Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Sep 5.
Published in final edited form as: J Health Psychol. 2009 Sep;14(6):761–770. doi: 10.1177/1359105309338976

How Do Social Norms Impact HIV Sexual Risk Behavior in HIV Positive Men Who Have Sex with Men: Multiple Mediator Effects

Michael H Miner 1, John L Peterson 2, Seth L Welles 3, Scott M Jacoby 4, BR Simon Rosser 5
PMCID: PMC3433849  NIHMSID: NIHMS401060  PMID: 19687113

Abstract

This study examines mediation of the association between social norms and unsafe sexual behavior. Self-report data were collected from 675 HIV-infected men enrolled in a study exploring interventions for HIV risk behavior. Unsafe sex included any unprotected anal sex with HIV-negative or HIV status unknown partners in the last three months. Norms for condom use indirectly influenced unsafe sex through condom self-efficacy and/or safer sex intentions. Additionally, sexual behavior discontrol influenced unsafe sex regardless of other individual or social factors. Our results suggest that interventions consider the combined effects of condom self-efficacy, safer sex intentions, and sexual behavior control.

Keywords: Theory of Planned Behavior, unsafe sex, social norms, self-efficacy, intentions

INTRODUCTION

The HIV/AIDS epidemic in the United States continues to exert a devastating toll on men who have sex with men (MSM), reflected in their over-representation among cases of HIV morbidity and mortality (CDC, 2005). In the last two decades, there has been an increased emphasis in understanding the psychological factors that influence the practice of safer or unsafe sex. Surprisingly, little research has examined the predictors of unprotected anal sex among HIV-positive men who have sex with men. In results from the Seropositive Urban Men's Study and the Seropositive Urban Men's Intervention Trial, self-efficacy, personal responsibility, substance use, mental health and contextual influences, such as engaging in sex in public venues, correlated with unprotected anal sex (Wolitski, Parsons & Gomez, 2004). Data from HIV-positive MSM indicate, however, that most contextual variables are mediated by condom use self-efficacy (O'Leary et al., 2005) and the consistent findings support the exploration of variables such as perceived lack of behavioral control, social norms, self-efficacy, and intentions as predictors of unprotected sexual behavior in HIV-positive men who have sex with men (Chesney et al., 2003; Parsons, Halkitis, Wolitski & Gomez, 2003). Thus, this line of research indicates the application of the theory of reasoned action (TRA, Fishbein & Ajzen, 1975) and the theory of planned behavior (TPB, Ajzen, 1985; Ajzen, 1988)), which are specific applications of social cognitive theory (Bandura, 1982). Theory of Reasoned Action posits that use of condoms, or doing any other behavior, is related to one's intentions to do so. These intentions are influenced by attitudes toward the behavior and perceived social norms that support such behavior. Meta-analyses have found substantial support for TRA components as predictors of condom use among a range of samples, including men who have sex with men (MSM, Albarracín, Johnson, Fishbein, & Muellerleile, 2001; Albarracín, Kumkale, & Johnson, 2004). Also, Sheeran and Taylor (1999), found that when the variables posited by TRA were controlled, perception of behavioral control was independently and substantially associated with intentions to use condoms. This finding supports TPB as a model for understanding safer sex, in that not only are attitudes and norms important for the formation of behavioral intentions as posited by TRA, but it is also necessary for individuals to believe that they have control over the behavior. That is, there are no barriers to condom use (Ajzen, 1985, 1991). Sheeran and Taylor's (1999) findings have been substantially replicated in other meta-analyses that have explored TRA and TPB (Albarracín, et al. 2001; Albarracín, et al., 2004).

As in other populations (Basen-Engquish, 1992; Redding & Rossi, 1999; Svenson, Ostergren, Merlo & Rastam, 2002; Wulfert & Wan, 1995), studies exploring predictors of unsafe sex with MSM have identified diverse psychosocial factors associated with HIV risk. Specifically, Wulfert, Wan & Backus (1996) found that self-efficacy mediated the effects of outcome expectancies and peer-group norms on engaging in unprotected anal intercourse and Semple, Patterson and Grant (2000) found that, at least for casual partners, self-efficacy for condom use and for negotiation were associated with decreased rates of unprotected anal intercourse. In a study of young gay-identified men in Amsterdam, Janssen, De Wit, Stroebe and Van Griensven (2000) found that the best predictor of consistent condom use with steady partners was intention, while the best predictor for condom use with casual partners was perceived behavioral control.

Also, studies have shown that the adoption of safer sex practices is associated with perceptions of normative support for HIV risk-reduction activities among general community samples of gay men. Gay men who perceive strong social norms for condom use engage in less risky sexual behavior than those who perceive weak social norms for condom use (e.g., Hart & Peterson, 2004; Kelly et al., 1997).

