been concentrated gay community-attached men in metropolitan areas. The main founding principles of these surveys are: feasibility of recruitment, format and language suitable for self-administered interviews, consistency and stability over time and across jurisdictions, systematic collection of main indicators measuring HIV/STIrelated practices, community participation, and commitment by researchers, funding bodies and communitybased organisations to collect data without interruption. This has been a joint project of the National Centre in HIV Epidemiology and Clinical Research and the National Centre in HIV Social Research (both affiliated with the University of NSW in Sydney), funded by the state and federal government (Departments of Health or their counterparts) and conducted in close partnership with the state community organisations (AIDS councils and People Living with HIV/AIDS). Reporting has been regular and state-specific. Comparisons of behaviours across jurisdictions are also reported in the annual reports of trends in behaviour. 12 Population, sampling and recruitment The number of homosexual men in Australia is not readily available and can only be estimated. 10 However, gay communities and gay men are not hard to reach in Australia, mainly due to their historically strong advocacy for gay and HIV-related issues, their role in shaping the national response to the HIV epidemic, and active participation in HIV prevention and research. 13 Therefore, behavioural surveillance is gay community-based, which is reflected in the name of the GCPS. GCPSs are repeated cross-sectional surveys and employ convenience time and location sampling in four types of venues: gay social events, social venues (e.g. bars, clubs and gyms), sex-on-premises venues and sexual health clinics with substantial gay clientele. The same venues, if still active, have been retained for recruitment each time, and about the same ratio of participants from each type of venue has been maintained in each jurisdiction to keep maximum stability in the sample composition. Sample sizes vary depending on the size of communities where surveys are conducted. 10 The average annual sample size was 2642 men (range: ) in NSW, 1907 ( ) in Vic., 1428 ( ) in Qld, 595 ( ) in SA, 894 ( ) in WA and 296 ( ) in ACT. There have been variations in the sample age composition; for example, men in Qld surveys were younger (mean: 33 years; s.d.: 11.0 years) than in Vic. (mean: 34.8; s.d.: 10.1), NSW (mean: 35.6; s.d.: 9.5) and the smaller states. As expected, the predominant majority of participants identified themselves as gay (90%) or bisexual (7%); therefore, the samples are representative only of the gay identified and gay community-attached men, where HIV/STI epidemics are mostly concentrated. 1 Two factors the involvement of community-based organisations and the general sense of generating knowledge for the better health of gay men 14 have contributed to the popularity of the surveys in gay communities and the high participation rates (70% to 85% over time). Each survey lasts about a week and is scheduled around a major community event (for example, the Gay and Lesbian Mardi Gras festival in Sydney). Trained recruiters distribute and collect questionnaires. The surveys are self-administered, voluntary and anonymous. They have been approved by the Human Research Ethics Committee of the University of NSW, Australia. Survey content and indicators Standard survey questionnaires include ~60 70 questions. The questionnaire has mainly remained unaltered since 1998 in all jurisdictional surveys. Only a very small number of questions undergo changes to fill in temporal information gaps. Core behavioural surveillance topics and available indicators are presented in Tables 1 and 2, respectively. The consistency of data collection across all jurisdictions and over time has enabled comparative analyses of trends for most indicators since All behaviours are self-reported. Participants are not asked directly about engaging in risky practices. Rather, they are asked to report on the nature of their sexual practices in the previous 6 months, e.g. whether they had never, occasionally or often engaged (as a receptive or insertive partner) in anal intercourse with a condom, without a condom with ejaculation Table 1. Core topics covered by all Australian Gay Community Periodic Surveys Core topics Number of questions Socio-demographic characteristics (including age, ethnic or Aboriginal background, 6 7 A occupation, employment and residential location) Sexual identification 1 Social engagement with gay community (number of and time spent with gay friends) 2 Sexual partnerships (including awareness of regular partners HIV status) 9 Sexual practices (by type of partners: regular or casual) 25 Respondents HIV testing, knowledge and disclosure of serostatus 9 Sexual health testing 1 Illicit drug use (including amyl (poppers), Ecstasy, marijuana, Viagra or Cialis, speed, cocaine, Crystal Meth, 3 LSD (lysergic acid diethylamide), GHB (gamma-hydroxy butyrate), Special K, heroin, steroids and other ) Occasional topical questions 5 10 A In NSW, men are also asked about their country of birth.

