Safer Sex Knowledge Base

Table of Contents

• Definition of terms
• Assumptions
• Recommendations
• References

Definition of Terms

There is a low risk of infection when prospective, cohort-style studies have failed to demonstrate a statistically significant relationship between the behavior and infection, but case reports continue to suggest a correlation. There is a high risk of infection when prospective cohort-style study has established a relationship and the risk is deemed substantial by the Subcommittee.

Anilingus

oral-anal contact; rimming.

At risk for HIV infection

non-monogamous since 1979, or in sexual relationship with a partner who has been non-monogamous since 1979, and without persistently negative anti-HIV antibody tests for a period of at least six months.

Barrier Precautions

use of a latex condom and water-based lubricant for vaginal or rectal sex.

Brachio-anal/brachio-rectal contact

finger(s) or whole-hand in rectum contact, fisting.

Brachio-vaginal contact

finger(s) or whole-hand in vagina contact, fisting.

Coprolagnia

sexual behavior involving feces, scat.

Deep kissing

oral-oral contact where the tongue is inserted and/or oral fluids are exchanged, French kissing.

Fellatio

oral-penile contact with or without ejaculation, sucking dick.

Informed client

an individual who possess all of four attributes

1. competency,
2. awareness of HIV infection status for both self and partner(s),
3. knowledge of the relative risks of sexually transmitted disease (STD) transmission for preferred sexual behaviors, and
4. knowledge of the potential personal and social outcomes of STDs for which s/he is at risk.

Plastic wrap

commercially available wraps for food preservation. Plastic wrap designed for micro-wave use is also acceptable for safer sex purposes.

Tribadism

Vaginal-vaginal frottage.

Urolagnia

sexual behavior involving urine, water sports.

Assumptions

The major focus of a Safer Sex teaching model must be the infected individual's responsibility to protect others. The corresponding focus for the uninfected must be the individual's responsibility for self-protection, but the client must also be assessed for whether they are at risk of harm (e.g. assault) should they attempt to protect themselves; and, if so, referred to a Social Services professional.

A Safer Sex Program will be most useful to the educator and the client if it is simply and succinctly written and provides useful information and advice about the risk of specific sexual behaviors.

The array of sexual behaviors reviewed for risk must be comprehensive and not limited as a product of the educator's morality, personal taste or sexual preference.

Sexuality is a Quality of Life issue and is not expendable for anyone, including the HIV-infected. A Safer Sex Program must establish standards which the provider and client can use to develop a sexual practice style which provides the client with a personally acceptable sexual experience.

Sexual behaviors can be classed as no risk, low risk, and high risk; and theoretical or known based on scientific literature and case studies. Behaviors characterized in the knowledge base as low risk or theoretical present a risk low enough that whether to engage in the behavior should be left to the informed client's individual choice.

The relative risks of specific sexual behaviors based on whether the individual is the insertive or receptive partner cannot be addressed if the standards are to remain simple and succinct. The presumption is made therefore that where there is a theoretical or known risk of infection, the risk is always bi-directional.

The Safer Sex needs of the HIV-infected, especially those with CD4+ counts of less than 200, are different from those of the general population and must be specially addressed.

Knowledge Base and Recommendations

After review of the peer-reviewed literature, community-based organization educational materials, case reports and expert presentation and discussion, the Subcommittee acknowledges and concludes:

The most important Safer Sex principles are:

• Use barrier precautions for vaginal and rectal intercourse, but withdraw prior to climax and ejaculate outside of partner.
• The only absolutely Safe Sex is abstinence from any behavior which exposes another person to bodily fluids. Touching and "dry" kissing carry no risk of HIV transmission.

The most important teaching principle is:

Sexual expression and transmission of HIV in various populations differs substantially along trajectories of age, race, culture and gender1.

• To be effective, educational programs must be knowledgeable about different practices and sensitive to different personal expressions and values.
• Penile|rectal2,3,4 and penile|vaginal5,6 transmission accounts for the largest number of sex-related HIV infections. In bivarate analyses of large cohort studies, no other behavior achieved statistical significance as an independent mode of infection. In multivarate analyses, brachio|rectal contact and enema use were found to correlate with a higher risk of infection (see below), but no other behavior achieved significance. Most sexual transmissions occur in persons who engage in receptive rectal or vaginal intercourse without barrier precautions.
• There is no situation in which a lack of barrier precautions for vaginal or rectal sex between partners at risk for HIV infection is defensible from a public health standpoint.

