Sexual Risk Behaviors among Men Who Have Sex with Men Using Erectile Dysfunction Medications

Abstract

This study examined relationships between use of the phosphodiesterase type-5 (PDE-5) inhibitors (erectile dysfunction medications) sildenafil (Viagra^®, Pfizer, New York, NY), tadalafil (Cialis^®, Eli Lily, Indianapolis, IN), and/or vardenafil (Levitra^®, Bayer, Berlin, Germany), substance use, perceptions of risk, and sexual behavior in men who have sex with men (MSM). MSM (N = 342) attending a gay pride festival completed a brief survey assessing sexual behavior, risk perceptions, and substance use, including the use and the source of PDE-5 inhibitors. More than a quarter of the sample (26.3%, n = 89) reported having ever used a PDE-5 inhibitor. Those reporting use of PDE-5 inhibitors had higher rates of sexual risk behaviors and differed in their assessment of the risk of HIV transmission for unprotected anal sex. Users who received PDE-5 inhibitors from their doctors did not report sexual behaviors that differed significantly from those who received PDE-5 inhibitors from nonphysician sources. In a sequential logistic regression analysis, recent PDE-5 inhibitor use was associated with unprotected anal sex after accounting for the influence of age, education, ethnic identity, and substance use. Many MSM users of erectile dysfunction drugs report behaviors that may place their and others' health at risk. Interventions to reduce risk among MSM PDE-5 inhibitor users should be explored.

Introduction

D uring the 1980s and 1990s the number of HIV infections decreased among men who have sex with men (MSM).¹ In recent years, accumulating evidence suggests that HIV and other sexually transmitted infections (STIs) are on the rise in this group.² For example, several recent outbreaks of syphilis and gonorrhea have been documented among MSM.^3

–6 Other recent studies report a high incidence of STIs in HIV-positive MSM, suggesting unsafe sexual practices that run the risk of exposing others to HIV.^7,8

Factors facilitating increases in high-risk sexual behavior among MSM are complex and may include such things as HIV optimism due to effective treatments and safe-sex fatigue.^9,10 Another potential factor that has received attention lately is the use of phosphodiesterase type-5 (PDE-5) inhibitors, such as sildenafil, in a manner that could increase the likelihood of infections in the person taking the drug or their partner(s).^{11

–23} PDE-5 inhibitors are a class of drugs used to treat erectile dysfunction in men. These drugs work by blocking the action of an enzyme that results in an increase in cyclic gaunosine monophosphate and subsequent smooth muscle relaxation and erection of the penis.^24,25

Within a couple of years after the approval of the first PDE-5 inhibitor, evidence began to emerge of potentially problematic use of PDE-5 inhibitors. There is evidence that young college-age men and women are increasingly using PDE-5 inhibitors without any underlying impairment in sexual functioning.^11,12 Men who report compulsive online sexual behaviors also appear to be at-risk for misuse of PDE-5 inhibitors.¹³ In data published by Sherr and colleagues,¹⁴ MSM in Britain reported high rates of recreational sildenafil use, fueling concerns that sildenafil might be facilitating high-risk sexual behavior among MSM. Research later published by Mansergh and colleagues¹⁵ documented the use of sildenafil in gay and bisexual men attending circuit parties in the United States. Subsequent studies have continued to show links between sildenafil use, other substance use, and high-risk sexual behaviors in MSM.^{14

–23} Self-described motivations for use include “adding to the fun” and “to have sex for hours.”¹⁹

It should be noted that use of PDE-5 inhibitors for treating erectile dysfunction (ED) and the subsequent improvement in quality of life among those suffering from this condition is clearly warranted. Studies suggest that the prevalence of ED among HIV-positive men is much higher than in the general population; yet, relatively few HIV-positive men appear to be receiving treatment for ED.^26,27 Failure to treat ED in seropositive men may be related to other problem behaviors such as an increased likelihood of engaging in unprotected receptive anal intercourse (URAI) and antiretroviral mediation adherence difficulties.^28,29 However, the link between ED and antiretroviral medication adherence problems is less clear.³⁰ The ethical considerations around prescribing PDE-5 inhibitors to HIV-positive individuals are indeed complex.^29,31

On the other hand, there do appear to be valid concerns that PDE-5 inhibitor use may be facilitating and increasing sexual risk among MSM. In 2006, the National Institute of Mental Health organized a conference designed to identify gaps in knowledge regarding the use of PDE-5 inhibitors in MSM and to spur additional research into the phenomenon.³² One issue raised at this conference was that the published research has been limited in geographic scope, with most studies in the United States collecting data from men residing in urban areas on the East or West Coast.^15

–19 Additional studies from other geographic areas and venues would improve the base of knowledge regarding the prevalence and nature of PDE-5 inhibitor use. A second concern was the lack of information about the use of the more recently approved PDE-5 inhibitors tadalafil and vardenafil.