3 274 Sexual Health I. B. Zablotska et al. Table 2. Available indicators: Australian Gay Community Periodic Surveys ART, antiretroviral therapy; STI, sexually transmissible infections; UAIR, unprotected anal intercourse with regular partners; UAIC, unprotected anal intercourse with casual partners Indicators Base A Trend over time available since Testing for HIV and STI (in last 12 months before the survey) Proportion of men who had never been tested for HIV 1 B 1998 Proportion of tested men who had their last test for HIV in 12 months before the survey 1 C 1998 Proportion of men who were tested for STI other than HIV in 12 months before survey 1 D 2003 Proportion of men on combination ART 1 E 2003 Proportion of men with undetectable viral load (on ART and not on ART) 1 D 2003 HIV serostatus and its disclosure (in last 6 months before the survey) Proportion of men who reported being HIV-positive 1 F 1998 Proportion of men who knew serostatus of their regular partners Match of HIV status between regular partners (HIV-positive and HIV-negative seroconcordant, serodiscordant and not concordant) Proportion of men who disclosed their serostatus to all casual sex partners Proportion of men who reported that they disclosed their HIV status to all or none of their casual partners Proportion of men who reported that some or all of their casual partners had disclosed their HIV status Sexual partnerships (in last 6 months before the survey) Proportion of men currently in regular relationships Proportion of relationships that are monogamous Proportion of men currently having casual sex partners Number of partners Proportion of men who have agreements with their regular partners about sex within the relationship G Proportion of men who have agreements with their regular partners about sex outside the relationship H Sexual practices (in last 6 months before the survey) Proportion of men who engaged in any UAIR Proportion of men who always used condoms during anal intercourse with regular partners Positioning in anal intercourse with regular partners (insertive only, receptive only and both) Proportion of men who engaged in any UAIC Proportion of men who always used condoms during anal intercourse with casual partners Positioning in anal intercourse with casual partners (insertive only, receptive only and both) Proportion of men who received post exposure prophylaxis Where men found sex partners (by venue type) Proportion of men engaging in group sex Proportion of men using party drugs for the purpose of sex Drug use (in last 12 months before the survey) Proportion of men who used specific drugs Frequency of injecting drug use (any) A 1 all men; 2 men with regular partners only; 3 men with casual partners only. B Men recruited from sexual health clinics have usually had a much higher prevalence of HIV than men recruited in other venues and are therefore traditionally excluded from these analyses. C Men previously tested for HIV and not HIV-positive. D Excluding men recruited from sexual health clinics. E HIV-positive men. F Men previously tested for HIV. G Clear (spoken) agreements with regular partners about anal sex within the relationship which include one of the options: No anal sex at all, All anal sex is with a condom, or Anal sex can be without a condom. H Clear (spoken) agreements with regular partners about anal sex outside the relationship (i.e. with casual partners) which include one of the options: No sex [with casual partners] at all, No anal sex at all, All anal sex is with a condom, or Anal sex can be without a condom. and without a condom with withdrawal before ejaculating. The same questions are repeated for regular and casual sex partners. Indicators of unprotected anal intercourse with regular (UAIR) and casual partners (UAIC) are usually constructed as binary variables ( any versus none for each), based on this series of questions. Indicators of drug use measure the use of specific drugs, injection of any illicit drugs and use of drugs in the context of group sex in the past 12 months. HIV testing indicators include the proportion of all participants who report having ever been tested for HIV, and the proportion of HIV-negative men who were tested for HIV in the past and who report having had their last test within a year before a survey. The STI testing indicator is based on the question Which of the following sexual health tests did you have in the last 12 months: anal, throat or penile swabs; urine sample; and blood test other than for HIV? (answers are yes v. no, plus the frequency for each).

