Sexual decision making and safer sex behavior among young female injection drug users and female partners of IDUs.

A current public health priority is strengthening women's ability to protect themselves from HIV and other sexually transmitted diseases (STDs) and to negotiate safer sex behaviors with their heterosexual partners. The number of AIDS cases among women has steadily increased (Centers for Disease Control and Prevention [CDC], 1999), and approximately half of all new cases occur in persons under the age of 25, with disproportionate numbers among young women and ethnic minorities (CDC, 2000). Since 1994, heterosexual transmission has been the most common mode of HIV transmission in women (CDC, 1999) and has accounted for 38% of all female AIDS cases reported in 2000 (CDC, 2000).

Through December 2000, at least 57% of all accumulated AIDS cases among women were attributed either to their own injection drug use (IDU; 41%) or to sex with an IDU partner (16%), compared with only 24% due to either of these exposures among men (CDC, 2000). Young women who inject drugs may be at even higher risk than males for a number of reasons. Female injection drug users (IDUs) are more likely than males to also have a sex partner who injects (Booth, 1995; Davies, Dominy, Peters, & Richardson, 1996), with needle sharing and sexual risk taking co-occurring (Baker, Kochan, Dixon, Wodak, & Heather, 1994; Booth, 1995). Female IDUs also have higher overall levels of sexual risk taking including having sex with three or more partners, exchanging sex for money or drugs, or having a partner who injects drugs or has sex with males (Sly, Quadagno, Harrison, Eberstein, & Riehman, 1997). Thus, young IDU women or women who have an IDU sex partner are considered to be at potentially high risk for HIV due to both their sexual and their drug-related risk behaviors.

Moreover, young people who are not in school and are homeless and/or who are involved in the street economy are known to have higher levels of IDU compared with their peers (1.5-2% vs. 5-39%; Johnston, O'Malley, & Bachman, 2000; Kipke, O'Connor, Palmer, & MacKenzie, 1995). In addition, compared with IDUs who begin injecting at an older age, IDUs who initiate use during adolescence show an overall higher drug risk profile (Battjes, Leukefeld, & Pickens, 1992), engage in more high risk sexual behaviors (Nelson, Vlahov, Solomon, Cohn, & Munoz, 1995), and are more vulnerable to HIV, Hepatitis C, other STDs, and infectious diseases (Bailey, Camlin, & Ennett, 1998; Clements, Gleghorn, Garcia, Katz, & Marx, 1997; DeMatteo et al., 1999; Ennett, Federman, Bailey, Ringwalt, & Hubbart, 1999; Kipke, Montgomery, Simon, Unger, & Johnson, 1997).

Finally, engaging in unprotected sexual intercourse puts women at risk not only for STDs including HIV but also for unintended pregnancy. Despite the availability of highly effective methods of contraception, almost a third (30.8%) of births and one half of pregnancies in the United States are unintended (Henshaw, 1998). Pregnancies among adolescents and young women ages 20 to 24 are particularly likely to be described as unintended (Brown & Eisenberg, 1995). Moreover, the realities of street life--including substance abuse, poor access to health care, food, and shelter, and basic hygienic and personal security needs--put female street youth at increased risk of unintended pregnancies (Ensign & Santelli, 1998; Green, Ennett, & Ringwalt, 1999; Green & Ringwalt, 1998; Noell, Rhode, & Ochs, 1997) and the serious negative consequences associated with these pregnancies (Brown & Eisenberg, 1995). Despite these trends, few studies have been conducted among young injection-drag-using populations, especially those that are street involved, and in the existing studies young women tend to be underrepresented.

