Contraceptive injections and HIV transmission risk - what happens now?

Roger Pebody
Published: 26 October 2011
Professor Jared Baeten, pictured presenting at the International AIDS Society conference in July. ©IAS/Marcus Rose/Worker's Photos

With new data suggesting that injectable contraceptives may double the risk both of acquiring and passing on HIV, how will this affect women’s contraceptive choices? What are the implications for family planning policy in countries with a high burden of HIV?

Panellists on a teleconference on the topic, organised by AVAC last week, agreed that the data do not yet provide definitive answers and that healthcare providers need to avoid frightening women away from contraceptive methods they know and trust.

Experts are mindful that the HIV-related risks need to be balanced with contraception’s benefits for maternal and child health. Family planning helps to prevent unintended pregnancies and the number of unsafely performed abortions, thereby reducing maternal deaths, disabilities and infertility. It can prevent high-risk pregnancies among adolescents, older women, women in poor health and women who have had many births or births spaced too closely together. Because it helps women to space births, child mortality rates are lower; mothers have more time to breastfeed, improving infant health; and women have more time to recover physically and nutritionally between births.

Moreover, preventing unwanted pregnancies in women with HIV is also one component of strategies to reduce mother-to-child HIV transmission.

At the International AIDS Society’s conference in Rome in July, Dr Renee Heffron of the University of Washington presented results from an analysis of data from the Partners in Prevention cohort in seven African countries. The results, reported on Aidsmap at the time, showed that HIV-negative women who were in a relationship with an HIV-positive man had twice the risk of acquiring HIV if they used hormonal contraception. Furthermore, HIV-positive women had twice the risk of transmitting HIV to their male partners if they used hormonal contraception.

Whilst there was considerable interest in the study from conference delegates, it was only when the findings were published earlier this month in the journal Lancet Infectious Diseases that the mainstream media paid attention to the issue. Some news reports have been alarmist.

News reports have also tended to ignore the long history of (inconclusive) research into the possible links between contraceptives and HIV transmission. Some studies have found a link, others have not. The studies have all been observational (rather than randomised control trials), meaning that the results could be biased by factors the researchers did not consider or measure. Also, the studies have tended to be secondary analyses of data that were collected for other purposes (in the case of Partners in Prevention, a study of the effect of herpes treatment on HIV transmission).

Professor Jared Baeten of the University of Washington stressed that the Partners in Prevention researchers came to this issue with an open mind, determined to be “as careful and conscientious with the data as possible”.

“We were frankly quite disappointed to see that we had a doubling of HIV risk,” Baeten said. “We analysed the data several ways, to be sure that we had confidence in the results.”

One strength of the study is that the participants were all couples in which one partner had HIV, and the other did not. Only HIV transmissions which could be genetically linked to the primary partner were included. Rates of self-reported condom use were similar between those exposed to hormonal contraceptives and those who were not. The statistical analysis was sophisticated.

On the other hand, critics say that the researchers don’t know enough about participants’ use of contraception, which is all based on self-report. The investigators did not ask participants for the brand names of their contraceptives, they have no data on adherence to contraception and they did not take into account switches between methods. Moreover, it is possible that there were differences in the sexual behaviour of hormonal contraceptive users (compared to non-users) that were not fully captured by the researchers.

The World Health Organization (WHO) is convening an expert meeting in January in order to see whether it needs to revise the contraceptive guidance it provides in parts of the world where HIV is a major issue. Until now, WHO has concluded that overall, the weight of scientific evidence does not indicate that hormonal contraceptives increase the risk of acquiring or transmitting HIV.

January’s meeting will consider new systematic reviews which will assess all the relevant studies on the topic. Depending on the reviews’ findings, WHO may decide to revise its guidance to women at risk of HIV infection and to women with diagnosed HIV.

Previous research has tended to concentrate on women’s risk of acquiring HIV, rather than passing it on. The most recent study is the first one to look at transmission from HIV-positive contraceptive users to their male partners. It also identified a biological mechanism for the increased risk of onward transmission – higher viral loads in genital secretions. Nonetheless, other studies will be needed to confirm or refute these findings.

And Dr Charles Morrison of the non-governmental organisation FHI 360 – which provides both HIV and family planning services – argues that because observational studies have inherent limitations, researchers need to set up a randomised controlled trial. This could, for example, randomise women either to use contraceptive injections (the method thought most likely to raise the HIV risk) or to use an alternative contraceptive, such as an intrauterine device or an implant. Women’s HIV acquisition rates would then be compared between the two groups.

Such a study would take several years to provide results. Some experts also question whether women would be willing to take part and have their contraceptive choice determined by the randomisation process.  

In trying to make trade-offs between family planning and HIV prevention, health officials will also need to consider the connections between pregnancy and HIV transmission. Another analysis of the Partners in Prevention data set, by the same study team, suggests that HIV-positive women have double the risk of passing on HIV when they are pregnant than at other times.

Therefore if the concerns about HIV transmission led women with HIV to stop using contraceptive injections and they did not switch to alternative methods, they would be more likely to get pregnant - and thus more likely to pass on HIV.

Taken together, the findings on pregnancy and contraception do point to hormonal changes having an influence on HIV transmission.

Contraceptive injections lead to a large surge of hormones at one moment, so they may have more impact on HIV transmission than other hormonal methods which provide a lower dose.

Indeed, an important aspect of the most study is that four-fifths of those using hormonal contraceptives were using contraceptive injections, with the remainder taking oral pills. While women weren’t asked which injection they had received, by far the most popular injectable in the countries studied is DMPA (Depo-Provera), a progestogen-only contraceptive which protects against pregnancy for three months at a time.

The associations between injections, HIV acquisition and HIV transmission were statistically significant, whereas they were not for oral contraceptives. Moreover, the study has no data in respect of implants, patches or hormonal intrauterine devices.

It may therefore be unhelpful to lump all hormonal methods together and say that ‘hormonal contraception’ raises the risk of HIV transmission. The new data may suggest that the problem is specifically with DMPA injections.

However in many African countries, DMPA is one of the most widely used contraceptives. Reasons for its popularity may include its long-lasting protection (particularly when healthcare facilities are difficult to access), the ability of women to take it without the knowledge of their sexual partners, a low failure rate, widespread availability and the preferences of healthcare providers.

“We've allowed the method mix to shrink,” Professor Helen Rees of the Wits Reproductive Health and HIV Institute said. “Women’s choices have become limited to injectable contraceptives and combined oral contraceptives.”

She argued that healthcare providers urgently need to expand the range of contraceptive options that are available to women – including lower dose hormonal products, intrauterine devices, implants and condoms. Women need to have alternatives and options available before a phasing out of DMPA could be considered.

Jared Baeten argued for an integration of family planning and HIV services, with “the best counselling that we can provide to women on safe reproductive choices and safe HIV choices”. This should include explaining the possibility of an increased HIV risk with hormonal contraceptive use, especially injectable DMPA, and encouraging women to employ dual protection - effective contraception plus condoms to reduce the HIV risk.