Hence, while there is some support for TPB as a model for unsafe sex within HIV-positive MSM, this support is not consistent. Additionally, there is evidence that the relationships between the variables posited by both TRA and TPB are moderated by social status and social power. That is, those populations that lack social power show a stronger correlation between perceived behavioral control and intentions, and between perceived behavioral control and actual condom use. Moreover, samples with greater representation of ethnic minorities, along with younger male samples, showed stronger associations between social norms and intentions (Albarracín, Gillette, Ho, Earl, Glasman, & Durantini, 2005). In a small sample (n=96), Godin, Savard, Kok, Fortin, and Boyer (1996) examined the associations between TPB variables and condom use in a sample of seropositive MSM. These authors found partial support for the effects of TPB variables, specifically that perceived behavioral control predicted intentions both to use condoms with anal intercourse and to engage in sexual behaviors other than anal intercourse. However, the results of this study indicated that perceived behavioral control had the most influence over behavior, with intentions being unrelated to subsequent behavior. The authors further reported that perceived behavioral control over condom use was most influenced by contextual barriers, such as not having a condom at hand at the time of intercourse and falling in love. The results reported in the extant literature leave it unclear whether variables derived from TPB will similarly predict unsafe sex in HIV-positive MSM as shown in other populations. The risks of engaging in unprotected anal intercourse are different for HIV-negative and HIV-positive MSM, thus it seems reasonable to expect that variables such as norms, self-efficacy, and intentions will interact differentially to affect sexual decisions.

Given prior results, we examined whether social norms for condom use are related to unprotected anal sex with serodiscordant partners among an ethically diverse sample of HIV-positive MSM. Specifically, we examined whether the association between social norms and unsafe sex is mediated by condom self-efficacy and intentions to practice safer sex. Additionally, our investigation builds on the extant literature to examine possible effects of social support, which may influence social norms to practice safer sex (Albarracín, et al., 2005), and to introduce two variables, sexual comfort and sexual behavior discontrol, which have not been investigated within the context of TPB. The latter two variables are seen as salient omissions in the current literature, which does not consider the influence of sexuality variables as the predictors of unsafe sex (Robinson, Bockting, Rosser, Miner, & Coleman, 2002). Sexual comfort, which is defined in terms of comfort with one's sexuality, their sexual performance and their body image, has been shown to effect sexual decision-making. Specifically, sexual comfort has been implicated in the willingness of individuals to assert themselves and express their needs in sexual situations (Abraham & Sheeran, 1994; Robinson et al., 2002; Stoner et al., 2008). Thus, sexual comfort is likely to influence whether social norms, self-efficacy or behavioral intentions are expressed within the sexual context that leads to safer or unsafe sex.

The current literature indicates that the associations found for behavioral control are similar whether perceived behavioral control or self-efficacy is measured (Albarracín, et al., 2004). We have chosen to measure both condom use self efficacy, the belief in the ability to use condoms across multiple situations, and perceived sexual behavioral discontrol, the perceived inability to control sexual behaviors across multiple situations, in order to better understand the concept of behavioral control as related to unsafe sexual behavior. Behavior discontrol has shown not only an association with unprotected anal intercourse (Benotsch, Kalichman & Kelly, 1999; Kalichman, Adair, Rompa, Multhauf, Johnson, & Kelly, 1994; Kalichman & Rompa, 1995), but this association seems to be related to those with behavioral discontrol showing increased levels of anal sexual behavior, regardless of whether or not a condom is used (Miner, Coleman, Center, Ross & Rosser, 2007). Thus, behavioral discontrol is a broader concept than condom use self-efficacy and it is likely that both beliefs influence whether or not an individual engages in unsafe sex. The extant literature appears to have confused the concepts by failing to measure perceived inability to control sexual behavior per se.

Thus, this study will test an augmented TPB model which assumes that social norms are not only mediated by self-efficacy and intentions, but are also influenced by social support. That is, social norms will only effect behavior to the extent that an individual receives support from their social system. The more isolated an individual is from a social system, the less likely social norms are to influence behavior (Okun, Ruehlman, Karoly, Lutz, Fairholme, & Schaub, 2003). Additionally, we expect that sexual comfort and sexual behavior discontrol will have direct effects on unsafe sexual behavior and indirect effects, through condom use, on self efficacy.