In the absence of an effective vaccine or cure for AIDS, efforts to reduce the risk of HIV infection among women have focused on behavioral risk reduction. However, to design effective interventions for women, more empirical research is needed to understand what factors influence women's safer sex behavior. Despite the growing proportion of HIV infections among women that are due to heterosexual exposure, until recently much of the theory and research on HIV behavior change has been individualistic in conceptualization, often ignoring the social and cultural context of women's sexual behaviors (Amaro, 1995; Amaro & Raj, 2000; Wingood & DiClemente, 2000). These theories, such as the health belief model (Becker, 1974), theory of reasoned action (Azjen & Fishbein, 1977), social cognitive theory (Bandura, 1994), transtheoretical stages of change model (Prochaska & DiClemente, 1983, 1984), and AIDS risk reduction model (Catania, Kegeles, & Coates, 1992), assume that the individual has total control over behavior and do not address the diverse contextual factors related to gender (e.g., power differentials, gender roles, relationship types) that likely influence sexual behavior (Amaro, 1995). Unlike other risky behaviors such as smoking, drinking, and drag use, the transmission of HIV through sexual intercourse involves an interaction with another person. In addition, broader socio-cultural factors such as inequities in pay, discrimination based on gender, and inequalities in heterosexual relationships contribute to a set of life conditions that could greatly impact a woman's risk for HIV (Amaro, 1995; Amaro & Raj, 2000; Wingood & DiClemente, 2000). Consensus is mounting among theorists and researchers that HIV risk in women must be seen within the larger context in which women live. According to Amaro (1995, p. 445), "risk of HIV infection in women cannot be separated from the unequal status of women in American society and the resulting differences in power between men and women."

Similarly, in a recent article, Wingood and DiClemente (2000) applied and extended the theory of gender and power developed by Connell (1987) to define the exposures and risk factors that put women at increased vulnerability to HIV. They provided a theoretical framework for understanding the risk factors for HIV among women by examining the three major structures that characterize the gendered relationships between men and women: the sexual division of labor, the sexual division of power, and the structure of cathexis. Because the structure of cathexis emphasizes affective and normative components of relationships, Wingood and DiClemente (2000) refer to it as the structure of affective attachments and social norms. Of relevance to this paper is their conceptualization of how the sexual division of power influences women's risk of HIV. They defined behavioral risk factors to include the lack of perceived control over or power to avoid unhealthy behaviors and limited perceived control over condom use.

Male condoms are currently the only widely accepted method for preventing the transmission of HIV and other STDs for sexually active individuals. Condom use is a unique health behavior because it involves either an implicit or explicit agreement between both partners. However, because condoms are male-directed, use of this method requires the participation, or at least cooperation, of the male partner. In other words, women do not wear condoms, men do. Thus, to protect themselves from HIV/STDs, sexually active women must successfully influence the behavior of their male partners. This need creates a dynamic that may be complicated by interpersonal power and relationship factors. Because men wear condoms, power in sexual decision making may be a more significant determinant of safer sex behavior for women than for men.

It has been suggested, and some studies indicate, that women perceive that they do not have power in their relationships and, therefore, have difficulty negotiating safer sex strategies with their male partners (e.g., Amaro, 1995; Amaro & Gornemann, 1992; Fullilove, Fullilove, Haynes, & Gross, 1990; Gomez & Marin, 1996; Wingood & DiClemente, 2000; Wingood, Hunter-Gamble, & DiClemente, 1993; Worth, 1989). Several studies suggest, however, that many women believe that they do have power in their relationships and that they are involved in making decisions about sexual behavior and contraceptive and condom use (e.g., Cabral, Pulley, Artz, Brill, & Macaluso, 1998; Eldridge, St. Lawrence, Little, Shelby, & Brasfield, 1995; Harrison, Kay, Dixon, Peters, & Moore, 1996; Harvey, Beckman, & Doty, 1999; Harvey, Bird, Galavotti, Duncan, & Greenberg, in press; Kline, Kline, & Oken, 1992; Sobo, 1995; Soet, Dudley, & Dilorio, 1999; Wyatt et al., 2000).