Despite the effective barrier which latex condoms represent, penile|rectal and penile|vaginal transmissions do occur, albeit more rarely, among those persons who practice rectal and vaginal intercourse using barrier precautions9, most likely due to product failure7.

• Demonstration and return demonstration of the correct use of latex condoms and lubricants must be made part of Safer Sex education.
• Clients must also be counseled that latex condoms are not complete protection and abstinence from rectal and vaginal intercourse will lower their risk of infection to near zero.

Penile-rectal intercourse is not confined to male homosexuals but is practiced with substantial frequency by heterosexual couples8, 9,10.

• Women must be educated in ways to decrease their risk of infection by anal as well as vaginal routes.

Penile-rectal and penile-vaginal transmissions also occur more frequently in the setting of concurrent rectal or vaginal irritation, infection or disease11,12,13,14,15. Genital ulcerative diseases (GUD) such as trichomonas, chlamydia, syphilis, gonorrhea, herpes and condyloma are among the infections which predict a substantially increased risk of transmission. Localized irritation caused by chemical irritants (e.g. lubricants containing high concentrations of nonoxynol-9) may also increase the likelihood of infection should barriers fail 16.

• Persons should refrain from rectal or vaginal intercourse in the setting of concurrent rectal or vaginal irritation or disease.
• Personal and social programs of STD/GUD detection, treatment and prevention will decrease HIV transmission.

Infection of intact epithelial mucosal cells has been demonstrated in vitro; these cells may serve as the primary site of infection17.

• Trauma is not a necessary component to transmission of HIV. If barrier precautions are not used for vaginal and rectal intercourse, even the most gentle and atraumatic event may result in HIV transmission.

Preparatory rectal douching or enema increases the likelihood of infection should barrier precautions fail8,18,19.

• Rectal douching or enema use preparatory to rectal intercourse should be discouraged.

A higher rate of tampon use was noted among infected wives as compared with uninfected wives of HIV infected hemophiliacs20. Tampon use, especially "superabsorbant" tampon use, has been shown to result in dryness, desquamation and other cellular abnormalities which may increase the likelihood of infection21. In one study the older wives of infected men were more likely to become infected than younger women22; this finding may reflect aging-related changes in vaginal mucosa that result in greater tissue fragility.

• The generous use of a water-based lubricant is recommended for all occasions of vaginal intercourse.
  * Persons should refrain from vaginal intercourse in the setting of vaginal dryness or irritation.
• Alternative methods of absorption (e.g. sanitary napkins) should be tried if tampon-related vaginal irritation or dryness occur.

Oral-oral transmission by deep kissing is theoretically possible7, 8, 9, 23. Oral secretions have been demonstrated to contain HIV23. Oral secretions have been demonstrated to inactivate HIV in vitro24,25.

• Persons should refrain from kissing when either partner is experiencing oral bleeding or disease.
• "Dry" kissing does not transmit HIV (no risk).

Penile-oral transmission by fellatio is theoretically possible and the case for its occurrence is stronger than that for kissing since the amount of virus is markedly greater in ejaculate than in oral fluids26,27,28. Oral transmission from fellatio with ejaculate has been the subject of controversial case reports detailing suspected transmission in seven men33, 29,30,31,32,33.

The reliability of sexual behavior histories taken from clients under circumstances of assessment for route of HIV exposure has been questioned by some authors9, 34. One important question is whether a newly infected client may find it so difficult to admit to non-barrier protected intercourse (a community censored behavior) when confronted by the health care worker that the client would mislead the interviewer as part of a pattern of coping with guilt and difficult feelings. All studies reviewed reported substantial cohorts of persons continuing to engage in penile|oral behaviors, and who deny engaging in vaginal or rectal intercourse, who remain persistently seronegative7, 8, 9, 23, 35.

There is a theoretical risk associated with non-barrier protected fellatio from virus in pre-ejaculate since pre-ejaculate contains lymphocytes17.

• Fellatio with ejaculation into the mouth carries a low risk of HIV infection.
• The informed client who wishes to accept the theoretical risk of non-barrier protected fellatio from virus in pre-ejaculate should not permit ejaculation into the mouth.
• Persons who wish the greatest possible protection should use latex condoms, dental rubber dams, or plastic wrap as barriers.
• Persons should refrain from fellatio in the setting of oral or penile bleeding or disease.