4 Behavioural surveillance in Australia Sexual Health 275 Analyses We described and evaluated the Australian surveillance system and also simultaneously showed its trend findings. To identify trends, we used data available from 1996 to Where appropriate, the comparisons of proportions were performed using Pearson s c 2 -test for independence with a Type I error of 5%. Indicators, where appropriate, were directly standardised to the national male population in 5-year age groups using Australian Bureau of Statistics mid-year estimates of the adult male population. 15 Trends over time assessed using c 2 -test for trend. All analyses were executed in STATA 10.0 (College Station, TX, USA). Results Gay social engagement, relationships and sex partners Between 40% and 55% of GCPS participants reported that they spend a lot or most of their free time with their gay friends. In recent years, there has been a substantial decline in gay social engagement (number of and time spent with gay friends) and in the use of gay social venues to find sex partners. The internet has increasingly become the leading meeting place. 16 In 2009, 70% of men who had sex partners in the preceding 6 months reported finding partners online (a 50% increase from 2001, P > 0.001), including 10% who used the internet exclusively (see Fig. 1). Future research and prevention approaches will need to adapt to these changes. In 2009, ~64% of respondents reported being in regular relationships. About 90% of the men in regular relationships of more than 6 months had been tested for HIV, and 84% knew the serostatus of their regular partners. In the last decade, there has been an increase in the proportion of men in HIV-negative seroconcordant relationships. 17 About three-quarters of men with regular partners had negotiated an agreement about sex within their relationship, and the likelihood of having negotiated such agreements has remained stable over time. However, the nature of agreements has changed, with fewer men requiring consistent condom use with casual partners (P-trend <0.001). 17 Some of these changes in negotiated agreements represent an increase in the potential risk of HIV transmission. About 69% of respondents reported having had casual sex partners in the preceding 6 months (5.6% reported 50 or more partners). A high number of sexual partners in a short period of time is an indirect indicator of HIV risk. State-specific trends in the proportion of HIV-negative and men of unknown serostatus who reported having had 50 or more casual sexual partners in the preceding 6 months indicate a decline in this indicator in NSW since 2003 (from 7.7% to 4.3%, P-trend <0.01), but similar rates and no significant changes in other jurisdictions. Among HIVpositive men, this indicator is persistently higher than among HIV-negative men (P < 0.001). In the Australian GCPS, it is left to participants to distinguish between regular and casual partners. As this distinction is not clear-cut, multiple regular partnerships are possible. Similarly, different categories of casual partners are not distinguished. However, it has been found that ~50% of respondents had known their casual partners and had had sex with these partners before. 18 Further, men are more inclined to engage in UAIC if they previously had sex with a casual partner (often independent of HIV status). Sexual practices Since 2001, trends in risky sexual practices among gay men in Australia have been consistent with trends observed in other developed countries The prevalence and increasing trends in UAIR were generally similar across Australia (see Table 3). The prevalence of any UAIR was highest among men in Fig. 1. Age-standardised time trends in the relative use of the internet and physical venues in search for sex partners by gay men who reported having had sex partners in the 6 months before surveys: Australian Gay Community Periodic Surveys, (Data about the use of the internet were collected from 2001 onwards.)

