Joa 23#3 2000

JAIDS Journal of Acquired Immune Deficiency Syndromes
23:272–278 2000 Lippincott Williams & Wilkins, Inc., Philadelphia
Correlates of Risky Behaviors Among Young and Older Men Having Sexual Relations With Men in Montre´al, *Annie Dufour, *†Michel Alary, ‡Joanne Otis, ‡Roger Noe¨l, §࿣Robert S. Remis, *Benoît Maˆsse, †Raymond Parent, ࿣Bruno Turmel, ¶Rene´ Lavoie, #Roger LeClerc, **Jean Vincelette, and the ´ pide´miologie de l’Universite´ Laval, Centre Hospitalier Affilie´ Universitaire de Que´bec, Que´bec; †Centre de Sante´ Publique de Que´bec, Beauport, Que´bec; ‡De´partement de Sexologie, Universite´ du Que´bec a` Montre´al, Montre´al, Que´bec; §Department of Public Health Sciences, University of Toronto, Toronto, Ontario; ࿣Direction de la Sante´ Publique de Montre´al-Centre, Montre´al, Que´bec; ¶Action Se´ro-Ze´ro, Montre´al, Que´bec; ¶COCQ-Sida, Montre´al, Que´bec; and **Service de Microbiologie, Centre Hospitalier de l’Universite´ de Montre´al, Campus Saint-Luc, Montre´al, Que´bec, Canada Objective: To assess characteristics associated with 1) age and 2) recent unpro-
tected anal sex with casual partners among men having sexual relations with men(MSM) participating in the Omega Cohort, Montre´al, Que´bec, Canada.
Methods: The Omega Cohort is a study of the incidence and psychosocial deter-
minants of HIV infection among MSM living in Montre´al. MSM complete a ques-tionnaire and are tested for HIV every 6 months.
Results: Thirteen percent of young MSM (<30 years of age, n ס 355) and 12% of
older MSM (Ն30 years of age, n ס 455) reported recent unprotected anal sex withcasual partners. The predictors of this latter behavior were: not living with a malesexual partner, unprotected anal sex with regular partner, >5 casual partners, alcohol/drug use before anal sex, and having difficulties with procedures needed for safe sex.
Among young MSM, additional predictors were: to have been living in Montre´al forless than 1 year and to have exchanged money for sex. Among older MSM, additionalpredictors were: female sexual partners, unprotected anal sex with an HIV-infectedpartner, and feeling invulnerable to AIDS.
Conclusion: Young Omega participants do not have more risky behaviors than
older participants. Some predictors of recent risk behaviors with casual partners weredifferent between the two groups. Prevention programs should be adapted conse-quently.
Key Words: MSM—Risk behaviors—Unprotected anal sex—HIV
In the early 1980s, several cohort studies have as- cross-sectional studies have documented high levels of sessed HIV incidence and associated factors among men risky behaviors among young MSM (4–7), several cohort having sexual relations with men (MSM) (1–3). How- studies recruiting young MSM were implemented at the ever, given that most men in these studies were recruited more than 10 years ago, we have limited information on In the United States, cohort studies among young younger MSM. To resolve this problem and also because MSM have observed an HIV prevalence of 18% and 9%in San Francisco (8,9), 9% in New York City (HIV in-cidence was 2/100 person-years [p-y]) (10), and 2% in Address correspondence and reprint requests to Michel Alary, Boston (12). According to the Vanguard Project imple- ´ pide´miologie, CHA, Pavillon St-Sacrement, mented in Vancouver, British Columbia, Canada in 1995 1050 Chemin Ste-Foy, Que´bec, Que´bec, G1S 4L8 Canada.
Manuscript received August 15, 1999; accepted December 10, 1999.
(11), HIV incidence was 2/100 p-y at the end of 1997.
CORRELATES OF RISK BEHAVIORS AMONG MEN WHO HAVE SEX WITH MEN Moreover, 4% of young MSM from a cohort study in Amsterdam in the Netherlands were HIV positive at en- The age-based dichotmization at 30 years is interesting because it creates two separate groups according to the pre- and post-AIDS pe- In the province of Que´bec, Canada, little is known riods. The MSM <30 years old began their sexual life, generally, after about the HIV epidemic among MSM. The only specific AIDS had become a major health problem. MSM or 30 years or olderbegan their sexual life before AIDS and were obliged to adjust their study among MSM in Montre´al was carried out in 1989 sexual lifestyle to this new reality. However, the analyses were also (14). The Omega cohort was implemented in 1996 to performed using age as a continuous variable and a similar final re- assess HIV incidence and associated determinants gression model such as when using the two age groups was obtained.
