In countries such as the United States, where incidence of the human immunodeficiency virus (HIV) is relatively low and the availability of highly active antiretroviral treatment can prolong life, focusing prevention efforts on HIV-positive persons to limit HIV transmission may be a more efficient strategy than directing prevention efforts toward those at risk. Community acceptance of such an approach is growing, and recent national guidelines articulate ways to work with those who are already HIV positive.1-3 This new prevention paradigm, along with an HIV epidemic among injection drug users (IDUs) that may be increasingly fueled by unsafe sex,4,5 means that prevention with HIV-positive individuals should, correspondingly, include a focus on the sexual behaviors of IDUs.
By 2001, in the United States, HIV transmission via IDU or heterosexual contact accounted for half of all HIV infections among those people living with HIV, and IDUs were estimated to be the second largest risk group among those living with AIDS.6 Although the likelihood of acquiring HIV during heterosexual contact is much higher for women than men, HIV can be sexually transmitted to men.7,8 In addition, HIV-positive drug-using women are less likely than non-drug-using women to be receiving highly active antiretroviral treatment9 and therefore are more likely to have advanced disease and higher viral loads, which in turn increases the risk of HIV transmission.10
Studies of the sexual behavior of HIV-positive women with serodiscordant male partners suggest that these encounters present opportunities for HIV transmission as condom use is insufficient and anal sex is not rare.7,11,12 The few studies involving HIV-positive drug-using women suggest that whereas condom use is more frequent among HIV-positive than HIV-negative women, it remains insufficient13-16 and may be even less likely among IDU women.17-19 Other studies suggest that IDU women may feel less responsible for protecting their partner after disclosing their HIV status20 and highlight challenges faced by women in convincing their male'partners to use condoms.21,22
Relationship dynamics may also influence HIV risk. Among HIV-positive women, unprotected sex is more likely with male partners who are also HIV positive,11,13,16,23 varies by whether a partner is considered a main or casual partner,11 and is more likely when the male partner desires children.24,25 The impact that other relationship dynamics, such as disclosing HIV status, have on sexual risk is not clear.20,26 Thus, it is important to understand both personal and relationship aspects as these may shape sexual risk. In the present analysis, we characterized the sexual behavior of HIV-positive IDU women during encounters with HIV-negative or unknown serostatus male partners. We identified how 3 domains of women s lives-individual, relationship, and social-were associated with inconsistent condom use.
Baseline data were analyzed from the Intervention for Sero-Positive IDUs-Research and Evaluation (INSPIRE) study, a randomized trial of a 10-session behavioral intervention to reduce risky sexual and injection behaviors and improve health care use and adherence among HIV-positive IDUs. The study, described in detail elsewhere,27 enrolled 1161 men and women in Baltimore, Miami, New York, and San Francisco. Individuals were eligible for the study if they were at least 18 years old, able to communicate in English, confirmed as HIV positive by means of an oral specimen (OraSUre, OraSUre Technologies, Inc., Bethlehem, PA), and reported both IDU in the past 12 months and sex with an opposite sex partner in the past 3 months. Of the 426 women enrolled, 86.8% (370/426) were sexually active at baseline; of these, 35.9% (133/370) had sex exclusively with HIV-positive men and are not analyzed here. We performed analyses for 144 and 148 women who had sex with main and casual partners, respectively, who were of negative or unknown HIV serostatus. Fifty-three women had both types of partners. Some women (n = 45) had sex with both men and women; however, given homogeneity with respect to the outcome and other correlates, modeling was done without adjustment for this distinction.