In addition, previous research has provided relatively little information on where men are obtaining these drugs. Nor does there appear to be much published information on whether the source of these drugs is associated with different levels of sexual risk behavior. Finally, previous work has not examined if men who use PDE-5 inhibitors differ in their perceptions of risk.

The present study builds on previous research by addressing these issues. Our study gathered information on the use of sildenafil, tadalafil, and vardenafil, the source of these PDE-5 inhibitors, the use of other recreational drugs, estimations of risk, and sexual risk behavior in MSM attending a gay pride celebration in Denver, Colorado. We hypothesized that: (1) a heterogeneous sample of MSM attending a gay pride celebration would exhibit similar patterns of use that have been demonstrated in previous studies; (2) sildenafil, tadalafil, and varvenafil would be significantly associated with high-risk sexual behaviors and substance use; (3) men using PDE-5 inhibitors would assess transmission risk differently than men who do not use these medications, and (4) the source of the PDE-5 inhibitor would be associated with high-risk sexual behaviors.

Methods

Participants, setting, and procedures

To investigate the association between PDE-5 inhibitors and sexual risk behavior, 378 people attending a gay pride festival in Denver, Colorado, were recruited to complete self-administered surveys. Participants were asked to complete a 9-page survey concerning HIV and AIDS as they walked through the festival grounds where retail venders and community organizations occupied display booths, one of which was rented for the purposes of this study. Participants were told that the survey was about sexual relationships, contained personal questions about their sexual history and substance use, was anonymous, and required approximately 15 minutes to complete. Research assistants attempted to recruit all men walking through the vending area. Over 70% of men approached agreed to complete the survey. Participants' names were not collected with the survey at any time. Participants were offered $2 for completing the survey, and an additional $2 donation was made to a local charity. All research procedures were approved by the Institutional Review Board of the University of Colorado Denver. Recruitment methods such as these have been used extensively in prior research and have been shown to yield samples that are roughly representative of methods that use more sophisticated sampling procedures.^33
–35

This festival was chosen as the site for the survey because of the more than 250,000 people who attend this annual event.³⁶ Previous research has shown that men who attend gay pride festivals report significant rates of high-risk sexual behaviors.³⁷ Almost 60% of all AIDS cases in Colorado have been reported in Denver County and more than 70% of Colorado's HIV infections have occurred among men who have sex with men.³⁸

Measures

Participants completed self-administered anonymous surveys that included measures of demographic information, PDE-5 inhibitor use, other substance use, perceptions of risk, and sexual practices.

Demographics

Participants were asked their age, years of education, income, ethnicity, home zip code, whether they self-identified as gay, bisexual, or heterosexual, relationship status, whether they had been tested for HIV antibodies, and if so the results of their most recent HIV test.

PDE-5 inhibitor use

Participants were asked to indicate “yes” or “no” if they had ever used sildenafil, tadalafil or vardenafil. Participants were asked how they obtained the PDE-5 inhibitors: from a doctor, a friend, a sex partner, a stranger, the Internet, or some other source. Participants also reported the amount of recent (past 3 months) PDE-5 inhibitor use.

Substance use

Participants were asked questions concerning the frequency of use of alcohol, marijuana, nitrite inhalants (poppers), cocaine, methamphetamine, ecstasy, ketamine, GHB, and rophynol in the previous 3 months. Street names were included for many of the substances. The frequency of use was measured using a 4-point Likert scale (1 = “none” to 4 = “at least once a week”). Participants were also asked the number of times they used alcohol in conjunction with sexual activity and the number of times they had used drugs in conjunction with sexual activity in the previous 3 months. This measure is similar to measures use in our previous research.^39,40