5 276 Sexual Health I. B. Zablotska et al. Table 3. Age-standardised proportion of men who reported having had unprotected anal intercourse with regular (UAIR) and casual partners (UAIC), by year and state: Australian Gay Community Periodic Surveys, State where survey was conducted UAIR (among respondents who reported having had regular partners) NSW Vic Qld WA SA ACT UAIC (among respondents who reported having had casual partners) NSW Vic Qld WA SA ACT seroconcordant HIV-positive relationships and lowest in relationships that were not seroconcordant (on average 82% and 48%, respectively, in 2009). Unlike UAIR, trends in UAIC differed across Australian jurisdictions: during , there has been a decline in UAIC in NSW (c 2 -test for trend, P < 0.01) and continuing increases elsewhere in Australia. 22 Similar patterns were observed in the trends in HIV diagnoses among gay men across jurisdictions (see Table 3). While a combination of factors have contributed to these jurisdictional patterns (including testing patterns, the size of gay communities, baseline HIV and STI prevalence, prevention approaches, policies and resources for HIV prevention), UAIC among HIV-negative gay men has been identified as the main predictor of trends in HIV diagnoses, with a possible 2-year lag between the changes in UAIC and the following changes in HIV diagnoses. 23 HIV testing and prevalence In 2009, close to 90% of respondents reported they had been tested for HIV (see Table 4 and Prestage et al. 24 ). Nationally, this indicator was stable over time. These testing rates among homosexual men are higher than in middle and low income countries 25 and comparable to 26 or higher than in similar settings in developed countries. Currently, 60% to 70% of non- HIV-positive men report having their last test in the last year (68.2% in NSW, 66.0% in Vic. and 63.3% in Qld), as recommended by current Australian guidelines, 30 and there have been increases in this indicator in recent years. Self-reported HIV prevalence in the community-based samples of homosexual men recruited at non-clinic sites varies across Australian states and territories. Historically, the highest prevalence was observed in NSW, but the levels have become similar in recent years, at least in the eastern states. Prestage et al. reported that there was a marked decline in agestandardised HIV prevalence in NSW (from 14.2% in 1998 to 9.0% in 2007 (P-trend <0.001), and small changes in other states, predominantly in younger men. 10,31 About 67% of HIVpositive men are currently on antiretroviral therapy, and 82% of these have an undetectable viral load. It has been reported that some men in regular relationships use viral load in negotiating sexual practices. 32 STI testing Overall, STI testing was less common than testing for HIV and rates have remained stable since 2003 (see Table 5 and Zablotska et al. 33 ), even though the use of some tests (e.g. swabs and urine samples) increased significantly (P-trend <0.001 for each). Sexual behaviours (higher number of partners, having casual partners and engaging in UAIC with them) were associated with STI testing. However, until 2009, 33% of men were not tested for STIs in the last year before the survey. HIV-negative men were tested for STIs less often than HIV-positive men (prevalence ratio = 0.56; 95% confidence interval: ). Table 4. Age-standardised proportion of men who have ever been tested for HIV (excluding men recruited from sexual health clinics), by year and state: Australian Gay Community Periodic Surveys, State where survey was conducted NSW Vic Qld WA SA ACT

6 Behavioural surveillance in Australia Sexual Health 277 Table 5. Age-standardised proportion of men who had sexual health check-ups in the last 12 months (excluding men recruited from sexual health clinics), by year and state: Australian Gay Community Periodic Surveys, A State where survey was conducted NSW Vic Qld WA SA ACT A Data about sexual health testing were collected from 2003 onwards. STI testing among HIV-negative men has improved significantly but remains inadequate for STI control and HIV prevention. 33 Discussion Our review of Australian behavioural surveillance among gay men demonstrates that this surveillance tool, although concise and parsimonious, has provided a wealth of HIV and STI behavioural surveillance data, and has been influential in STI policy development and prevention. The efforts made to sustain regular data collection in six Australian states and territories with respect to sexual, drug use, and HIV and STI testing practices have paid off in the availability of key indicators since Available indicators have allowed us to explain the role of specific sexual behaviours and prevention strategies in the HIV epidemic. Specifically, UAIC among gay men who are not HIV-positive has been identified and used as a warning sign for the changes in HIV diagnosis trends in Australia. In most Australian states, the HIV epidemic resembles epidemics among gay men in other developed counties, where increases in HIV incidence have been observed since ~ ,34,35 The lack of increase in UAIC and the HIV diagnoses in NSW has raised interest. This unusual scenario in Australia highlights the value of behavioural surveillance in explaining HIV surveillance trends. Australia is neither the first country to establish an HIV behavioural surveillance system, nor has it been the only country to collect behavioural data for a long period of time. 4 In some European countries, behavioural surveillance studies have been conducted since the late 1980s mid-1990s (e.g. France, Netherlands and the UK). 