Vaginal-oral transmission by cunnilingus is theoretically possible. The presence of virus in vaginal fluids has been demonstrated36, 37,38.

• Persons who wish the greatest protection should use dental rubber dams or plastic wrap as barriers.
• Persons should refrain from cunnilingus in the setting of vaginal or oral bleeding or disease and during menstruation.

Vaginal-vaginal transmission by tribadism is theoretically possible. The presence of virus in vaginal fluids has been demonstrated41, 42, 43. The ability of HIV to infect intact mucosal cells has been demonstrated in vitro22.

• Persons who wish the greatest protection should use dental rubber dams or plastic wrap as barriers.
• Persons should refrain from tribadism during menstruation or when vaginal infection or disease is present.

Anal-oral transmission during anilingus is theoretically possible since anal secretions and fecal matter may contain lymphocytes. Anal- oral transmission is the mode of infection for other pathogens associated with substantial morbidity.

• Persons should refrain from fecal|oral contact to avoid enteric infections.
• Persons who wish the greatest protection should use dental rubber dams or plastic wrap as barriers.
• Persons should refrain from anilingus in the setting of concurrent rectal-anal or oral bleeding or disease.

Fecal-oral transmission during coprolagnia is theoretically possible since anal secretions and fecal matter may contain lymphocytes. Fecal-oral transmission is the mode of infection for other pathogens associated with substantial morbidity.

• Persons should refrain from fecal-oral contact to avoid enteric infections.
• Persons should refrain from fecal-oral contact in the setting of rectal or oral bleeding or disease.

Urino/oral transmission during urolagnia, in particular immediately following ejaculation while ejaculate is still in the urethra, is theoretically possible. Virus has been recovered from urine39.

• Persons should avoid oral contact with a partner's urine immediately following ejaculation.
• Persons should avoid oral contact with a partner's urine in the setting of concurrent urinary or oral bleeding or disease.

Brachio-vaginal and brachio-rectal transmission during brachio-rectal and brachio-vaginal contact is theoretically possible and brachio-rectal contact was found to be a significant transmission mode by multivarate analysis in two studies24, 40.

• Persons who wish the greatest protection should use latex gloves and a water-based lubricant.
• Persons should refrain from these behaviors in the setting of concurrent bleeding or disease of hands, vagina or anus/rectum.

Breast milk has been demonstrated to contain virus41. Breast milk has been demonstrated to infect infants42.

• Partners should avoid ingestion of breast milk from an infected person.

HIV infection requires contact of bodily fluids with mucous membranes or blood stream43, 44.

• HIV cannot be transmitted by contact of ejaculate or any other bodily fluid with intact skin (no risk).

There is a theoretical risk of infection during piercing, tattooing and scarification if infected bodily fluids contact disrupted skin surfaces.

• Persons undergoing these procedures should have the skin prepared with an antimicrobial (e.g. chlorhexidine or povidone iodine) and partner should use latex gloves.
• All equipment which pierces the skin should be autoclaved or otherwise sterilized to a standard known to kill spores.

There is a theoretical risk of cross-infection from sexual appliances since they may harbor traces of infected bodily fluids.

• Appliances which are not made of latex or other impervious substances should be kept solely for the use of one person.
• Appliances which are shared, or used with multiple partners, must be cleaned with a 10% solution of household bleach in water and thoroughly rinsed. (Lemon-scented bleach is more aesthetically pleasing than the usual smell of hypochlorite.)

Latex condoms are an effective barrier against HIV45. Heat, pressure and age degrade latex condoms and lead to product failure46. Condoms made from biologic products (e.g. lamb parenchyma) leak virus47. Latex condoms may rupture when roughly handled or handled with fingernails48, 49. The polyurethane women's condom is impermeable to HIV50.

• Education about selection, shelf-life and storage of these products must be given priority in a Safer Sex Program.
• Latex condoms and lubricants should be purchased at a drug store where product reliability, rotation and freshness is more likely than at a "boutique."
• Only latex condoms should be used for vaginal and rectal intercourse.
• Condoms must be handled with care to prevent rupture by fingernails and other sharp objects, and some care must be taken to avoid rough usage.
• The polyurethane women's condom represents a potential alternative to the conventional penile condom for vaginal intercourse.