Sexual practices

Sexual behavior was assessed by asking participants to report the number of times they had engaged in anal intercourse, as the insertive and receptive partner, as well as the number of times they used or did not use condoms during anal intercourse in the past 3 months. We were particularly interested in unprotected anal intercourse because of the high risk that this behavior poses for HIV transmission. Participants also recorded the number of sexual partners with whom they had engaged in each behavior in the previous 3 months. In addition, participants indicated if they had ever had a sexually transmitted infection. Consistent with our prior work, open response formats were used for the sexual behavior measures to reduce response bias and to minimize measurement error.⁴¹ Measures similar to these have been found to be reliable in self-reported sexual behavior assessments and to yield aggregate indices of risk that are comparable to those obtained by finer-grained partner-by-partner sexual behavior assessments.^42,43

Estimates of risk

Participants responded to items that asked them to estimate the risk of HIV transmission in two scenarios. In both scenarios, an HIV-positive man and an HIV-negative man engaged in a single act of unprotected anal sex with ejaculation. In the first scenario, the HIV-positive man was the insertive partner; in the second scenario, the HIV-negative man was the insertive partner. Participants were asked to estimate the likelihood that the HIV-negative man would contract HIV based on his sexual activities described in the scenario (top or bottom). For both scenarios, response choices were: less than 1%, 1%–5%, 6%–10%, 11%–20%, and greater than 20%.

Data quality assurances and statistical analyses

All surveys were examined for inconsistencies and invalid responses. Missing data were omitted from analyses, resulting in slightly different ns for various statistical tests. The primary outcome variable in our sequential logistic regression was a sum of the unprotected insertive and receptive anal acts in the previous three months. Because distributions of sexual behavior were highly skewed, nonparametric analyses were used as recommended by Hays.⁴⁴ Lifetime use of PDE-5 inhibitors was measured categorically. Two-tailed significance levels were used for all tests.

Results

Participants

A total of 378 individuals completed questionnaires. Because we specifically wanted to focus on men who have sex with men (MSM), participants who self-identified as heterosexual and reported no sexual contact with a man in the 3 months prior to the study (n = 27), or who reported their gender as female (n = 3) were eliminated from further analyses. In addition, five participants were eliminated for problematic or random responding (e.g., circling all possible response choices). Of the 378 individuals who completed a survey, 342 (90%) were included in the final analyses.

Among the 342 participants, the mean age was 34.7 years (standard deviation [SD] = 11.6), and the average years of education was 14.6 (SD = 2.1). Ninety percent of participants self-identified as gay and 10% as bisexual. The majority of the sample was white (75%), with the remainder being Latino (8%), African American (6%), Asian American (4%), Native American (2%), or other/mixed ethnic heritage (5%). The majority of participants indicated that they live in Colorado (88%). Twenty-four percent of participants had annual incomes below $16,000, 26% had incomes between $16,000 and $30,000, 23% had annual incomes between $31,000 and $45,000, and 27% had incomes above $45,000. The majority (92%) reported having been tested for HIV antibodies; of those, 80% tested HIV negative, 16% HIV positive, and 4% did not know their test results. Seven percent reported a lifetime history of injection drug use.

Sexual activity

In the present study, 75% (n = 258) of the respondents were sexually active. Of the men reporting sexual activity, 66% (n = 170) reported unprotected anal intercourse (UAI) in the previous 3 months. In the portion of the sample reporting UAI, 54% indicated that they had only 1 sexual partner and 46% reported sexual activity with 2 or more partners in the last 3 months.

PDE-5 inhibitor use

More than a quarter of the sample (26.3%, n = 89) reported having ever used sildenafil, tadalafil, or vardenafil. A smaller portion of the sample (10.1%) reported using one of these medications in the previous 3 months. Among men who had used a PDE-5 inhibitor, sildenafil was the most commonly used medication (92.1%), followed by tadalafil (13.5%), and vardenafil (8.9%). Twelve men reported having used two or more PDE-5 inhibitors over their lifetimes.

Demographic factors and PDE-5 inhibitor use

Overall, PDE-5 inhibitor users ranged in age from 20 to 76. Men reporting that they had ever used PDE-5 inhibitors were significantly older (M = 39.3 years, SD = 10.9) than those reporting no use (M = 33.1 years, SD = 11.4, t [338] = 4.42, p < 0.001). A significantly higher percentage of HIV-positive respondents reported use of PDE-5 inhibitor (39.2%) than HIV-negative participants (25.9%; χ² [1, N = 307] = 3.74, p = 0.05). PDE-5 inhibitor users also had higher educational attainment (M = 15 years, SD = 2.1) than nonusers (M = 14.4 years, SD = 2.1; Mann-Whitney z = 2.28, p < 0.05) and overall higher income (Mann-Whitney z = 2.74, p < 0.01). PDE-5 inhibitor users and nonusers did not differ with regards to race, ethnicity, or sexual orientation (gay versus bisexual).

