Hormonal contraception and HIV risk: no increased risk in South African cohort

This article is more than 12 years old. Click here for more recent articles on this topic

During a study of 5567 South African women, looking at the effect of hormonal contraception on HIV acquisition, 270 women became HIV-infected. The researchers report that use of hormonal contraception did not increase the risk, overall, of becoming HIV positive, compared to non-hormonal contraception, in the advance online edition of AIDS.

There was a modest but non-significant trend towards an increased risk of getting HIV in women who used an injectable hormonal contraceptive (DMPA).

Hormonal contraception included combined oral contraceptives (COC) and injectable progestin (DMPA [Depo-Provera] or norethisterone enanthate [Net-En]).



A chemical messenger which stimulates or suppresses cell and tissue activity. Hormones control most bodily functions, from simple basic needs like hunger to complex systems like reproduction, and even the emotions and mood.

not significant

Usually means ‘not statistically significant’, meaning that the observed difference between two or more figures could have arisen by chance. 


A product (such as a gel or cream) that is being tested in HIV prevention research. It could be applied topically to genital surfaces to prevent or reduce the transmission of HIV during sexual intercourse. Microbicides might also take other forms, including films, suppositories, and slow-releasing sponges or vaginal rings.


In everyday language, a general movement upwards or downwards (e.g. every year there are more HIV infections). When discussing statistics, a trend often describes an apparent difference between results that is not statistically significant. 


A serodiscordant couple is one in which one partner has HIV and the other has not. Many people dislike this word as it implies disagreement or conflict. Alternative terms include mixed status, magnetic or serodifferent.

This analysis, of a subset of data from a randomised, double-blind, placebo controlled trial (Carraguard Phase 3 Efficacy) that looked at whether vaginal use of a microbicide prevented HIV transmission, did find a modest increased risk with the use of DMPA, notably, among younger women (16 to 24 years of age; aHR: 1.68 95% CI: 0.96-2.94).

Over 150 million women worldwide use hormonal contraception. Approximately two-thirds use COCs and the remaining third use injectable progestin. Use of injectable progestin among women, especially younger women, in South Africa is increasing. Condom use, in sub-Saharan Africa, remains low among married women as well as those using effective forms of contraception.

Previous prospective studies about hormonal contraceptive use and the risks of getting HIV are inconsistent. Some studies have suggested that use of hormonal contraception, in particular DMPA, may increase the risk of getting HIV, while others do not.

The authors cite the reanalysis of the hormonal contraception and risk of HIV acquisition (HC-HIV) study, considered the most rigorous and largest of its kind. Among those using DPMA there was a 50% increased risk of getting HIV but not among those using COCs. Yet among women under 24 years of age use of either DPMA or COC increased the risk.

The authors chose to look at the use of hormonal contraception and the risks of getting HIV among 5567 women aged 16 to 49 who participated in the Carraguard study in three sites in South Africa: KwaZulu-Natal, Gauteng and the Western Cape, between March 2004 and March 2007.

The participants were HIV-negative, not pregnant and sexually active (defined as having had at least one sex act in the previous three months). After enrolment, participants were seen at one month, three months and then every three months after that, for a minimum of nine months and a maximum of 24 months. All were interviewed about their contraceptive use and sexual behaviours, were given a pelvic exam and, every three months, were tested for HIV.

Contraceptive use was encouraged throughout the study. All sites provided male latex condoms and female condoms on request.

Of the 5567 women, 36, 39 and 25% were from the Western Cape, Gauteng and the KwaZulu-Natal sites, respectively.

At enrolment 29% were using DPMA, 21% Net-En, and 9% COCs, with 41% in the non-hormonal group (comprising a combination of women who used male and female condoms, sterilisation, diaphragm, traditional methods or were not using any form of contraception).

Non-hormonal users and DMPA had the highest (35%) and lowest (11%) reported condom use, respectively.

Of the 270 women (3.7 per 100 woman-years) who became HIV infected, incidence was 2.8, 4.6, 3.5 and 3.4 per 100 woman-years, p=0.09 among those in the COC, DMPA, Net-En and non-hormonal groups, respectively.

The authors cite the recent publication of the secondary analyses of almost 3800 HIV sero-discordant couples from the two-year seven African country Partners in Prevention study data that showed double the risk of getting HIV and twice the risk of transmitting HIV, if infected, with the use of hormonal contraception (notably DMPA).

The authors note that, while their findings of a moderate, but not significant, increased risk with the use of DPMA and getting HIV is in keeping with the HC-HIV study and other recent reports, it contrasts with the Partners in Prevention study findings published in 2011.

The difference may be attributable to a difference in exposure risk in the study populations. The Partners in Prevention study analysis evaluated the risk of infection according to contraceptive mode among women with HIV-positive partners (who could have been at risk of infection upon each occasion of sexual intercourse).

In contrast, in this microbicide trial cohort, women may have been exposed to HIV on fewer occasions, if at all. The contraceptive analysis of the Carraguard trial cohort could not control for the HIV status of regular partners.

If this explanation were correct, the greater exposure risk in the Partners in Prevention study would tend to amplify any trend towards an elevated risk of contracting HIV when using DMPA.

As with other studies, the authors found no increased risk with Net-En. While differences between DMPA and Net-En were small, nonetheless there was a slightly higher risk for DMPA use. The authors note that comparing dosages between different progestins is not appropriate. A single dose of DMPA protects against pregnancy for three months while Net-En does so for two.

The authors suggest that because the main agent (medroxyprogesterone) of DMPA binds to the glucocorticoid receptor of many cells of the immune system, it may have a greater immunosuppressive effect (compared to progesterone).

Strengths of the study include:

  • having a large number of incident HIV infections and a large number of women, of whom 37% were under 25 years of age

  • using different forms of contraception to provide enough statistical power to show differences among contraception groups

  • testing every three months, which allowed for accurate timing of HIV infection. 

The authors note that the marginal structural modelling (MSM) approach used is becoming a preferred method of analysis of longitudinal data subject to time-dependent confounding.

Limitations included the fact that hormonal contraception was not available at all sites throughout the study, so comparison of self-reported use with staff provision data could not be made. The validity of self-reports on condom use and sexual behaviour of male partners is unclear.

The authors conclude that women should be counselled about dual protection; that, in particular, younger women and those using DPMA should be counselled about the increased risks for HIV; and that they should be encouraged to use condoms in addition to contraception.

The authors call for the need to expand women’s options for effective contraception to include non-hormonal as well as low-dose hormonal methods; and state that “to resolve this important public health question, we need to conduct a randomised trial to clarify the relationship between hormonal contraception and getting HIV”.


Morrison CS et al. Hormonal contraception and the risk of HIV acquisition among women in South Africa. AIDS 25, doi: 10.1097/QAD.0b013e32834fa13d, 2012.