Several studies have shown that a majority of persons who know they are HIV infected reduce or at least temporarily refrain from activities that would create risk to others. 1,2 A significant minority of infected individuals report that they continue to engage in unprotected behavior with their sexual partners, however. For example, between 29% and 39% of HIV-infected men who have sex with men (MSM) in community samples say that they have engaged in unprotected sex during time periods ranging from the past 3 months to the past year, 3,4 including men in known HIV-serodiscordant relationships. 5 Similar or higher proportions of HIV-infected injection drug users (IDUs), homeless men, and heterosexual men and women indicate that they have unprotected sex even though they are aware of their positive HIV serostatus. 6–10
Most of these conclusions are inferred from smaller samples drawn from a single geographic area/site. Several other gaps in the literature are also addressed in the current study, including the following: (1) studies have tended to focus primarily on the sexual risk behavior of HIV-seropositive MSM, whereas the behavior of women, heterosexual men, and IDUs has received less attention; (2) most studies have defined unprotected sex as the occurrence of any anal, vaginal, or oral intercourse, even though oral sex carries much lower HIV transmission risk than anal and vaginal intercourse, and this lack of specificity has limited the ability to estimate the number of infections caused by the behavior reported; (3) the context of potentially unsafe sex (eg, partner status, disclosure of HIV status) is infrequently described; and (4) many studies were conducted before the advent of highly active antiretroviral therapy (HAART), which may have influenced perceptions about the seriousness of the disease and the need to maintain safer sex practices. 11–16 These methodologic issues restrict the conclusions that can be reached from past research about the extent to which HIV-infected persons currently engage in activities that could pose risk for viral transmission.
The purpose of the current study was to determine the prevalence and frequency of unprotected high-risk sexual and drug injection behaviors in a large sample of persons living with HIV infection and to assess the associated risk of HIV transmission to their partners while addressing the limitations of prior research. This information was used to estimate the number of sexual partners who would acquire HIV as a consequence of sexual contact with study participants.
Participants and Recruitment
The data reported in this paper are from baseline interviews that were used as a screening assessment for a randomized controlled trial (RCT) of a coping and sexual risk reduction intervention for HIV-infected persons. Assessment interviews lasting between 2 and 4 hours were conducted in private settings in research offices, community-based organizations, and clinics in 4 cities: Los Angeles, Milwaukee, New York, and San Francisco. The sample reported in this article consists of 3723 HIV-positive persons: 1918 MSM, 978 women, and 827 heterosexual men.* Persons were recruited and consented first for the survey only. If eligible, based on survey responses, a participant was informed of the opportunity to enroll in the RCT.
Men were considered MSM if they reported any sexual contact with other men in the past 3 months regardless of self-identified sexual orientation or additional sexual contact with women. Men were placed in the heterosexual category if they had only had sexual contact with women in the past 3 months regardless of self-identified sexual orientation. When no sexual activity in the past 3 months was reported, we relied on the men’s self-identified orientation. Because sex with an IDU has been identified as an important vector for HIV transmission, in a separate analysis, we examined the sexual behavior of those who reported injecting illicit substances in the past 12 months, aggregating across the 373 IDUs who were women, MSM, and heterosexual men.
Recruitment and screening of potential intervention study participants took place in clinics and community agencies serving HIV-positive clients by staff trained to explain the project, recruit, and screen clients served by the organization. Brochures, posters, and staff descriptions about the study were used to advertise the study. In addition, advertisements were placed in newspapers and magazines serving HIV-positive and MSM populations. Persons learning of the study by word of mouth were also eligible for screening. Interested persons who provided verbal consent were briefly screened by project personnel and, if they wished to participate, scheduled for a baseline interview at their preferred study site.
Participants were required to be at least 18 years of age, to provide written informed consent and medical documentation of their HIV-positive serostatus, to be free of severe neuropsychologic impairment or psychosis (assessed on a case-by-case basis by senior project personnel), and not to be currently involved in another behavioral intervention study related to HIV risk behavior reduction. Study procedures were approved by the Institutional Review Boards of the University of California at Los Angeles, the Medical College of Wisconsin, the New York State Psychiatric Institute, and the University of California at San Francisco.
Individual private assessment interviews were conducted by trained interviewers using a combination of Audio Computer-Assisted Self-Interviewing (ACASI) and Computer-Assisted Personal Interviewing (CAPI) procedures. Participants were compensated $50 for completing the baseline interview.