Since this analysis was concerned with HIV transmission, the dependent variable was self-reported vaginal and/or anal sex acts with male partners who were known to the participant as HIV negative or of unknown serostatus (herein referred to as serodiscordant). The analysis was separately conducted for main versus casual partners because condom use may vary by this dimension. A main partner was defined as a boyfriend, spouse, or significant other. Casual partners were defined as a friend, acquaintance, or a stranger who was not a commercial sex partner. The dependent variable was dichotomized into women reporting either inconsistent (at least 1 episode of condom-unprotected vaginal and/or anal sex) or consistent condom use (referent group). This cut point was used because only consistent condom use is associated with reduced HIV transmission among serodiscordant pairs.7 This approach also made it possible to identify characteristics of women most in need of support for limiting the sexual spread of HIV. To facilitate comparison with other reports,28 we also present data on the proportion of vaginal and anal sex acts that were protected with condoms.
All variables were collected via self-report through audio-computer-assisted-structured interviewing (ACASI) (Questionnaire Development System, NOVA Research Company, Bethesda, MD). The recall period for all behaviors was 3 months prior unless otherwise noted. Individual-level variables evaluated included demographics (age, race/ethnicity, education, income, employment, incarceration, sexual orientation), sexual activity (number of partners, frequency of sex), years since HIV diagnosis and injecting drugs, illicit drug use (crack/cocaine; heroin, both crack/cocaine and heroin, other, none), sharing of injection paraphernalia (yes/no), the use of alcohol or drug during sex (always/usually, sometimes, never), and whether a woman had traded sex for money or drugs (yes/no).
We also considered 5 psychologic constructs. For these variables, no referent timeframe was specified except where noted. Psychologic distress experienced in the week before interview was measured using the sum of the anger (6 items), depression (7 items), and hostility (5 items) subcomponents of the Brief Symptoms Inventory (higher score indicated greater distress; Cronbach alpha-a measure of a scale's reliability-was 0.91).29 Self-efficacy for safe sex was assessed via a 9-item scale, which inquired about participants' confidence to use condoms during different situations (higher score indicated greater self-efficacy; Cronbach alpha among women with their main and casual partners were each 0.95). Women were asked the self-efficacy scale questions with respect to each of her sexual partner(s). Among women with multiple partners, scale scores among women's main and among casual partner(s) were positively and highly correlated. Given this, mean scale scores for each partner type were used. Condom beliefs were measured using the 4-item, hedonistic outcomes expectancies scale (higher score indicated more negative beliefs about condoms; Cronbach alpha = 0.88).30 Personal responsibility was measured via a 7-item scale that inquired about perceived responsibility to limit the spread of HIV during a variety of situations (higher score indicated greater responsibility; Cronbach alpha = 0.83).31 The construct of empowerment was assessed using a 28-item scale measuring perceived ability to influence one's environment generally, not specific to partner nor context (higher score indicated greater empowerment; Cronbach alpha = 0.76).32
The 4 relationship-level variables evaluated were partner specific. For women with more than 1 partner within a category (ie, 2 casual partners), scores were correlated within partner type and mean scale score was used for each partner type. Mutual disclosure was categorized into whether both partners in a coupling revealed their HIV status (yes) versus not. Disclosure patterns among women with multiple casual partnerships were inconsistent, and therefore this variable was not considered for the casual partner models. Partner norms about condom use was measured by inquiring whether women (1) perceived her partner thought a condom should be used and (2) felt it was important to adhere to that expectation. Combined, these measured normative beliefs about condom use within the partnership (higher score indicated stronger normative beliefs supporting condom use).33 Expected partner reactions were measured via 8 questions asking whether one s partner would be mad, supportive, distrustful, or want to break up the relationship in response to a request for condom use (higher score indicated more favorable reactions to requests for condom use; Cronbach alpha = 0.82 and 0.73 for main and casual partner(s), respectively). Physical violence was measured using women's reports of victimization by her male partner with the Conflict Tactics Scale (higher score indicated greater victimization; Cronbach alpha = 0.85).34 This scale was only administered in reference to steady partner(s).