Furthermore, of those studies that have explored these issues, only a few have empirically examined the impact of relationship power and sexual decision making on women's safer sex behavior, and findings from these studies are mixed and inconclusive. Some studies suggest that when women have power in their sexual relationships, it increases the likelihood that condoms will be used (e.g., Harvey et al., in press; Pulerwitz, Gortmaker, & DeJong, 2000; Wingood & DiClemente, 1998). Other studies indicate that lack of power in interpersonal relationships is not a major obstacle to condom use for women (e.g., Cabral et al., 1998; Soler et al., 2000). Differences in findings may in part be attributed to how relationship power is measured. For example, in a recent study with women at high risk for HIV/STDs (Harvey et al., in press), we found that our general measure of relationship power (i.e., in general who in the relationship has more power) did not predict actual condom use. Our domain-specific measure of who decides to use condoms, however, did predict use, with women who dominated or shared condom use decision making being more likely to report recent use. In the present study, we use a similar measure to assess sexual decision making in the relationship.

In addition to power in sexual decision making, the type of relationship a women is involved in may also be an important determinant of safer sex behavior. Previous research has documented that women's perception of HIV risk and their practice of safer sex behavior varies by type of relationship. More specifically, women tend to perceive steady long-term partners as less risky than casual partners or partners of short duration. Accordingly, they tend to engage in safer sex behaviors and to use condoms more with casual partners or partners of short duration than with long-term partners or partners in committed relationships (summarized in Misovich, Fisher, & Fisher, 1997).

Additional research is needed to improve our understanding of the role that relationship power and decision making and relationship type play in women's safer sex behaviors. Few studies to our knowledge have examined the role of control over sexual decision making and safer sex behaviors among young female IDUs and female partners of IDUs (e.g., Kline et al., 1992). The present analyses are part of a larger study that used a social network approach to help provide a better understanding of the social contexts, gender issues, and risk and protective behaviors of young IDUs, their sexual partners, and their social network members. Because prior research on sexual decision making and its relationship to contraceptive and condom use is limited and inconclusive, the overall purpose of this analysis was to examine the relationships among these variables among young female IDUs and female partners of IDUs. We have chosen to limit the focus of our study to women because the sexual division of power is hypothesized to negatively influence women's, not men's, risk of HIV. Accordingly, the objectives of this study are (a) to examine perceptions of sexual decision making among a sample of young female injection drug users (IDUs) and female partners of IDUs, (b) to examine the association between partner characteristics (e.g., main vs. other) and sexual decision making, and (c) to explore the relationship between sexual decision making and condom and contraceptive use.

METHOD

Sample

Respondents were recruited as part of a larger project that explored the social contexts and risk behaviors of young IDUs, their sexual partners, and their social network members. Recruitment procedures are discussed in detail elsewhere (Montgomery et al., in press). In brief, youths were eligible to participate if they were between 15 and 23 years old and reported injecting drugs and/or having sex with an IDU in the last 30 days. Purposive sampling techniques were used to recruit 193 youths (the "core respondents") from needle exchange service centers, drop-in centers for homeless people, and street-based sites in Southern California. Ethnographic research revealed where IDUs could be contacted. Trained interviewers screened all youths at each recruitment site who appeared to be eligible for the study based on age. To maximize the number of female core respondents, we gave young women a higher priority in our recruitment efforts. Interviewers screened females first when they approached a group of youths, and only females were screened on certain designated weeks. If eligible young people agreed to participate, either they were taken to a private location such as a cafe, park, or an office at a community-based agency for the consent procedures and an interview, or they were scheduled for an appointment at a future time. In addition, we used snowball sampling to recruit 127 members of the core respondents' social networks (the "referral respondents"; there was no age requirement for referral respondents). For this study, we combined the core and referral respondents (N = 320), of whom 126 (39%) were female. We further limited the sample to female respondents who reported having one or more male sex partners in the past 30 days (N = 94). These respondents are described below.

Procedure

Eligible individuals who agreed to participate in the study were interviewed in a quiet location. Prior to administering the survey, the interviewers discussed the study protocol and obtained written informed consent. A trained team of interviewers who were familiar with the target population and similar to them with respect to age and ethnicity conducted the interviews. The interviews lasted between 60 and 90 minutes. At the completion of the interview, respondents were given $20 in gift certificates from a local grocery store in compensation for their time. The study procedures were approved by the Institutional Review Board of the Childrens Hospital Los Angeles.