METHODS

Participants

The sample includes 675 HIV-positive men recruited in six major metro areas across the United States to participate in a weekend seminar-based intervention to reduce HIV sexual risk behavior: Seattle (n=114), Washington, D.C. (n=71), Boston, (n=64), New York (n=177), Los Angeles (n=146), and Houston (n=103). Participants were eligible if they reported being HIV positive, were at least 18 years of age, spoke English, had sex with at least one man in their lifetime, and had anal sex (insertive or receptive) without a condom at least once in the last 12 months. Participants were recruited by Community Based Organizations in each city who offered services for HIV-positive MSM. Recruitment included posting on agency websites, flyers in venues frequented by MSM, advertisements in newspapers that target MSM, coordination with health care providers, and, when appropriate, recruitment of agency clients. This study was reviewed and the procedures approved by the Committee for the Protection of Human Subjects at the University of Minnesota.

Procedures

The data presented are baseline data collected between January 29, 2005 and April 22, 2006 prior to participants’ assignment to intervention condition. Data were collected through a self-report inventory administered in group settings, where aid was available from research staff for participants with problems reading or understanding questions.

Predictor Measures

Sexual Comfort

This is a 6-item Likert scale that measures comfort with sexuality and one's body (Marin, Gomez, Tschann & Gregorich, 1997). For example: How would you feel teaching your partner what feels pleasurable to your during sex? Items are rated from 1, very uncomfortable to 4, very comfortable. High scores indicate more sexual comfort. The reliability in this sample was α = .83.

Social Norms

This is a 3-item Likert scale that measures the participants’ perceptions of their friends’ attitudes toward using condoms (Marin, Gomez, Tschann & Gregorich, 1997). For example: How many of your closest men friends do you think use condoms with every partner? Items are rated from 1, almost none of my men friends to 5, almost all of my men friends. High scores indicate attitudes that support using condoms in multiple contexts. The reliability in this sample was α = .77.

Sexual Behavior Discontrol

This construct was measured by the 13-item Control subscale from the Compulsive Sexual Behavior Inventory and consists of items rated from 1, never, to 5, very frequently, that measure a sense of lack of control over one's sexual behavior (Coleman, Miner, Ohlerking & Raymond, 2001). For example: In the past 3 months, how often have you had trouble controlling your sexual urges? High scores indicate more lack of control over sexual behavior. The reliability in this sample was α = .92.

Social Support

This is a 6-item Likert scale that measures the amount of support that participants receive from their social environment (Zimet, Dahlem, Zimet & Farley, 1988). For example: I can count on my friends when things go wrong. Items are rated from 1, disagree strongly to 5, agree strongly and ask about friends and about a “special person.” High scores indicate more social supports. The reliability in this sample was α = .91.

Condom Self Efficacy

This is a 14-item Likert scale that measures self-efficacy with respect to using condoms in multiple situations and settings (Marin, Gomez, Tschann & Gregorich, 1997). For example: Can you continue to insist on using a condom with a person who threatens to leave if you use it? Items are rated as 1, definitely not to 4, definitely yes. High scores indicate more self-efficacy. The reliability in this sample is α = .95.

Intent to Practice Safer Sex

This is a one-item measure that was developed for this study and asks “The next time you have sex with someone, how likely are you to bring up the need to practice safer sex?” It is rated on a 1 to 7 scale where 1 is not at all likely to practice safer sex and 7 is very likely to practice safer sex.

Outcome Measure

Unsafe sex

The outcome or dependent variable is defined as any unprotected anal sex (insertive or receptive) with a partner of discordant or unknown HIV infection status during the last three months. We chose to define our unsafe measure in this manner so that we could control for sero-sorting and other potentially effective strategies for limiting transmission of HIV (Miner, Robinson, Hoffman, Albright & Bockting, 2002). The frequencies of unsafe intercourse were calculated by first summing the number of acts of unprotected anal sex with primary partners who were of unknown or negative HIV infection status with the number of acts of unprotected anal sex with secondary partners who were of unknown or negative HIV status. This variable was then dichotomized as having no incidents of unprotected anal sex in the last three months (0) or at least one incident of unprotected anal sex in the last three months (1) because of the severe skew in the frequency distribution.

Data Analysis

Data analysis proceeded in three steps. First, we compared those engaging in unsafe sex with those not engaging in such behavior on each of the predictor variables using Mann-Whitney U. Then an intercorrelation matrix was computed of the predictor variables, and finally, a hierarchical logistic regression was used to explore the associations between predictor variables and unsafe sex. The order in which variables were entered into the logistic regression was determined by the associations described in TPB. Mediation was determined using Baron and Kenny's (1986) criteria. That is, the presence of a direct effect on unsafe sex as determined by significant first order (unadjusted) odds ratios, coupled with a significant association between predictor and mediator variables, and the decrease from a significant unadjusted odds ratio (OR) to an non-significant OR with the addition of the mediator variable.

