Article Text

PDF

Original article
Willingness to use HIV pre-exposure prophylaxis and the likelihood of decreased condom use are both associated with unprotected anal intercourse and the perceived likelihood of becoming HIV positive among Australian gay and bisexual men
  1. Martin Holt1,
  2. Dean A Murphy1,2,
  3. Denton Callander1,
  4. Jeanne Ellard1,
  5. Marsha Rosengarten3,
  6. Susan C Kippax4,
  7. John B F de Wit1,5
  1. 1National Centre in HIV Social Research, The University of New South Wales, Sydney, Australia
  2. 2Australian Federation of AIDS Organisations, Sydney, Australia
  3. 3Department of Sociology, Goldsmiths, University of London, London, UK
  4. 4Social Policy Research Centre, The University of New South Wales, Sydney, Australia
  5. 5Department of Social and Organizational Psychology, Utrecht University, Utrecht, the Netherlands
  1. Correspondence to Dr Martin Holt, National Centre in HIV Social Research, The University of New South Wales, Sydney NSW 2052, Australia; m.holt{at}unsw.edu.au

Abstract

Objectives To investigate willingness to use HIV pre-exposure prophylaxis (PrEP) and the likelihood of decreased condom use among Australian gay and bisexual men.

Methods A national, online cross-sectional survey was conducted in April to May 2011. Bivariate relationships were assessed with χ2 or Fisher's exact test. Multivariate logistic regression analysis was performed to assess independent relationships with primary outcome variables.

Results Responses from 1161 HIV-negative and untested men were analysed. Prior use of antiretroviral drugs as PrEP was rare (n=6). Just over a quarter of the sample (n=327; 28.2%) was classified as willing to use PrEP. Willingness to use PrEP was independently associated with younger age, having anal intercourse with casual partners (protected or unprotected), having fewer concerns about PrEP and perceiving oneself to be at risk of HIV. Among men who were willing to use PrEP (n=327), only 26 men (8.0%) indicated that they would be less likely to use condoms if using PrEP. The likelihood of decreased condom use was independently associated with older age, unprotected anal intercourse with casual partners (UAIC) and perceiving oneself to be at increased risk of HIV.

Conclusions The Australian gay and bisexual men the authors surveyed were cautiously optimistic about PrEP. The minority of men who expressed willingness to use PrEP appear to be appropriate candidates, given that they are likely to report UAIC and to perceive themselves to be at risk of HIV.

Statistics from Altmetric.com

Introduction

Pre-exposure prophylaxis (PrEP) for HIV involves prescribing antiretroviral drugs to HIV-negative people to protect them from HIV infection. Positive results from the iPrEx trial of daily tenofovir and emtricitabine use by men who have sex with men (MSM) prompted the Centers for Disease Control and Prevention to issue interim prescription and monitoring guidance for PrEP use in the USA.1 2 PrEP is not currently available in other countries, including Australia where our research was conducted.

US studies of gay, bisexual and other MSM during 2004–2009 found low levels of awareness of PrEP (from 16% to 25%) and very low levels of ‘informal’ or non-prescribed use of antiretroviral drugs for prevention (≤2%).3–9 Up to 2007, there was no evidence of informal PrEP use among HIV-negative men in Australia,10 although community-based surveys conducted in 2011 suggest informal PrEP use is at similar levels to those found in the USA (≤2%).11 12

Interest in using PrEP has also been assessed. US studies have found that 44%–80% of gay and bisexual men are interested in PrEP, depending on its efficacy.6 7 9 13 Younger, less educated men, those who dislike using condoms and those on lower incomes have been found to be more interested in PrEP, and men have indicated that they would be more willing to use PrEP if they do not have to pay for it.6 7 9 13 A recent study in Peru found that having to pay, partial efficacy and potential side effects diminished interest in using PrEP.14 ‘Risk compensation’ (decreased condom use) as a result of PrEP has also been investigated.15 One study found that, among men who reported unprotected anal intercourse and substance use and who were willing to use PrEP, 36% would be less likely to use condoms if they were taking PrEP.7 Another study found that, in an online survey of MSM, 7% would be less likely to use condoms if they were taking PrEP.16

Although PrEP is not currently available in Australia, reporting of positive trial results is likely to have raised awareness of the technology. Since the iPrEx study,1 the Partners PrEP study (of PrEP use by HIV-negative partners within heterosexual serodiscordant relationships) and the TDF2 trial (of PrEP use among HIV-negative men and women) have found efficacy levels of over 60%.17 18 We therefore set out to assess which Australian gay and bisexual men were willing to use PrEP and the potential impact on condom use in order to inform responses by HIV prevention programmes.

