lack of circumcision and increased risk for HIV infection (Moses et al., 1994). In a prospective study of men at high risk for STDs, those who were not circumcised were 8 times as likely to become infected with HIV than circumcised men (Cameron et al., 1989). Another study of gay men suggested that uncircumcised men were twice as likely to be infected with HIV compared to circumcised men (Kreiss and Hopkins, 1993). As a result of these studies, some have proposed that male circumcision be considered an intervention to prevent HIV infection. Several studies have found associations between lack of circumcision and other STDs, including chancroid (Aral and Holmes, 1990). It has been hypothesized that lack of circumcision increases risk for STDs because (a) the cells that line the fold of skin that is removed by circumcision are prone to trauma or infection, (b) this fold of skin may serve as a reservoir for pathogens, and (c) this fold of skin may increase the likelihood that infections will go undetected (Aral and Holmes, 1990).
Vaginal douching seems to increase risk for pelvic inflammatory disease (Forrest et al., 1989; Wolner-Hanssen, Eschenbach DA, Paavonen J, Stevens CE, et al., 1990;). In one study, compared to women who did not douche, women who douched during the previous 3-month period were twice as likely to have clinical pelvic inflammatory disease (Scholes et al., 1993). The risk for pelvic inflammatory disease seems to increase with greater frequency of douching (Wolner-Hanssen, Eschenbach DA, Paavonen J, Stevens CE, et al., 1990; Scholes et al., 1993).
Certain sexual practices such as receptive rectal intercourse predispose to STDs. As mentioned in Chapter 2, STDs such as HIV infection and hepatitis B virus infection are more easily acquired by rectal intercourse than by vaginal intercourse. This may be because the bleeding and tissue trauma that can result from rectal intercourse facilitate invasion by pathogens. Other sexual practices, such as sex during menses and "dry sex," also predispose to acquisition of an STD.
The influence of hormonal contraceptives on acquisition and transmission of STDs is not fully defined. However, several studies have found oral contraceptive use to be associated with increased risk of acquiring chlamydial infection (Critchlow et al., 1995) but with decreased risk of developing pelvic inflammatory disease among women with chlamydial infection (Wolner-Hanssen P, Eschenbach DA, Paavonen J, Kiviat N, et al., 1990; Kimani et al., 1996). Some, but not all, studies have found an association of oral contraceptives with increased risk of HIV acquisition (Cates, in press). A recent study in Kenya has demonstrated that use of oral contraceptives or injectable progesterone among women with HIV-1 infection is associated with increased shedding of HIV-1 DNA from the cervix (Mostad et al., 1996). In one animal model study, monkeys with progesterone implants were several times more likely to become infected with the simian immunodeficiency virus than monkeys who did not have such implants (Marx et al., 1996). More study is indicated, but these data raise the