36 However, data collection in Australia has been uninterrupted and its methodology consistent for more than a decade. Such consistency and coverage of all Australian states and territories with meaningfully-sized gay communities have positioned Australia to understand the nature and course of its HIV epidemic better. Since their inception in 1996, Australian GCPSs have served as a significant behaviour monitoring tool for HIV/STI prevention and policy in Australia. The surveillance alert about the increases in UAIC in NSW in the late 1990s was probably the most important single example of the use of this tool. This alert triggered a substantial mobilisation of resources, partnerships and prevention activities in NSW, and resulted in a decline in UAIC since 2001 and stabilisation in HIV diagnoses since Another important example is the change in STI testing recommendations. By 2006, GCPSs had demonstrated that highly sexually active gay men were at increased risk of an STI. Although, on average, STI testing coverage among men with a large number of partners was high, it appeared these men tested far less frequently than recommended. 30 Surveillance findings triggered a shift in STI testing guidelines. In NSW, these were changed to recommend that men with a larger number of sexual partners in a 6-month period should consider testing biannually. 30 The AIDS Council of NSW developed a specific education campaign to promote this policy. Recent surveillance data indicate increases in the rates and frequency of STI testing, including among highly sexually active men. 33 Australian GCPSs have some limitations and challenges. First, because these are repeated cross-sectional behavioural surveys, which do not follow the same individuals over time and do not include testing, they allow only ecological analyses of behavioural indicators and HIV diagnoses from other sources. Second, there are some limitations to the generalisability of data collected by GCPSs. As is the case for most behavioural surveillance studies, GCPS samples are recruited using convenience sampling methodology and they represent only gay community-attached and mostly gay-identified homosexual men. Since this population group is contributing most HIV diagnoses in Australia, we believe that the Australian GCPSs provide a fair picture of behaviours involved in HIV transmission among gay men in Australia. Third, changes in gay communities over time, such as age-specific declines in the use of traditional gay social venues or increased use of the internet, are common. They pose some challenges for maintaining sufficient recruitment and may require some new creative recruitment strategies in the future. Fourth, the sample sizes depend on the source population and availability of recruitment venues in locations where surveys are conducted. Therefore, some surveys (e.g. in ACT, WA and SA) are limited in the scope of analyses and provide only key indicators. Finally, the understanding of sexual practices among gay men (for example, serosorting, strategic positioning in casual sex) requires complex data that cannot be collected through brief self-administered questionnaires. We have dealt with this challenge by conducting several topic-specific GCPSs in NSW and by using data from other concurrent studies. 37 Our findings are similar to those reported in other countries. For example, most behavioural surveys in European countries 4 and in Northern America use the same sampling approaches, have similar recruitment challenges, and face similar issues of representativeness and interpretation. Our key behavioural indicators (particularly UAIC) differ from

7 278 Sexual Health I. B. Zablotska et al. how they are derived in other countries (any UAIC in the last 6 months in Australia versus in the last 12 months in European 4 and US surveys 26 ), but general country-specific trends show similar patterns: an increase in risk-taking behaviours among homosexual men since ~2001, which coincides with increases in HIV diagnoses. 