In this study, a casual partner was defined as someone with whom The objectives of this study were to assess risk behav- the participant had sex only once (a “one-night stand”), someone he didnot intend to see again; if he had a subsequent sexual encounter with iors associated with age (<30 versus Ն30 years old) him, it was by chance. The variable “unprotected anal sex with casual among MSM participating in the Omega Cohort in Mon- partners” had a value of zero when the participant avoided unprotected tre´al and to assess demographic, sexual, and psychoso- anal sex with casual partners, avoided anal sex, or did not report any cial factors associated with the practice of unprotected casual partner. It had a value of one when condoms were not used anal sex with casual partners during the previous 6 systematically during anal sex with casual partners. The analyses werealso performed using only MSM who avoided unprotected anal sex with casual partners as the comparison group and essentially the samefinal logistic regression model was obtained. Moreover, because the predictors of unprotected receptive and insertive anal sex were similar,these two sexual behaviors were considered together in the analyses.
Background
The Omega Cohort project is an ongoing prospective study of HIV incidence and associated determinants among MSM living in the Mon- Psychosocial variables were developed using several different mod- tre´al metropolitan area. Recruitment started in October 1996 and was els such as the AIDS Risk Reduction Model (15), the Protection Mo- carried out through the following strategies: a bilingual publicity cam- tivation Theory (16), and the Theory of Planned Behavior (17,18).
paign in print general and gay media; collaboration with gay commu- Health concerns: Participants responded to statements related to nity organizations and medical clinics where the study was discussed some health concerns using a 5-point Likert scale (1 ס not at all–5 ס with eligible men; individual contacts with eligible men during gay very often). After factor analysis was performed, four different factors pride events; and use of posters and leaflets placed at strategic sites in measuring health concerns were constructed: “surrounding oneself with gay venues and clinics. The study population consists of MSM aged 16 friends,” “taking care of one’s health,” “find ways to reduce stress,” years and older who have had sex with another man at least once in the preceding year and are HIV negative or do not know their serostatus at Social concerns: Participants responded to statements related to baseline. Interviews are mainly carried out at a community organization some social concerns using a 5-point Likert scale (1 ס not at all–5 ס called Centre des Gais et Lesbiennes de Montre´al (Montre´al Gay and a lot). Four factors were constructed: “loneliness”; “job, study and Lesbian Center) but also at one of three private medical clinics and a money problems, concerns about future and difficulty finding support when necessary”; “heavy responsibilities”; and “difficulties in accept-ing one’s sexual orientation”. For each factor, a low score corresponds Data Collection
to a high social concern whereas a high score corresponds to a lowsocial concern.
At the first interview, after informed written consent, participants Attitudes toward AIDS: Participants responded to statements related complete the questionnaire and blood is collected t test for HIV, hepa- to their attitudes toward AIDS using a 5-point Likert scale (1 ס titis B, and syphilis. Three weeks after the first appointment, partici- never–5 ס very often). Five factors were constructed: “anxiety toward pants return to receive their test results and posttest counseling. Par- AIDS”; “feeling invulnerable to AIDS”; “avoiding thinking about ticipants who test positive fr HIV are excluded from the study and AIDS”; “taking AIDS as a banality” and “underestimation of the im- referred to appropriate services. Participants return every 6 months to complete a follow-up questionnaire and to be retested for HIV and Attitudes toward condom use during anal sex: Participants re- sponded to statements related to their attitudes toward condom useduring anal sex using a 5-point Likert scale (1 ס not at all–5 ס verymuch). Three factors were constructed: “sense of security,” “feeling of Statistical Analyses
trust between partners,” and “interference with eroticism.” For each The prevalence odds ratio (POR) was used as the measure of asso- factor, a low score corresponds to a negative attitude toward condom ciation. In addition, ␹2 for trend and Fisher’s exact test were used for use whereas a high score corresponds to a positive attitude toward categoric variables and the Mann-Whitney U-test for continuous vari- ables. We performed logistic regression analysis to identify variables Perceived behavioral control toward condom use during anal sex: independently associated with age and with the practice of unprotected Participants responded to statements related to their perceived behav- anal sex with casual partners. We calculated an adjusted R with 95% ioral control toward condom use during anal sex using a 5-point Likert likelihood ratio–based confidence intervals (CI) for variables that re- scale (1 ס not difficult at all–5 ס very difficult). Five factors were mained in the final models. All analyses were performed using SAS, constructed: “difficulties in handling safe sexual encounter”; “difficul- version 6.12 (SAS Institute, Cary, NC, U.S.A.).