RESULTS

The average age of participants was 42 years (range 18 years to 69 years). Forty-five percent of the sample was African American, 25% White, 23% Latino/Hispanic, with another 7% reporting other race/ethnicity groups (including Asian/Pacific Islanders, mixed race) Participants tended towards lower levels of education: 40% of the sample had a high school diploma or less education, 38% had attended some college or technical training, and 22% had at least a Bachelor's degree. Participants’ average annual income was $16,357 , with about 50% of the participants earning $10,000 or less annually. Eighty-one percent of the sample identified as gay/same gender loving. Participants’ median CD4 count at last check-up was 429/mm3 and those with a detectable plasma HIV RNA level had a median of 2,900 copies/mm3. Forty-two percent of the sample reported an undetectable viral load. Seventy-five percent of participants reported current use of anti-retroviral therapy (ART).

We first evaluated the distributions of each variable. Descriptive statistics are presented in Table 1. The distributions for intent to practice safer sex and sexual comfort indicated that a binary split was most appropriate, while the distributions for condom self efficacy, safer sex norms, social support, and sexual behavior discontrol indicated that quartile splits were most appropriate (see Table 1). Computation of the dichotomous unsafe sex variable resulted in 297 participants (44%) defined as having engaged in unsafe sex over the last three months, while 378 participants (56%) had never engaged in unsafe sex over the last three months. Participants engaging in unsafe sex were significantly different from those not engaging in unsafe sex on all predictor variable except sexual comfort.

Table 1.

Descriptive Statistics: Predictor Variables

Unsafe Sex last 3 months (N=297) No Unsafe Sex last 3 months (N=378) p-value
Mean (sd) Median (IQR) Mean (sd) Median (IQR)
Intent to Practice Safer Sex 13.8 (4.7) 14 (11, 18) 16.5 (4.2) 17 (17, 21) <.001
Condom Self-efficacy 55.0 (14.2) 57 (45, 67) 63.1 (13.3) 67 (57, 74) <.001
Safe Sex Norms 8.6 (2.8) 9 (7, 10) 9.5 (3.0) 9 (8, 12) <.001
Social Support 21.4 (6.7) 23 (16, 27) 22.9 (6.7) 24 (19, 29) .001
Sexual Comfort 20.6 (3.7) 22 (19, 24) 20.8 (3.6) 22 (19, 24) .175
Sexual Behavior 30.4 (11.0) 28 (22, 38) 25.4 (10.4) 23 (17, 31) <.001
Discontrol
Open in a new tab

IQR = Interquartile range

Secondly, we explored the correlations between potential predictor variables. No significant associations were found between any socio-demographic variables and unsafe sex, including age, race/ethnicity, education, employment status, or income. As can be seen in Table 2, the correlations, are generally of low magnitude, although statistically significant due to our large sample size. The magnitude of the relationship between condom self efficacy and intent to practice safer sex is large (r2=.35), while the magnitudes of the relationships between condom self-efficacy and safe sex norms (r2=.13) and between safe sex norms and intent to practice safer sex (r2=.10) are medium (Cohen, 1992, p.157).

Table 2.

Correlation Matrix: Predictor Variables

Condom Self Efficacy Safe Sex Norms Social Support Sexual Comfort Sexual Behavior Discontrol Intent to Practice Safer Sex
Condom Self -- .36** .20** .21** -.21** .59**
Efficacy
Safe Sex -- .14** .15** -.05 .31**
Norms
Social -- .17** -.16** .22**
Support
Sexual -- - 21** .14**
Comfort
Sexual -- -.17**
Behavior
Discontrol
Intent to --
Practice
Safer Sex
Open in a new tab
**

Significant at p=<.0001

Finally, hierarchical logistic regression was used to evaluate the associations of psychosocial factors with the dependent variable—the binary occurrence of unsafe sex. Table 3 presents the step by step results of the analysis. The first three variables, sexual comfort, social support, and behavioral discontrol were entered to control for possible confounding effects. The addition of social norms to this block added significantly to the prediction of unsafe sex (χ2=5.6, df=1, p=.018). Also, as can be seen from table 3, the association between social norms and unsafe sex was significant and negative (B=-0.18, p=.019). However, in the final model (Step 4), which adds both self-efficacy and intentions to practice safer sex, social norms does not make a significant independent contribution to predicting unsafe sex, and the odds ratio associated with social norms is essentially 1.0. The change between the initial OR and the OR at Step 4 for safer sex norms when both self-effiacy and safer sex intentions were entered into the equation was modest (.14). The mediation effect was then broken down by first investigating the change in OR with the addition of self-efficacy to the model (Step 3). The decrease in OR with the addition of self-efficacy was .12. When intentions was entered after safer sex norms, rather than self efficacy, the decrease in OR was .10. Thus, the mediation was substantially the same for both variables, and slightly lower than the two variables combined. Both self-efficacy and intentions were significant predictors of unsafe sex in the final model (see table 3). The effect of behavioral discontrol on unsafe sex was unaffected by the addition of other variables to the model.