Sources of PDE-5 inhibitors

The majority of men reporting PDE-5 inhibitor use received them from their physician (55.1%). The remainder reported other sources, with 32.6% reporting receiving them from a friend, 7.9% purchased them on the Internet, and 6.7% obtained them from a sex partner. No participants reported receiving a PDE-5 inhibitor from a stranger. A small number (4.5%) indicated that they obtained PDE-5 inhibitors from other sources including during foreign travel to Mexico and as part of a research study.

PDE-5 inhibitor use and sexual risk

Men reporting the use of PDE-5 inhibitors reported significantly more UAI acts (M = 6.78, SD = 12.89) in the past 3 months than those who have never used PDE-5 inhibitors (M = 3.87, SD = 9.54, Mann-Whitney z = 2.26, p < 0.05). As seen in Table 1, men who have used PDE-5 inhibitors also had significantly more male partners, more total partners (male + female), greater numbers of men with whom they had unprotected receptive (URAI) and insertive (UIAI) anal intercourse, greater numbers of URAI and UIAI acts, and a higher number of times having sex after using drugs. In addition, PDE-5 inhibitor users were twice as likely to report having ever had a sexually transmitted infection other than HIV (36%) than nonusers (18%; χ² [1, N = 338] = 11.69, p < 0.01).

Table 1.

Sexual Behavior Among Users and Nonusers of Phosphodiesterase Type-5 Inhibitors

	Users of PDE-5 inhibitors (n = 89)		Nonusers (n = 253)
	Mean	SD	Mean	SD	z^a	P
Unprotected anal intercourse (# of acts)	6.78	12.89	3.87	9.54	2.26	0.048
Unprotected receptive anal intercourse (# of acts)	3.65	10.22	2.20	7.08	1.84	0.066
Unprotected insertive anal intercourse (# of acts)	3.12	8.11	1.69	5.18	2.40	0.016
Unprotected insertive anal partners	0.85	1.40	0.64	1.53	2.17	0.030
Unprotected receptive anal partners	1.18	4.80	0.54	1.05	2.13	0.033
Total number male partners	2.55	5.55	1.62	2.60	1.98	0.048
Total number sexual partners (male & female)	2.64	5.54	1.74	2.60	1.98	0.048
Number of sexual acts on drugs	1.13	5.60	0.44	2.19	2.63	0.009
Number of sexual acts on alcohol	1.22	3.51	1.84	7.73	ns	ns

Mann-Whitney test. Reporting period = past 3 months.

PDE-5, phosphodiesterase type-5; ns, not significant.

Sources of PDE-5 inhibitors and sexual behavior

Men who obtained PDE-5 inhibitors from a physician were significantly older (M = 42.4 years, SD = 11.2) than men who obtained these medications from another source (M = 35.5 years, SD = 9.4; t [86] = 3.052, p < 0.01). In contrast to our hypothesis, users who received PDE-5 inhibitors from their doctors did not report sexual behaviors that differed significantly from those who received PDE-5 inhibitors from other sources (all p s for behaviors shown in Table 1 >0.10).

PDE-5 inhibitor use and substance use

Although PDE-5 inhibitor users reported a significantly greater number of sexual acts while under the influence of drugs, differences in substance use between PDE-5 inhibitor users and nonusers were modest in this sample. PDE-5 inhibitor users were significantly more likely to report use of poppers over the past 3 months (25.3%) than nonusers of PDE-5 inhibitors (9.7%; %; χ² [1, N = 335] = 11.95, p < 0.01). There was a trend for PDE-5 inhibitor users to have higher rates of marijuana use (37%) than nonusers of PDE-5 inhibitors (27%, p < 0.10). For the other substances assessed, users of PDE-5 inhibitors reported substance use rates comparable to the rates of nonusers of PDE-5 inhibitors.