Demographic Characteristics and Health Status Indicators
Demographic characteristics and health status indicators were assessed using CAPI. Items included participant age, race/ethnicity, gender, self-identified sexual orientation, relationship status, educational level, employment status, and income. In addition, health status indicators, including self-reported most recent CD4 count, HIV viral load, 17 number of months living with HIV infection, and current use of HAART, were assessed.
Sexual Behavior: Partner by Partner and Global Assessment
Informed by the results of earlier qualitative studies at the initiation of the project, a detailed ACASI interview adapted from the Sexual Risk Behavior Assessment Schedule (SERBAS)† was developed to assess sexual behavior. The interview had a 3-month retrospective recall period and included separate but equivalent versions of close-ended questions for heterosexual men, homosexual men, and women.‡ All participants were asked if they engaged in any sexual activity during the past 3 months with men, women, or both and the number of partners of each gender. Based on responses to these items and the gender of the participant, the interview only asked pertinent questions about sexual behavior.
Individually, for up to 5 sexual partners of each gender, the participant defined the relationship between them (main partner; someone with whom you had sex for love or fun but not as a main partner; someone with whom you had sex once but not again; sex with someone for drugs, money, or a place to stay; or someone who forced you to have sex). For each of the first 5 partners of each gender and beginning with the most recent sexual partner, participants were asked (1) how many times vaginal, anal, and oral sex took place, including, if appropriate, whether anal and oral contact was insertive or receptive; (2) the number of times they used condoms (using separate questions for male and female condoms, when appropriate) from the beginning to the end of penetration; (3) when the participant was the insertive partner for a given sex act, he was asked the number of times internal ejaculation occurred with no protection; (4) if they knew the most recent HIV test result (positive, negative, or unknown) of each sex partner they identified; and (5) if the partner was aware of the participant’s HIV status; if so, participants were given options for how the partner learned this information. Disclosure was operationalized as the participant having told the partner directly that he/she was HIV-positive.
For participants who reported more than 5 partners of either gender, partner-by-partner questions were followed by global partner questions that assessed the total number of times they engaged in specific protected and unprotected activities as well as the other questions mentioned previously, including numbers of partners of different HIV serostatus, for all other partners in the past 3 months (beyond the first 5 partners).
Needle-Sharing Items and Needle Exchange
Questions from the Risk Assessment Questionnaire, the core instrument used in the National Institute on Drug Abuse (NIDA)-supported National AIDS Demonstration Research (NADR) project, were adapted to assess injection practices and injection-related risk behaviors 19 using ACASI. These items were used to determine type(s) of drugs injected; frequency of drug injection; patterns of new, cleaned, reused, and shared needle use; patterns of syringe exchange program use; and presence and nature of equipment sharing and cleaning in the past 3 months. In the current report, only indicators of injection frequency and needle sharing were examined.
Transmission Risk Modeling
For each study participant, the probability, Pi, that he or she would transmit HIV to sex partner i was estimated using the equation:
The total number of new infections expected among the sex partners of a study participant equals the sum of the transmission probabilities for each of his or her partners. 20 In this equation, a, b, c, and d denote the number of sex acts with the partner (unprotected and condom-protected insertive and receptive anal intercourse for male-to-male sex, unprotected and protected insertive anal and vaginal intercourse for heterosexual male participants, and unprotected and protected receptive anal and vaginal intercourse for female participants), and αa, αb, αc, and αd represent the associated per-act transmission probabilities. The following transmission probabilities were used in the analysis: 0.0006 for unprotected receptive anal and vaginal intercourse, 0.001 for unprotected insertive vaginal intercourse, and 0.02 for unprotected insertive anal intercourse 21,22; condoms were assumed to reduce the probability of HIV transmission by 90%, taking into account breakage, slippage, incorrect usage, and other potential sources of failure. 23,24 The μ parameter represents the reduction in transmissibility because of HAART, which was set to 50% if the study participant reported receiving HAART, based on recent estimates of the impact of HAART on HIV transmission among MSM in San Francisco, 25 and to 0 otherwise. The λ parameter denotes the probability that the partner was already infected with HIV: λwas set to 1 if the participant reported that the partner was HIV-positive, to 0 if the partner was HIV-negative, and to an estimate of the prevalence of HIV among MSM or heterosexuals, as appropriate, in the study city 26 if the partner’s HIV status was not known. Sensitivity analyses were conducted to determine how the parameter values used in the model affected the estimated number of new infections associated with study participants’ sexual activities.
