Objectives: Concurrent partnerships accelerate dissemination of STIs. Most investigations of the features of concurrent partnerships have focused on higher risk subpopulations.
Goal: To assess condom use and the duration of concurrent sexual partnerships among men in the United States.
Study Design: Analysis of concurrent sexual partnerships among men in the 2002 National Survey of Family Growth. We classified pairs of concurrent partnerships into 3 types: transitional, contained, and experimental concurrency, and assessed the duration of overlap. We also report the distribution of condom use at the last sexual intercourse with neither, one or both concurrent partners of each pair and characteristics of men more likely to have used condoms with neither sex partner.
Results: The duration of overlap was <1 month in 32%, 1 to 3 months in 19%, and >12 months in 25% of concurrency pairs. Half (55%) of the pairs (whites, 64%; blacks, 41%) involved unprotected sex at the last sexual intercourse with at least 1 partner. The 35% of men who were more likely to use condoms with neither sex partner at the last sexual intercourse were older, white (48%), married/cohabitating (55%), and during the previous 12 months were incarcerated (49%), or used crack/cocaine (51%).
Conclusions: Although blacks generally experience higher rates of STIs and HIV, in this representative sample of men in the United States, blacks in concurrent partnerships seemed to use the only available protection (condoms) against infection (apart from abstinence) more than other racial/ethnic groups. Continued investigation of features of sexual partnership patterns is critical for curbing STI and HIV transmission.
This analysis of male respondents in the 2002 National Survey of Family Growth describes features of concurrent sexual partnerships including duration, patterns, and condom use.
From the *Department of Medicine, School of Medicine and †Department of Epidemiology, School of Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
We thank Anjani Chandra, Ph.D., for helpful consultations concerning the National Survey of Family Growth (Cycle 6).
Supported by grant from National Institute of Allergy and Infectious Diseases Ruth L. Kirschstein National Research Service Award 2 T32 AI007001-29 (to I.A.D) and National Institute of Child Health and Human Development 1R21HD054293-01A1 (to A.A.A).
Correspondence: Irene A. Doherty, PhD, 130 Mason Farm Rd. Suite 2118 CB 7030, Chapel Hill, North Carolina 27599-7030. E-mail:email@example.com
Received for publication, May 27, 2008, and accepted October 15, 2008.
STRUCTURAL CHARACTERISTICS OF SEXUAL NETWORKS including the timing of sexual partnerships influence the incidence rate of sexually transmitted infections (STIs) in a population. Concurrent sexual partnerships (partnerships that overlap in time) accelerate population dissemination of STIs, including HIV.1,2 A number of characteristics of concurrent partnerships that influence their effect on transmission rates have been described quantitatively3–5 and qualitatively.6,7 Multiple concurrent partnerships and frequent sexual contact8,9 over extended periods expedite transmission, whereas consistent condom use with all partners impedes transmission.
Men are more likely than women to have concurrent partnerships,10–14 but most investigations of the features of concurrent partnerships among men have focused on higher risk subpopulations.3,15,16
In this article, we extend our previous analysis of concurrent sexual partnerships among male respondents in the 2002 National Survey of Family Growth (NSFG), a nationally representative survey of men from the general population in the United States.11 We estimated that approximately 11% of men in the United States had concurrent sex partners during the 12 months preceding the interview and estimated crude and adjusted prevalence odds ratios of factors associated with concurrency. This report describes the types and duration of concurrent partnerships and condom use in those partnerships among the 430 male respondents with one or more concurrent partnerships. We also identify correlates of using condoms with neither concurrent partner.
MATERIALS AND METHODS
The National Center for Health Statistics conducted the sixth cycle of the NSFG during 2001 to 2002 to assess marriage, cohabitation, sexual activity, contraception, and childbearing trends.17 The survey comprises a multistage household probability sample, representative of men and women aged 15 to 44 in the United States. Non-Hispanic blacks, Hispanic/Latinos, and adolescents were oversampled. The face-to-face portion of the survey was administered in respondents’ homes by trained female interviewers using computer-assisted personal interviewing technology. Sensitive questions were self-administered via audio computer-assisted selfinterviewing software (ACASI) to increase reporting of proscribed behaviors. The survey includes 4928 male respondents.
