Heard, Isabelle MD*; Potard, Valérie MS†; Costagliola, Dominique MD†; Kazatchkine, Michel D. MD*
HIV infection impacts sexual practices, reproductive health, and contraceptive choice. 1 The proportion of HIV-infected women desiring a child in the United States was shown to be lower than within the general population. 2 Seropositive women were also less likely than seronegative women to have a subsequent pregnancy after learning of their HIV status. 3 Although the HIV epidemic is growing among women, knowledge of contraception and access to family planning remain limited among HIV-positive women. 1 The choice of contraception in HIV-positive women is constrained by the need to prevent sexual transmission of HIV as well as unwanted pregnancies and potential infection with HIV in infants. Correct and consistent use of male condoms prevents HIV transmission to the male partner, 4–6 and the risk of seroconversion appears to be halved for consistent compared with inconsistent condom use among HIV-discordant couples. 7 For women, condoms have been recognized as a means to reduce the risk of acquiring other sexually transmitted infections (STIs) or being superinfected with HIV. Although they are associated with a high degree of protection against HIV transmission, 5,6 they provide less protection against pregnancy. Unintended pregnancy rates range between 3% and 14% between those who use condoms correctly and typical condom users. 8 Oral contraceptives are highly effective to prevent pregnancy, with 0.1–5% of unintended pregnancies among those who use oral contraceptives correctly and typical users. 8 Oral contraceptives, intra-uterine devices (IUDs), and sterilization are highly effective methods of contraception. However, they do not prevent HIV transmission. Dual-function contraceptives that simultaneously prevent HIV transmission as well as unwanted pregnancies might be the most appropriate contraceptive method for HIV-positive women. 9 Nevertheless, correct and consistent use of condoms is seldom achieved and different studies have shown that couples using 2 contraceptive methods were less likely than single-method users to be consistent condom users. 10,11
Our objective was to describe the contraceptive methods used in a cohort of HIV-positive sexually active women who knew the serologic status of their steady partners. Women were prospectively followed between 1993 and 2001. Specifically, we investigated the chosen method of contraception depending on its relative efficacy to prevent HIV transmission, STIs, and unwanted pregnancies.
MATERIALS AND METHODS
All women attending the gynecology outpatient clinic of the HIV departments of Hôpital Européen Georges Pompidou and Hôpital Cochin, Paris, between June 1993 and December 2001 were approached to participate in a gynecologic survey. Eligibility criteria for enrollment in the study included documented HIV-positive status and consenting to the protocol. There were no exclusion criteria. Among the 884 eligible women, 879 agreed to enroll in the study, and 5 declined because of fear of loss of confidentiality by participating in a cohort of HIV-infected patients. The protocol was approved by the French National Agency for AIDS Research (ANRS). All women were interviewed using a standardized questionnaire to obtain a demographic, sexual, gynecologic, and obstetrical history. At each semiannual visit, women were interviewed about sexual relationships with steady male partners since their last visit, the HIV status of the steady partner, and on methods of contraception that they had used, including IUD, hormonal contraceptives, condoms, or other methods (vaginal sponge, spermicides, tubal ligation, or natural family planning methods). Women were also asked about the frequency of condom use during sexual intercourse. Possible responses regarding condom use were: never, some times, always, or no sex. Sexual activity was not quantified. Because women who used a contraceptive method other than condoms may still use condoms to prevent HIV transmission, we analyzed answers to the questionnaire with regard to condom use as well as to contraception. The most recent values of CD4+ cell counts and HIV RNA plasma viral load within 6 months of the gynecologic examinations were obtained from clinical records.
For the present study, analyses were restricted to the 575 women aged <50 years who had no history of hysterectomy or confirmed menopause, and who reported at ≥1 visits that they had sex with a steady partner of known HIV status. Our objective was to describe self-reported contraceptive use in HIV-discordant and -concordant couples. Contraceptive use was classified into 2 groups according to the degree of protection against unintended pregnancy. Effective methods of contraception include tubal ligation, vasectomy of the male partner, injectable and implantable contraception, oral contraception, and IUD. During the study period, tubal ligation and vasectomy were not authorized in France (until July 2001) and implantable contraception was not available. As no women reported injectable contraception, effective contraception (EC) was restricted to oral contraceptives and IUDs. Consistent use of male condoms as the only form of contraception was considered to be a less effective method of contraception. Inconsistent use of condoms was not considered to be an EC method, and visits during which only inconsistent condom use and no other contraceptive use was reported were combined with visits during which no contraceptive use at all was reported.