Measures

We developed the interview guide specifically for this study. Questions were close-ended and existing scales were adapted for use with this population. The guide included two sections: one that consisted of questions about the respondents' characteristics and behavior, and one that included questions about the respondents' social network members (i.e., the individuals with whom they had spent the most time, injected drugs, and/or had sex in the past month) and the respondents' behavior with those network members. This analysis focuses on participants' responses to questions concerning their own sexual decision making with male sex partners (i.e., male partners with whom they had had sex during the past month). The measures used in the study are described below.

Respondent characteristics. Respondents were asked questions to assess their age, ethnicity, sexual orientation, education, homelessness, childhood physical and sexual abuse, involvement in prostitution, history of foster home or group home placement, and drug use. They were also asked to report whether a doctor, psychologist, therapist, or psychiatrist had ever told them that they had a mental health or psychological problem. In addition, we collected data on whether they had ever been diagnosed with a mental disorder, had ever been tested for HIV, or had ever tested positive for HIV.

Condom use self-efficacy. To measure self-efficacy regarding condom use, we asked respondents three questions: (a) "How confident are you that you could ask a sex partner to use a condom?"; (b) "If you had been having sex with someone for a while and not using condoms, how confident are you that you could ask your partner to start using one?"; and (c) "How confident are you that you would be able to refuse sex if you wanted to use a condom and your partner did not?" We scored responses on a 4-point scale ranging from not confident at all to very confident. We created a dichotomous measure of condom use self-efficacy by categorizing respondents according to whether they reported being very confident in response to all three questions versus being somewhat confident, not very confident, or not confident at all in response to one or more of the questions.

Partner-specific relationship factors. For each of their male sex partners, respondents were asked to report relationship type and the length of sexual relationship. The question used to measure relationship type was "Would you describe [name] as a steady or main partner (that is, someone that you have sex with on a regular basis)?" Response categories were yes and no. Length of sexual relationship was assessed with the question "How long have/had you and [name] been having sex?" Response categories ranged on a 7-point scale from one day to more than one year. In addition, respondents reported the likelihood (on a 4-point scale ranging from not likely at all to very likely) that their male sex partner(s) would (a) get upset, (b) get violent, or (c) refuse to have sex if the respondents asked them to use condoms.

Sexual decision making. For each of their male sex partners, respondents were asked the following questions: (a) "When you and [name] have sex, who makes decisions about condom use?"; (b) "When you and [name] have sex, who makes decisions about birth control?"; and (c) "When you and [name] have sex, who makes decisions about when to have sex?" Responses were self, partner, both, and neither. Because of the small sample size and distribution of responses to these questions, we collapsed the responses into fewer categories for analysis of the relationship between participation in sexual decision making and condom and contraceptive use. Specifically, we dichotomized the responses into self or both and partner or neither. In the results section, we refer to these two groups as I/we and he/neither, respectively. We combined the responses in this manner because we wanted to measure whether or not a woman perceived that she was an active participant in decision making regarding condom use, contraceptive use, and when to have sex, regardless of her partner's involvement.

Condom and contraceptive use. For each of their male sex partners, respondents were asked, "Of the times you and [name] had vaginal sex in the past month, how often did you use a condom?" Response categories were none of the time, less than half the time, half the time, more than half the time, and every time. For some analyses, we dichotomized this variable by combining the last four response categories. The new measure was labeled used condoms in the past month and response categories were no (i.e., none of the time) and yes (i.e., less than half the time or more frequently). (Because only seven respondents reported having had anal sex with their male partners in the past month, and none reported using condoms with anal sex in the past month, we did not include condom use during anal sex in our measurement of condom use.) Respondents were also asked, "Of the times you and [name] had vaginal sex in the past month, how often did you do or use something to prevent pregnancy?" Response categories also ranged from none of the time to every time. For some analyses, we created a dichotomous measure: used birth control in the past month (response categories no and yes). (1)