Demographic Characteristics of the Sample
Table 1 shows the number of participants recruited in each city and presents demographic and background characteristics of the sample categorized into MSM, women, and heterosexual male subgroups. In all groups, the mean age of participants was in the early 40s. Sixty percent of women and heterosexual men were African American, and approximately 20% were Hispanic. The MSM sample was ethnically diverse, with approximately one third of its members African American, one third white, and one third men of other ethnicities, predominantly Hispanic. Most members of the MSM sample had completed at least some college, whereas the majority of women and heterosexual men had a high school education or less. Most participants were unemployed. Approximately 60% of women and 49% of heterosexual men were currently in a primary relationship, although most were not married; 36% of MSM had a current primary partner. The sample reflects the demographic characteristics of the present HIV epidemic in the United States within each subgroup. 27
Consistent with the history of the HIV epidemic in the United States, on average, MSM in the sample had been living longest with HIV, whereas women had been infected most recently. Three fourths of the sample reported a current HAART regimen. 28
Sexual and Injection-Related Transmission Risk Behaviors
Table 2 shows the prevalence of sexual behaviors and injection-related behaviors during the past 3 months. Seventy-four percent of women and heterosexual men and 81.5% of MSM were sexually active in the past 3 months. Fifty-nine percent of MSM, 21.2% of women, and 27.5% of heterosexual men had multiple sexual partners during this time period. The mean number of sexual partners in the past 3 months was 5.8 for MSM and 1.7 for both women and heterosexual men. There was considerable variability in participants’ reported total number of sexual partnerships, with some persons reporting large numbers of partners.
Similar percentages of MSM, women, and heterosexual men indicated that they had sex during the past 3 months with a partner whose serostatus was known to be HIV-negative (MSM: 30.1%, women: 28.7%, heterosexual men: 30.1%; see Table 2) or whose serostatus was not known (MSM: 30.9%, women: 20.3%, heterosexual men: 20.9%). Overall, 44.1% of women, 44.0% of heterosexual men, and 49.0% of MSM participants had recent sex with partners who were of either HIV-negative or unknown serostatus.
As expected, vaginal intercourse was reported by the large majority of women (67%) and heterosexual men (70%). Eighteen percent of heterosexual men and 15% of women also reported engaging in heterosexual anal intercourse. Among MSM, 49.4% had insertive anal intercourse and 52.6% had receptive anal intercourse with same-sex partners during the past 3 months. With respect to unprotected behavior, 44.7% of MSM, 36.5% of women, and 34.0% of heterosexual men said that they had engaged in unprotected anal or vaginal sex during the previous 3-month period.
Occurrence of Unprotected Intercourse in Relation to Partner Serostatus and Serostatus Disclosure
From the perspective of viral transmission to uninfected persons, a critical question is whether unprotected vaginal or anal intercourse acts occur with partners who are not themselves infected. Taking into account only the reports of activities occurring between participants and up to 5 of their partners of each gender who were described in detail, 15.6% of MSM, 19.0% of women, and 13.1% of heterosexual men engaged in unprotected vaginal or anal intercourse with partners who were HIV-negative or whose serostatus was unknown (see Table 2). Some participants, especially MSM, had more than 5 partners in the past 3 months. Approximately 14.7% of MSM said that they had engaged in unprotected intercourse with some of these additional partners, although our global partner assessment section could not detect the specific relation between partner status and prevalence of unprotected sex.
Table 3 includes the mean and median frequencies of unprotected vaginal or anal intercourse by partner serostatus; these values constitute the average levels of risk behavior among participants who reported any occurrence of that behavior during the past 3 months. The MSM who reported unprotected intercourse with HIV-negative partners in the past 3 months did so a mean of 6.2 times and a mean of 3.8 times with partners of unknown serostatus in this time period (see Table 3). Among women, there was a mean of 13.2 acts with seronegative partners and a mean of 14.5 acts with partners whose serostatus was not known. Heterosexual men had a mean of 10.1 unprotected acts with seronegative partners and a mean of 9.2 unprotected intercourse acts with female partners of unknown serostatus during the past 3-month period.
Overall, the majority of participants reported disclosing their serostatus to all their sexual partners (Table 4). Only among MSM was the percentage of disclosure to HIV-positive partners higher than to HIV-negative partners and partners of unknown serostatus.