Among the 2529 visits of the 575 women during the study period, 1561 were analyzed and 968 were excluded from the analysis. Reasons for exclusion were: no steady partner since the last visit (455 visits); unknown serostatus of the male partner (263 visits); pregnancy (142 visits); sexual abstinence (95 visits); or use of contraception other than hormonal contraceptives, IUD, or condoms (e.g., vaginal sponge, spermicides, and natural family planning, 13 visits).
Statistical analyses were performed using SAS software (version 8, SAS Institute, Cary, NC). Analyses were performed using univariate and multivariate generalized estimating equation (GEE) models. GEE models allow analysis of all visits of all women accounting for the dependence between visits of a given woman. 12 A block diagonal correlation matrix was used to specify the nature of the correlation among repeat observations within women. We used an exchangeable correlation structure in which the correlation between observations made on a given woman at different times was assumed to be constant. We first compared the frequency of EC use according to the HIV serostatus of the male partner, then compared the frequency of consistent condom use according to the serostatus of the partner, using a univariate GEE model with a logit link function for a binomial distribution. We then performed separate analyses depending on the HIV status of the male partner. To study the factors associated with EC use, we considered age, calendar periods (1993–1995, 1996–1997, 1998–1999, 2000–2001), route of infection, married or cohabitating, ethnic origin, condom use (always vs. never or sometimes, i.e., inconsistent), HIV treatment (no treatment, mono- or bi-therapy, highly active antiretroviral therapy [HAART]), CD4 cell count (<200/μL vs. ≥200/μL), and HIV viral load (<500 copies/mL vs. ≥500 copies/mL). Only variables with P value <0.1 in univariate analyses were included in the multivariate analyses. In addition, time periods, age, and antiretroviral therapy were forced into the model. We used the Wald test to evaluate the impact of these variables. We further studied the frequency of use of condoms according to calendar period using a multivariate GEE model with a logit link function for a binomial distribution. Among women whose steady partner was HIV seronegative, the model was adjusted for age, EC use, and antiretroviral treatment. Among women whose steady partner was HIV seropositive, the model was adjusted for age, EC use, antiretroviral treatment, and route of infection.
Women included in the study were aged between 19–49 years, with a median age of 32 years. Three hundred fifty women (62%) were white, 170 (30%) were black African or Caribbean, and the remaining 55 had various ethnic backgrounds. Most women had been infected through heterosexual contact (73%) or injection drug use (23%). At baseline, 67% of the women reported to be living with a partner, and 63% had at least one child.
Four hundred twenty-nine women reported having a relationship with an HIV-negative partner and 190 with an HIV-positive partner at ≥1 visits. The total exceeded 575 because of changes in the relationships during the study period. Contraceptive methods that were analyzed included EC or the consistent use of condoms as the only form of contraception. Contraceptive use was found to depend on the partner's HIV status. Women with an HIV-positive partner reported the use of contraception at 69% of the visits, compared with 91% of the visits among women with a HIV-negative partner (P = 0.0001). The use of EC was 2.1 times more frequent (95% CI = 1.5–2.9, P < 0.001) among women with a HIV-positive partner than among women with an HIV-negative partner (29.6 and 14.2%, respectively) (Table 1). The consistent use of condoms as the only form of contraception was 6.1 times less likely (95% CI = 0.1–0.2, P < 0.001) among women with an HIV-positive partner than among women with an HIV-negative partner (47.2 and 87.2%, respectively;Table 1).
Among women with an HIV-negative partner reporting the use of contraception, the use of condoms alone was reported at 84% of visits, EC alone at 4% of visits, and dual methods (both EC and condoms) at 12% of visits.
In multivariate analysis of study factors associated with EC use, the use of EC decreased with the introduction of HAART after 1998 (P = 0.02). EC use was increased in inconsistent or noncondom users compared with consistent users (27.3 and 12.3%, respectively; OR 2.0, 95% CI = 1.3–3.3, P = 0.003) (Table 2). EC use was negatively associated with condom use (OR 2.1, 95% CI = 1.4–3.3, P = 0.0005).