Definition of Concurrency
For each of up to 4 female sex partners, respondents reported the month and year of the first and last sexual intercourse and whether they used condoms during the most recent sexual intercourse. We identified concurrent partnerships as partnerships from the 12 months preceding the interview where the first sexual intercourse with one partner occurred in a calendar month preceding the calendar month of last intercourse with an earlier partner.10,11
Features of Concurrency
We classified pairs of concurrent partnerships into 3 types (Fig. 1). In transitional concurrency (described qualitatively by Gorbach et al.),6 the second partnership begins before an earlier one ends and continues after the first relationship ends. In contained concurrency, a partnership lasts for at least 1 month, beginning and ending during the course of another partnership. Experimental concurrency is defined as overlapping concurrent partnership pairs in which the dates of first and last (or only) sex with 1 partner occurred within the same month during the course of a second partnership (our definition of experimental concurrency is more restrictive than the 1 used by Gorbach et al.6).
Our method of calculating the duration of concurrency depended upon its type (Fig. 1). Because respondents reported only the month and year in which each sexual partnership began and ended, the durations are expressed in month-long increments. The duration of a transitional concurrent partnership pair was the period from the month in which the later partnership began to the month in which the earlier partnership ended. The duration of a contained concurrent partnership pair was the duration of the contained/shorter partnership. The duration of experimental concurrency was zero months. Based on the distribution, durations were categorized as follows: <1-month, 3-month intervals for durations between 1 to 18 months, 19 to 24 months, 25 to 36 months, and ≥37 months. For each partnership pair we coded condom use at last sexual intercourse as use with both partners, 1 partner, or neither partner.
Demographic variables including age, race/ethnicity, and marital status were collected during the face-to-face portion of the interview. The age categories presented here match those from our previous analysis of this dataset.11 Responses for the following variables (all with respect to the 12 months before the interview) were obtained during the ACASI portion of the interview: binge drinking (consumption of ≥5 alcoholic beverages in 1 day, at least once a month), marijuana use at least once a month, any use of crack or cocaine, being intoxicated on alcohol or drugs at least once during sex, reporting that at least 1 female sex partner had other partners during the partnership with the respondent, incarceration, and treatment for an STI. ACASI variables were also used for ever being diagnosed with herpes simplex virus, human papillomavirus, or syphilis.
We conducted all analyses using Stata version 9.0 (College Station, TX). Analyses were weighted with the FINALWGT variable included in the NSFG dataset and accounted for the sampling design.17 Some analyses were carried out at the respondent level; these analyses include only the subset of respondents identified in our previous analysis11 as having had at least 1 pair of concurrent partnerships. The present analysis excluded 12 men who reported race/ethnicities other than white, non-Hispanic black, or Hispanic. Respondents with 3 or 4 sex partners could have more than 1 pair of concurrent partnerships. We assessed each concurrent pair for each respondent to identify men who had at least 1 concurrent pair in which he used condoms at the last intercourse with neither partner. We then used generalized linear models to identify correlates of this behavior. Because lack of condom use was not a rare outcome (prevalence >10%), odds ratios (e.g., from logistic regression) cannot be interpreted as estimates of risk ratios.18,19 Therefore, the model was specified with a Poisson distribution and the log link function [weighted by the probability weight provided in the NSFG dataset (FINALWGT)]. We examined risk characteristics and behaviors that were either correlated with having concurrent partnerships in our previous analyses11 or hypothesized a priori as related to unprotected sex or STIs. All factors associated with using condoms with neither partner in bivariate analyses were initially included in the multivariable model. We checked for near colinearity. We also assessed if race/ethnicity modified the effect of incarceration on condom use. Blacks have disproportionate incarceration rates20 and in the full sample, the prevalence of recent incarceration among blacks (12.5%) was more than twice that among whites (5.0%) (Table 1, ref11.) Because other studies of condom use in the context of concurrent sex partners have been conducted among adolescents and young adults (e.g.,3,15,16), we performed stratified analyses by age groups 15 to 17, 18 to 30, and 31 to 44 years.
Other analyses were carried out with the pair of concurrent partnerships as the unit of analysis. We conducted weighted tabular analysis to describe the types and duration of concurrent partnership pairs and condom use patterns within these pairs.