A total of 17,144 unprotected intercourse acts were reported in the sample during the past 3 months. Of these, 11,561 (67%) acts occurred with HIV-positive seroconcordant partners, 3049 (18%) acts occurred with HIV-negative partners, and 2202 (13%) acts occurred with partners whose HIV status was not known to the participant. The majority of unprotected acts (13,264 of 16,757 [78%]) took place with persons described as main or steady partners, although the HIV status of the steady partner was reportedly negative or unknown for nearly one third of these acts. Partner type information was elicited only for up to the first 5 sexual partners, and unprotected activities that occurred with additional partners are not included in this partner type summary.
Needle-Sharing Risk Behavior Among Injection Drug Users in the Sample
As the bottom panel in Table 2 shows, 173 MSM (9.0%), 79 heterosexual men (9.6%), and 52 women (5.3%) said that they had injected illicit drugs during the past 3 months. Between 15.4% and 19.1% of IDUs said they had lent their needles to someone else in the past 3 months. Sexual risk behaviors among the 304 participants who reported injecting drugs in the past year were most similar to those of the MSM sample: 78.5% of IDUs (n = 293) were sexually active, 49.9% (n = 186) had multiple partners, and 52.0% (n = 194) engaged in unprotected intercourse during the past 3 months. Finally, 70 (18.7%) HIV-infected IDUs reported engaging in unprotected vaginal or anal intercourse with seronegative partners or partners of unknown serostatus during this time period.
Based on the sexual behaviors reported by study participants, 30.4 at-risk (HIV-negative or serostatus unknown) sexual partners would be expected to acquire HIV as a result of their sexual interactions with study participants during the 3-month assessment period. As detailed in Table 5, the partners of MSM accounted for 24.2 (79.7%) of these new HIV infections, the partners of heterosexual men for 4.8 (15.8%), and the partners of women for 1.4 (4.5%). The mean number of new infections per study participant also was greater for MSM (12.6 per 1000 participants) than for heterosexual men (5.8 per 1000 participants) or women (1.4 per 1000 participants). These estimates were sensitive to the per-act transmission probability values used in the analysis: doubling these probabilities increased the total number of infections to 58.4, whereas halving them decreased the number of secondary infections to 15.6. The results also were sensitive to the presumed effectiveness of HAART at reducing the probability of HIV transmission, ranging from 18.9 new infections (assuming that HAART completely eliminates transmission risk) to 44.4 new infections (assuming that HAART has no effect on transmission risk). The results were not especially sensitive to changes in the condom effectiveness parameter or to the prevalence of infection in study cities.
In this large ethnically and geographically diverse multisite sample of persons living with HIV infection, most participants were sexually active, but more than half of MSM and approximately two thirds of women and heterosexual men did not report engaging in unprotected vaginal or anal intercourse with any of their recent sexual partners. Among those who did engage in unprotected behavior, more than two thirds of unprotected acts occurred with HIV-positive partners. These findings are heartening because they indicate that most persons aware of their own HIV infection refrain from sexual activities likely to transmit the virus to others who are not infected. Further, most HIV-infected persons disclosed their positive HIV status to all partners of any HIV serostatus. Although there were other response options to indicate that the partner knew about the participant’s HIV serostatus (eg, “we met at an HIV-positive support group”), we examined the most unambiguous measure of disclosure: directly telling the partner. Thus, these disclosure rates are likely conservative, because additional partners were undoubtedly aware of the participant’s HIV status through the other means described.
Our findings also indicate several patterns of major public health concern. First, although most participants reported safer behavior, between 13% and 19% of participants engaged in unprotected vaginal or anal intercourse during the past 3 months with partners who were HIV-negative or whose HIV status was not known. The majority of these acts took place with steady or main partners, although more than 20% of unprotected intercourse occurred with other types of partners. These findings may underestimate the actual prevalence of high-risk sex taking place with uninfected partners, because for participants with more than 5 partners of either gender, we were not able to obtain data on each additional partner’s HIV status individually. If time frames longer than “the past 3 months” were examined, even higher transmission risk behavior prevalence might be found. Further, it is reasonable to question whether many participants knew the HIV serostatus of sexual partners with whom they were not well acquainted (eg, described as “someone I had sex with once but don’t plan to again,” or “a person I had sex with for money drugs, et cetera”). Thus, at least some of these partners who the participants reported as being HIV-positive may have in fact been HIV-negative and vice versa.