Social-level variables included women's perception of peer support for condom use with her partners, which was measured by questions similar to perceived partner support (normative beliefs variable). Although this variable was asked in reference to each partner, it did not measure whether peers knew the HIV status of the subject or her partner(s). Social support was measured using 5 items adapted from the Barerra method for assessing social support in community settings (higher score indicated greater perceived social support; Cronbach alpha = 0.87).35 Domains included intimate interaction and directive guidance (ie, listened to personal problems and gave advice), positive social interaction, and instrumental support (ie, would lend money or help).
Bivariate associations between independent variables and the outcome were assessed using χ2 and t tests, as applicable. There was no association between the dependent variable and recruitment city in bivariate or multivariate analysis, nor did inclusion of recruitment city affect other variables in the model. Therefore, final models did not include recruitment city. Logistic regression models were separately built to examine the outcome among sex with main versus casual partners because some women had both partner types, and because reasons for condom use may vary by partner type.28 Modeling was performed by first considering demographic variables, then variables significant at P < 0.15 were added in groups starting first with individual-level variables, then those at the relationship-level variables, and finally the social-level variables. We used the −2 log-likelihood value and the deviance statistic to guide modeling decisions. We obtained the model of best fit by first obtaining a final set of individual-level variables, and then considered whether each successive set of variables (relationship; social) improved the fit.
Women in this sample were typically impoverished minorities with an average age of 40 years (Table 1). Most women had an annual income under $5000, a history of incarceration, and most were unemployed; about three quarters were self-identified as heterosexual and one fifth as bisexual. On average, women had known they were HIV positive for 8 years and had injected drugs for 17.5 years.
One hundred forty-four women reported sex with at least 1 serodiscordant main partner and 148 women with at least one such casual partner. Among these couplings, 64% and 47% of women knew the HIV status of their main and casual partners, respectively. Women had an average of 1 main and 2 casual serodiscordant partners (Table 1). Women engaged in a variety of sexual behavior with these partners: all but 1 had vaginal sex, about two thirds had received oral sex whereas 30% and 40% had anal sex with their main and casual partners, respectively. Most women engaged in a variety of sexual acts: 28% engaged in all 3 types of sex and 40% in 2 types of sex (data not shown in table). Regarding vaginal and/or anal sex, on average condoms were used half of the time with main partners and two thirds of the time with casual partners; 60% of women were inconsistent condom users during vaginal and/or anal sex with their serodiscordant main partners, and 53% were so with their casual partners.
In bivariate analysis, inconsistent condom use with main partners was associated with alcohol or drug use during sex, greater psychologic distress, lower self-efficacy to use condoms, more negative beliefs about condoms, and lower sense of personal responsibility to protect others from HIV (Table 2). Additionally, all of the relationship and social-level variables examined were associated with inconsistent condom use with main partners (Table 3). Regarding serodiscordant casual partners, inconsistent condom use was associated with greater psychologic distress and lower self-efficacy for using condoms, personal sense of responsibility for limiting HIV transmission, normative support for condoms within the partnership, and perceived social support (Tables 2 and 3).
In multivariate analysis, 3 individual-level and 2 relationship characteristics remained associated with inconsistent condom use with serodiscordant main partners (Table 4). Compared with consistent condom users, inconsistent condom users had lower self-efficacy to use condoms, more negative beliefs about condoms, and believed they had less responsibility for limiting the spread of HIV to others. Women who believed their main partners were unsupportive of condoms were more likely to be inconsistent condom users. Inconsistent condom use was also more likely in relationships without mutual disclosure of both partners HIV status. Regarding serodiscordant casual partners, 3 personal characteristics were associated with inconsistent condom use. Women with lower self-efficacy for condom use, higher psychologic distress, and those who had traded sex for money or drugs were more likely to be inconsistent condom users with casual partners.
