Hormonal contraception is widely used and plays a critical role in women's health, education, employment, economic development and autonomy.
But there is considerable controversy about the role of hormonal contraception in HIV acquisition. There have also been concerns over its potential role in disease progression. A session at IAS 2012 explored the evidence to date.1
A prospective cohort study of 3790 couples from East and southern Africa, the Partners in Prevention HSV/HIV Transmission Study recently found that HIV negative women in partnerships with HIV positive men and using injectable contraceptives had a 2-fold increase in HIV acquisition risk (aHR 2.05, 95% CI 1.04 4.04, p=0.04).2 These results gained a great deal of international attention. Following an expert consultation and a systematic review of the existing data, WHO issued a statement upholding its previous guidance on hormonal contraceptive use and re-emphasised the importance of condom use for dual protection.3
Renee Heffron presented the results from multiple additional sensitivity analyses examining the robustness of the Partners in Prevention findings.4 She noted that the findings from this and other observational studies have been inconsistent, are hard to interpret and might have behavioural and/or biological explanations.
In the primary analysis previously reported the investigators used Cox proportional hazards regression, adjusting for age, male partner's viral load, and time-dependent unprotected sex and pregnancy, to compare HIV incidence in women using injectable hormonal contraceptives to women reporting no hormonal method.
The more recent sensitivity analyses examined the possible effects of behaviour, miss-classification of women who switched hormonal contraception during follow-up and the effect of depot medroxyprogesterone acetate (DMPA) compared to other hormonal contraception methods. As information about the brand of hormonal contraception was not collected in the study, in order to isolate the effect of DMPA, data from South African women were censored, as this is the only country with study sites where other methods of hormonal contraception are used.
Dr Heffron reported a persistent increase in HIV acquisition risk of approximately 2-fold across the analyses. When limited to only DMPA users, injectable hormonal contraception was associated with an almost 4-fold increase in risk of HIV acquisition (see Table 1).
Table 1. Results from sensitivity analyses of HIV incidence in women using injectable contraception versus no hormonal contraception
|Number of sex acts||2.06||1.04 4.07||0.04|
|Male report of unprotected sex*||2.03||0.95 4.35||0.07|
|Women who reported unprotected sex*||2.29||0.70 - 7.53||0.17|
|Restricting to periods prior to HC method switch||2.62||0.93 7.33||0.07|
|DMPA users||3.39||1.38 11.22||0.01|
Adjusted for age, baseline viral load of HIV-positive partner and time dependent report of pregnancy
Dr Heffron noted that some analyses had diminished power for precision due to decreased sample size of the sub groups but the adjusted hazard ratios always suggested a trend towards increased risk. She concluded that these results must be balanced with the unequivocal benefits of injectable contraception. She added that more high quality studies of hormonal contraception and HIV risk are needed and stressed the importance of integrating reproductive health and HIV prevention programmes.
Following this presentation Chelsea Polis from USAID, Office of Population and Reproductive Health, Washington showed results from a systematic review of the epidemiological literature on the association between hormonal contraception and HIV acquisition, including articles published or in press by December 15, 2011.5
The investigators identified twenty relevant studies, eight of which met minimum quality criteria. Only one study found a statistically significant association between use of oral contraceptive pills and HIV acquisition.
None of the studies reported statistically significant associations between use of norethisterone enantate (Net-En) and HIV acquisition, but the investigators noted that these data were limited. For DMPA, the data available at the time of review neither established a clear association nor definitely ruled out the possibility of an effect on HIV acquisition.
In order to look at disease progression in HIV positive women, Sharon Phillips from the WHO showed findings from a similarly conducted systematic review.6
This review revealed twelve reports of eleven studies meeting the inclusion criteria. One randomised controlled trial found increased risk of a composite endpoint of CD4 decline or death in hormonal contraceptive users versus copper IUD users. Ten observational studies reported no increased risk of HIV disease progression, as determined by death, time to CD4 below 200, time to initiation of antiretroviral therapy, increased viral load, or decline in CD4 with use of hormonal contraception versus non-use.
The investigators noted that although one randomised controlled trial found that hormonal contraceptive use was associated with increased risk of HIV disease progression it had important methodological shortcomings. None of the cohort studies found an association between hormonal contraceptive use and HIV disease progression compared with non-use of hormonal contraceptives.
They concluded that, "the preponderance of evidence indicates that HIV positive women can use hormonal contraceptive methods without concerns related to HIV disease progression".
In the first systematic review conducted for the WHO consultation and presented here, findings are mixed and conflicting. None of the studies reviewed were designed to assess HIV transmission risk and, like all observational findings have many potential confounders.
Eight of 20 relevant studies reviewed met minimum quality criteria -- related to use of multivariate analysis, low loss to follow up and clear definition of hormonal contraceptive use. Only one study of sex workers in Kenya showed an association between oral contraception and 46% elevated HIV transmission risk.7 The Heffron et al study described above showed elevated point estimates but this was not significant.
No studies reported statistically significant associations between use of norethisterone enatate (Net En) and elevated transmission risk, but only three gave estimates specific to this method.
Three of eight studies had statistically significantly increased relative risks with DMPA or unspecified injectable methods, including double the risk in Heffron 2012, 73% increase (and narrower confidence intervals) in Baeton 2007 and 48% increase in risk with marginal structural modelling in Morrison 2010 (but not with Cox proportional model in a previous analysis).8,9 Other studies had non-significant increased and decreased relative risks. This heterogeneity did not appear to be explained by differences in HIV incidence, purpose of data collection, size of study population, approach to pregnancy, number of seroconverters or overall statistical approach. Differences in manner of handling condom use, length of inter-survey interval and analysis of serodiscordant couples need further consideration and concerns about the potential for residual confounding remain.
The systematic review and the WHO expert consultation concluded that the body of evidence on injectable contraceptives does not establish a clear casual association with HIV acquisition nor does it rule out the possibility of an effect.
The medical eligibility criteria (MEC) for use of contraception offers guidance on the safety of use of different methods for women and men with specific characteristics or known medical conditions and is used by policy-makers, programme managers, and the scientific community, to support national programmes in preparing service delivery guidelines. The MEC did not change following the WHO consultation and there are no restrictions on the use of any hormonal contraceptive method for women living with HIV or at high risk of HIV, although the following clarification was added: "Some studies suggest that women using progestogen-only injectable contraception may be at increased risk of HIV acquisition, other studies do not show this association. A WHO expert group reviewed all the available evidence and agreed that the data were not sufficiently conclusive to change current guidance. However, because of the inconclusive nature of the body of evidence on possible increased risk of HIV acquisition, women using progestogen-only injectable contraception should be strongly advised to also always use condoms, male or female, and other HIV preventive measures. Expansion of contraceptive method mix and further research on the relationship between hormonal contraception and HIV infection is essential. These recommendations will be continually reviewed in light of new evidence".
It is important to note however that a numbering system such as this one cannot reflect nuance, uncertainty (and the complexity of discussion) involved.
Women living with HIV and activists involved in the expert consultation and a subsequent one to discuss the associated operational considerations have stressed the critical importance of clear information for women using hormonal contraceptives and health workers.