A total of 430 respondents had 668 concurrent partnership pairs. Most of these respondents (71%) had 1 pair of concurrent partnerships; 15% had 2; and 14% had 3 pairs of concurrent partnerships (Table 1). As noted in our previous analysis,11 the demographic and behavioral characteristics of the male respondents with concurrent partnerships differ significantly from the complete sample of respondents (n = 4654, excluding respondents who were not white, black, or Hispanic) Of the 430 men who had concurrent partnerships, 52%, 27%, and 21% were white, black, and Hispanic, respectively. The majority (83%) were neither married nor cohabitating as compared with 48% of the complete sample. More than half (54%) believed that one or more of their female sex partners was not monogamous during the previous 12 months, compared with 11% of the full sample. In the previous 12 months, 66% had been intoxicated on alcohol or drugs during sexual intercourse, 46% engaged in binge alcohol consumption (defined as consuming ≥5 alcoholic beverages in 1 day) at least once a month, 28% smoked marijuana at least once a month, and 20% had used crack or cocaine.
Features of Concurrent Partnerships
About one-fourth (23%) of the concurrent partnership pairs were transitional, 46% were contained, and 31% were experimental (Table 2). Half of the concurrency pairs (51%) (including experimental pairs) overlapped for a period of 3 months or less. Respondents used condoms during the last sexual intercourse with both partners in 45% of concurrency pairs, with only 1 partner in 27%, and with neither partner in 28% (Table 1).
In 43%, 44%, and 48% of experimental, contained, and transitional partnership pairs, respectively, condoms were used during the last sexual intercourse with both partners (Fig. 2A). In contrast, condom use during the last sexual intercourse with neither partner was less likely for experimental partnership pairs (19%) than contained (32%) and transitional (30%) pairs (P value < 0.04, not shown). Condom use with neither partner was more prevalent in long-term concurrent partnership pairs (Fig. 2B). For example, condoms were used with neither partner in 28% of the partnership pairs that overlapped for 1 to 3 months, compared with 55% that overlapped for 25 to 36 months. Use of condoms with both partners was also less prevalent in long overlapping partnership pairs (Fig. 2B). For example, 64% of concurrent partnership pairs of 10 to 12 months duration involved condom use with both partners, compared with 30% of those that overlapped for 25 to 36 months.
Correlates of Respondents Who Used Condoms With Neither Concurrent Partner
About one-third (35%) of respondents had a pair of concurrent partnerships in which the respondent used a condom with neither partner during the last sexual intercourse (Table 3). Older men, including 50% of men aged 40 to 44 years, were more likely to use condoms at last sex with neither partner. The prevalence ratio of using condoms with neither partner was 1.19 [95% confidence interval (CI) (1.08, 1.32)] for each 5-year increase in age. For example, condom use with neither partner was 1.19 times as likely among 30 year-olds as 25 year-olds. Compared with white men (48%), black (17%)[PR = 0.35 (0.21, 0.59)] and Hispanic (24%) [PR = 0.50 (0.31, 0.82)] men were significantly less likely to use condoms with neither partner. Condom use with neither partner was more likely among men who were married or cohabitating [PR 1.81 (1.21, 2.70)], had been incarcerated during the previous 12 months [PR 1.58 (1.07,2.32)], or used crack or cocaine [PR 1.66 (1.13, 2.46)]. Condom use was not associated with drug or alcohol intoxication during sexual intercourse, frequent binge alcohol consumption, frequent marijuana use, or the respondent’s belief that at least 1 of his female sex partners had other partners. Having a history of STI diagnosis and receiving treatment for STIs in the past 12 months were not associated with condom use, which may be due in part to the low prevalence of these factors.
In the multivariable model, the 95% CI for the associations between condom use with neither partner and crack or cocaine use [PR 1.46, (0.96, 2.22)] fell slightly below statistical significance, but the crude associations persisted for age, race/ethnicity, marital status, and recent incarceration.
There was no association between race/ethnicity and incarceration among men with concurrent partnerships who used condoms with neither partner. The interaction terms involving these 2 variables were not significant in the multivariable model and the prevalence of incarceration in the past 12 months was 18% for all 3 racial/ethnic groups (data not shown). Stratified analyses by age groups did not substantially deviate from the results using the full sample (not shown).
Concurrent Partnerships, Condom Use, and Race/Ethnicity
Because the prevalence of condom use with neither partner at the last sexual intercourse was substantially higher among white respondents, we examined the distribution of condom use patterns in concurrent partnership pairs stratified by respondents’ race/ethnicity (Table 4). Condoms were used with neither partner in 43% of white men’s concurrency pairs, compared with 11% of black and 19% of Hispanic men’s concurrency pairs (P value <0.0001, not shown). Furthermore, the proportion of concurrent partnership pairs that involved condom use with both partners was markedly lower among white men’s partnerships (36%) than among those of black men (60%).