PDE-5 inhibitor use and estimates of risk

Users and nonusers of these medications did not differ in their perceptions of the risk of HIV transmission when an HIV-positive man is the insertive partner during unprotected anal sex with an HIV-negative man. Users of PDE-5 inhibitors had significantly lower perceptions of the risk of HIV transmission when an HIV-negative man is the insertive partner during unprotected anal sex with an HIV-positive man (M = 2.88, SD = 1.52), relative to nonusers of PDE-5 inhibitors (M = 3.31, SD = 1.37; Mann-Whitney z = 2.32, p < 0.05).

Multivariate test of association between demographic factors, use of substance in conjunction with sex and PDE-5 inhibitors

We conducted a sequential logistic regression analysis to examine the independent association of PDE-5 inhibitors, after controlling for factors that have been commonly shown to be related to sexual risk. Demographic features, use of substances in conjunction with sexual activity, and use of medications intended for erectile dysfunction were used to predict participation in unprotected anal intercourse, as shown in Table 2. Demographic characteristics were entered as control variables on the first step and were significant predictors of UAI, χ² (3, N = 327) = 18.6, p < 0.001. On the first step, education (p < 0.05) and ethnic identity (p < 0.01) were significant factors and age showed a trend toward significance (p = 0.052). Substance use (alcohol and/or drugs) while having sex was a significant predictor of UAI after controlling for demographic factors, χ² (1, N = 327) = 24.6, p < 0.001. Use of a PDE-5 inhibitor was entered on the final step and significantly added to the prediction of unprotected anal sex, χ² (1, N = 327) = 4.68, p < 0.05, indicating that the use of medications intended for erectile dysfunction is significantly associated with unprotected anal intercourse, even after controlling for the effects of demographic characteristics and substance use while having sex.

Table 2.

Sequential Logistic Regression Analysis Predicting Unprotected Anal Intercourse versus No Unprotected Anal Intercourse

Variable and step	OR	CI	P
1. Age	0.98	(0.96, 1.0)	0.052
Education	0.87	(0.78, 0.97)	0.011
Ethnic identity	0.49	(0.29, 0.83)	0.008
2. Anal Sex after using substances	3.49	(2.10, 5.81)	<0.001
3. PDE-5 Inhibitor Use	1.84	(1.05, 3.21)	0.032

OR, odds ratio; CI, confidence interval; PDE-5, phosphodiesterase type-5.

Discussion

The results are consistent with previous work that has found the use of PDE-5 inhibitor medications to be associated with sexual risk behavior among MSM.^{15

–23} Sildenafil remains the most popular PDE-5 inhibitor but smaller numbers of MSM have also tried tadalafil and vardenafil. Partly in response to this concern and in response to pressure from some researchers, the manufacturers of PDE-5 inhibitors are now including cautionary information regarding STI and HIV risk in medication advertisements and in the patient inserts.^21,45
–47

Men who use medication intended to treat erectile dysfunction are engaging in sexual behavior that places themselves and others at risk for transmission of HIV and other STIs significantly more than those who have not used a PDE-5 inhibitor. Interestingly, men receiving these medications through their doctors are engaging in the same frequencies of high-risk sexual behaviors as those who are getting PDE-5 inhibitors from other sources. Many studies suggest that communication between doctor and patient may significantly influence patient health behavior.⁴⁸ Physicians should consider assessing for use of PDE-5 inhibitors while gathering the medical history of their patients. This is especially important for HIV-positive patients, as previous work has indicated that PDE-5 inhibitors have potential pharmacologic interactions with medications for the treatment of HIV infection.⁴⁹ Furthermore, when receiving a request to prescribe PDE-5 inhibitors, in addition to assessing for the physiological need for PDE-5 inhibitors, physicians may want to inquire about their patients' motivations and expectations regarding the use of these medications. This may open the door for counseling both those to whom doctors prescribe PDE-5 inhibitors and other users about safer-sex practices. In addition, educators and others involved in HIV and STI prevention may wish to discuss use of these drugs with target populations and how such drugs may facilitate sexual risk.

In contrast to many of the previous studies, PDE-5 inhibitor users in this sample reported modestly increased rates of the use of other substances. As might be expected, poppers, a second substance that facilitates sexual activity, was used with greater frequency among PDE-5 inhibitor users. The co-use of these two substances may be another topic important for physicians to address, given that the simultaneous use of poppers and PDE-5 inhibitors is contraindicated and could lead to a life-threatening drop in blood pressure.⁵⁰ The reasons for the lower co-occurrence of PDE-5 inhibitor use and the use of illegal substances in this sample are unclear; however, this may reflect a cultural difference between MSM on the east and west coast versus MSM in the Rocky Mountain region. Further research would be needed to confirm these results.

