Increased rates of HIV infection were seen among women using depot medroxyprogesterone acetate (Depo-Provera or DMPA), whereas other methods of contraception including oral contraceptive pills or norethisterone enanthate (Net-En, an injectable progestin steroid) did not, according to a new meta-analysis.

Pooled data from 10 studies conducted in sub-Saharan Africa showed DMPA use was associated with a 40% higher HIV risk (hazard ratio 1.40, 95% CI 1.10-1.57) versus nonhormonal or no contraception, whereas 10 studies examining oral contraceptive pills (pooled HR 1.00, 95% CI 0.86-1.16) and five studies examining norethisterone enanthate (pooled HR 1.10, 95% CI 0.88-1.37) were not associated with increased HIV risk, according to Lauren Ralph, MPH, division of epidemiology of the School of Public Health at the University of California at Berkeley, and colleagues.

Ralph and colleagues wrote that controversy over DMPA has already caused some sub-Saharan African countries to consider withdrawing this form of birth control from family planning programs, which would leave women in these countries with fewer contraceptive options.

“The moderate elevation in risk for DMPA is not enough to justify a complete withdrawal of DMPA from women’s contraceptive options in most settings,” said lead study author Ralph in a press release. “This is likely to lead to more unintended pregnancies and their associated maternal and infant morbidity and mortality.”

For the meta-analysis, the authors reviewed 26 studies, searching PubMed for the terms “hormonal contraception,” “HIV/acquisition,” injectables,” “progestin” and “oral contraceptive pills.” The review included articles published in English after Dec. 1, 2011, as well as relevant abstracts from International AIDS Society meetings and the Conference on Retroviruses and Opportunistic Infections from 2011 to 2014.

Of these, 12 studies were designed to examine the link between hormonal contraception and HIV. The remaining studies sampled women from the general population at health centers and family planning clinics. The median age of participants in the studies ranged from 25 to 40 years.

The link between DMPA and HIV risk was also seen in these general population studies (pooled HR 1.31, 95% CI 1.10-1.57). But because the 10 studies from the primary analysis were associated with a greater level of heterogeneity (I²=42.5%, CI 0%-72%) than the general population studies (I²=27.3%, CI 0%-67%), most results of the primary studies were considered not applicable to the general population.

Two of 12 studies in the primary analysis examined high-risk populations, such asserodiscordant partners and commercial sex workers. Those two high-risk groups were associated with the greatest increases in HIV risk (HR 3.93, 1.37-11.2, and HR 1.73, 1.10-1.57, respectively). However, the high heterogeneity (I²=54.0%, CI 0%-88.7%) between these two studies meant their data couldn’t be pooled, the authors wrote.

Ralph told MedPage Today that there were only “a limited number” of studies that met the criteria for analyzing populations of high-risk women. “Thus, there remains uncertainty for this important subgroup of women,” she said.

While some of these findings may not be statistically significant, they may have clinical significance, including in developed countries. Diane Harper, MD, MPH, MS , professor, department chair for family and geriatric medicine at the University of Louisville School of Medicine in Kentucky, who was not involved with the analysis, suggested that it could help primary care and ob/gyn physicians who treat high-risk patients in the U.S., including serodiscordant couples, commercial sex workers, and drug addicts.

“I think there are valid populations in the U.S. for which we do need to pay attention and I think that the study very clearly says to me that there are so many other options of birth control that do not have this increased risk of HIV,” said Harper. “In these populations we should not be using Depo-Provera, that it just really is not a wise clinical choice.”

In an accompanying editorial, Christopher J. Colvin and Abigail Harrison, division of social and behavioral sciences of the School of Public Health and Family Medicine at the University of Cape Town in South Africa, called the environment surrounding any proposed trial about DMPA and HIV “polarizing” and arguments on both sides “rhetorical” and “generic.”

Colvin and Harrison wrote that studies such as this “synthesize and evaluate the existing observational data and identify many ways in which further modeling, epidemiological studies, behavioral and clinical research, and basic biological silence could work synergistically to increase our knowledge.”

Harper saw this limitation as one of the strengths of the study.

“What makes the paper so strong is that it has been able [to get these results] without increasing costs and doing another trial and putting more women at risk for HIV and repeating this in another population,” she said.

Another limitation cited by the authors was potential “confounding effects of misreported condom use,” especially since “many study populations were drawn from HIV prevention trials in which condom use is strongly encouraged and women may feel pressure to report socially acceptable behaviors,” they wrote. But they added that over-reporting of condom use was likely the same in all groups studied, including the reference samples.