Depo-Provera and HIV transmission: the jury’s still out by C. Sathyamala

The link between HIV transmission and the use of Depo-Provera®, a three-monthly injectable contraceptive, has raised concerns for its use in populations with high prevalence of HIV. The WHO is convening a Guideline Development Group at the end of this month to review the status of Depo-Provera based on evidence from a new study. Considering that its results are not unequivocal, sufficient time should be allowed for response from scholars, health authorities of the concerned countries, and civil society representatives to facilitate an informed decision.


From 29-31st July, 2019, the World Health Organization (WHO) is convening a Guideline Development Group to review the existing recommendations for the use of contraceptive methods for women ‘at high risk of HIV’ (World Health Organization  2019a). This review will specifically focus on the results of a randomized ‘clinical’ trial, the Evidence for Contraceptive Options and HIV Outcomes (ECHO) published in the Lancet last month (ECHO Trial Consortium 2019). The ECHO trial, carried out in four African countries (South Africa, Kenya, Swaziland and Zambia), was designed  to ‘settle’ the two-decade long contestation whether or not the three-monthly injectable contraceptive Depo-Provera® increased the possibility of HIV transmission in women users.

Depo-Provera is not a new contraceptive. However, it has been dogged by controversy since the 1960s when the first application for licensing as a contraceptive was made by the American pharmaceutical Upjohn. Depo-Provera was considered hazardous because of its potential carcinogenic, teratogenic, and mutagenic effects (Multinational Monitor 1985; Sathyamala 2000). However, after its approval in 1992, evidence began to accumulate that it increased the possibility of HIV transmission in women users. In 2012, WHO cautioned that women using it ‘should be strongly advised to also always use condoms’ (emphasis as in original) (World Health Organization 2012). Because of the uncertainty in observational studies, the ECHO consortium was formed in 2012 to design a randomized trial.

Despite reservations expressed by others (see for instance, Ralph et al., 2013) on the ethics of randomization and subjecting participants knowingly to a contraceptive method that could increase HIV transmission, the ECHO consortium went ahead with the trial in 2015.  Contrary to expectations, it was successful in recruiting sufficient numbers of women and randomizing them to one of the three contraceptive methods, viz., Depo-Provera, a levonorgestrel sub-dermal implant, and a copper intrauterine device; the latter two acting as the control arms for testing Depo-Provera’s association with HIV. It has been argued that the randomization was made possible by exercising undue influence, and by concealing the real nature of the clinical trial during recruitment (Sathyamala 2019). Therefore, the ECHO trial has violated a central tenet of the Helsinki declaration by privileging the pursuit of knowledge over the interests of the girls/women trial participants from Africa (ibid).

Importantly, in the ECHO trial the sample size was calculated to estimate an increased risk of 45% or above only; by choosing a high cut off point, a decision was made not to detect lower risks (Hofmeyr et al., 2018). Hence, although the trial showed that Depo-Provera increased HIV transmission risk by 23 to 29% as compared with the sub-dermal implant, the finding is being dismissed as of no consequence due to a lack of statistical significance (ECHO Trial Consortium 2019). This makes little sense either for the individual girl/woman or from a public health perspective. A modeling exercise showed that even a 1.2 fold increase of HIV with Depo-Provera could lead to 27,000 new HIV infections per year, a number higher than that due to maternal deaths that would be averted by its use as a contraceptive (Butler et al., 2013). Moreover, the ECHO team, even while admitting to the inadequate statistical power to detect risks lower than 30%, affirmed that ‘for individual women at very high HIV risk, … even a relatively small effect might be important in contraceptive and HIV prevention decision making’ (ECHO Trial Consortium 2019 , p.8). Considering that one of the surprising findings of the ECHO trial was the ‘alarmingly high’ HIV incidence in the trial population, observed ‘throughout the course of the trial’ (ECHO Trial Consortium 2019, p.9),  it could well mean that Depo-Provera should not be used in populations with high prevalence and transmission rates of HIV.

Yet, a systematic misinformation campaign has been launched in the media[1] fed by the WHO’s own statement that the trial found no link between the methods studied and HIV (World Health Organization 2019b). This indicates that a decision has already been made to declare Depo-Provera as safe for use in high risk populations. In reality, the ECHO trial findings, instead of ‘settling’ the issue, have only added to the uncertainties. It is therefore essential that the Guideline Development Group of the WHO that is meeting at the end of this month does not reach a decision in haste by the end of August as notified. Instead, sufficient time should be given to scholars, health authorities of the concerned countries, and civil society representatives to analyze the results of the ECHO trial and work out the policy implications.[2] Only can then a truly informed decision be made on an issue that is going to affect the lives of millions of women worldwide.

[1] See for instance, Guardian (2019), Walker (2019).

[2] The ECHO team has kindly agreed to share the data of the trial beginning 3 months following the publication of the article ending 36 months (ECHO Trial Consortium 2019, p.10)


Image Credits: ZaldyImg on Flickr


sathya 5.7.19About the author:

C. Sathyamala is a public health physician and an epidemiologist with a PhD in Development Studies from the ISS. Currently she is an academic researcher at the ISS.  

One comment

  1. WHO must wait on usage of these contraceptives that were on the Echo study due to unanswered questions that are still hanging.

Comment anonymously or log in