One implication of this finding is that use of a PDE-5 inhibitor is associated with risk even in situations where it is not part of a pattern of polysubstance use. In our analyses, use of illicit drugs in conjunction with sexual activity was associated with HIV risk behavior but use of a PDE-5 inhibitor was also associated with risk even after controlling for the effects of substance use.

Consistent with previous research, both PDE-5 inhibitor users and nonusers overestimated the risk of HIV transmission as the result of a single unprotected sexual act.^51
–53 Men with a lifetime history of PDE-5 inhibitor use perceived the risk of HIV transmission differently than nonusers. Men who use these medications reported lower perceptions of risk of HIV transmission when an HIV-negative man is the insertive partner during unprotected anal sex with an HIV-positive man. Both groups perceived the risk of transmission as higher when the HIV-positive man is the insertive partner and the two groups did not differ in their perceptions of this risk. Differences in risk perceptions may be a cognitive factor that underlies the use of substances that facilitate risk. It may be useful for future studies to examine in greater detail the role of cognitive factors in decision making regarding the problematic use of drugs that facilitate sexual activity. Such research might include an in-depth examination of the expectations of those using these drugs.

The data for this study were collected from a convenience sample of men attending a gay pride festival in Denver, Colorado; generalization to other regions and populations may not be warranted. The use of a cross-sectional study design limits drawing causal conclusions concerning the associations between PDE-5 inhibitor use and high-risk sexual behavior. In addition, the surveys relied on self-reported behavior, potentially leading respondents to over or underreport risk behaviors. We did not assess participants' underlying sexual functioning. Including measures of sexual functioning in future research on PDE-5 inhibitor use may shed additional light on use by individuals with sound etiologic reasons for using the drugs.

This study corroborates other studies documenting the use and abuse of drugs used to treat erectile dysfunction.^{11

–20} Much of the published behavioral research on this class of drugs has been limited to the use of sildenafil. This study contributes additional information to the behavioral literature around the use of vardenafil and tadalafil as well as information regarding the sources of PDE-5 inhibitors and some preliminary information about differences in perceptions of risk between men who use and do not use these medications.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Centers for Disease Control and Prevention. CDC HIV/AIDS Fact Sheet: HIV/AIDS among Men Who Have Sex With Men. Atlanta: US Department of Health and Human Services, July 2005.

Centers for Disease Control and Prevention. Increases in HIV diagnoses—29 states, 1999–2000. MMWR Morb Mortal Wkly Rep, 2003; 52:1145–1148.

Centers for Disease Control and Prevention. Outbreak of syphilis among men who have sex with men—Southern California, 2000. MMWR Morb Mortal Wkly Rep, 2001; 50:117–120.

Centers for Disease Control and Prevention. Primary and secondary syphilis among men who have sex with men—New York City, 2001. MMWR Morb Mortal Wkly Rep, 2002; 51:853–856.

Centers for Disease Control and Prevention. Primary and secondary syphilis—United States, 2000–2001. MMWR Morb Mortal Wkly Rep, 2002; 51:971–973.

Rietmeijer

, Patnaik

, Judson

, Douglas

. Increases in gonorrhea and sexual risk behaviors among men who have sex with men: A 12-year trend analysis at the Denver Metro Health Clinic. Sex Transm Dis, 200; 30:562–567.

Erbelding

, Chung

, Kamb

, Irsin

, Rompalo

. New Sexually transmitted diseases in HIV-infected patients: Markers for onging HIV transmission behavior. J Acquir Immune Defic Syndr, 2003; 33:247–242.

Scheer

, Chu

, Klausner

, Katz

, Schwarcz

. Effects of highly active antiretroviral therapy on diagnoses of sexually transmitted diseases in people with AIDS. Lancet, 2001; 357:432–435.

Van de Ven

, Prestage

, Crawford

, Grulich

, Kippax

. Sexual risk behaviour increases and is associated with HIV optimism among HIV-negative and HIV-positive gay men in Sydney over the 4 year period to February 2000. AIDS, 2000; 14:2951–2953.

10.