ties in getting condoms”; “difficulties in using a condom when very JAIDS Journal of Acquired Immune Deficiency Syndromes, Vol. 23, No. 3, March 1, 2000 aroused sexually or the partner appeared safe”; “difficulties in using a between young MSM and older MSM with regard to condom when they did not want to disappoint the partner or were afraid other sexual behaviors. Table 1 presents data on safe sex of him”; and “difficulties in using a condom when they were under the separately by type of partner, type of anal sex, and age influence of alcohol/drugs”. For each factor, a low score corresponds toa high perceived behavioral control toward condom use during anal sex whereas a high score corresponds to a low perceived behavioral control Among MSM who reported only one casual partner, 5% (n ס 2 of 37) of young MSM and 4% (n ס 2 of 46)of older MSM reported unprotected anal sex during the previous 6 months (p ס 1.00). Among MSM who re-ported between 2 and 5 casual partners, these proportions Sociodemographic Characteristics
were 10% (n ס 11 of 111) and 6% (n ס 7 of 113),respectively (p ס .34). Among MSM who reported be- As of the end of 1997, 810 MSM with a mean age of tween 6 and 19 casual partners, these proportions were 33 years (median ס 32; range 16–73) completed their 28% (n ס 19 of 69) and 21% (n ס 22 of 104), respec- baseline interview in the Omega Cohort. Most (79%) tively (p ס 0.37) whereas for >19 casual partners, it was were born in the province of Que´bec, were living on the 40% (n ס 14 of 35) and 24% (n ס 22 of 90), respec- island of Montre´al (90%), were single (74%) and well tively (p ס .12). The proportion of MSM who reported educated (44% attended college). Half the participants unprotected anal sex increased as the number of casual reported a yearly income of less than $CDN 20,000 partners increased (p ס .036, ␹2 for trend). Table 2 (mean of $CDN 24,000 compared with $CDN 30,000 shows the variables that remained associated with age in among the general population of men in Montre´al, ac- cording to census data [19]), 45% worked full-time at thetime of interview, 20% studied full time, and 17% re- Factors Associated With the Practice of
ceived welfare or unemployment insurance benefits.
Unprotected Anal Sex With Casual Partners During
the Previous 6 Months
Risky Behaviors Associated With Age
Overall, 46 (13%) MSM <30 years of age reported In the univariate analysis comparing young MSM unprotected anal sex with casual partners during the pre- (<30 years, n ס 355) with older MSM (Ն30 years, n ס vious 6 months whereas 54 of those aged 30 years or 455) with regard to risk behaviors, fewer young MSM more did so (12%). In a logistic regression analysis in- reported sex in bathhouses (34% versus 52%; p Յ .001) cluding the two age groups and in which the interactions and having had more than 5 casual partners during the with age were tested systematically (Table 3), variables previous 6 months (30% versus 44%; p Յ .001) whereas that remained associated with the practice of unprotected a greater proportion reported anal sex with their regular anal sex with casual partners among 810 Omega partici- partners during the same period (68% versus 54%; p Յ pants were: not living with a male sexual partner, had .001). There was no statistically significant difference unprotected anal sex with regular partners during one’s TABLE 1. Safe sex in the previous 6 months among 810 men who have sex with men (MSM) participating in the
a Includes MSM who avoided anal sex or did not have sexual partners during the previous six months.
b Mann Whitney U-test comparing safe sex between young and older MSM, for each type of sexual partners (regular or casual) and each type of anal sex (insertive or receptive).
c Numbers do not total 355 because of missing values.
d Numbers do not total 455 because of missing values.