In this analysis, we extend our previous analysis that reported the prevalence and correlates of concurrent partnerships among 15 to 44 year old men in the 2002 National Survey of Family Growth,11 a representative sample of the US household population. In the present investigation, we analyzed features of concurrent partnerships that further elucidate how concurrent partnerships contribute to the spread of STIs. Brief sexual partnerships that occurred during another relationship (i.e., experimental concurrency) comprised 32% of the concurrent partnership pairs. About half of the concurrent partnership pairs overlapped for less than 3 months, whereas 25% overlapped for more than 12 months. More than half of the pairs (55%) involved unprotected sex with at least 1 partner at the last sexual intercourse, with notable differences by race/ethnicity; the greatest proportion of such reports occurred among whites (64%) and the lowest among blacks (41%). Condom use with neither partner during the last sexual intercourse occurred most often in concurrent partnership pairs that overlapped for more than 18 months.
Our analysis has several limitations. Partnership dates were limited to the month and year of first and last intercourse, so a partnership that “lasted for 12 months” could have consisted of just 2 episodes of sexual intercourse that were a year apart. Partnership questions, including the months of first and last intercourse, were restricted to the respondent’s wife/cohabitating partner and the last 3 other sex partners, so men with additional partners may have had concurrent partnerships that went undetected. The respondent’s belief that his female sex partner(s) may have had other partners during their relationship is a proxy measure susceptible to inaccurate reporting, as some individuals may be unaware of their partners’ behaviors.15 Because all partnership data used for these analyses, including condom use, were collected during the face-to-face portion of the interview, responses were subject to social desirability that would likely underestimate the occurrence of concurrency and unprotected sexual intercourse. Furthermore, questions about condom use at last sexual intercourse do not provide information on consistency of use during the partnership. Nonetheless, this measure of condom use is used widely because it increases accuracy of recall for people who have many partners.21
Compared with multiple sequential partnerships, concurrent partnerships play a pivotal role in accelerating the transmission of HIV and STIs through sexual networks.22 Epidemiologic studies document that having concurrent partnerships is associated with individual risk of acquiring infection and transmitting disease to others.23–27 Mathematical models demonstrate that the prevalence of concurrent partnerships is positively correlated with the incidence and prevalence of bacterial and viral infections.28–32 Furthermore, seemingly negligible increases in the average number of concurrent partners may have profound effects on the growth of sexual networks and consequent STI or HIV transmission.1,31
Other than complete abstention from sexual intercourse, consistent condom use with all sex partners confers the greatest protection against STIs.33 Our previous analysis of the complete sample of male NSFG respondents indicates that men who did not use condoms during the last sexual intercourse with any of their partners in the past 12 months were more than twice as likely to have concurrent partnerships as men who used condoms with all partners.11 We show here that in 55% of these concurrent partnership pairs, respondents did not use condoms at last sex with at least 1 partner. Men who had been incarcerated were more likely to use condoms with neither partner, which is consistent with other findings that inmates engage in risky sexual behaviors after release.34–36 Crack and cocaine use is associated with acquiring STIs, including HIV (e.g.,23,37–39). Men who used crack or cocaine within the previous 12 months were also more likely to use condoms with neither partner.
Although black respondents were 3 times as likely as white respondents to have had a concurrent partnership in the previous 12 months,11 they were considerably less likely [PR 0.38 (0.23, 0.63)] than whites to have unprotected sex with both concurrent partners during the most recent sexual intercourse. Eleven percent of black men’s concurrent partnership pairs involved condom use with neither partner, compared with 43% and 19% of white and Hispanic men’s partnership pairs, respectively. Even though report of condom use at the last sexual intercourse may not accurately reflect overall participation in protected sex,21 in this representative sample of men in the United States, black men seemed to use the only available protection against infection at least as often as the majority population. These results are consistent with findings from The Longitudinal Adolescent Health Study (Add Health),40 a representative cohort of over 19,000 US youth recruited from public schools in 1994 to 1995 and followed through 2003. As both adolescents (Wave 1) and young adults (Wave 3), the behavioral risk profiles for HIV and STI among blacks were similar to whites and sometimes less risky, yet their STI rates were higher.41,42 Likewise, the prevalence of HIV infection is substantially higher among black men who have sex with men than among white men who have sex with men and cannot be explained by differences in the frequency of high risk sexual behaviors.43,44 These findings underscore the need for prevention measures that supplement condom use promotion.