This analysis indicated that there are widespread opportunities for the sexual transmission of HIV from women who inject drugs to at-risk men. Women in this sample had known they had HIV for almost a decade, were having sex about twice a week, and at least half had unprotected with at-risk men. Also of note, and in contrast to other reports,36 was the large proportion of women engaging in anal sex with casual partners. That 40% of this sample had anal sex outside steady partnerships is troubling given the potential that multiple partnering may have on the risk of spreading disease.37 The present analysis is one of only a few to specifically evaluate condom use during heterosexual couplings while taking into account the serostatus of both partners. Focusing on at-risk partners is important because sexual behaviors may differ by partner serostatus24 and because it is these discordant sexual encounters that propagate the sexual transmission of HIV. This is only the second study to specifically focus on HIV-positive IDU women with their serodiscordant partners and the first study to include a substantial number of women. A better understanding of sexual risks among IDUs is important because IDUs often feel vulnerable to HIV from their injection, but not their sexual, behaviors.38 Yet as evidenced here and in other reports, sexual risk taking among IDUs continues and is contributing to the spread of HIV among IDUs.39,40
Condom use and selective partnering are 2 strategies for reducing the risk of HIV transmission. Whereas the amount of condom use in the present sample was higher than that observed among other samples of at-risk women,18 it was on par with that reported from other studies of HIV-positive people.41 Another secondary prevention strategy that HIV-positive people have used is to couple with other HIV positives, but we saw that only one third of women in this sample exclusively did so. The present study echoes the need to find innovative ways to help a substantial proportion of HIV-positive women engaged in protected sex for the many years they live with HIV infection.
In this sample, factors associated with inconsistent condom use differed by partner type. Condom use in steady partnerships was influenced by attitudes and beliefs of the HIV-positive woman and, to some extent, by her partner. Consistent with the literature,31,42-44 women who felt confident about their ability to use condoms during a variety of challenging situations, who felt a greater personal responsibility to limit HIV transmission, and who held fewer negative beliefs about condoms were 2 to almost 4 times more likely to use them. In addition, most women felt responsible for limiting HIV transmission, and this sentiment was associated with safer sex behavior indicating that HIV-positive IDU women may be receptive to messages encouraging them to play a greater role in curbing HIV transmission. The strongest predictor of inconsistent condom use in steady partnerships was whether each member in the couple had disclosed their HIV status. We were unable to determine how patterns of disclosure (ie, she told/he did not or vice versa) were associated with condom use. Further, it is unclear from this cross-sectional analysis whether it is the act of sharing information that helps to reduce risk, or whether disclosure is a marker for a safer type of relationship. Nevertheless, we do know that most of women in this sample had disclosed their HIV status to their steady male partner (data not shown), indicating that when men are informed of risk, they may be more likely to use a condom. Further, prospective investigation into the dynamics of disclosure is warranted as current literature does not provide a clear understanding of how it relates to sexual risk.20,26
In contrast to experiences with steady partners, women's behavior with casual partners was not influenced so much by beliefs as it was by women's personal situations. For example, psychologic distress and trading sex for money or drugs were strongly associated with inconsistent condom use. Together these characteristics suggest that women's need to support their addictions, and attendant distress associated with such a lifestyle, may have trumped the potentially protective effect of social support that has been found in other studies.45-47 Our findings underscore the importance of more immediate life priorities in determining HIV risk behavior.48 Our findings also suggest that interventions addressing the life circumstances of HIV-positive women may be necessary for reducing HIV risk to their casual partners.
Several study limitations should be noted. These findings derive from a convenience sample of women recruited from clinic and community sources in 4 US cities; hence, generalizations to other IDU women cannot be made with certainty. In particular, the present findings cannot be generalized to HIV-negative IDU women, who likely have more risky sexual and drug-using behaviors.15 However, the multicenter sample and use of a sample derived from treatment- and community-based sources extends information from other studies. Whereas veracity of self-report among IDUs is a common concern,49 this study used computerized data collection methods which have been shown to enhance reporting of sensitive risk behaviors among IDU samples.50 We used means of scale data collected across multiple partnerships, which may have attenuated statistical power. However, many scaled measures produced statistically significant associations, which diminished this concern. We had only limited measures available to assess relationship and social factors, and thus we were unable to comprehensively evaluate their impact on the outcome. Of particular note was our inability to assess the impact of disclosure on condom use among casual partnerships. Finally, the analysis was cross sectional; hence, time order could not be established. Strengths of this study were the use of psychometric scales51 that performed well among the present sample52 and the ability to focus on sexual encounters with partners known or suspected to be at risk for HIV.