JAIDS Journal of Acquired Immune Deficiency Syndromes, Vol. 23, No. 3, March 1, 2000 CORRELATES OF RISK BEHAVIORS AMONG MEN WHO HAVE SEX WITH MEN TABLE 2. Prevalence odds ratio (POR) and 95% confidence interval (CI) of variables associated with
young age (<30 years) among 810 men who have sex with men (MSM) participating in the Omega Cohort * Adjusted for other variables listed in the table.
lifetime, had >5 casual partners during the previous 6 protected anal sex with regular partners during the pre- months, used alcohol or drugs before anal sex with ca- vious 6 months, sex with female partners during the pre- sual partners, and had lower perceived behavioral control vious year, unprotected anal sex with an HIV-infected with regard to one’s capacity to handle safe sexual en- partner during one’s lifetime, and feeling invulnerable to counter. Among young MSM, additional variables asso- ciated with the practice of unprotected anal sex with It is interesting to note that the other variables related casual partners were: not being a practicing Roman to perceived behavioral control toward condom use dur- Catholic, to have been living in Montre´al for Յ1 year, ing anal sex (in men of all ages, difficulties in using a and to have ever given money for sex. Among older condom: “when very aroused sexually or the partner ap- MSM, additional variables associated with the practice peared safe,” “when they did not want to disappoint the of unprotected anal sex with casual partners were: un- partner or were afraid of him,” “when they were under TABLE 3. Prevalence odds ratio (POR) and 95% confidence interval (CI) of variables associated (with
multivariate analysis) with the practice of unprotected anal sex with casual partners during the previous six months among 810 men who have sex with men (MSM) participating in the Omega Cohort a Adjusted for other variables listed in the table.
NA, not applicable because there is a significant interaction between the variable and age.
JAIDS Journal of Acquired Immune Deficiency Syndromes, Vol. 23, No. 3, March 1, 2000 the influence of alcohol/drugs”; in younger men: “diffi- behaviors. This may be explained by the fact that MSM culties in getting condoms”) were associated to unpro- who prefer risky practices also seek many partners (20) tected anal sex with casual partners in the univariate or because as the number of episodes of sex increases, analysis (data not shown). However, they were not sig- the likelihood of having unprotected sex also increases.
nificant in the multivariate model because they were As reported elsewhere (4,5,20,25), frequent alcohol or highly linked together and with the variable “difficulties drug use before anal sex with casual partners (to be under in handling safe sexual encounter.” In addition, among the influence of alcohol or drugs during half or more of men aged <30 years, very young age was associated with anal intercourse) was strongly predictive of unprotected risky behaviors. Indeed, among a small group of 32 men anal sex. Whether this association between alcohol or aged <20 years, 8 (25%) reported the practice of unpro- drug intake is causal, this behavior could be used as a tected anal sex with casual partners compared with 12% marker for high-risk sexual activity to target MSM who (38 of 316) among those aged 20 to 29 (p ס .05, Fisher’s are more likely to practice unprotected anal sex (9,27).
Prevention strategies should be reinforced in locationswhere alcohol is sold, such as bars, clubs, and other DISCUSSION
public venues. The link between alcohol use and unsafe Overall, risk behaviors of Omega participants during sex should also be made more explicitly in prevention the previous 6 months did not differ according to age.
This is somewhat surprising because most North Ameri- MSM who reported risky sex with casual partners also can studies have reported that risk behaviors such as reported more frequent difficulties in using safe sex. This unprotected anal sex were associated with younger age is related to a low perceived behavioral control toward condom use. These MSM may feel emotional discomfort MSM who were living with a male sexual partner at using condoms, communicating openly about sex or be the time of interview were less likely to have had unpro- less confident about the negotiation of safe sex (7,9,28).
tected anal sex with casual partners. Some of these MSM Prevention strategies should aim to help MSM in Mon- may be monogamous or may have decided not to have tre´al to improve the communication skills necessary for risky sex outside their couple relationship to avoid con- dom use with their regular partner. Many Omega partici- Among young MSM, having lived in Montre´al for 1 pants reported that they and their regular partners are year or less was a determinant of unprotected anal sex getting tested for HIV before stopping condom use.
with casual partners. MSM who just arrived in Montre´al However, it has been reported that some MSM who are may be less aware of the prevention messages. Some of practicing unprotected anal sex only with their regular these young men may be willing to have new sexual partners have serial or concomitant regular partners (22) experiences without regard to the risks of HIV infection.
and/or do not know their own serostatus or that of their This situation is also reflected by the fact that very young MSM (<20 years) also reported more risky behaviors.