The unprotected sexual behaviors reported with HIV-negative and unknown serostatus partners are a significant public health concern. The mathematic modeling analyses indicated that more than 30 new infections could be expected among the sexual partners of study participants in the 3-month assessment period. (These analyses assumed, based on limited empiric evidence, that the risk of HIV transmission was reduced by 50% for study participants on HAART. If HAART completely suppressed HIV transmission, 19 new infections would be expected, whereas 44 new infections would be expected if HAART had no effect on the transmission of HIV.) Significantly, if the behaviors reported during the assessment period were maintained for 10 years, 671 new infections would occur, even if the number of partners did not increase during this time. The male partners of MSM were at particularly high risk of HIV acquisition because of the substantial transmission risk associated with unprotected anal intercourse, which was not uncommon in the study population.
Second, a significant minority of participants did not disclose their serostatus to partners before sexual activity. Although findings are mixed regarding the relation between disclosure and risk behavior in a range of contexts, it is possible that engendering disclosure and open communication could contribute to a reduction in sexual risk behaviors in some contexts and thus reduce transmission to those who are HIV-negative. Further research is necessary to identify determinants of risk behavior and disclosure of HIV status in different sexual contexts depending on relationship status and perceived HIV status of the partner.
A third major public health concern is for the health of the HIV-positive participants. More than 44% of HIV-positive MSM and more than one third of heterosexual men and women engaged in unprotected vaginal or anal intercourse, and a large proportion of them had multiple partners even in the short time frame of 3 months. Thus, and regardless of whether their partners were of concordant HIV serostatus, these individuals are exposed to a wide variety of sexually transmitted pathogens. Moreover, unprotected intercourse between HIV-positive partners carries the potential for transmitting and contracting different and possibly drug-resistant HIV strains. 29–31 HIV-positive MSM and IDUs may be at particularly high risk, because approximately twice as many reported having multiple partners in the past 3 months compared with others in the sample.
Two limitations of the study are noteworthy. First, the data consist of self-reported behavior. We used several techniques frequently employed in health behavior research to minimize biased reporting (eg, emphasizing data confidentiality, computerized assessment of sensitive behaviors) and recall errors (eg, partner-by-partner assessment of sexual behavior, up to 5 partners of each gender). Given the rates of transmission risk behavior reported, it appears that participants felt sufficiently comfortable to report their behavior candidly. Second, because it is impossible to obtain a comprehensive list of HIV-positive persons from which to randomly sample and because we were recruiting for a randomized behavioral intervention trial, we interviewed a convenience sample from multiple sources in each city (eg, infectious disease clinics, AIDS service organizations, bars, advertisements in periodicals). Nevertheless, by using a large number and range of recruitment sites, we were able to obtain a sample demographically representative of the HIV epidemic in each subgroup in the United States. 27 In comparison to another large multicity sample of HIV-infected adults, the HIV Cost and Services Utilization Study (HCSUS) cohort, 32 our study included more members of ethnic minority and low socioeconomic status groups as well as individuals who were not receiving treatment and may thus more closely approximate the HIV-infected population in the United States.
The findings of this study underscore the need to integrate ongoing effective HIV risk reduction counseling better within medical and other care programs serving persons living with HIV infection. For many years, HIV services were conceptualized primarily as prevention for the uninfected and as health care for those with the disease. This conceptualization is inadequate. Just as health, social, and other care service needs must be met to help at-risk uninfected persons successfully reduce their vulnerability to HIV, it is also essential to recognize that risk reduction behavior change assistance is needed by those HIV-infected persons who engage in risky behavior. 2 Primary care clinicians need to recognize the importance of sexuality in the lives of their patients and the reality that many of them do engage in specific behaviors that carry risk to their own health as well as to the health of some of their partners. It may be important for providers to initiate discussions around such issues to address the relative risks of certain activities, helping the patient to consider his or her own health as well as responsibility for potential transmission of HIV to partners.
It may also be important to reinforce safer sex practices among people who report using protection to maintain the behavior over time. Although brief posttest counseling is ordinarily given at the time individuals learn of their positive serostatus, this alone is not sufficient to assist many people in making, and then consistently sustaining, risk reduction steps for many years. Although some patients are successful on their own in refraining from risky practices with minimal encouragement, others are likely to require much more intensive and ongoing behavior change assistance. Continued transmission risk behavior among those with HIV has a negative impact on public health and carries significant personal risks as well. By integrating prevention with medical and social care services, it will be possible for persons with HIV infection to live longer and healthier lives and to avoid behaviors that could result in virus transmission to others and their own exposure to additional sexually transmitted infections and treatment-resistant strains of HIV.