Table 3.

Step by Step Results of Hierarchical Logistic Regression Predicting Unsafe Sexual Behavior

Factor Beta p-value Odds Ratio 95% CI
Step 1 <.001
Sexual Comfort 0.204 .234 1.23 0.87, 1.72
Social Support -0.165 .031 1.18a 1.01, 1.37
Behavioral 0.435 <.0001 1.54 1.32, 1.81
Discontrol
Step 2 .018
Sexual Comfort 0.233 .183 1.26 0.90, 1.78
Social Support -0.150 .053 1.16a 1.00, 1.35
Behavioral 0.432 <.001 1.54 1.32, 1.80
Discontrol
Safe Sex Norms -0.180 .019 1.20a 1.03, 1.39
Step 3 <.001
Sexual Comfort 0.348 .053 1.42 0.99, 2.02
Social Support -0.990 .211 1.10a 0.94, 1.06
Behavioral 0.366 <.001 1.44 1.23, 1.70
Discontrol
Safe Sex Norms -0.100 .208 1.10a 0.95, 1.29
Self-efficacy -0.382 <.0001 1.47a 1.23, 1.74
Step 4 <.001
Sexual Comfort 0.367 .045 1.44 1.01, 2.07
Social Support -0.078 .334 1.08a 0.92, 1.26
Behavioral 0.357 <.001 1.43 1.21, 1.68
Discontrol
Safe Sex Norms -0.061 .454 1.06a 0.91, 1.24
Self-efficacy -0.235 .015 1.26 1.05, 1.53
Intentions -0.744 <.0001 2.10b 1.43, 3.10
Open in a new tab
a

Odds Ratios reflect the odds of unsafe sex associated with each lower quartile score.

b

Odds Ratio reflects the odds of unsafe sex with behavioral intentions below the scale median.

DISCUSSION

There is considerable evidence that the Theory of Reasoned Action (TRA) and the Theory of Planned Behavior (TPB) explain unsafe sexual behavior in several different populations (Albarracín et al. 2001; 2004). This study extends prior research regarding the factors that make up TPB by considering the effect of behavioral control and the role of sexual comfort, an important aspect of the sexual health model (Robinson, et al., 2002). The prior research is also extended by contextualizing the measure of unsafe sex to account for sero-sorting, as well as condom use. We not only examined whether beliefs in the ability to use condoms (self-efficacy), and intentions to use condoms, mediate the association between condom social norms and unsafe sex with HIV-positive MSM and their serodiscordant sexual partners, but also the involvement of sexual comfort, social support, and perceived control over sexual behavior in influencing social norms, self-efficacy, intentions, and unsafe sexual behavior. Our results are consistent with TPB in that social norms mediated the association between social support and unsafe sexual behavior such that those who perceived high social support, tended to have stronger social norms for condom use and, in turn were less l ikely to engage in unsafe sex. Similarly, those men who perceived stronger social norms for condom use were more likely to hold stronger beliefs in their ability to use condoms. And those men who held stronger beliefs in their ability to use condoms engaged in less unsafe sex. Moreover, the effect of social norms on unsafe sex was diminished once the effect of self-efficacy was taken into account. The change in association, while significant, was of modest magnitude.

The association between men's perceptions of safer sex norms and occurrence of unsafe sex was also influenced by their intentions to practice safer sex. We found that men who perceived stronger norms for condom use had stronger intentions to practice safer sex, and those with stronger intentions engaged in less unsafe sex. Again, the association between intentions and safer sex norms was modest and the change in the association between safer sex norms and unsafe sex was modest when accounting for intentions.

In addition, we found that men who experienced an inability to control their sexual urges and impulses engaged in more unsafe sex than those who had behavioral control. This finding is consistent with others (Benotsch, et al., 1999; Kalichman, et al., 1994; Kalichman & Rompa, 1995). This lack of behavioral control was independent of social norms and had a small association with intentions to use condoms. Perceived sexual discontrol was modestly and negatively related to condom use self efficacy. Thus, while there is some overlap between the perception of ability to control sexual behavior and the ability to use condoms when having sex, these concepts are not the same and both appear to have influences on unsafe sexual behavior that are independent of each other. Thus, behavioral control, at least within the context of unsafe sexual behavior of HIV-positive MSM, appears to have two components, perceptions of one's ability to control their sexual behavior and perceptions of one's ability to use condoms. It appears that both act directly on the likelihood of an individual practicing safer sex, while perception of sexual control has a small effect on perceived ability to use a condom.