Petroleum- and mineral oil-based lubricants degrade latex and cause condom rupture51. The wording "water-based" has been confused by some consumers to mean "washes away easily with water" leading to inappropriate product choice and condom rupture12.

• Only water-based lubricants should be used for vaginal and rectal intercourse.
• Only lubricants labeled "for use with latex condoms or diaphragms" should be used for vaginal and rectal intercourse.

Nonoxynol-9 (N-9) is an effective viricidal when used in conjunction with a latex condom52, 53,54,55, but may cause mucosal irritation17. The failure of N-9 to prevent HIV transmission when used without barrier precautions has been documented17. Clinical consensus suggests N-9 concentrations for inactivation of HIV need not exceed 5%. Clinical consensus favors discontinuing a N-9 lubricant should irritation occur. The consensus feeling is that the processes of local irritation will result in a greater risk of infection than that the risk of using a lubricant which does not contain N-9.

• A small amount of water-based lubricant containing N-9 should be placed in the inside tip of the latex condom before application; or the condom should be pre-lubricated with N-9.
• N-9 is never a substitute for a latex condom or other appropriate barrier.
• Concentrations for use as a viricidal in vaginal and rectal sex should not exceed 5%.
• Should irritation develop, the lubricant should be discontinued and a water-based lubricant without N-9 substituted.

Reduction in inhibition control is associated with unsafe sexual behaviors56.

• Safer Sex programs must contain advice against combining alcohol/drugs and sexuality.

To further decrease the possibility of infection the client should be advised to decrease their number of sexual partners. The Subcommittee offers two behavioral definitions of this advice; which to use might be based on a variety of individual factors including the client's willingness and ability to alter personal sexual behavior:

• the individual should decrease the number of partners with whom they engage in any sexual conduct, or * the individual should decrease the number of partners with whom they engage in vaginal or rectal intercourse;

HIV-infected persons with less than 200 CD4+ lymphocytes are at substantially greater risk of infection with opportunistic organisms57. Clinical consensus in the group was that enteric infections are more difficult to treat and more debilitating in the HIV-infected individual regardless of CD4+ count.

• Persons with fewer than 200 CD4+ lymphocytes should avoid contact with oral and rectal fluids by reducing the number of partners with whom they engage in deep kissing, and using dental rubber dams or plastic wrap as a barrier during cunnilingus and anilingus.
• All HIV-infected clients should be advised that enteric infections may be more debilitating and risky for them and to decrease their risk of enteric infection by using barriers for all occurrences of cunnilingus and anilingus.

A number of authorities have concluded that disclosure of HIV infection status to a new sexual partner should occur prior to sexual conduct58, 59. The Subcommittee endorses two strategies for providing potential partners of HIV- infected persons with information prior to potential risk:

The HIV-infected individual should be advised to inform every potential partner of their infection prior to sexual contact.

The HIV-infected individual should be encouraged to participate in community-based social programming for HIV-infected persons where all attending are known to be infected.

References

1. Fullilove, M.T. et al. Race/gender issues in the sexual transmission of AIDS. AIDS Clinical Review 1990 25-62.

2. Schechter, M.T. and others. The Vancouver lymphadenopathy-AIDS study: HIV seroconversion in a cohort of homosexual men. CMAJ 1986 Dec 15;135(12):1355-60.

3. Kingsley, L.A. and others. Risk factors for seroconversion to human immunodeficiency virus among male homosexuals. Lancet 1987 Feb 14;1(8529):345-9.

4. Detels, R., and others. Seroconversion, sexual activity, and condom use among 2915 HIV seronegative men followed for up to 2 years. J Acquir Immune Defic Syndr 1989;2(1):77-83.

5. Padian, N. Heterosexual transmission of acquired immunodeficiencysyndrome: International perspectives and national projections. Rev Infect Dis 1987 Sep-Oct;9(5):947-60.

6. Peterman, T.A. and others. Risk of human immunodeficiency virus transmission from heterosexual adults with transfusion-associated infections. JAMA 1988 Jan 1;259(1):55-8.

7. Voeller, B. Persistent Condom Breakage. Poster W.A.P. 99 presented at the Fifth International Conference on AIDS, Montreal. June 1989.