Methods

Participants and procedures

Data were collected as part of the PrEPARE Project, a study of Australian gay and bisexual men's attitudes to PrEP. The study design was approved by the University of New South Wales Human Research Ethics Committee (ref. HREC 11034). A national online survey of gay and bisexual men was constructed using NETQ software (NetQuestionnaires Nederland BV) and conducted in April to May 2011. The survey, containing 61 questions (some with multiple subitems), was advertised through Facebook, gaydar.com.au and email lists. Participants were eligible if they were at least 18 years old, male, lived in Australia and were gay, bisexual or other MSM. No incentive was offered for participation.

Participants were not given specific details about PrEP trial results or efficacy. Instead, participants were shown the following definition: ‘Pre-exposure prophylaxis means ‘prevention before exposure’. PrEP is when HIV-negative people take anti-HIV drugs before sex to try to reduce the chance of infection. Although there have been some promising trial results, PrEP is still being tested and is not currently available in Australia through the health system. PrEP is different from post-exposure prophylaxis (PEP), which is available in Australia’.

Measures

Questions assessing demographics, sexual practices with men, relationships, HIV testing and HIV status were adapted from existing research.11 12 HIV status was assessed with the question, ‘Based on your last test, what is your HIV status?’ (Don't know/no results yet; HIV negative; HIV positive). For men who indicated that they were HIV negative or did not know their status, the perceived likelihood of becoming HIV positive was assessed with the question, ‘How likely do you think it is you will become HIV-positive?’ (scored from 1=very unlikely to 5=very likely).19

Fifty-three attitudinal items developed by the research team were included in the survey, all scored from ‘strongly disagree’ (1) to ‘strongly agree’ (5). These items addressed attitudes to medicines (seven items), PrEP (25 items) and condoms (13 items). Eight previously validated items about HIV treatment optimism were included.20 The following only describes items used in this paper. Nine items, including ‘I enjoy using condoms’, ‘Condoms are uncomfortable’ (reverse scored) and ‘Condoms make sex more exciting’, formed a reliable scale (Cronbach's α=0.88) measuring Personal Experience in Using Condoms. A mean of these nine items formed the scale score (from 1=very negative to 5=very positive experience). Participants scoring ≥4 were regarded as having a positive experience with condoms (for each scale used in the analyses, we used a cut-off point of ≥4 to identify those men who expressed high agreement with the subject of the scale). Two items, ‘I'm comfortable talking about using condoms with partners’ and ‘It is easy to suggest using condoms with partners’, formed a reliable scale (α=0.76) measuring Confidence in Discussing Condoms with Partners. A mean of these two items formed the scale score (from 1=very unconfident to 5=very confident). Participants who scored ≥4 were regarded as confident in discussing condom use.

Two items, ‘I am worried about the side effects of PrEP drugs’ and ‘I would be worried about taking PrEP on an ongoing basis’, formed a reliable scale (α=0.72) measuring Concern About Using PrEP. A mean of these two items formed the scale score (from 1=very unconcerned to 5=very concerned about PrEP). Participants who scored ≥4 were regarded as concerned about PrEP use.

Primary outcomes of interest

A seven-item scale (α=0.81) measuring Willingness to Use PrEP was constructed using the following items:

  1. I would be willing to take PrEP to prevent getting HIV.

  2. I would take pills before and after sex if it would prevent me getting HIV.

  3. I would take a pill every day if it would prevent me getting HIV.

  4. I am going to take PrEP as soon as it becomes available.

  5. I would never need to take PrEP (reverse scored).

  6. I would be willing to pay for PrEP.

  7. I would take PrEP even if it wasn't 100% effective.

Each item was scored from 1 to 5. A mean of these items formed the scale score (from 1=very unwilling to 5=very willing to use PrEP). Men who scored ≥4 were regarded as willing to use PrEP.

A two-item scale (α=0.75) measuring the Likelihood of Decreased Condom Use if Using PrEP was constructed using the items ‘I would still use condoms if I was taking PrEP’ (reverse scored) and ‘I would be less likely to use condoms if I was taking PrEP’. Each item was scored from 1 to 5, and a mean of these two items was calculated to create the scale score (from 1=very unlikely to 5=very likely to decrease condom use). Men who scored ≥4 were regarded as likely to decrease condom use when using PrEP.

Data analysis

Before analysis, the database was inspected manually for duplicate or suspicious entries, for example, those with identical IP addresses, responses, free text or email addresses. SPSS V.19.0 was used for statistical analysis. Factor analysis was used to identify reliable scales among the attitudinal items. Only scales with high internal consistency reliability (α≥0.7) were used in the analyses. χ2 tests were used to assess bivariate relationships between independent variables, Willingness to Use PrEP and the Likelihood of Decreased Condom Use if Using PrEP (in cases where there was an expected cell frequency of <5, Fisher's exact test was used). Statistical significance was set at the p<0.05 level. Multivariate logistic regression analysis was used to determine which of the variables identified in the bivariate analyses were independently associated with Willingness to Use PrEP and the Likelihood of Decreased Condom Use. All variables with a significant bivariate relationship to the outcome measure were initially entered into the multivariate model. Variables that were non-significant in the initial model were excluded to construct the final multivariate model.