Sowell

, Lindsey

, Spicer

. Group sex in gay men: Its meaning and HIV prevention implications. J Assoc Nurses AIDS Care, 1998; 9:59–71.

11.

Delate

, Simmons

, Motheral

. Patterns of use of sildenafil among commercially insured adults in the United States: 1998–2002. Int J Impotence Res, 2004; 16:313–318.

12.

Aldridge

, Measham

. Sildenafil (Viagra) is used as a recreational drug in England. BMJ, 1999; 318:669.

13.

Graham

, Polles

, Gold

. Performance enhancing, non-prescription use of erectile dysfunction medications. J Addict Dis, 25:61–68.

14.

Sherr

, Bolding

, Maguire

, Elford

. Viagra use and sexual risk behavior among gay men in London. AIDS, 2000; 14:2051–2053.

15.

Mansergh

, Colfax

, Marks

, Rader

, Guzman

, Buchbinder

. The circuit party men's health survey: Findings and implications for gay and bisexual men. Am J Public Health, 2001; 91:953–958.

16.

Kim

, Kent

, Klausner

. Increased risk of HIV and sexually transmitted disease transmission among gay or bisexual men who use Viagra, San Francisco 2000–2001. AIDS, 2002; 15:1425–1428.

17.

Chu

, McFarland

, Gibson

et al. Viagra use in a community recruited sample of men who have sex with men, San Francisco. J Acquir Immune Defic Syndr, 2003; 33:191–193.

18.

Spindler

, Scheer

, Chen

et al. Viagra, methamphetamine, and HIV risk: Results from a probability sample of MSM, San Francisco. Sex Transm Dis, 2007; 34:586–691.

19.

Pantalone

, Bimbi

, Parsons

. Motivations for recreational use of erectile enhancing medications in urban gay and bisexual men. Sex Transm Infect, 2008; 84:458–462.

20.

Crosby

, DiClemente

. Use of recreational Viagra among men having sex with men. Sex Transm Infect, 2004; 80:466–468.

21.

Paul

, Pollack

, Osmond

, Cantania

. Viagra (sildenafil) use in a population-based sample of U.S. men who have sex with men. Sex Transm Infect, 2005; 32:531–533.

22.

Jackson

. PDE-5 inhibitors and HIV risk: Current concepts and controversies. Int J Clin Pract, 2005; 59:1247–1248.

23.

Swearingen

, Klausner

. Sildenafil use, sexual risk behavior, and risk for sexually transmitted diseases, including HIV infection. Am J Med, 2005; 118:571–577.

24.

Rosen

, McKenna

. PDE-5 inhibition and sexual response: Pharmacological mechanisms and clinical outcomes. Annu Rev Sex Res, 2002; 13:36–88.

25.

Padma-Nathan

, Christ

, Adaikan

. et al. Original research – Phosphodiesterase inhibitor therapy for ED: Pharmacotherapy for erectile dysfunction. J Sex Med, 2004; 1:128–140.

26.

Ende

, Lo Re

, DiNubile

, Mounzer

. Erectile dysfunction in an urban HIV-positive population. AIDS Patient Care STDs, 2006; 20:75–78.

27.

Lallemand

, Salhi

, Linard

, Giami

, Rozenbaum

. Sexual dysfunction in 156 ambulatory HIV-infected men receiving highly active antiretroviral therapy combinations with and without protease inhibitors. J Acquir Immune Defic Syndr, 2002; 30:187–190.

28.

Cachay

, Mar-Tang

, Matthews

. Screening for potentially transmitting sexual risk behaviors, urethral sexually transmitted infection, and sildenafil use among males entering care for HIV infection. AIDS Patient Care STDs, 2004; 18:349–354.

29.

Trotta

, Ammassari

, Murri

et al. Self-reported sexual dysfunction is frequent among HIV-infected persons and is associated with suboptimal adherence to antiretrovirals. AIDS Patient Care STDs, 2008; 22:291–299.

30.

Crum-Cianflone

, Barvo

, Hale

et al. Erectile dysfunction and hypogonadism among men with HIV. AIDS Patient Care STDs, 2007; 21:9–19.

31.

Kell

, Sadeghi-Nejad

, Price

. An ethical dilemma: Erectile dysfunction in the HIV-positive patient: To treat or not to treat. Int J STD AIDS, 2002; 13:355–357.

32.