MSM who had ever had unprotected anal sex with Organizations involved in HIV prevention should in- regular partners were much more likely to report unpro- crease prevention messages targeting young MSM living tected anal sex with casual partners than MSM who outside urban centers, such as in suburbs where there are avoided this behavior. Many previous studies have found fewer resources for MSM. Peer counselling strategies, such an association between past risk behaviors and pre- such as Montre´al’s “Project 10,” should be implemented sent risk behaviors and with relapse of risk behaviors in bars or in schools where a gay leader who endorses (20–22,25). However, by targeting only MSM who re- and recommends safe sex could serve as a role model and port current unsafe behaviors, we may miss many men at offers support to young MSM (27,29–31).
risk for future unsafe sex (22). Organizations involved in Young MSM who had paid for sex were also more HIV prevention should target MSM at the beginning of likely to have had unprotected anal sex with casual part- their sexual lives, that is, before they acquire risky habits.
ners than MSM who avoided this behavior. These men Prevention strategies should be implemented in schools, may represent a group of MSM who take risks in general particularly those located outside downtown Montre´al, to and thus are willing to have unprotected anal sex with increase the awareness of youth about safe sex guidelines Among older men, unprotected anal sex with casual As reported elsewhere (26), MSM who reported un- partners was also associated with unprotected anal sex protected anal sex with casual partners also reported with an HIV-positive partner. These MSM may take more casual partners than MSM who avoided such risk risks in general or feel somewhat invulnerable to HIV JAIDS Journal of Acquired Immune Deficiency Syndromes, Vol. 23, No. 3, March 1, 2000 CORRELATES OF RISK BEHAVIORS AMONG MEN WHO HAVE SEX WITH MEN because they did not become infected in spite of previous APPENDIX
behaviors. This is supported by the fact that feeling in- The Omega Study Group is composed of Michel Bouchard, Michel vulnerable to AIDS was also a predictor of unprotected Chaˆteauvert, Andre´ Desmarais, Fre´de´ric Doutrelepont, Micheline Du- anal sex with casual partners among older Omega par- puis, Yves Jalbert, Pierre Junod, Ralf Ju¨rgens, Roger Leblanc, E ticipants. These men may also incorrectly underestimate Lefebvre, Rene´ Le´gare´, Catherine Lowndes, Eva Nonn, Sylvie Savard, their personal vulnerability to AIDS (21). Some MSM Robert Steinman, Re´jean Thomas, and Anne Vassal.
could correctly identify practices risky for HIV transmis- Acknowledgments: This research was supported by the Na-
sion but are unable to interpret the implications of their tional Health Research and Development Program (NHRDP), own risky sexual behaviors (27). It could be an indication Health Canada (grant no. R6605-4639-AIDS). Additional fund- that these MSM have not reached the first step required ing was also provided by the Centre Que´be´cois de Coordination for the adoption of safe sexual practices, which is the sur le Sida and through the AIDS Research Network of theFonds de la Recherche en Sante´ du Que´bec (grant no.
recognition of one’s sexual behaviors as being at risk 960071.06). Dufour was recipient of a training award from (15). Among Omega participants, at risk MSM perceived NHRDP (grant no. 6605-4857-47A). Alary is a research wrongly their sexual activities as relatively safe.
scholar of the Fonds de la Recherche en Sante´ du Que´bec (grant Among older men, to have had sex with a woman no. 970097). We would like to thank Omega participants, during the previous year was a predictor of unprotected Omega employees, collaborators, gay and HIV-related commu-nity organizations who have collaborated since the study’s in- anal sex with casual partners. Some of these men may ception and the Montre´al Regional Public Health Department have just have begun homosexual activity and have poor that provided hepatitis B vaccine without charge.
safe sex negotiation skills, fewer social norms, or lesssupport for safe sex (21). According to Fisher (32), “the REFERENCES
more comfortable gay men are with being gay, thegreater their involvement with gay social networks in 1. Kaslow RA, Ostrow DG, Detels R, Phair JP, Polk F, Rinaldo CR.
which AIDS-preventive behaviors and their practice of The Multicenter AIDS Cohort Study: rationale, organization, andselected characteristics of the participants. Am J Epidemiol 1987; safe sex.” Some MSM may also be bisexual and thus place their female partners at risk of HIV infection.