Taken together, these findings reveal that unsafe sexual behavior, defined as unprotected anal intercourse with serodiscordant partners, is influenced by a constellation of cognitive factors consistent with social cognitive theory and the Theory of Planned Behavior. That is, social support appears to influence the impact of social norms and also influences feelings of an inability to control sexual behavior. Norms and sexual behavioral control appear to have independent effects on condom use self efficacy and intentions to engage in unsafe sex. Consistent with the Theory of Planned Behavior, the independent association between social norms and unsafe sex is mediated by both condom use self efficacy and safer sex intentions. However, while those with a lack of behavioral control also have low condom use self-efficacy, lack of sexual behavioral control has both this indirect effect on unsafe sex and a direct effect.

These findings suggest that HIV-positive men in the United States who engage in unsafe sex with sexual partners of a different HIV status than their own warrant greater interventions that strengthen their behavioral control, self-efficacy and intention to use condoms. Weak beliefs in the ability to control sexual behavior or use condoms, and weak intentions to actually engage in safer sex practices, may have contributed to the unsafe sex between HIV positive MSM and their seordiscordant partners. Notably, these effects of behavioral control, self-efficacy and intentions may have impact even among men whose communities support safe sex norms. Hence, social interventions that improve social support and sustain safer sex norms may be insufficient to reduce HIV risk behaviors without equal, if not greater, effort to change men's condom beliefs and intentions. Additionally, a lack sexual behavioral control is associated with sexual risk behavior even after controling for safer sex norms. Community interventions may be more effective if they include both social diffusion approaches to change norms and increase perceived social support, as well as, small group approaches to reinforce ability and intentions.

Despite the promising evidence of our findings, a few caveats should be noted in our study. Our participants were not a random sample of HIV-positive men, but a convenience sample of HIV-positive MSM who agreed to participate in a weekend long HIV prevention trial. The data were collected via self-report and the sexual behavior data were retrospective over a three-month period. However, these limitations are typical of studies of sexual behavior, especially in baseline assessments of community interventions. Both random designs and longitudinal analyses will be employed to examine possible effects of the intervention in our larger study. Moreover, our baseline findings provide evidence that individual factors -- behavioral control, condom self efficacy and intentions to practice safer sex -- may affect how contextual factors, such as social norms, influence unsafe sex behavior between HIV-positive men and their serodiscordant sexual partners. These findings should be replicated in community-based studies, which draw random, representative samples of HIV-positive MSM and follow them longitudinally.

Acknowledgments

This research was funded by the NIMH Center for Mental Health Research on AIDS grant number 1 R01 MH064412-01A2 to B.R. Simon Rosser (PI). This paper was made possible by a partnership between the University of Minnesota and the following Community-Based Organizations (CBOs): Gay City Health Project (Seattle, WA), Whitman-Walker Clinic (Washington, D.C.), Fenway Community Health (Boston, MA), Gay Men's Health Crisis (New York, NY), AIDS Project Los Angeles and The Black AIDS Institute (Los Angeles, CA), and Legacy Community Health Services (Houston, TX). The authors wish to thank Antony Stately, Ph.D. and Anne Marie Webber-Main, Ph.D. for reviewing and editing this manuscript.

Contributor Information

Michael H. Miner, Department of Family Medicine and Community Health, University of Minnesota

John L. Peterson, Department of Psychology, Georgia State University

Seth L. Welles, Department of Epidemiology and Biostatistics, Drexel University

Scott M. Jacoby, Department of Family Medicine and Community Health, University of Minnesota

B.R. Simon Rosser, Department of Epidemiology and Community Health, University of Minnesota..