RESULTS

Analysis

As noted above, for this study, we limited the sample to those respondents who reported having one or more male sex partners in the past month (N = 94). Seventy-five (80%) reported only one male sex partner in the past month, 13 (14%) reported two male sex partners, and 6 (6%) reported three male sex partners. Table 1 shows the distribution of respondents by number and type of male sex partner(s). For each of the 19 respondents who reported two or three male sex partners, we randomly selected one of their male sex partners and analyzed data for that partner only, so that we did not violate assumptions of independence.

We generated frequency distributions and descriptive statistics for all variables of interest. We conducted bivariate analyses to examine the relationships between participation in sexual decision making and relationship type, relationship length, condom use, and birth control use. We conducted two multiple logistic regressions to examine the relationship between participation in sexual decision making and condom and birth control use in the past month, having adjusted for selected potential covariates (i.e., age, education, number of male partners, condom use self-efficacy, relationship type, and length of sexual relationship).

Respondent Characteristics

The characteristics of the respondents are presented in Table 2. As shown, nearly half were between the ages of 18 and 20. The majority were White, had less than a high school education, and were currently homeless. Roughly half said that they were heterosexual and half said they were bisexual; it is important to note, however, that only 6% reported having a female sex partner during the past month (female partner data not shown in table). Nearly 40% reported that they had experienced childhood physical abuse, and almost 30% had said that they experienced childhood sexual abuse. Most (60%) reported that they had been told by a health care provider that they had a mental problem. The vast majority said that they had been tested for HIV, but none reported being HIV positive. Most respondents reported that they were currently using IV drugs (73%), and over 80% said they were using non-IV drugs. The majority of respondents indicated that they were confident that they could ask a partner to use a condom and could refuse sex if the partner did not want to use one. (As noted in the Method section, this measure was not about a specific partner.)

Respondents' Relationships With Male Sex Partners

As shown in Table 3, the majority of respondents considered their male sex partners to be "main" partners. Roughly a third of respondents reported having sex with their partners for 1 month or less, a third said that they had been having sex between 1 and 6 months, and a third reported they had been having sex for more than 6 months. The vast majority of respondents did not perceive that their partners would have negative reactions to requests for condom use. Almost all reported having had vaginal sex with their partners in the past month, but few reported having recent anal sex with their partners. Nearly half indicated that they had used condoms during vaginal sex with their partners in the past month. Over two thirds reported that they had used birth control with their partners in the past month (60% of whom indicated that they most often used condoms, alone or with another method, as their birth control method).

Participation in Sexual Decision Making

The majority of respondents perceived that they were active participants in sexual decision making. Over 70% reported that they participated in decisions about condom use: About a quarter (25%) indicated that they made decisions about condom use alone, and nearly half (48%) said they made such decisions with their partners. Only 8% said that their partners made decisions about condom use alone. Nearly one fifth (19%) reported that neither they nor their partners made decisions about condom use.

Over three quarters of respondents reported that they made decisions about birth control, either alone (47%) or with their partners (32%). None of the respondents reported that their partners made decisions about birth control without their involvement. Over one fifth (22%) said neither they nor their partners made decisions about birth control.

Finally, 85% indicated that they participated in deciding when to have sex; two thirds (67%) said that they made those decisions with their partners, whereas less than one fifth (18%) said that they made them alone. Although 13% said that their partners made decisions alone about when to have sex, only 2% reported that neither they nor their partners made decisions about when to have sex.

Relationship Characteristics and Participation in Sexual Decision Making

Participation in sexual decision making was unrelated to relationship type (main or non-main; see Table 4). Relationship type was unrelated to self-reported participation in decisions about condom use ([chi square][1, n = 91] = 0.56, p = .45), birth control ([chi square][1, n = 92] = 1.74, p = .19), or when to have sex (Fisher's Exact Test, p = .06, n = 93).