1,19 In conclusion, this review of the Australian behavioural surveillance system highlights important lessons for HIV/STI behavioural surveillance among homosexual men, particularly the need for painstakingly consistent and dedicated data collection over time and across jurisdictions. Investment in systematic HIV/STI behavioural surveillance appears to result in a better understanding of the HIV epidemic, the availability of a warning system and better targeted prevention strategies. Contributors IBZ formulated the research issue and design of this analysis, and assumed principal responsibility for data collection, analysis and preparation of the paper. SK, AEG, GP and MH contributed to the study design, data collection, interpretation of results and preparation of the paper. All authors have seen and approved the final version of this paper. Funding Gay Community Periodic Surveys are funded by the New South Wales Health Department in New South Wales, by the Victoria Department of Human Services in Victoria, by Queensland Health in Queensland, by the Department of Health Western Australia in Western Australia, by the Department of Health South Australia in South Australia, and by the AIDS Council of ACT in Australian Capital Territory. The National Centre in HIV Epidemiology and Clinical Research is funded by the Australian Government Department of Health and Ageing. The views in this publication do not necessarily represent the position of the Australian government. Conflicts of interest None declared. Acknowledgements The authors would like to acknowledge the key community partners the Australian Federation of AIDS Organisations, the National Association of People Living with HIV/AIDS, the AIDS councils, and organisations of people living with HIV in each state and territory where GCPS are conducted for being instrumental in the establishment of the behavioural surveillance system in Australia and being a part of the partnership in HIV response. We are grateful to the Health Departments and their counterparts in each state and territory for their committed funding of the studies in support of the behavioural surveillance in Australia. Many thanks also go to all study participants for sharing their life experiences with the research team. References 1 Grulich AE, Kaldor JM. Trends in HIV incidence in homosexual men in developed countries. Sex Health 2008; 5: doi: / SH van Griensven F, de Lind van Wijngaarden JW, Baral S, Grulich A. The global epidemic of HIV infection among men who have sex with men. Curr Opin HIV AIDS 2009; 4: doi: /coh. 0b013e32832c3bb3 3 UNAIDS/WHO Working Group of Global HIV/AIDS and STI Surveillance. Guidelines for second generation HIV surveillance. Geneva, UNAIDS; Elford J, Jeannin A, Spencer B, Gervasoni JP, van de Laar MJ, Dubois-Arber F. HIV and STI behavioural surveillance among men who have sex with men in Europe. Euro Surveill 2009; 14: Zaba B, Slaymaker E, Urassa M, Boerma JT. The role of behavioral data in HIV surveillance. AIDS 2005; 19: S doi: /01. aids Garnett GP, Garcia-Calleja JM, Rehle T, Gregson S. Behavioural data as an adjunct to HIV surveillance data. Sex Transm Infect 2006; 82(Suppl. 1): i doi: /sti McDonald AM, Crofts N, Blumer CE, Gertig DM, Patten JJ, Roberts M, et al. The pattern of diagnosed HIV infection in Australia, AIDS 1994; 8: doi: / Kippax S, Crawford J, Davis M, Rodden P, Dowsett G. Sustaining safe sex: a longitudinal study of a sample of homosexual men. AIDS 1993; 7: doi: / Kippax S, Crawford J, Rodden P, Noble J. Predictors of unprotected male-to-male anal intercourse with casual partners in a national sample. Aust J Public Health 1995; 19: doi: /j tb00362.x 10 Prestage G, Ferris J, Grierson J, Thorpe R, Zablotska I, Imrie J, et al. Homosexual men in Australia: population, distribution and HIV prevalence. Sex Health 2008; 5: doi: /sh Rehle T, Lazzari S, Dallabetta G, Samoah-Odei E.. Second-generation HIV surveillance: better data for decision-making. Bull World Health Organ 2004; 82: Imrie J, Frankland A. HIV/AIDS, hepatitis and sexually transmissible infections in Australia: Annual report of trends in behaviour Sydney: National Centre in HIV Social Research, University of New South Wales; Altman D. Power and community: organizational and cultural responses to AIDS. London: Taylor and Francis; Bernard D, Kippax S, Baxter D. Effective partnership and adequate investment underpin a successful response: key factors in dealing with HIV increases. Sex Health 2008; 5: doi: / SH Australian Bureau of Statistics. Population by age and sex, Australian states and territories, June 2009, catalogue no Canberra: Australian Bureau of Statistics; Zablotska I, Holt M, Prestage G. Changes in gay men s participation in gay community life: implications for HIV prevention, surveillance and research. AIDS Behav doi: /s Prestage G, Jin F, Zablotska I, Grulich A, Imrie J, Kaldor J, et al. Trends in agreements between regular partners among gay men in Sydney, Melbourne and Brisbane, Australia. AIDS Behav 2008; 12: doi: /s Zablotska IB, Grulich AE, de Witt J, Prestage G. Casual sexual encounters among gay men: familiarity, trust and unprotected anal intercourse. AIDS Behav 2011; in press. doi: /s Elford J. Changing patterns of sexual behaviour in the era of highly active antiretroviral therapy. Curr Opin Infect Dis 2006; 19: doi: /01.qco e1 20 Marcus U, Voss L, Kollan C, Hamouda O. HIV incidence increasing in MSM in Germany: factors influencing infection dynamics. Euro Surveill 2006; 11: Sullivan PS, Hamouda O, Delpech V, Geduld JE, Prejean J, S le C, et al. Reemergence of the HIV epidemic among men who have sex with men in North America, Western Europe and Australia, Ann Epidemiol 2009; 19: doi: / j.annepidem