REFERENCES

  1. Abraham CS, Sheeran P. Modeling and modifying your heterosexuals’ HIV-prevention behaviour; a review of theories findings and education implications. Patient Education and Counseling. 1994;23:173–186. doi: 10.1016/0738-3991(94)90033-7. [DOI] [PubMed] [Google Scholar]
  2. Ajzen I. From intentions to action: A theory of planned behavior. In: Kuhl J, Beckman J, editors. Action Control: From Cognition to Behavior. Springer Verlag; Berlin, Germany: 1985. pp. 11–39. [Google Scholar]
  3. Ajzen I. Attitudes, Personality, and Behavior. Dorsey Press; Homewood, IL: 1988. [Google Scholar]
  4. Albarracín D, Gillette JC, Ho M-H, Earl AN, Glasman LR, Durantini MR. A test of major assumptions about behavior change: A comprehensive look at the effects of passive and active HIV-prevention interventions since the beginning of the epidemic. Psychological Bulletin. 2005;131:856–897. doi: 10.1037/0033-2909.131.6.856. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Albarracín D, Johnson BT, Fishbein M, Muellerleile PA. Theories of reasoned action and planned behavior as models of condom use: A meta-analysis. Psychological Bulletin. 2001;127:142–161. doi: 10.1037/0033-2909.127.1.142. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Albarracín D, Kumkale GT, Johnson BT. Influences of social power and normative support on condom use decisions: a research synthesis. AIDS Care. 2004;16:700–723. doi: 10.1080/09540120412331269558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bandura A. Self-efficacy mechanism in human agency. American Journal of Psychology. 1982;37:122–147. [Google Scholar]
  8. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology. 1986;51:1173–1182. doi: 10.1037//0022-3514.51.6.1173. [DOI] [PubMed] [Google Scholar]
  9. Basen-Engquist K. Psychosocial predictors of “safer sex” behaviors in young adults. AIDS Education and Prevention. 1992;4:120–134. [PubMed] [Google Scholar]
  10. Benotsch EG, Kalichman SC, Kelly JA. Sexual compulsivity and substance use in HIV seropositive men who have sex with men: Prevalence and predictors of high-risk sexual behaviors. Addictive Behaviors. 1999;24:857–868. doi: 10.1016/s0306-4603(99)00056-8. [DOI] [PubMed] [Google Scholar]
  11. Centers for Disease Control and Prevention HIV/AIDS Surveillance Report, 2004. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, Atlanta, 16. 2005 Retrieved from http://www.cdc.gov/hiv/stats/hasrlink.htm.
  12. Chesney MA, Koblin BA, Barresi PJ, Husnick MJ, Celum CL, Colfax G, et al. An individually tailored intervention for HIV prevention: Baseline data from the EXPLORE study. American Journal of Public Health. 2003;93:933–938. doi: 10.2105/ajph.93.6.933. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Coleman E, Miner MH, Ohlerking F, Raymond N. Compulsive sexual behavior inventory: A preliminary study of reliability and validity. Journal of Sex and Marital Therapy. 2001;27:325–332. doi: 10.1080/009262301317081070. [DOI] [PubMed] [Google Scholar]
  14. Fishbein M, Ajzen I. Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research. Addison-Wesley; Reading, MA: 1975. [Google Scholar]
  15. Godin G, Savard J, Kok G, Fortin C, Boyer R. HIV seropositive gay men: Understanding adoption of safe sexual practices. AIDS Education and Prevention. 1996;8:529–545. [PubMed] [Google Scholar]
  16. Halkitis PN, Wilton L, Parsons JT, Hoff C. Correlates of sexual risk taking among HIV-positive men in seroconcordant primary partner relationships. Psychology, Health and Medicine. 2003;9:99–113. [Google Scholar]
  17. Hart T, Peterson JL, CITY Study Team Predictors of risky sexual behavior among young African American men who have sex with men. American Journal of Public Health. 2004;94:1122–1124. doi: 10.2105/ajph.94.7.1122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Janssen M, De Wit J, Stroebe W, Van Griensven F. Educational status and risk of HIV in young gay men. Journal of Health Psychology. 2000;5:487–499. doi: 10.1177/135910530000500407. [DOI] [PubMed] [Google Scholar]
  19. Kalichman SC, Adair V, Rompa D, Multhauf K, Johnson J, Kelly J. Sexual sensation seeking: Scale development and predicting AIDS-risk behavior among homosexually active men. Journal of Personality Assessment. 1994;62:385–397. doi: 10.1207/s15327752jpa6203_1. [DOI] [PubMed] [Google Scholar]
  20. Kalichman SC, Rompa D. Sexual sensation-seeking and sexual compulsivity scales: Reliability, validity, and predicting HIV risk behavior. Journal of Personality Assessment. 1995;65:586–601. doi: 10.1207/s15327752jpa6503_16. [DOI] [PubMed] [Google Scholar]
  21. Kelly JA, Murphy DA, Sikkema KJ, McAuliffe TL, Roffman RA, Solomon LJ, et al. Randomized, controlled, community-level HIV prevention intervention for sexual risk behaviour among homosexual men in U.S. cities. Lancet. 1997;350:1500–1505. doi: 10.1016/s0140-6736(97)07439-4. [DOI] [PubMed] [Google Scholar]