Results

Participant characteristics

The online survey was completed by 1283 men. It took between 18 and 52 min to complete. No duplicate or suspicious entries were identified. HIV-positive men (n=122) were excluded from this analysis, leaving a sample of 919 HIV-negative and 242 untested/unknown status men (n=1161). Details of the sample are shown in table 1. The sample was aged 18–69 years (M=31.6, SD=10.9). Twenty-one men (1.8%) identified as Aboriginal or Torres Strait Islander. The majority of the sample lived in New South Wales (37.1%), Victoria (25.8%) or Queensland (17.7%), with the remainder from the other five states and territories.

Table 1

Participant characteristics and Willingness to Use PrEP

Over four-fifths of the sample (80.5%) had been tested for HIV at least once, with over half the sample (59.4%) reporting an HIV test within the last 12 months. Over half the sample (51.7%) had a regular male partner at the time of the survey. Over two-thirds of the sample (68.8%) had at least two male sex partners in the 6 months prior to the survey. Condoms were more likely to be consistently used with casual male partners than with regular male partners; 56.0% reported using condoms consistently with casual partners and 28.9% reported using condoms consistently with regular partners (among men who reported anal intercourse with those partners). Only 3.5% of the sample thought it was likely that they would become HIV positive.

Based on scale scores, the majority of the sample (89.9%) reported a negative or neutral personal experience in using condoms. However, nearly three-quarters of the sample (73.6%) were confident in discussing the use of condoms with sexual partners. Just over half the sample (52.1%) was concerned about using PrEP.

A minority of men (14.3%) had received PEP, a course of antiretroviral drugs to prevent HIV infection after a suspected exposure to HIV. Only six men (0.5%) reported the previous use of antiretroviral drugs as informal/non-prescribed PrEP. These men all identified as gay, were aged 19–42 years (M=28.8; SD =7.89) and lived in Melbourne (n=3), Sydney (n=2) or Brisbane (n=1). Most (n=5) reported some unprotected anal intercourse with casual partners (UAIC) in the 6 months prior to survey.

Willingness to use PrEP

The mean score on the outcome measure, Willingness to Use PrEP, was 3.55 (SD=0.70). A score of ≥4 indicated willingness to use PrEP. Based on this classification, the majority of the sample (n=834; 71.8%) were unwilling to use or neutral about PrEP and just over a quarter of the sample (n=327; 28.2%) were willing to use PrEP.

Details of participant characteristics and their relationships with willingness to use PrEP are shown in table 1. Sexual identity, the HIV status of the participant's regular partner and use of PEP were unrelated to interest in PrEP. A range of variables were associated with willingness to use PrEP. Multivariate logistic regression analysis initially identified six variables independently associated with willingness to use PrEP, including prior use of antiretroviral drugs as PrEP. Because of an unusually large CI, this variable was excluded, resulting in the final model shown in table 2. This model accurately predicted willingness to use PrEP in 72% of cases. Younger men, men who had anal sex with casual partners (protected or unprotected), those with fewer concerns about PrEP and men who perceived themselves to be more likely to become infected with HIV were more willing to use PrEP.

Table 2

Multivariate logistic regression model of factors associated with Willingness to Use PrEP (n=1161)

Likelihood of decreased condom use if using PrEP (among men willing to use it)

We assessed the likelihood of decreased condom use among men willing to use PrEP (participants scoring ≥4 on the willingness to use PrEP scale; n=327). The mean score on the outcome measure, the Likelihood of Decreased Condom Use if Using PrEP, was 2.27 (SD=1.01). We considered men who scored ≥4 on this measure as likely to decrease condom use if using PrEP. Based on this classification, the majority of men willing to use PrEP indicated that they would maintain their condom use (n=301; 92.0%), while 26 men (8.0%) indicated that they would be less likely to use condoms.

Details of participant characteristics and their relationship to the likelihood of decreased condom use if using PrEP are shown in table 3. Multivariate logistic regression analysis identified three variables independently associated with the likelihood of decreased condom use (see table 4). This model accurately predicted the likelihood of decreased condom use in 93% of cases. Older men, those who perceived themselves to be at more risk of becoming HIV positive and those who had engaged in UAIC were more likely to indicate that they would be less likely to use condoms if they were taking PrEP.

Table 3

Participant characteristics and Likelihood of Decreased Condom Use if Using PrEP (n=327)

Table 4

Multivariate logistic regression model of factors associated with Likelihood of Decreased Condom Use if Using PrEP (n=327)

Discussion

Our survey found that the majority of Australian gay and bisexual men were cautious about PrEP and a minority were interested in using it. Very few men (0.5%) had previously used antiretroviral drugs as PrEP, similar to the levels of use reported in US studies but lower than that found in gay community surveys in Melbourne and Sydney.3 5–9 11 12 We currently know little about the circumstances of this informal PrEP use.