As medications enhance and extend HIV-positive people's lives, the need intensifies to identify innovative, effective, and realistic prevention strategies to help HIV-positive people limit the sexual transmission of HIV. Realistic efforts are those that acknowledge the life-long commitment HIV-positive people must make in altering their sexual behavior. The present study suggests that prevention efforts focused on HIV-positive IDU women need to address personal factors and the contextual aspects of sex.
The authors thank Yingfeng Wu for data analysis. The INSPIRE Study Group includes the following people: Carl Latkin, Amy Knowlton, and Karin Tobin (Baltimore); Lisa Metsch, Eduardo Valverde, James Wilkinson, and Martina Devarona (Miami); Mary Latka, David Vlahov, Phillip Coffin, Marc Gourevitch, Julia Arnsten, and Robert Gern (New York); Cynthia Gomez, Kelly Knight, Carol Dawson Rose, Starley Shade, and Sonia Mankenzie (San Francisco); David Purcell, Yuko Mizuno, Scott Santibanez, Richard Garfein, and Ann O'Leary (CDC); and Lois Eldred and Kathleen Handley (HRSA). The authors also acknowledge the following people for their contributions to this research: Susan Sherman, Roeina Marvin, Joanne Jenkins, Donny Gann, and Tonya Johnson (Baltimore); Clyde McCoy, Rob Malow, Wei Zhao, Lauren Gooden, Sam Comerford, Virginia Locascio, Curtis Delford, Laurel Hall, Henry Boza, and Cheryl Riles (Miami); George Fesser, Carol Gerran, and Diane Thornton (New York); Caryn Pelegrino, Barbara Garcia, Jeff Moore, Erin Rowley, Debra Allen, Dinah Iglesia-Usog, Gilda Mendez, Paul Lum, and Greg Austin (San Francisco);Gladys Ibanez, Hae-Young Kim, Toni McWhorter, Jan Moore, Lynn Paxton, and John Williamson (CDC); and Lee Lam, Jeanne Urban, Stephan Soroka, Zilma Rey, Astrid Ortiz, Sheila Bashirian, Marjorie Hubbard, Karen Tao, Bharat Parekh, and Thomas Spira (CDC Laboratory).
1. Advancing HIV
prevention: new strategies for a changing epidemic-United States, 2003. MMWR Morb Mortal Wkly Rep
2. Incorporating HIV
prevention into the medical care of persons living with HIV
. Recommendations of CDC, the Health Resources and Services Administration, the National Institutes of Health, and the HIV
Medicine Association of the Infectious Diseases Society of America. MMWR Recomm Rep
3. Janssen RS, Valdiserri RO. HIV
prevention in the United States: increasing emphasis on working with those living with HIV
. J Acquir Immune Defic Syndr
. 2004;37(suppl 2):S119-S121.
4. Kral AH, Bluthenthal RN, Lorvick J, et al. Sexual transmission of HIV
-1 among injection drug users in San Francisco, USA: risk-factor analysis. Lancet
5. Strathdee SA, Sherman SG. The role of sexual transmission of HIV
infection among injection and non-injection drug users. J Urban Health
6. Centers for Disease Control and Prevention. HIV
/AIDS Surveillance Report, 2003. Department of Health and Human Services, Centers for Disease Control and Prevention 15, 2004.