The authors thank Ellen Stover, PhD, and Willo Pequegnat, PhD, at the National Institute of Mental Health (NIMH) for their technical assistance in developing the study and Christopher M. Gordon, PhD, and Dianne Rausch, PhD, at the NIMH for their support of this research. Thanks are also extended to the Psychosexual Core of the HIV Center for Clinical and Behavioral Studies at New York State Psychiatric Institute and Columbia University, especially Heino Meyer-Bahlburg, Terry Dugan, and Theresa Exner for collaborating with us in developing the sexual behavior interview and Terry Dugan for training the interviewers; to the assessors in each city who conducted the interviews; to our clinic- and community-based organization collaborators; to all other support staff involved in the project; and to the men and women who participated in the interviews.
This project was conducted by the NIMH Healthy Living Project Team, which comprises the following individuals: Research Steering Committee (site principal investigators and NIMH staff collaborator): Mary Jane Rotheram-Borus, PhD (University of California at Los Angeles, Los Angeles, CA), Jeffrey A. Kelly, PhD (Medical College of Wisconsin, Milwaukee, WI), Anke A. Ehrhardt, PhD (New York State Psychiatric Institute and Columbia University, New York, NY), Margaret A. Chesney, PhD (University of California at San Francisco, San Francisco, CA), and Willo Pequegnat, PhD (NIMH, Bethesda, MD); Co-Principal Investigators, Investigators, and Collaborating Scientists: Naihua Duan, PhD, Martha Lee, PhD, Marguerita Lightfoot, PhD, Rise B. Goldstein, PhD, MPH, Fen Rhodes, PhD, Robert Weiss, PhD, Richard Wight, PhD, Tyson Rogers, MA, Lennie Wong, PhD, and Philip Batterham, MA (University of California at Los Angeles, Los Angeles, CA), Lance S. Weinhardt, PhD, Eric G. Benotsch, PhD, Michael J. Brondino, PhD, Sheryl L. Catz, PhD, Cheryl Gore-Felton, PhD, and Steven D. Pinkerton, PhD (Medical College of Wisconsin, Milwaukee, WI), Robert H. Remien, PhD, A. Elizabeth Hirky, PhD, Robert M. Kertzner, MD, Sheri B. Kirshenbaum, PhD, Lauren E. Kittel, PsyD, Robert Klitzman, MD, Bruce Levin, PhD, and Susan Tross, PhD (New York State Psychiatric Institute and Columbia University, New York, NY), Stephen F. Morin, PhD, and Mallory O. Johnson, PhD (University of California at San Francisco, San Francisco, CA), Don C. DesJarlais, PhD (Beth Israel Medical Center, New York, NY), and Hannah Wolfe, PhD (St. Luke’s Roosevelt Medical Center, New York, NY); Site Project Coordinators: Willy Singh, MPH, and Daniel Hong, MA (University of California at Los Angeles, Los Angeles, CA), Kristin Hackl, MSW, and Margaret Peterson, MSW (Medical College of Wisconsin, Milwaukee, WI), Joanne Mickalian, MA (University of California at San Francisco, San Francisco, CA); and NIMH Staff Support: Ellen Stover, PhD, Christopher M. Gordon, PhD, and Dianne Rausch, PhD (NIMH, Bethesda, MD).
