360 • CID 2007:45 (1 August) • HIV/AIDS
H I V / A I D SI N V I T E D A R T I C L E
Kenneth H. Mayer, Section Editor
The Influence of Hormonal Contraceptive Use on HIV-1
Transmission and Disease Progression
Jared M. Baeten,1Ludo Lavreys,2and Julie Overbaugh3,4
Research Center, Seattle, Washington
2Epidemiology, University of Washington, and the Divisions of
3Human Biology and
4Public Health Sciences, Fred Hutchinson Cancer
Women account for nearly one-half of new human immunodeficiency virus type 1 (HIV-1) infections worldwide, including
the majority of infections in Africa. Biological and epidemiological studies suggest that hormonal contraceptive use could
influence susceptibility to HIV-1, as well as infectivity and disease progression for those who become infected. However, not
all studies have shown this relationship, and many questions remain. Safe and effective contraceptive choices are essential
for women with and at risk for HIV-1 infection. Thus, understanding the effect, if any, of hormonal contraception on HIV-
1 disease among women is a public health priority.
Worldwide, 17.3 million women are infected with HIV-1 .
Heterosexual intercourse is responsible for most HIV-1 trans-
missions to women, and the majority of infected women are
of reproductive age.
Hormonal forms of contraception, primarily oral contra-
ceptive pills and the injectable depot medroxyprogesterone ac-
etate (DMPA), are used by 1100 million women . Hormonal
contraceptives have physiological actions beyond pregnancy
prevention—both beneficial health effects and risks of adverse
health consequences—some of which were unanticipatedwhen
these products were developed [3, 4].
During the past ∼15 years, a number of biological and ep-
idemiological studies have suggested that the use of hormonal
contraception may influence susceptibility to HIV-1, infectivity
of HIV-1, and progression of HIV-1 disease, although this has
not been consistently seen in all studies. A conclusive relation-
ship between hormonal contraception and adverse HIV-1 out-
comes would be of considerable public health importance, be-
cause effective family planning services are central to initiatives
to slow population growth, promote economic development,
and improve the health of women and children worldwide.
Received 10 January 2007; accepted 4 April 2007; electronically published 18 June 2007.
Reprints or correspondence: Dr. Jared Baeten, Seattle HIV Prevention Trials Unit, 901 Boren
Ave., Suite 1300, Seattle, WA 98104 (email@example.com).
Clinical Infectious Diseases2007;45:360–9
? 2007 by the Infectious Diseases Society of America. All rights reserved.
Epidemiological and laboratory studies provide some insights
into possible biological mechanisms by which hormonal con-
traception could influence HIV-1. In a series of studies, Marx
and colleagues demonstrated that progesterone treatment of
female macaques increased susceptibility to vaginalinoculation
with simian immunodeficiency virus (SIV) , effects that
could be reversed by pretreatment with estrogen [6, 7]. Indeed,
progesterone has been used to increase susceptibility to SIV
among macaques in studies of candidate HIV-1 vaccines and
microbicides [8, 9]. The mechanism for these effects has not
been determined, although the authors of the original study
postulated that vaginal thinning may play a role. If altered
vaginal epithelial structure is central to the effect of progester-
one on increasing susceptibility to SIV in macaques, thesefind-
ings may have less relevance to humans, because studies of the
effect of DMPA among women have not shown thinning to
the extent seen in the nonhuman primate studies [10–12].
Cervical ectopy, or extension of the endocervical columnar
epithelium onto the exocervical face, has been associated with
hormonal contraceptive use [13, 14] and with increased sus-
ceptibility to HIV-1 . Ectopy is a common feature of the
immature cervix, which may contribute to thedisproportionate
risk of HIV-1 faced by young African women .
Genital tract infections may also mediate a relationship be-
tween hormonal contraception and susceptibility to HIV-1.Ep-
idemiological studies repeatedly have shown that users of hor-
monal contraceptives have increased risk of cervical chlamydial
infection [14, 17–20], which agreeswithanimalstudiesshowing
by guest on February 2, 2016
HIV/AIDS • CID 2007:45 (1 August) • 361
that estrogen and progesterone enhance growth of Chlamydia
trachomatis . Limited animal and human data suggest that
use of hormonal contraceptives increases susceptibility to gen-
ital herpes [22, 23]. DMPA may decrease vaginal colonization
by hydrogen peroxide–producing Lactobacillus species ,
which are protective against HIV-1 .
On a cellular level, hormonal contraceptives have been as-
sociated with cervical and vaginal inflammation [14, 25], in-
creased genital tract expression of the HIV-1 coreceptor CCR5
[26, 27], and mucosal and systemic immune responses that
could mediate susceptibility to HIV-1 [28–30]. Finally, hor-
mones may directly enhance the replication of the virus itself
, which could affect both earlyinfectionandthesubsequent
Collectively, there are data to argue biological plausibilityfor
hormonal contraception to influence HIV-1, especially initial
susceptibility to the virus. Given the multiple mechanisms pro-
posed, the effect may be multifactorial or indirect. However, it
is important to recognize that most of these potentialbiological
mechanisms are proposed on the basis of limited data, and it
is difficult to discern which, if any, are actually relevant to the
risk of HIV-1 transmission.