Concurrent sexual partnerships are an aspect of sexual networks with important effects on the potential for STI dissemination. Consistent condom use by persons who engage in concurrent partnerships could mitigate their transmission impact. Our analyses of a representative sample of US men indicate that a substantial proportion of concurrent partnership pairs are unprotected. Understanding the nature, variety, and dynamics of concurrent partnership patterns is critical for developing disease control measures to curb STI and HIV transmission.
1. Morris M, Goodreau S, Moody J. Sexual Networks, Concurrency, and STD/HIV. In: Holmes KK, Sparling PF, Piot P, et al., eds. Sexually Transmitted Diseases. New York: McGraw-Hill, 2007: chap 7.
2. Morris M, Kretzschmar M. Concurrent partnerships and the transmission dynamics in networks. Soc Networks 1995; 17:299–318.
3. Doherty IA, Minnis A, Auerswald CL, et al. Concurrent partnerships among adolescents in a Latino community: The mission district of San Francisco, California. Sex Transm Dis 2007; 34:437–443.
4. Kraut-Becher JR, Aral SO. Gap length: An important factor in sexually transmitted disease transmission. Sex Transm Dis 2003; 30:221–225.
5. Foxman B, Newman M, Percha B, et al. Measures of sexual partnerships: Lengths, gaps, overlaps, and sexually transmitted infection. Sex Transm Dis 2006; 33:209–214.
6. Gorbach PM, Stoner BP, Aral SO, et al. “It takes a village”: Understanding concurrent sexual partnerships in Seattle, Washington. Sex Transm Dis 2002; 29:453–462.
7. Adimora AA, Schoenbach VJ, Martinson FE, et al. Social context of sexual relationships among rural African Americans. Sex Transm Dis 2001; 28:69–76.
8. Britton T, Nordvik MK, Liljeros F. Modelling sexually transmitted infections: The effect of partnership activity and number of partners on R0. Theor Popul Biol 2007; 72:389–399.
9. Nordvik MK, Liljeros F. Number of sexual encounters involving intercourse and the transmission of sexually transmitted infections. Sex Transm Dis 2006; 33:342–349.
10. Adimora AA, Schoenbach VJ, Bonas DM, et al. Concurrent sexual partnerships among women in the United States. Epidemiology 2002; 13:320–327.
11. Adimora AA, Schoenbach VJ, Doherty IA. Concurrent sexual partnerships among men in the United States. Am J Public Health 2007; 97:2230–2237.
12. Atkins DC, Baucom DH, Jacobson NS. Understanding infidelity: Correlates in a national random sample. J Fam Psychol 2001; 15:735–749.
13. Treas J, Giesen D. Sexual infidelity among married and cohabiting Americans. J Marriage Fam 2000; 62:48–60.
14. Whisman MA, Gordon KC, Chatav Y. Predicting sexual infidelity in a population-based sample of married individuals. J Fam Psychol 2007; 21:320–324.
15. Drumright LN, Gorbach PM, Holmes KK. Do people really know their sex partners? Concurrency, knowledge of partner behavior, and sexually transmitted infections within partnerships. Sex Transm Dis 2004; 31:437–442.
16. Riehman KS, Wechsberg WM, Francis SA, et al. Discordance in monogamy beliefs, sexual concurrency, and condom use among young adult substance-involved couples: Implications for risk of sexually transmitted infections. Sex Transm Dis 2006; 33:677–682.
17. Lepkowski J, Mosher W, Davis K, et al. National Survey of Family Growth, Cycle 6:Sample design, weighting, imputation,and variance estimation. Vital Health Stat 2006; 2.
18. Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic Research Principles and Quantitative Methods. New York: Van Nostrand Reinhold, A Division of International Thomson Publishing Inc, 1982.
19. Greenland S. Model-based estimation of relative risks and other epidemiologic measures in studies of common outcomes and in case-control studies. Am J Epidemiol 2004; 160:301–305.
20. Pettit B, Western B. Mass imprisonment and the life course: Race and class inequality in US incarceration. Am Sociol Rev 2004; 69:151–169.