  22. Marin BV, Gomez CA, Tschann JM, Gregorich SE. Condom use in unmarried Latino men: A test of cultural constructs. Health Psychology. 1997;16:458–467. doi: 10.1037//0278-6133.16.5.458. [DOI] [PubMed] [Google Scholar]
  23. Marks G, Burris S, Peterman TA. Reducing sexual transmission of HIV from those who know they are infected: The need for personal and collective responsibility. AIDS. 1999;13:297–306. doi: 10.1097/00002030-199902250-00001. [DOI] [PubMed] [Google Scholar]
  24. Miner MH, Coleman E, Center BA, Ross M, Rosser BRS. The compulsive sexual behavior inventory: Psychometric properties. Archives of Sexual Behavior. 2007;36:579–587. doi: 10.1007/s10508-006-9127-2. [DOI] [PubMed] [Google Scholar]
  25. Miner MH, Robinson BE, Hoffman L, Albright CL, Bockting WO. Improving safer sex measures through the inclusion of relationship and partner characteristics. AIDS Care. 2002;14:827–837. doi: 10.1080/0954012021000031895. [DOI] [PubMed] [Google Scholar]
  26. Okun MA, Ruehlman L, Karoly P, Lutz R, Fairholme C, Schaub R. Social support and social norms: Do both contribute to predicting leisure-time exercise. American Journal of Health Behavior. 2003;27:493–507. doi: 10.5993/ajhb.27.5.2. [DOI] [PubMed] [Google Scholar]
  27. O'Leary A, Maibach E, Ambrose TK, Jemmott JB, III, Celentano DD. Social cognitive predictors of condom use behavior change among STD patients. AIDS and Behavior. 2000;4:309–316. [Google Scholar]
  28. O'Leary A, Wolitski RJ, Remien RH, Woods WJ, Parsons JT, Moss S, et al. Psychosocial correlates of transmission risk behavior among HIV-seropositive gay and bisexual men. AIDS. 2005;19:S67–S75. doi: 10.1097/01.aids.0000167353.02289.b9. [DOI] [PubMed] [Google Scholar]
  29. Parsons JT, Harlkitis PN, Wolitski RJ, Gómez CA, SUMS Study Team Correlates of sexual risk behaviors among HIV-positive men who have sex with men. AIDS Education and Prevention. 2003;15:383–400. doi: 10.1521/aeap.15.6.383.24043. [DOI] [PubMed] [Google Scholar]
  30. Redding CA, Rossi JS. Testing a model of situational self-efficacy for safer sex among college students: Stage of change and gender-based differences. Psychology and Health. 1999;14:467–486. [Google Scholar]
  31. Robinson BE, Bockting WO, Rosser BRS, Miner MH, Coleman E. The sexual health model: Application of a sexological approach to HIV prevention. Health Education Research. 2002;17:43–57. doi: 10.1093/her/17.1.43. [DOI] [PubMed] [Google Scholar]
  32. Semple SJ, Patterson TL, Grant I. Partner type and sexual risk behavior among HIV positive gay and bisexual men: Social cognitive correlates. AIDS Education and Prevention. 2000;12:340–356. [PubMed] [Google Scholar]
  33. Sheeran P, Taylor S. Predicting intentions to use condoms: A meta-analysis and comparison of the theories of reasoned action and planned behavior. Journal of Applied Social Psychology. 1999;29:1624–1675. [Google Scholar]
  34. Stoner SA, Norris J, George WH, Morrison DM, Zawacki T, Davis KC, Hessler DM. Women's condom use assertiveness and sexual risk-taking: Effects of alcohol intoxication and adult victimization. Addictive Behaviors. 2008;33:1167–1176. doi: 10.1016/j.addbeh.2008.04.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Svenson GR, Östergren PO, Merlo J, Råstam L. Action control and situational risks in the prevention of HIV and STIs: Individual, dyadic, and social influences on consistent condom use in a university population. AIDS Education and Prevention. 2002;14:515–531. doi: 10.1521/aeap.14.8.515.24111. [DOI] [PubMed] [Google Scholar]
  36. White KM, Terry DJ, Hogg MA. Safer sex behavior: The role of attitudes, norms, and control factors. Journal of Applied Social Psychology. 1994;24:2164–2192. [Google Scholar]
  37. Wolitski RJ, Parsons JT, Gomez CA, SUMS Study Team. SUMIT Study Team Prevention with HIV-seropositive men who have sex with men: lessons from the Seropositive Urban Men's Study (SUMS) and the Seropositive Urban Men's Intervention Trial (SUMIT). Journal of Acquired Immune Deficiency Syndromes. 2004;37:S101–S109. doi: 10.1097/01.qai.0000140608.36393.37. [DOI] [PubMed] [Google Scholar]
  38. Wulfert E, Wan CK. Condom use: A self-efficacy model. Journal of Health Psychology. 1995;12:346–353. doi: 10.1037//0278-6133.12.5.346. [DOI] [PubMed] [Google Scholar]
  39. Wulfert E, Wan CK, Backus CA. Gay men's safer sex behavior: An integration of three models. Journal of Behavioral Medicine. 1996;19:345–366. doi: 10.1007/BF01904761. [DOI] [PubMed] [Google Scholar]
  40. Zimet GD, Dahlem NW, Zimet SG, Farley GK. The multidimensional scale of perceived social support. Journal of Personality Assessment. 1988;52:30–41. doi: 10.1080/00223891.1990.9674095. [DOI] [PubMed] [Google Scholar]

RESOURCES