Individuals' Perceptions About Their Sex Partners' Risk Behavior

Journal of Sex Research, Nov, 1998 by Jonathan M. Ellen, Eric Vittinghoff, Gail Bolan, Cherrie B. Boyer, Nancy S. Padian

A goal of most sexually transmitted disease (STD) and human immunodeficiency virus (HIV) prevention programs is to reduce the prevalence of high risk sexual behaviors, such as inconsistent condom use (Pulley, McAlister, Kay, & O'Reilly, 1996; Wasserheit & Aral, 1996; Wight, 1993). One potential strategy to reduce the prevalence of high risk sexual behaviors is to encourage individuals to have open discussions with their sex partners about their partners' previous and current private risk behaviors, such as number of previous sex partners and current and previous drug use (Catania, Coates, Golden, et al., 1994; Catania, Coates, & Kegeles, 1994; Kelly & Kalichman, 1995; Lewis, Malow, & Ireland, 1997; Sieving et al., 1997). Decision-making theories such as the Theory of Planned Behavior and the Health Belief Model suggest that perceptions of personal risk may affect sexual behavior (Ford & Norris, 1 995; Hale & Trumbetta, 1996; Wulfert, Wan, & Backus, 1996; Mahoney, Thombs, & Ford, 1995; Gielen, Faden, O'Campo, Kass, & Anderson, 1994). Empirical research illustrates the fact that individuals' perceptions about their sex partners' risk behaviors may influence their perceptions of personal risk (Kelly & Kalichman, 1995; Lear, 1995; Moore & Rosenthal, 1992; Prohaska, Albrecht, Levy, Sugrue, & Kim, 1990). For example, Prohaska et al. (1990) found that perceptions of sex partners' past sexual behaviors are associated with perceptions of higher risk. Our own experimental work has shown that perceptions of a sex partner's infection status affects perception of risk for STDs and HIV (Ellen, Orr, Langefeld, & Boyer, 1997). Information about the accuracy of individuals' perceptions about their sex partners' previous or current private risk behaviors might help guide the development of more effective STD and HIV prevention programs. However, it is difficult to determine accuracy of perceptions about sex partners behaviors since sex partners' self-reported private risk behaviors are hard to verify. To the extent that main sex partners honestly and reliably report their risk behaviors, the amount of agreement between individuals and their partners may represent a reasonable estimate of the accuracy of individuals' perceptions about their main sex partners' risk behaviors.

Studies have compared the extent of agreement between individuals' perceptions and their sex partners' reports of shared behaviors such as sexual intercourse and condom use (Coates et al., 1988; Ellish, Weisman, Celentano, & Zenilman, 1996; Lagarde, Enel, & Pison, 1995; Padian, 1990; Padian, Aral, Vranizan, & Bolan, 1995; Seage, Mayer, Horasburgh, Cai, & Lamb, 1992; Upchurch et al., 1991; Wells et al., 1994). These studies suggest that the extent of agreement between sex partners shifts with the behavior and study population. For example, Ellish et al. (1996) found that among a sample of heterosexual couples recruited from an STD clinic, there was poor agreement between sex partners about their use of a condom the first time they had vaginal sexual intercourse (kappa score = 0.41). In contrast, among male homosexual couples recruited from a non-STD clinic setting there was substantial agreement about whether condoms were ever used for anal intercourse (kappa score = 0.80).

Only one study specifically examined the accuracy of couples' perceptions about sexual behaviors that they did not necessarily engage in together (i.e., private behaviors), such as sex with other partners and drug use (Seal, 1997). In this sample of couples recruited from a university, the percent agreement between individuals' perceptions of their sex partners' risk behaviors and their sex partners' self-reported behavior was generally low. For example, only 24% of women and 41% of men agreed with their sex partners' self-report of having had an extra-dyadic sex partner. In addition, only 50% of women and 31% of men agreed with their sex partners' self-reported history of an STD.

The objective of this study was to determine the agreement between an individual's perceptions about their main sex partner's private risk behaviors and their main sex partner's self-reported private risk behaviors. This study adds to the current literature by focusing on a population at high risk for STDs and HIV (i.e., an STD clinic sample), and by examining the extent of agreement about STD/HIV-related risk behaviors such as injection drug use and crack use.

METHODS

Sample

This study was conducted between July 1991 and January 1992 at San Francisco's only public STD clinic that has the approval of our institution's review board. This convenience sample consisted of heterosexual clients and their main sex partners. Heterosexual clients age 18 or older who had a current main partner were eligible. Main partner was defined as the client's current "steady" partner. In order to participate, eligible clients and their main sex partners had to provide verbal consent to an interview and physical examination.

Consecutive clients presenting to the clinic during scheduled recruitment times were approached and screened for eligibility. The partner was recruited after obtaining consent to contact the main partner from the participant seen at the clinic. In cases where the main partner accompanied the client to the clinic, they were recruited by a research assistant (60% of eligible couples). In cases where the main partner had not accompanied the client to the clinic, the research assistant contacted and recruited the main partner by telephone using approved protocols for partner notification. Recruitment continued until 100 couples were enrolled.