7. de Vincenzi I. A longitudinal study of human immunodeficiency virus transmission by heterosexual partners. European Study Group on Heterosexual Transmission of HIV
. N Engl J Med
8. Pinkerton SD, Abramson PR, Kalichman SC, et al. Secondary HIV
transmission rates in a mixed-gender sample. Int J STD AIDS
9. Celentano DD, Galai N, Sethi AK, et al. Time to initiating highly active antiretroviral therapy among HIV
-infected injection drug users. AIDS
10. Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med
11. Heard I, Potard V, Costagliola D, et al. Contraceptive use in HIV
. J Acquir Immune Defic Syndr
12. Skurnick JH, Kennedy CA, Perez G, et al. Behavioral and demographic risk factors for transmission of human immunodeficiency virus type 1 in heterosexual couples: report from the Heterosexual HIV
Transmission Study. Clin Infect Dis
13. Diaz T, Schable B, Chu SY. Relationship between use of condoms and other forms of contraception among human immunodeficiency virus-infected women
. Supplement to HIV
and AIDS Surveillance Project Group. Obstet Gynecol
14. Dolezal C, Ehrhardt AA, Meyer-Bahlburg HF, et al. Sexual risk behavior
changes among HIV
+ and HIV
− female injecting drug users over 4 years. Women Health
15. Wilson TE, Koenig L, Ickovics J, et al. Contraception use, family planning, and unprotected sex: few differences among HIV
-infected and uninfected postpartum women
in four US states. J Acquir Immune Defic Syndr
16. Timpson SC, Williams ML, Bowen AM, et al. Condom use
behaviors in HIV
-infected African American crack cocaine users. Subst Abus
17. Buchacz K, van der SA, Saul J, et al. Sociodemographic, behavioral, and clinical correlates of inconsistent condom use
heterosexual couples. J Acquir Immune Defic Syndr
18. Wilson TE, Massad LS, Riester KA, et al. Sexual, contraceptive, and drug use behaviors of women
and those at high risk for infection: results from the Women
's Interagency HIV
19. Campsmith ML, Nakashima AK, Jones JL. Association between crack cocaine use and high-risk sexual behaviors after HIV
diagnosis. J Acquir Immune Defic Syndr
20. Parsons JT, Missildine W, Van OJ, et al. HIV
serostatus disclosure to sexual partners among HIV
-positive injection drug users. AIDS Patient Care STDs
21. Bedimo AL, Bennett M, Kissinger P, et al. Understanding barriers to condom usage among HIV
-infected African American women
. J Assoc Nurses AIDS Care
22. Hetherington SE, Harris RM, Bausell RB, et al. AIDS prevention in high-risk African American women
: behavioral, psychological, and gender issues. J Sex Marital Ther
23. Lansky A, Nakashima AK, Jones JL. Risk behaviors related to heterosexual transmission from HIV
-infected persons Supplement to HIV
/AIDS Surveillance Study Group. Sex Transm Dis
. 2000;27: 483-489.
24. Crepaz N, Marks G. Towards an understanding of sexual risk behavior
in people living with HIV
: a review of social, psychological, and medical findings. AIDS
25. Kirshenbaum SB, Hirky AE, Correale J, et al. "Throwing the dice": pregnancy decision-making among HIV
in four U.S. cities. Perspect Sex Reprod Health
26. Kalichman SC, Rompa D, Luke W, et al. HIV
transmission risk behaviours among HIV
-positive persons in serodiscordant
relationships. Int J STD AIDS
27. Purcell DW, Metsch LR, Latka M, et al. Interventions for seropositive injectors-research and evaluation: an integrated behavioral intervention with HIV
-positive injection drug users to address medical care, adherence, and risk reduction. J Acquir Immune Defic Syndr
28. Mills S, Saidel T, Magnani R, et al. Surveillance and modelling of HIV
, STI, and risk behaviours in concentrated HIV
epidemics. Sex Transm Infect
. 2004;80(suppl 2):ii57-ii62.