8. Voeller, B. AIDS and heterosexual anal intercourse. Arch Sex Behav 1991 Jun;20(3):233-76.

9. Greenblatt, et al. Behaviors associated with incident HIV infection among homosexual men in San Francisco. Paper presented at the IV International Conference on AIDS, Stockholm. 1988.

10. Sion, F.S. et al. The importance of anal intercourse in transmission of HIV to women. Paper presented at the IV International Conference on AIDS, Stockholm. 1988.

11. Simonsen, J.N. and others. Human immunodeficiency virus infection amongmen withsexually transmitted diseases: Experience from a center in Africa. N Engl J Med 1988 Aug 4;319(5):274-8.

12. Alexander, N.J. Sexual transmission of human immunodeficiency virus: Virus entry into the male and female genital tract. Fertil Steril 1990 Jul;54(1):1-18.

13. Moss, G.B. and Kreiss, J.K. The interrelationship between human immunodeficiency virus infection and other sexually transmitted diseases. Med Clin North Am 1990 Nov;74(6):1647-60.

14. Jessamine, P.G. & Ronald, A.R. Chancroid and the role of genital ulcer disease in the spread of human retroviruses. Med Clin North Am 1990 Nov;74(6):1417-31.

15. World Health Organization. Consensus Statements on HIV Transmission. Lancet February 18, 1989 396.

16. Kreiss, J. et al. Efficacy of nonoxynol-9 in preventing HIV transmission. (Abstract M.A.O. 36). Presented at the Fifth International Conference on AIDS,Montreal, June 4 to9, 1989.

17. Adahi, A. Koenig, S, Gendelman, H.E., Daughtery, D. Gattoni-Celli, S.Fauci, A.S., Martin, M.A. Productive persistent infection of human colorectal cell lines with humanimmunodeficiency virus. J Virol 1987 Jan;61(1):209-13.

18. Winkelstein, W. et al. Sexual practices and risk of infection by the human immunodeficiency virus, the San Francisco men's health study. JAMA 1987 Jan 16;257(3):321-5.

19. Jeffries, E. et al. The Vancouver lymphadenopathy AIDS study: 2. Seroepidemiologyand HTLV-III antibody. CMAJ 132 1373-1377. 1984.

20. Webster A. et al. Cytomegalovirus infection and progression towards AIDS in haemophiliacs with human immunodeficiency virus infection. Lancet 2(8654) 63. 1989.

21. Raudrant, D. et al. Study of the vaginal mucous membrane following tampon utilization; aspect on colposcopy, scanning electron microscopy and transmission electron microscopy. Eur J Obstet Gynecol Reprod Biol 1989 Apr;31(1):53-65.

22. Peterman, T. A. and others. Risk of human immunodeficiency virustransmission from heterosexual adults with transfusion-associated infections. JAMA 1988 Jan 1;259(1):55-8.

23. Levy, J.A. & Greenspan, D. HIV in saliva. Lancet II 1248. 1988.

24. Fox, P.C., and others. Saliva inhibits HIV-1 infectivity. Journal ofthe American Dental Association, 116(6):635-7. May 1988.

25. Fox, P.C. and others. Salivary inhibition of HIV infectivity: functional properties and distribution in men, women, and children. Journal of the American Dental Association, 118(6):709-711

26. Schechter, M.T. and others. Can HTLV-III be transmitted orally? Lancet,1986; 2:77-83.

27. Mayer, K.H. and others. Human immunodeficiency virus and oral intercourse. AIM,107:428-429. 1987.

28. Lifson, A.R. and others. HIV seroconversion in two homosexual men after receptive oral intercourse with ejaculation: implications for counseling concerning safe sexual practice. AJPH, 80(12):1509-1511. December 1990.

29. Rozenbaum, W. et al. HIV transmission by oral sex. Lancet 1988 Jun 18;1(8599):1395.

30. Dassey, D. E. HIV and orogenital transmission. Lancet II(8618) 1023. 10/29/88.

31. Detels, R. & Visscher B. HIV and orogenital transmission. LancetII (8618) 1023. 10/29/88.

32. Rozenbaum, W. et al. HIV transmission by oral sex. IV International AIDS Conference, 12-16 June 1988, Stockholm. Abstract no. 4562.

33. Rozenbaum, W. et al. HIV and orogenital transmission. Lancet II 1023- 1024. 1988.

34. Catania, J.A. et al. Methodological problems in AIDS behavioral research: Influences on measurement error and participation bias in studies of sexual behavior. Psychological Bulletin 108(3) 339-362. 1990.