Rosen

, Catania

, Ehrhardt

et al. The Bolger conference on PDE-5 inhibition and HIV risk: Implications for health policy and prevention. J Sex Med, 2006; 3:960–975.

33.

Gore-Felton

, Kalichman

, Brondino

, Benotsch

, Cage

, DiFonzo

. Childhood sexual abuse and HIV risk among men who have sex with men: Initial test of a conceptual model. J Fam Violence, 2006; 21:263–270.

34.

Grov

, Parsons

, Bimbi

. Sexual risk behavior and venues for meeting sex partners: An intercept survey of gay and bisexual men in LA and NYC. AIDS Behav, 2007; 11:915–926.

35.

Halkitis

, Parsons

. Recreational drug use and HIV-risk sexual behavior among men frequenting gay social venues. J Gay Lesb Soc Serv, 2002; 14:19–38.

36.

The Center. Pridefest. 2008. http://pridefestdenver.org/index.html. 2008 December 26

37.

Kalichman

, Gore-Felton

, Benotsch

, Cage

, Rompa

. Trauma symptoms, sexual behaviors, and substance abuse: Correlates of childhood sexual abuse and HIV risks among men who have sex with men. J Child Sex Abus, 2004; 13:1–15.

38.

Colorado Department of Public Health and Environment. HIV and AIDS in Colorado: Surveillance Report (Revised December 2007) 2000. www.cdphe.state.co.us/dc/HIVandSTD/HIV_STDSurv/MonitortheEpi.pdf. 2008 May 6.

39.

Benotsch

, Kalichman

, Cage

. Men who have met sex partners via the Internet: Prevalence, predictors, and implications for HIV prevention. Arch Sex Behav, 2002; 31:177–183.

40.

Benotsch

, Seeley

, Mikytuck

, Pinkerton

, Nettles

, Ragsdale

. Substance use, medications for sexual facilitation, and sexual risk behavior among traveling men who have sex with men. Sex Transm Dis, 2006; 33:706–711.

41.

Benotsch

, Mikytuck

, Ragsdale

, Pinkerton

. Sexual risk and HIV acquisition among MSM travelers to Key West, Florida: A mathematical modeling analysis. AIDS Patient Care STDs, 2006; 20:549–556.

42.

Kauth

, St. Lawrence

, Kelly

. Reliability of retrospective assessments of sexual hiv risk behavior: A comparison of biweekly, three-month, and twelve-month self-reports. AIDS Educ Prev, 1991; 3:207–214.

43.

Pinkerton

, Benotsch

, Mikytuck

. When do simpler sexual behavior data collection techniques suffice? An analysis of consequent uncertainty in HIV acquisition risk estimates. Eval Rev, 2007; 31:401–412.

44.

Hays

. Statistics, 4th. Fort Worth, TX: Holt, Rinehart, and Winston, 1988.

45.

Pfizer. Viagra (sildenafil citrate) tablets [patient information]. Revised October 2007. www.pfizer.com/pfizer/download/uspi_viagra.pdf. 2009 January 2.

46.

Bayer Pharmaceuticals Corporation. Levitra (vardenafil HCL) tablets [patient information], revised May 2008. www.univgraph.com/bayer/inserts/levitra.pdf. 2009 January 2.

47.

Eli Lilly and Company. Cialis (tadalafil) tablets [patient information], revised January 7, 2008. http://pi.lilly.com/us/cialis-ppi.pdf. 2009 January 2.

48.

Ong

LML

, de Haes JCJM Hoos

, Lammes

. Doctor–patient communication: A review of the literature. Soc Sci Med, 1995; 40:903–918.

49.

Piscitelli

, Gallicano

. Interactions among drugs for HIV and opportunistic infections. N Engl J Med, 2001; 44:984–996.

50.

James

. Viagra warning re: “poppers” and notice re: protease inhibitors. AIDS Treat News, 1998; 294:1.

51.

Cohen

, Bruce

. Sex and mortality: Real risk and perceived vulnerability. J Sex Res, 1997; 34:279–291.

52.

Katz

, Gerberding

. Postexposure treatment of people exposed to the human immunodeficiency virus through sexual contact or injection-drug use. N Engl J Med, 1997; 336:1097–1100.

53.

Pinkerton

, Holtgrave

, Bloom

. Cost-effectiveness of post-exposure prophylaxis following sexual exposure to HIV. AIDS, 1998; 12:1067–1078.