Among women with HIV-positive partners using a contraceptive method, condom use alone was reported at 57% of the visits, EC alone at 31% of visits, and dual methods at 12% of visits. In multivariate analysis, we found that EC use, with or without condoms, decreased with increasing age (P = 0.02) and increased among inconsistent or noncondom users (OR 1.9; 95% CI = 1.1–3.3, P = 0.02) (Table 3). Similar to the serodiscordant couples, women who were not using EC were more likely to use condoms than were those using EC (OR 2.3, 95% CI = 1.4–3.8; P = 0.0007).
After adjusting for age, condom use, and antiretroviral therapy, multivariate analysis showed that the use of EC decreased by two-thirds over the last decade (P = 0.02) in serodiscordant couples, whereas it remained stable in seroconcordant couples (Tables 4 and 5). In contrast, the use of condoms remained constant during the study period, independent of the HIV status of the partner (Tables 4 and 5).
In the present study, we report on contraceptive use in a cohort of HIV-positive women during the past decade. We observed that the reported contraceptive use depended on the partner's HIV status. The results of the study should, however, be interpreted with caution since the data were collected as self-reports about contraceptive use, sexual behavior, and the partner's serostatus.
Among women with an HIV-negative partner, we observed a high rate of contraceptive use (91%), higher than that observed in a national survey performed in 1994 on contraceptive use among French women aged 20–49 (57%). 13 Among serodiscordant couples, dual protection including the simultaneous use of condoms and EC is the most appropriate means to prevent both HIV transmission to the male partner and unplanned pregnancies. In our study population, dual protection was reported in only 12% of the visits. Women reported a high frequency of condom alone use (84%) in discordant relationships, whereas EC use was reported in only 4%. This observation suggests that most women chose a contraceptive method better suited to prevention of sexual HIV transmission than to prevention of unwanted pregnancy.
The rate of consistent condom use in our study was higher than that reported in previous studies of women with HIV or at high risk of infection. 5,11,14,15 Such high condom use may result from active promotion of condom use at each semi-annual visit, as counseling was previously shown to increase condom use in serodiscordant couples. 6 The unique factor related to not using condoms was the use of a more effective contraceptive method. It was previously reported that women who used a highly effective contraceptive method were less likely to use condoms, independent of their HIV status. 15–17 We observed a sustained decrease in the level of EC use from 23% in the period of 1993–1995 to 7% in the period of 2000–2001 among women reporting an HIV-discordant relationship. The reported choice of a contraceptive method that favored protection against HIV transmission, as well as the decreased use of EC, supports the hypothesis that HIV-positive women who engaged in a serodiscordant relationship may have had a desire for pregnancy. This is consistent with the dramatic improvement in the survival and quality of life of HIV-infected individuals associated with the widespread use of HAART. 18 This assumption is further supported by the increase in the incidence of pregnancies among HIV-positive women after the introduction of HAART (French Perinatal Cohort; personal communication, L. Mandelbrot). Chen et al 2 recently reported that HIV-positive women with an HIV-negative partner were twice as likely to expect children than women with an HIV-positive partner (OR, 11.0 and 5.3, respectively). In addition, antiretroviral treatment has dramatically reduced the risk of mother-to-child transmission of HIV, which is now 3%. 19
Among couples in which both partners were HIV positive, no contraception was reported at 31% of the visits. Although our study did not include quantitative assessment of sexual activity, we cannot rule out that sexual activity had decreased among these patients, due to the effects of debilitating illnesses or of sexual dysfunction related to HAART. 6,20,21 The use of an efficient contraceptive method might be perceived as less necessary among couples having less frequent intercourse. Counseling patients regarding the risk of unintended pregnancy and perinatal HIV transmission in this context should thus be seriously considered and family planning counseling should be reinforced.