21. Graham CA, Crosby RA, Sanders SA, et al. Assessment of condom use in men and women. Annu Rev Sex Res 2005; 16:20–52.
22. Morris M. Concurrent partnerships and syphilis persistence: New thoughts on an old puzzle. Sex Transm Dis 2001; 28:504–507.
23. Adimora AA, Schoenbach VJ, Martinson FE, et al. Heterosexually transmitted HIV infection among African Americans in North Carolina. J Acquir Immun Defic Syndr 2006; 41:616–623.
24. Fenton KA, Korovessis C, Johnson AM, et al. Sexual behavior in Britain: Reported sexually transmitted infections and prevalent genital Chlamydia trachomatis infection. Lancet 2001; 358:1851–1854.
25. Rosenberg M, Gurvey J, Adler N, et al. Concurrent sex partners and risk for sexually transmitted diseases among adolescents. Sex Transm Dis 1999; 26:208–212.
26. Koumans EH, Farley TA, Gibson JJ, et al. Characteristics of persons with syphilis in areas of persisting syphilis in the United States: Sustained transmission associated with concurrent partnerships. Sex Transm Dis 2001; 28:497–503.
27. Potterat JJ, Zimmerman-Rogers H, Muth SQ, et al. Chlamydia transmission: Concurrency, reproduction number, and the epidemic trajectory. Am J Epidemiol 1999; 150:1331–1339.
28. Chick SE, Adams AL, Koopman JS. Analysis and simulation of a stochastic, discrete-individual model of STD transmission with partnership concurrency. Math Biosci 2000; 166:45–68.
29. Doherty IA, Shiboski S, Ellen JM, et al. Sexual bridging socially and over time: A simulation model exploring the relative effects of mixing and concurrency on viral sexually transmitted infection transmission. Sex Transm Dis 2006; 33:368–373.
30. Ghani A, Garnett G. Risks of acquiring and transmitting sexually transmitted diseases in sexual partner networks. Sex Transm Dis 2000; 27:579–587.
31. Morris M, Kretzschmar M. Concurrent partnerships and the spread of HIV. AIDS 1997; 11:641–648.
32. Watts CH, May RM. The influence of concurrent partnerships on the dynamics of HIV/AIDS. Math Biosci 1992; 108:89–104.
33. Holmes K, Levine R, Weaver M. Effectiveness of condoms in preventing sexually transmitted infections. Bull World Health Organ 2004; 82:454–461.
34. Grinstead OA, Faigeles B, Comfort M, et al. HIV, STD, and hepatitis risk to primary female partners of men being released from prison. Women Health 2005; 41:63–80.
35. MacGowan RJ, Margolis A, Gaiter J, et al. Predictors of risky sex of young men after release from prison. Int J STD AIDS 2003; 14:519–523.
36. Stephenson BL, Wohl DA, Golin CE, et al. Effect of release from prison and re-incarceration on the viral loads of HIV-infected individuals. Public Health Rep 2005; 120:84–88.
37. Seña AC, Muth SQ, Heffelfinger JD, et al. Factors and the sociosexual network associated with a syphilis outbreak in rural North Carolina. Sex Transm Dis 2007; 34:280–287.
38. DeHovitz JA, Kelly P, Feldman J, et al. Sexually transmitted diseases, sexual behavior, and cocaine use in inner-city women. Am J Epidemiol 1994; 140:1125–1134.
39. Ellerbrock TV, Chamblee S, Bush TJ, et al. Human immunodeficiency virus infection in a rural community in the United States. Am J Epidemiol 2004; 160:582–588.
40. Klein JD. The National Longitudinal Study on Adolescent Health. Preliminary results: Great expectations. JAMA 1997; 278:864–865.
41. Hallfors DD, Iritani BJ, Miller WC, et al. Sexual and drug behavior patterns and HIV and STD racial disparities: The need for new directions. Am J Public Health 2007; 97:125–132.
42. Halpern CT, Hallfors D, Bauer DJ, et al. Implications of racial and gender differences in patterns of adolescent risk behavior for HIV and other sexually transmitted diseases. Perspect Sex Reprod Health 2004; 36:239–247.
43. Millett GA, Peterson JL, Wolitski RJ, et al. Greater risk for HIV infection of black men who have sex with men: A critical literature review. Am J Public Health 2006; 96:1007–1019.
44. Crosby R, Holtgrave DR, Stall R, et al. Differences in HIV risk behaviors among black and white men who have sex with men. Sex Transm Dis 2007; 34:744–748.