EPIDEMIOLOGICAL STUDIES OF HIV-1
More than 50 studies have explored whether use of hormonal
contraception is a risk factor for HIV-1 infection. Most were
cross-sectional, whereas 15 were prospective studies (table 1)
[16, 32–45]. A meta-analysis of 28 studies (7 prospective) pub-
lished in 1999 found that use of oral contraceptive pills was
1.19; 95% CI, 0.99–1.42), with greater risk among 7 studies
from Africa (OR, 1.65; 95% CI, 1.09–2.52) . Five pro-
spective studies published before 1999 examined the risk of
HIV-1 acquisition associated with use of DMPA: 2 found sta-
tistically significantly increased risks [40, 41], 2 found non–
statistically significantly increased risks [37, 42], and 1 dem-
onstrated no association . Nearly all of these studies were
conducted in Africa, where heterosexual transmission of HIV-
1 infection predominates.
Since 1999, 3 large prospective studies have been reported.
The first included 5117 women in Rakai, Uganda; 202 acquired
HIV-1 (incidence, 1.5 cases per 100 person-years) . This
cohort was followed from 1994 to 1999 for a community-
randomized trial of sexually transmitted disease (STD) treat-
ment for prevention of HIV-1 infection . Follow-up, in-
cluding measurement of current contraception and HIV-1
status, occurred at 10-month intervals. In unadjusted analysis,
hormonal contraceptive use was associated with increased risk
of acquisition of HIV-1 (risk ratio [RR], 1.56; 95% CI, 1.00–
2.33). However, after controlling for demographic factors and
measures of sexual behavior, no association remained (RR,
0.94). Similar results were found in separate analyses of oral
contraceptive pill and DMPA use.
In 2004, we reported the results of aprospectivestudyamong
basa, Kenya, between 1993 and 2003 . A principal aim of
this study was to describe risk factors for HIV-1 acquisition,
including hormonal contraception . At monthly follow-up,
HIV-1 status and contraceptive method were measured. There
were 248 seroconversions (incidence, 8.5 cases per 100 person-
years). Both oral contraceptive pills (hazard ratio [HR], 1.5;
95% CI, 1.0–2.1) and DMPA (HR, 1.8; 95% CI, 1.4–2.4) were
associated with greater risk of HIV-1, compared with nonuse
of contraception, in multivariate analyses adjusted for demo-
graphic characteristics, sexual behavior, condom use, and in-
The largest study to measure the risk of HIV-1 infection
associated with hormonal contraceptive use has recently been
published . The study recruited 4439 women from family
planning clinics in Uganda and Zimbabwe specifically to ad-
dress this question among a low-risk population. At 3-monthly
intervals, contraceptive methodwasrecordedandHIV-1testing
was performed. There were 213 HIV-1 seroconversions (inci-
dence, 2.8 cases per 100 person-years). Overall, neither oral
contraceptive pills (HR, 0.99; 95% CI, 0.69–1.42) nor DMPA
(HR, 1.25; 95% CI, 0.89–1.78) was associated with HIV-1 ac-
quisition. However, among women who were negative for her-
pes simplex virus type 2 (HSV-2) at study enrollment (48% of
the study population), both methods increased risk of HIV-1
acquisition (for oral contraceptive pills: HR, 2.85; 95% CI,
1.39–5.82; for DMPA: HR, 3.97; 95% CI, 1.98–8.00), a finding
that was robust in multiple sensitivity analyses.
It is difficult to reconcile the divergent results of studies ex-
amining the effect of hormonal contraceptive use on risk of
HIV-1 acquisition. Indeed, a 1998 review argued that substan-
tial differences among studies made summarization of the data
Several methodological issues are important to consider.
First, imprecision in measurement of the timing of hormonal
contraceptive use relative to HIV-1 acquisition could have in-
troduced bias . This is most problematic for the cross-
sectional studies, although several prospectivestudiesmeasured
contraceptive use only at study enrollment [38, 40] or infre-
quently during follow-up [37, 43]. Similarly, infrequent mea-
surement of HIV-1 infection status limited the ability to know
with precision which contraceptive method was used at the
time of HIV-1 acquisition.
Second, differences in sexual behaviors between users and
nonusers of hormonal contraception may have led to con-
by guest on February 2, 2016
HIV/AIDS • CID 2007:45 (1 August) • 369
86. World Health Organization (WHO). Hormonal contraception and
HIV: science and policy. Availableat:http://www.who.int/reproductive-
health/stis/hc_hiv/nairobi_statement.pdf. Accessed 1 May 2006.
87. Balkus J, Bosire R, John-Stewart G, et al. High uptake of postpartum
hormonal contraception among HIV-1-seropositive women in Kenya.
Sex Transm Dis 2007;34:25–9.
88. World Health Organization (WHO). Prevention of HIV in infants and
young children: review of evidence and WHO’s activities. Available at:
http://www.who.int/hiv/mtct/ReviewofEvidence.pdf. Accessed 13 Sep-
89. World Health Organization (WHO). Medicaleligibilitycriteriaforcon-
traceptive use: Available at: http://www.who.int/reproductive-health/
publications/mec/mec.pdf. Accessed 14 September 2006.
by guest on February 2, 2016