Some HIV-positive couples did not perceive the need to prevent HIV transmission when both partners were already infected. Among others, condom use was associated with the fear of HIV disease progression or desire for protection against other STIs or of transmission of new strains of HIV. 22 However, the roles of HIV superinfection in the progression of HIV disease remain unclear. 23 Although it was suggested that frequent exposure to HIV might cause more rapid progression to AIDS, 24 Goedert et al 25 reported no evidence for accelerated progression to AIDS among seropositive patients who practiced unprotected sex. In our study, condoms were used 6 times less frequently when the partner was HIV positive than when he was HIV negative, and consistent condom use was reported at only half of the visits among seroconcordant couples, as observed in other studies. 11,15 In the report by Diaz et al, 17 women with an HIV-positive partner were less likely to use condoms than women having an HIV-negative partner (48 and 55%, respectively). However, the difference was not significant.
We observed that the level of condom use remained stable over the past decade among both seroconcordant and serodiscordant couples. The sustained use of condoms despite availability and widespread use of HAART during that period supports the hypothesis that the decrease in HIV viral load did not impact on sexual behavior at least in heterosexual stable relationships. 26 This is consistent with several published studies indicating that sexual behavior in which the steady partner was at risk for HIV transmission remained stable since the introduction and widespread use of HAART. 27,28
Drug-drug interactions between antiretrovirals and steroid hormone contraceptives have been described. Nonnucleoside reverse transcriptase inhibitors and protease inhibitors are metabolized through the cytochrome P450 liver enzyme complex, as is ethinyl estradiol, the estrogen in combined oral contraceptives (COCs). In our study population, 26% of HIV-positive women on HAART whose partner was HIV positive used oral contraceptives. As women in an HIV-concordant relationship appeared less likely to use condoms, the potential risk of a decrease in COC efficacy should be considered and specific contraception counseling provided. 29,30
The use of a contraceptive method and condoms is a critical issue when discussing interventions in HIV-infected women at risk for transmitting HIV. Health care providers treating women with HIV often do not consider the reasons for using contraceptives. Our observations suggest that family planning should be reinforced among HIV-positive women in the current changing treatment environment. Counseling HIV-positive treated women on issues dealing with contraception should take into account the risk of drug interactions with hormonal methods and also review the need for possible use of alternative methods of contraception or dose adjustment for the interacting agent. Our data emphasize that contraception counseling should take into account the partner's serostatus. Education and counseling about reproductive issues should be reinforced in HIV-seroconcordant couples, since couples using no contraception may not necessarily intend to become parents. A thorough discussion on the relative effectiveness of contraceptive methods should be considered, emphasizing the importance of avoiding pregnancy and perinatal transmission of HIV, as well as that of using condoms to prevent transmission of HIV and STIs.
The authors thank Dr. Joel Palefsky (UCSF) for helpful comments.
1. Olaitan A, Madge S, Johnson M. Contraceptive use, and knowledge of postcoital contraception amongst HIV-infected women: a comparison with an HIV negative population. Br J Fam Plann. 1996;22:127–128.
2. Chen J, Phillips K, Kanouse D, et al. Fertility desires and intentions of HIV-positive men and women. Fam Plann Perspect. 2001;33:144–152.
3. Lindsay M, Grant J, Peterson H, et al. The impact of knowledge of human immunodeficiency virus serostatus on contraceptive choice and repeat pregnancy. Obstet Gynecol. 1995;85:675–679.
4. Davis K, Welker S. The effectiveness of condoms in reducing heterosexual transmission of HIV. Fam Plann Perspect. 1999;31:272–279.
5. De Vincenzi I. Longitudinal study of human immunodeficiency virus transmission by heterosexual partners. N Engl J Med. 1994;331:341–346.
6. Allen S, Meinzen-Derr J, Kautzman M, et al. Sexual behavior of HIV discordant couples after HIV counseling and testing. AIDS. 2003;17:733–740.
7. Nicolosi A, Musicco M, Saracco A, et al. Risk factors for woman-to-man sexual transmission of the human immunodeficiency virus. Italian Study Group on HIV Heterosexual Transmission. J Acquir Immune Defic Syndr. 1994;7:296–300.
8. Trussell J. Contraceptive efficacy. In: Hatcher RA, Trussell J, Stewart F, et al., eds. Contraceptive Technology. New York: Ardent Media; 1998:779–844.