2. Winkelstein W, Lyman DM, Padian N, et al. Sexual practices and Omega participants may not be representative of the risk infection by the human immunodeficiency virus: the San Fran-cisco Men’s Health Study. JAMA 1987;257:321–5.
gay community of Montre´al because participation was 3. Schechter MT, Boyko WJ, Jeffries E, Willoughby B, Nitz R, Con- voluntary and we did not attempt to recruit a random stance P. The Vancouver Lymphadenopathy–AIDS Study: 1. Per-sistent generalized lymphadenopathy. Can Med J Assoc 1985;132: sample of Montre´al’s MSM. However, because we used multiple recruitment strategies, the participants are di- 4. Hogg RS, Craib KJP, Willoughby B, Sestak P, Montaner JSG, versified. Despite this limitation, our results give several Schechter MT. Sociodemographic correlates for risk-taking behav-iour among HIV seronegative homosexual men. Can J Public insights for public health authorities to better target at 5. Valdiserri RO, Lyter D, Leviton LC, Callahan CM, Kinsgley LA, In summary, 13% of young MSM and 12% of older Rinaldo CR. Variables influencing condom use in a cohort of gayand bisexual men. Am J Public Health 1988;78:801–5.
MSM reported unprotected anal sex with casual partners.
6. Stall R, Barrett D, Bye L, et al. A comparison of younger and older Some variables that predict the practice of unprotected gay men’s HIV risk-taking behaviors: the Communication Tech- anal sex with casual partners such as unprotected anal nologies 1989 Cross-Sectional Survey. J Acquir Immune DeficSyndr 1992;5:682–7.
sex with regular partners, a high number of casual part- 7. Hays RB, Kegeles SM, Coates TJ. High HIV risk-taking among ners, frequent alcohol/drug use before sex, and having young gay men. AIDS 1990;4:901–7.
difficulties in handling safe sexual encounters are similar 8. Osmond DH, Page K, Wiley J, et al. HIV infection in homosexual and bisexual men 18 to 29 years of age: the San Francisco Young between the two age groups. However, two different pat- Men’s Health Study. Am J Public Health 1994;84:1933–7.
terns of at risk MSM emerge according to age groups.
9. Lemp GF, Hirozawa AM, Givertz D, et al. Seroprevalence of HIV Among young MSM, being <20 years old, to have been and risk behaviors among young homosexual and bisexual men:the San Francisco/Berkeley Young Men’s Survey. JAMA 1994; living in Montre´al for Յ1 year, and to ever have ex- changed money for sex are predictors of risk behaviors.
10. Dean L, Meyer I. HIV prevalence and sexual BEHAVIORS in a Among older MSM, bisexual practices, unprotected anal cohort of New York City gay men (aged 18–24). J Acquir ImmuneDefic Syndr Human Retrovirol 1995;8:208–11.
sex with an HIV-infected partner, and feeling invulner- 11. Martindale SL, Strathdee S, Miller ML, et al. Unsafe sex and able to AIDS are specific predictors. Montre´al organiza- elevated HIV incidence among young men having sex with men.
tions implicated in HIV prevention should use these re- Can J Infect Dis 1998;9(Suppl A):32.
12. Seage G, Mayer K, Lenderking WR, et al. HIV and hepatitis B sults to better target MSM who place themselves at risk infection and risk behavior in young gay and bisexual men. Public Health Rep 1997;112:158–67.
JAIDS Journal of Acquired Immune Deficiency Syndromes, Vol. 23, No. 3, March 1, 2000 13. De Wit JBF. The epidemic of HIV among young homosexual men.
23. Mansergh G, Marks G. Age and risk of HIV infection in men who AIDS 1996;10(Suppl 3):S21–5.
have sex with men. AIDS 1998;12:1119–28.
14. Remis RS, Najjar M, Pass C, Paradis G. Seroepidemiologic study 24. Elford J, Bolding G, Maguire M, Sherr L. Sexual risk behaviour of HIV infection and sexual behaviour among men attending a among gay men in a relationship. AIDS 1999;13:1407–11.
medical clinic in Montreal. V International Conference on AIDS, 25. Remafedi G. Predictors of unprotected intercourse among gay and bisexual youth: knowledge, beliefs, and behavior. Pediatrics 1994; 15. Catania J, Kegeles SM, Coates TJ. Toward an understanding of risk behavior: an AIDS risk reduction model. Health Educ Q 1990;17:53–72.