In contrast, women's participation in sexual decision making was associated with relationship length for two of the three decision domains. For decisions about birth control, t(90) = 2.26, p < .05, and decisions about when to have sex, t(91) = 2.25, p < .05, respondents who said that I/we made those decisions reported being in longer sexual relationships than did respondents who reported that he/neither made those decisions (see Table 4). Participation in decisions about condom use, however, was not related to relationship length, t(89) = 0.88, p = .38.

Participation in Sexual Decision Making and Condom Use

Results from bivariate analyses indicated that respondents who reported participating in sexual decision making also reported greater condom use. More specifically, the Mann-Whitney U Test indicated that respondents who said I/we made decisions about condom use reported using condoms with their partners more often than those who said he/neither made condom use decisions, U = 508.00, p < .01, n = 91. Among the 64 respondents who said I/we made decisions about condom use, 61% reported using condoms with their partners during the past month, compared with 25% of the 24 who said he/neither made condom use decisions, [chi square](1, n = 88) = 9.02, p < .01. In addition, 56% of the 77 respondents who said they participated in making decisions about when to have sex reported that they had used condoms with their partners during the past month, compared with only 15% of the 13 who indicated that they did not participate in those decisions, [chi square](1, n = 90) = 7.28, p < .01.

Table 5 presents the results from the multiple logistic regression analysis that assessed the relationship between participation in sexual decision making and condom use in the past month, adjusting for potentially confounding variables. As shown, even when adjusted for age, education, number of male partners, condom use self-efficacy, and relationship characteristics, respondents who reported that they participated in making decisions about condom use with their partners were 7 times more likely to report having used a condom with those partners in the past month than were those who indicated that they were not involved in such decisions. Furthermore, the adjusted odds ratios indicate that respondents who said that they participated in decisions about when to have sex were 19 times as likely to report using condoms with their partners in the past month than were those who indicated that they did not have a role in those decisions. In addition, when adjusted for the other variables included in the regression, women who reported that their sexual relationships were of longer duration were less likely than women who said they were in newer relationships to indicate that they used condoms with their partners.

Participation in Sexual Decision Making and Birth Control Use

Self-reported participation in sexual decision making was also associated with reports of birth control use. Respondents who said I/we made decisions about birth control reported using contraceptives with their partners more often in the past month than did those who said he/neither made these decisions, U = 359.50, p < .01, n = 89. In addition, 80% of the 69 respondents who said I/we made decisions about birth control reported using contraceptives with their partners during the past month, compared with only 30% of the 20 who said he/neither made these decisions, [chi square](1, n = 89) = 17.77, p < .001. In addition, 77% of the 64 respondents who reported participating in decisions about using condoms reported using contraceptives with their partners in the past month, compared with 50% of the 24 who indicated that they were not involved in these decisions, [chi square](1, n = 88) = 5.79, p < .05.

Adjusted odds ratios from the multiple logistic regression analysis that examined the relationship between participation in sexual decision making and birth control use are presented in Table 6. As shown, when adjusted for potentially confounding variables, respondents who reported that they participated in decisions about birth control were 20 times more likely to report that they had used birth control with their partners during the past month than were those who said that they were not participating in decisions about birth control. In addition, the odds ratios indicate that, when adjusted for the other variables in the regression, women who indicated that they had two or more male partners and women who reported being in longer relationships were less likely than other respondents to report that they had used birth control with their partners in the past month.

DISCUSSION

In this study we sought to explore women's perceptions of their role in sexual decision making and how that role influences safer sex behaviors including both condom and contraceptive use. Because previous studies indicate that young injection drug using populations, especially those that are street involved, are at increased risk for STDs and unintended pregnancies, we examined sexual decision making among a sample of young female IDUs and female partners of IDUs.