1. Higgins DL, Galavotti C, O’Reilly KR, et al. Evidence for the effects of HIV antibody counseling and testing on risk behaviors. JAMA
2. Weinhardt LS, Carey MP, Johnson BT, et al. Effects of HIV counseling and testing on sexual risk behavior: Meta-analysis of published research, 1985–1997. Am J Public Health (NY)
3. Simon PA, Thometz E, Bunch JG, et al. Prevalence of unprotected sex among men with AIDS in Los Angeles County, California, 1995–1997. AIDS
4. Kalichman SC, Kelly JA, Rompa D. Continued high-risk sex among HIV seropositive gay and bisexual men seeking HIV prevention services. Health Psychol
5. Remien RH, Wagner G, Dolezal C, et al. Factors associated with HIV sexual risk behavior in male couples of mixed HIV status. J Psychol Human Sex
6. Bluthenthal RN, Kral AH, Gee L, et al. Trends in HIV seroprevalence and risk among gay and bisexual men who inject drugs in San Francisco, 1988 to 2000. J Acquir Immune Defic Syndr
7. Singh BK, Koman JJ, Catan VM, et al. Sexual risk behavior among injection drug-using human immunodeficiency virus positive clients. Intl J Addictions
8. Kalichman SC, Belcher L, Cherry C, et al. Human immunodeficiency virus (HIV) seropositive homeless men: behavioral and social characteristics. J Soc Distress Homeless
9. van der Straten A, Gomez CA, Saul J, et al. Sexual risk behaviors among heterosexual HIV discordant couples in the era of post-exposure prevention and viral suppression therapy. AIDS
10. Rotheram-Borus MJ, Miller S. Secondary prevention for youths living with HIV. AIDS Care
11. Kelly JA, Hoffman RG, Rompa D, et al. Protease inhibitor combination therapies and perceptions of gay men regarding AIDS severity and the need to maintain safer sex. AIDS
12. Dilley JW, Woods WJ, McFarland W. Are advances in treatment changing views about high-risk sex?N Engl J Med
13. Ostrow DE, Fox KJ, Chmiel JS, et al. Attitudes towards highly active antiretroviral therapy are associated with sexual risk taking among HIV-infected and uninfected homosexual men. AIDS
14. Catz SL, Meredith KL, Mundy LM. Women’s HIV transmission risk perceptions and behaviors in the era of potent antiretroviral therapies. AIDS Educ Prev
15. Suarez TP, Kelly JA, Pinkerton SD, et al. The influence of a partner’s HIV serostatus and viral load on perceptions of sexual risk behavior in a community sample of gay and bisexual men. J Acquir Immune Defic Syndr
16. Remien RH, Wagner G. Carballo-Dieguez A, et al. Who may be engaging in high-risk sex due to medical treatment advances?AIDS
17. Kalichman SC, Rompa D, Cage M. Validity of self-reported cf4 lymphocyte count and viral load test results in people living with HIV-AIDS. Int J STD AIDS
18. McKinnon K, Cournos F, Meyer-Bahlburg HFL, et al. Reliability of sexual risk behavior interviews with psychiatric patients. Am J Psychiatry
19. Weatherby NL, Needle R, Cesari H, et al. Validity of self-reported drug use among injection drug users and crack cocaine users recruited through street outreach. Eval Prog Planning
20. Pinkerton SD, Abramson PR, Kalichman SC, et al. Secondary HIV prevention rates in a mixed-gender sample. Int J STD AIDS
21. Katz MH, Gerberding JL. The care of persons with recent sexual exposure to HIV. Ann Intern Med
22. Royce RA, Seña A, Cates W, Jr, et al. Sexual transmission of HIV. N Engl J Med
23. Davis KR, Weller SC. The effectiveness of condoms in reducing heterosexual transmission of HIV. Fam Plann Perspect
24. Pinkerton SD, Abramson PR. Effectiveness of condoms in preventing HIV transmission. Soc Sci Med
25. Porco TC, Martin JN, Page-Shafer KA, et al. HAART and the decline in HIV infectivity in San Francisco. AIDS Patient Care STDs
26. Holmberg SD. The estimated prevalence and incidence of HIV in 96 large US metropolitan areas. Am J Public Health
27. Centers for Disease Control and Prevention. HIV/AIDS surveillance report: midyear 2001 edition. MMWR
28. Johnson MO, Catz SL, Remien RH, et al. Decline in HIV infectivity following the introduction of highly active antiretroviral therapy:AIDS.
29. Blower SM, Aschenbach AN, Gershengorn HB, et al. Predicting the unpredictable: transmission of drug-resistant HIV. Nat Med
30. Hecht FM, Grant RM, Petropoulos CJ, et al. Sexual transmission of an HIV-1 variant resistant to multiple reverse-transcriptase and protease inhibitors. N Engl J Med
31. Jost S, Bernard MC, Kaiser L, et al. Brief report: a patient with HIV-1 superinfection. N Engl J Med
32. Bing EG, Burnam MA, Longshore D, et al. Psychiatric disorders and drug use among human immunodeficiency virus-infected adults in the United States. Arch Gen Psychiatry