Over the past 25 years, as the HIV epidemic has spread, a number of observational studies were undertaken to explore whether or not the use of contraceptive methods increases the risk of HIV acquisition.

No association between combined hormonal contraceptive methods—which include the use of two hormones (estrogen and progestogen) to prevent pregnancy from occurring—and HIV acquisition has been shown. Few studies have examined whether hormonal implants or intrauterine devices (IUDs) affect users’ risk of HIV acquisition.

Some studies suggest increased risk of HIV acquisition among users of DMPA. Data from recent studies included in the latest systematic review strengthen concerns about a possible increased risk of HIV acquisition associated with DMPA use.1 After reviewing the evidence, a WHO technical consultation concluded in March 2017 that the available evidence continues to indicate an association between use of progestogen injectables and an increased risk of acquiring HIV; however, it is unknown whether the associations seen in these observational studies were due to a true biological effect or the limitations of such studies.

Uncertainty remains because all of the data are from observational studies, in which women choose which contraceptive methods they use. Women who choose to use DMPA may be different from women who do not use DMPA — particularly those who do not use contraception at all — in important ways that affect their risk of HIV and are difficult to measure.  The WHO statement notes that ‘data from observational studies come with a high level of uncertainty’, and calls for randomised clinical trials to provide better information about possible causality.

Randomised clinical trials, such as ECHO, are the gold standard for producing reliable scientific evidence. When women have an equal chance of using any of the three contraceptive methods under study, sexual behaviours and other factors that might influence HIV risk are equally likely to occur across the groups of women who are randomly allocated by computer to use one of the study methods. As a result, with a randomised design there is more certainty that a true difference in HIV acquisition is being measured and that any differences can be more certainly attributed to the contraceptive method used.


[1] Polis CB, Curtis KM, Hannaford PC, et al. An updated systematic review of epidemiological evidence on hormonal contraceptive methods and risk of HIV acquisition in women. AIDS 2016;30(17): 2665-83.