Over a quarter of our sample indicated that they were willing to use PrEP. The level of interest among our participants appears to be lower than that found in the US studies,6 7 9 16 although this may relate to a difference in outcome measures; our measure of willingness to use PrEP took into account having to take pills every day, concerns about efficacy, payment and perceived need. We were therefore more focused on men who were very positive about the technology. Willingness to use PrEP was independently associated with being younger, having anal intercourse with casual partners (protected or unprotected), having fewer concerns about PrEP use and perceiving oneself to be at increased risk of HIV infection. Apart from younger age, these factors are different to those found in previous research.6 7 9 13 16

The vast majority (92.0%) of men who were willing to use PrEP indicated that they were likely to maintain their current level of condom use; only 8% indicated that they would be less likely to use condoms. This may indicate a reluctance to rely upon PrEP instead of condoms but more likely reflects that condom use was already inconsistent among these men (39% of men willing to use PrEP reported some UAIC and 50% reported some unprotected anal intercourse with regular partners). This suggests that men may desire to use PrEP in situations in which they currently practise unprotected sex, rather than every instance of anal sex. The level of potential ‘risk compensation’ identified here is much lower than that found in a recent US study of ‘high-risk’ MSM7 but is similar to that found in an online survey of US MSM.16 In our study, the likelihood of decreased condom use if using PrEP was independently associated with older age, UAIC and perceiving oneself to be at increased risk of HIV infection. This suggests that men whose practices already put them at risk of HIV would use condoms less if receiving PrEP (indicating risk compensation).15 This could place men at increased risk of HIV if the effectiveness of PrEP was reduced as a result of poor adherence.1 2

Our findings have a number of limitations. The cross-sectional design meant that we could only assess associations rather than causal relationships. Our online sample appears similar to community samples of Australian gay men11 12 19 but may not be representative of the broader population of Australian MSM or ‘high-risk’ MSM.7 It is important to note that we asked men to indicate their willingness to use PrEP when most of them probably knew little about the technology. We did not assess how the level of interest in using PrEP varied according to its efficacy in preventing HIV,13 instead giving participants a generic description of PrEP. Participants were therefore invited to speculate about what PrEP would be like to use and its potential impact on condom use without knowing details about its efficacy or guidelines about how it should be taken. These factors would probably affect interest in PrEP. Given that most participants indicated that they would maintain condom use if they were using PrEP, our measure for the likelihood of decreased condom use was probably susceptible to social desirability bias. We suspect that it is unlikely that men would use both condoms and PrEP at the same time, if they believed PrEP was as effective as condom use in preventing HIV. We would therefore expect condom use to decline more markedly among men using PrEP, unless it was selectively used for episodes of unprotected sex.

Despite these limitations, our findings have identified which Australian gay and bisexual men are currently willing to use PrEP. While only a minority of men appear willing to use PrEP, they appear to be good candidates for PrEP (or other interventions), given that they are particularly likely to report UAIC and to perceive themselves to be at risk of HIV infection (UAIC is associated with HIV incidence in Australia).21 The majority of men who are willing to use PrEP do not anticipate that it would significantly change their current level of condom use, although once again it is men who engage in UAIC and who perceive themselves to be at risk of HIV infection who are most likely to signal the potential for decreased condom use. Overall, there is a relatively small group of gay and bisexual men in Australia who would be willing users of PrEP; these men appear to be an appropriate target group, if PrEP is made available in Australia, given that their sexual practices put them at risk of HIV and they perceive themselves to be at risk of infection.

Key messages

  • A minority of Australian gay and bisexual men (28%) are willing to use HIV pre-exposure prophylaxis (PrEP), depending on efficacy and perceived need.

  • Willingness to use PrEP is particularly associated with unprotected anal intercourse with casual partners and the perceived likelihood of becoming HIV positive.

  • The likelihood of decreased condom use as a result of PrEP use appears to be low.

  • PrEP is not currently available in Australia but there appears to be a group of gay and bisexual men who would benefit from it.

Acknowledgments

Thanks to all the survey participants, Joanne Bryant and Limin Mao for their statistical advice and two anonymous reviewers for their feedback on the paper.

References

View Abstract

Footnotes

  • Funding This study was supported by a University of New South Wales GoldStar award. The National Centre in HIV Social Research is supported by the Australian Government Department of Health and Ageing.

  • Competing interests None.

  • Patient consent The study used an anonymous online survey to collect data. Starting and completing the survey was taken as evidence of consent.

  • Ethics approval The University of New South Wales Human Research Ethics Committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.