26. Kerr-Pontes LRS, Gondim R, Mota RS, Martins TA, Wypij D.
16. Van der Velde FW, Van der Pligt J. AIDS-related health behavior: Self-reported behaviour and HIV risk taking among men who have coping, protection motivation, and previous behavior. J Behav Med sex with men in Fortaleza, Brazil. AIDS 1999;13:709–17.
27. Siegel K, Mesagno FP, Chen JY, Christ G. Factors distinguishing 17. Godin G, Myers T, Lambert J, Calzavara LM, Locker D, and the homosexual males practicing risky and safer sex. Soc Sci Med Canadian AIDS Society. Understanding the intention of gay men and bisexual men to use condoms for insertive or receptive anal 28. Gold RS, Skinner MJ. Situational factors and thought processes sex. Can J Infect Dis 1993;4(Suppl):43B.
associated with unprotected intercourse in young gay men. AIDS 18. Godin G, Myers T, Lambert J, Calzavara LM, Locker D, and the Canadian AIDS Society. Sexual satisfaction, sexual risk-taking 29. Zimmerman MA, Ramirez-Valles J, Suarez E, de la Rosa G, Castro and behaviour change among gay and bisexual men. IX Interna- MA. An HIV/AIDS prevention project for Mexican homosexual tional Conference on AIDS, Berlin, Germany, 1993.
men: an empowerment approach. Health Educ Behav 1997;24: 19. Statistics Canada’s internet site, 1996 census. Statistical profile: income and work statistics for Montre´al (census metropolitan 30. Kelly JA, Murphy DA, Sikkema KJ, et al. Randomised, controlled, area), Quebec. Accessed November 17, 1999. Available at: www.
community-level HIV-prevention intervention for sexual-risk be- haviour among homosexual men in US cities. Lancet 1997;350: 20. Kelly JA, St-Lawrence JS, Brasfield TL. Predictors of vulnerabil- ity to AIDS risk behavior relapse. J Consult Clin Psychol 1991;59:163–6.
31. Chouinard N, Otis J, Pilote F, Ryan B. Le “Projet 10”: impact d’un 21. Kelly JA, Sikkema KJ, Winett RA, et al. Factors predicting con- groupe de soutien sur la vulne´rabilite´ des jeunes gais et bisexuels tinued high-risk behavior among gay men in small cities: psycho- a` l’infection au VIH. Satellite Symposium, VII Annual Canadian logical, behavioral, and demographic characteristics related to un- Conference for HIV Research, Que´bec City, Que´bec, Canada, safe sex. J Consult Clin Psychol 1995;63:101–7.
22. Ekstrand ML, Coates TJ. Maintenance of safer sexual behaviors 32. Fisher JD. Possible effects of reference group-based social influ- and predictors of risky sex: the San Francisco Men’s Health Study.
ence on AIDS-risk behavior and AIDS prevention. Am Psychol Am J Public Health 1990;80:973–7.
In retrospect, we believe that Stefania Paolucci and Fausto Baldanti, Servizio id Virologia, IRCCS Poli-clinico San Matteo, Pavia, Italy, should be listed as coauthors of the paper entitled “Efavirenz, Nelfiniavir,and Stavudine Rescue Combination Therapy in HIV-1–Positive Patients Heavily Pretreated With Nucleo-side Analogues and Protease Inhibitors,” published in the December 15, 1999 issue (volume 22, issue 5,pages 453–460). By developing a new sequencing assay and using samples which were stored in theirlaboratory, they made possible the drug resistance analysis described in the article. In addition, theyperformed this analysis using their own research grants (IRCCS Policlinico San Matteo, Ricerca Corrente1998, and Ministero della Sanita´, Istituto Superiore di Sanita´, Il Programma Nazionale AIDS, contract no.
30B.33). Thus, we ask that the article’s authorship now be listed as: Elena Seminar, Franco Maggiolo, PaolaVillani, Fredy Suter, Angelo Pan, Mario B. Regazzi, Stefania Paolucci, Fausto Baldanti, Carmine Tinelli,and Renato Maserati.
JAIDS Journal of Acquired Immune Deficiency Syndromes, Vol. 23, No. 3, March 1, 2000

Source: http://www.creces.uqam.ca/Page/Document/Dufour_Correlates_of_risky_behaviours.pdf