9. IPPF International Medical Advisory Panel. IMAP Statement on dual protection against unwanted pregnancy and sexually transmitted infections including HIV. IPPF Med Bull. 2000:34.
10. Critelli J, Suire D. Obstacles to condom use: the combination of other forms of birth control and short-term monogamy. J Am Coll Health. 1998;46:215–219.
11. Magalhaes J, Amaral E, Giraldo PC, et al. HIV infection in women: impact on contraception. Contraception. 2002;66:87–91.
12. Liang K, Zeger S. Longitudinal data analysis using generalized linear models. Biometrika. 1986;73:13–22.
13. Toulemon L, Leridon H. Contraceptive practices and trends in France. Fam Plann Perspect. 1998;30:114–120.
14. Buchacz K, van der Straten A, Saul J, et al. Sociodemographic, behavioral, and clinical correlates of inconsistent condom use in HIV-serodiscordant heterosexual couples. J Acquir Immune Defic Syndr. 2001;28:289–297.
15. Wilson TE, Massad LS, Riester KA, et al. Sexual, contraceptive, and drug use behaviors of women with HIV and those at high risk for infection: results from the Women's Interagency HIV Study. AIDS. 1999;13:591–598.
16. Centers for Disease Control and Prevention. Surgical sterilization among women and use of condoms: Baltimore, 1989–1990. MMWR Morb Mortal Wkly Rep. 1992;41:568–569, 575.
17. Diaz T, Schable B, Chu SY. Relationship between use of condoms and other forms of contraception among human immunodeficiency virus-infected women. Supplement to HIV and AIDS Surveillance Project Group. Obstet Gynecol. 1995;86:277–282.
18. Mocroft A, Brettle R, Kirk O, et al. Changes in the cause of death among HIV positive subjects across Europe: results from the EuroSIDA study. AIDS. 2002;16:1663–1671.
19. Mayaux M, Téglas J, Blanche S. Characteristics of HIV-infected women who do not receive preventive antiretroviral therapy in the French national cohort. J Acquir Immune Defic Syndr. 2003;34:338–343.
20. Sellmeyer DE, Grunfeld C. Endocrine and metabolic disturbances in human immunodeficiency virus infection and the acquired immune deficiency syndrome. Endocr Rev. 1996;17:518–532.
21. Collazos J, Martinez E, Mayo J, et al. Sexual dysfunction in HIV-infected patients treated with highly active antiretroviral therapy. J Acquir Immune Defic Syndr. 2002;31:322–326.
22. 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. 1998;339:307–311.
23. Goulder P, Walker B. HIV-1 superinfection, a word of caution. N Engl J Med. 2002;347:756–758.
24. Anzala OA, Nagelkerke NJ, Bwayo JJ, et al. Rapid progression to disease in African sex workers with human immunodeficiency virus type 1 infection. J Infect Dis. 1995;171:686–689.
25. Goedert JJ, Eyster ME, Biggar RJ, et al. Heterosexual transmission of human immunodeficiency virus: association with severe depletion of T-helper lymphocytes in men with hemophilia. AIDS Res Hum Retroviruses. 1987;3:355–361.
26. Lavoie R, Otis J, Leclerc R. Attitudes towards the triple-therapies and safer sex: what a mix. Paper presented at: XII International AIDS Conference; June 28–July 3, 1998; Geneva. Abstract 3427.
27. Desquilbet L, Deveau C, Goujard C, et al. PRIMO Cohort Study Group. Increase in at-risk sexual behaviour among HIV-1-infected patients followed in the French PRIMO cohort. AIDS. 2002;16:2329–2333.
28. van der Straten A, Gomez CA, Saul J, et al. Sexual risk behaviors among heterosexual HIV serodiscordant couples in the era of post-exposure prevention and viral suppressive therapy. AIDS. 2000;14:F47–F54.
29. Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. Report of the NIH Panel to Define Principles of Therapy of HIV Infection. MMWR Morb Mortal Wkly Rep. 1998;47/No. RR-05):1–47.
30. Delfraissy JF. Management of patients infected by the human immunodeficiency virus (HIV). Report of an expert group. Medecine-Sciences, Flammarion; 2002.
© 2004 Lippincott Williams & Wilkins, Inc.