8 Behavioural surveillance in Australia Sexual Health Zablotska IB, Prestage G, Grulich AE, Imrie J. Differing trends in sexual risk behaviours in three Australian states: New South Wales, Victoria and Queensland, Sex Health 2008; 5: doi: /sh Zablotska IB, Prestage G, Middleton M, Wilson D, Grulich AE. Contemporary HIV diagnoses trends in Australia can be predicted by trends in unprotected anal intercourse among gay men. AIDS 2010; 24: doi: /qad.0b013e32833bf74a 24 Prestage G, Jin F, Zablotska IB, Imrie J, Grulich AE, Pitts M. Trends in HIV testing among homosexual and bisexual men in eastern Australian states. Sex Health 2008; 5: doi: /sh Adam PC, de Wit JB, Toskin I, Mathers BM, Nashkhoev M, Zablotska I, et al. Estimating levels of HIV testing, HIV prevention coverage, HIV knowledge, and condom use among men who have sex with men (MSM) in low-income and middle-income countries. J Acquir Immune Defic Syndr 2009; 52(Suppl. 2): S doi: /qai.0b013e 3181baf Sanchez T, Finlayson T, Drake A, Behel S, Cribbin M, Dinenno E, et al. Human immunodeficiency virus (HIV) risk, prevention, and testing behaviors United States, National HIV Behavioral Surveillance System: men who have sex with men, November 2003 April MMWR Surveill Summ 2006; 55: Mayor S. HIV testing in UK hospitals is a lottery, studies warn. BMJ 2010; 340: c2255. doi: /bmj.c Mikolajczak J, Hospers HJ, Kok G. Reasons for not taking an HIV-test among untested men who have sex with men: an internet study. AIDS Behav 2006; 10: doi: /s Deblonde J, De Koker P, Hamers FF, Fontaine J, Luchters S, Temmerman M. Barriers to HIV testing in Europe: a systematic review. Eur J Public Health 2010; 20: doi: / eurpub/ckp Sexually Transmissible Infections in Gay Men Action Group (STIGMA). Sexually transmitted infection testing guidelines for men who have sex with men (MSM). Endorsed by the Australian Chapter of Sexual Health Medicine. STIGMA; Available online at: [verified February 2011]. 31 Prestage G, Jin F, Zablotska I, Imrie J, Kaldor JM, Grulich AE. Trends in HIV prevalence among homosexual and bisexual men in eastern Australian states. Sex Health 2008; 5: doi: / SH Prestage G, Mao L, Kippax S, Jin F, Hurley M, Grulich A, et al. Use of viral load to negotiate condom use among gay men in Sydney, Australia. AIDS Behav 2009; 13: doi: /s Zablotska IB, Imrie J, Bourne C, Grulich AE, Frankland A, Prestage G. Improvements in sexual health testing among gay men in Sydney, Australia, Int J STD AIDS 2008; 19: doi: / ijsa Hallett TB, Garnett GP. Has global HIV incidence peaked? Lancet 2006; 368: doi: /s (06) Clements MS, Prestage G, Grulich A, Van de Ven P, Kippax S, Law MG. Modeling trends in HIV incidence among homosexual men in Australia J Acquir Immune Defic Syndr 2004; 35: doi: / Stall RD, Hays RB, Waldo CR, Ekstrand M, McFarland W. The gay 90s: a review of research in the 1990s on sexual behavior and HIV risk among men who have sex with men. AIDS 2000; 14(Suppl. 3): S doi: / Zablotska IB, Imrie J, Prestage G, Crawford J, Rawsthorne P, Grulich A, et al. Gay men s current practice of HIV seroconcordant unprotected anal intercourse: serosorting or seroguessing? AIDS Care 2009; Available online at: smpp/title~db=all~content=t ~tab=issueslist~branches=21- v21 21: doi: / Lansky A, MacKellar D, Gallagher KM, Lin LS, Sullivan PS, Onorato IM. Evaluation of the new National HIV Behavioral Surveillance System (NHBS) is imperative. Sex Transm Dis 2006; 33: doi: /01.olq d4 39 MacKellar DA, Gallagher KM, Finlayson T, Sanchez T, Lansky A, Sullivan PS. Surveillance of HIV risk and prevention behaviors of men who have sex with men a national application of venue-based, time-space sampling. Public Health Rep 2007; 122(Suppl. 1): Paquette D, Kropp R, Cule S, Wylie J, Husbands W, Myers T, et al. HIV prevalence and associated risk behaviours among men who have sex with men (MSM) in Canada. XVII International AIDS Conference, Mexico City, Mexico, 10 July Abstract no. TUPE0252. Manuscript received 29 September 2010, accepted 27 January