35. Lyman, D. et al. Minimal risk of transmission of AIDS- associated retrovirus infection by oral-genital contact. JAMA 255 1703. 1986.

36. Archibald, D.W. et al. Antibodies to human immunodeficiency virus in cervical secretions from women at risk for AIDS. Journal of Infectious Diseases 156:1240. 1988.

37. Wofsy, C. B. et al. Isolation of AIDS-associated retrovirus from the genital secretions of women with antibodies to the virus. Lancet 1:527. 1986.

38. Vogt, M.W. et al. Isolation patterns of human immunodeficiency virus from cervical secretions during the menstrual cycle of women at risk for the acquired immunodeficiency syndrome. Annals of Internal Medicine 106:380. 1987.

39. Levy, J. A. et al. Infection by the retrovirus associated with the acquired immunodeficiency syndrome: clinical, biological, and molecular features. Annals of Internal Medicine 103:694-699. 1985.

40. Chmiel, J.S. et al. Factors associated with prevalent human immunodeficiency virus (HIV) infection in the multicenter AIDS cohort study. American Journal of Epidemiology 126 568-577. 1987.

41. Thirty, L. et al. Isolation of AIDS virus from cell-free breast milk of three healthy virus carriers. Lancet 2 891-892. 1985.

42. Ziegler, J. B. et al. Postnatal transmission of AIDS-associated retrovirus from mother to infant. Lancet 1 896-898. 1985.

43. Blattner, W. A. HIV epidemiology: Past, present, and future. The FASEB Journal 52340-2348. July 1991.

44. Rutherford, G. W. & Werdeger, D. The epidemiology of acquired immunodeficiency syndrome. Immunol Ser 44 1-36. 1989.

45. Rietmeijer, C.A.M., and others. Condoms as physical and chemical barriers against human immunodeficiency virus. Journal of the American Medical Association, 259:1851-3. March 25, 1988.

46. Voeller, B. Untitled letter. The Village Voice, 31(13). April 1, 1986.

47. Lytle, C.D and others. Virus leakage through natural membrane condoms. Sexually Transmitted Diseases, 58-62. April-June 1990.

48. Golombok, S. and others. Condom failure among homosexual men. Journal of Acquired Immune Deficiency Syndromes, 2:404-409. 1989.

49. Richters, J. and others. Low condom breakage rate in commercial sex. Lancet, December24/31, 1988:1487-1488.

50. Drew, W.L. and others. Evaluation of the virus permeability of a newcondom for women. Sexually Transmitted Diseases 17(2):110-112.

51. Voeller, B., and others. Mineral oil lubricants cause rapid deterioration of latex condoms. Contraception, 39(1):95-102. January 1989.

52. Voeller, B. Nonoxynol-9 and HTLV-III. Lancet 1986 May 17;1(8490):1153.

53. Hicks, D.R., et al. Inactivation of HTLV-III/LAV-infected cultures of normal human lymphocytes by nonoxynol-9 in vitro. Lancet 1985 Dec 21-28;2(8469-70):1422-3.

54. Hicks, D.R., et al. Inactivation of HTLV-III/LAV-infected cultures of normal human lymphocytes by nonoxynol-9 in vitro. Lancet 1985 Dec 21-28;2(8469-70):1422-3.

55. Scesney, S.M., and others. The impermeability of condoms to human immunodeficiency virus (HIV) and inactivation of HIV by the spermicide nonoxynol-9. Unpublished report of research. Available from: John L. Sullivan, M.D. Department of Pediatrics, University of Massachusetts Medical Center, 55 Lake Avenue North, Worcester, MA 01655.

56. Stall, R., and others. Alcohol and drug use during sexual activity and compliance with safe sex guidelines for AIDS: the AIDS behavioral research project. Health Education Quarterly, 13(4):359-371. Winter 1986.

57. Yarchoan, R. and others. CD4 count and risk for death in patients infected with HIV receiving antiretroviral therapy. Ann Intern Med 1991 Aug 1;115(3):184-9.

58. Goedert, J.J. What is safe sex? N Engl J Med 1987 May 21;316(21):1339-42.

59. Wofsy, C. B. Prevention of HIV transmission. Infectious Disease Clinics of North America 3:307-319, June 1988.

DT 9311
DOCN SSKB9301