29. Derogatis LR, Spencer PM. Brief Symptoms Inventory. Baltimore, MD: John Wiley & Sons, 1982.
30. O'leary A, Wolitski RJ, Remien RH, et al. Psychosocial correlates of transmission risk behavior among HIV
-seropositive gay and bisexual men. AIDS
31. Wolitski RJ, Bailey CJ, O'leary A, et al. Self-perceived responsibility of HIV
-seropositive men who have sex with men for preventing HIV
transmission. AIDS Behav
32. Rogers ES, Chamberlin J, Ellison ML, et al. A consumer-constructed scale to measure empowerment among users of mental health services. Psychiatr Serv
33. Fishbein M, Middlestadt SE, Trafimow D. Social norms for condom use
: implications for HIV
prevention interventions of a KABP survey with heterosexuals in the Eastern Caribbean. Adv Consum Res
. 1993; 20:292-296.
34. Straus M. Measuring intrafamily conflict and violence: the Conflict Tactics (CT) scales. J Marriage Fam
35. Barrera MA. A method for assessing social support networks in community survey research. Connections
36. Tross S. Women
at heterosexual risk for HIV
in inner-city New York: reaching the hard to reach. AIDS Behav
37. Laumann EO, Youm Y. Racial/ethnic group differences in the prevalence of sexually transmitted diseases in the United States: a network explanation. Sex Transm Dis
38. Grella CE, Campos M, Anglin MD. Relationship of HIV
testing and high-risk behaviors among clients in methadone maintenance treatment. AIDS Educ Prev
39. Nelson KE, Galai N, Safaeian M, et al. Temporal trends in the incidence of human immunodeficiency virus infection and risk behavior among injection drug users in Baltimore, Maryland, 1988-1998. Am J Epidemiol
40. Bogart LM, Kral AH, Scott A, et al. Sexual risk among injection drug users recruited from syringe exchange programs in California. Sex Transm Dis
transmission risk behavior among men and women
living with HIV
in 4 cities in the United States. J Acquir Immune Defic Syndr
42. Wingood GM, DiClemente RJ, Mikhail I, et al. A randomized controlled trial to reduce HIV
transmission risk behaviors and sexually transmitted diseases among women
living with HIV
: the WiLLOW Program. J Acquir Immune Defic Syndr
. 2004;37: S58-S67.
43. Wolitski RJ, Parsons JT, Gomez CA. 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). J Acquir Immune Defic Syndr
44. Dancy BL, Berbaum ML. Condom use
predictors for low-income African American women
. West J Nurs Res
45. Crosby RA, DiClemente RJ, Wingood GM, et al. HIV
/STD prevention benefits of living in supportive families: a prospective analysis of high risk African-American female teens. Am J Health Promot
46. Reilly T, Woo G. Social support and maintenance of safer sex practices among people living with HIV
/AIDS. Health Soc Work
47. Prado G, Feaster DJ, Schwartz SJ, et al. Religious involvement, coping, social support, and psychological distress in HIV
-seropositive African American mothers. AIDS Behav
48. Mizuno Y, Purcell D, Borkowski TM, et al. The life priorities of HIV
-seropositive injection drug users: findings from a community-based sample. AIDS Behav
49. Samuels JF, Vlahov D, Anthony JC, et al. Measurement of HIV
risk behaviors among intravenous drug users. Br J Addict
50. Macalino GE, Celentano DD, Latkin C, et al. Risk behaviors by audio computer-assisted self-interviews among HIV
-seropositive and HIV
-seronegative injection drug users. AIDS Educ Prev
51. McIver JP, Carmines EG. Unidimensional Scaling. Thousand Oaks: Sage Publications, 1981.
52. Gliem JA, Gliem RR. Calculating, interpreting and reporting Cronbach's alpha reliability coefficient for Likert-type scales. Midwest Research to Practice Conference in Adult, Continuing and Community Education, 2003. 10-8-2003.