It is noteworthy that the majority of the women in our study were White and reported having been tested for HIV and being HIV negative. Given that our sample consisted of homeless youths in Los Angeles, the low percentage of minorities is not surprising. Previous research with homeless youths in Hollywood found that White youths more frequently reported a history of injection drug use and symptoms of drug and/or alcohol abuse disorders than did youths from ethnic minority groups (Kipke et al., 1997). In addition, previous findings indicate that young people at risk for HIV infection who live in Los Angeles are more likely to get tested for HIV compared with similar groups of young people in other geographic areas (De Rosa, Montgomery, Hyde, Iverson, & Kipke, 2001).

Although it may seem unlikely that there were no HIV-positive young women in this sample, we have several reasons to believe that their self-reported HIV serostatus was accurate. First, geographic data from IDUs have repeatedly found lower seroprevalence in Los Angeles and other west coast samples than in other parts of the country, particularly east coast cities (Murrill et al., 2001). Second, women in our sample were mostly young (67% less than 21 years old), and HIV seroprevalence is typically lower among young IDUs, although their risk of seroconverting increases with time if they continue to inject (Battjes et al., 1992). Finally, many of these respondents were recruited from street- and agency-based sites that offered a range of HIV prevention services. We cannot conclude from our data that these services are helping to prevent HIV infection among young IDUs, but it is an optimistic possibility that cannot be ruled out.

Consistent with other studies (Cabral et al., 1998; Eldridge et al., 1995; Harrison et al., 1996; Harvey et al., 1999, in press; Kline et al., 1992; Sobo, 1995; Soet et al., 1999; Wyatt et al., 2000), the majority of women in this sample (over 70%) said that they do participate in making decisions about sexual behaviors, including contraceptive and condom use. Additional results corroborate the decision-making findings and, taken together, indicate that women in this population feel powerful in their sexual relationships. For example, these young women reported high levels of condom use self-efficacy and generally did not anticipate negative reactions from their sex partners if they asked to use condoms.

The extent to which women reported joint decision making, especially decisions about condom use (48%) and when to have sex (67%), is of particular interest. These results suggest that women are involved in these decisions, but so are men. Because sexual intercourse and condom use are examples of what Agnew (1999, p. 165) has defined as interdependent behaviors--"any action or activity that necessarily requires more than one person for enactment"--this finding is not surprising. On the other hand, independent behaviors, such as some contraceptive use, do not necessarily require the participation of both persons involved in a heterosexual relationship. For example, a woman can use contraceptives (e.g., birth control pills) without the knowledge of her sexual partner.

It is noteworthy that women who said that they participated in making decisions about condom use and about when to have sex were more likely to report having used condoms with their partners in the past month than were those who said that they were not involved in those decisions. Similarly, women who indicated that they participated in decisions about birth control were more likely to report using birth control with their partners. Our findings contribute to a growing body of literature that indicates that relationship power and involvement in sexual decision making are positively associated with women's safer sex behavior (e.g., Harvey et al., in press; Pulerwitz et al., 2000; Wingood & DiClemente, 1998). These results, in conjunction with the finding that the majority of women perceived that they participated in making sexual decisions, underscore the critical role that young female IDUs can play in protecting themselves from the negative consequences of unprotected sex. On the other hand, they demonstrate that women who do not control sexual decision making may be at higher risk for unintended pregnancy, HIV, and other STDs.

These findings are consistent with previous research on domain-specific measures of relationship power and decision making. That is, although general measures of power and decision making may not be related to condom use, measures of decision making about condom use, specifically, are associated with actual condom use. Thus, differences in measurement may explain, in part, the mixed results reported in the literature. Our findings emphasize the value of using domain-specific assessments of decision making in relationships and indicate that participation in decision making in different sexual domains may influence condom use.

The findings that condom and contraceptive use are negatively related to relationship duration support the results of previous studies (summarized in Misovich et al., 1997). Moreover, similar findings regarding the influence of relationship duration on patterns of unsafe sexual behavior have been reported in earlier studies of IDUs (summarized in Misovich et al., 1997). Women in longer relationships may be more likely to feel invulnerable to HIV/STDs and, as a result, may be less likely to engage in condom use. Misovich et al. (1997, p. 88) contended that the trust heuristic--"a decision rule that dictates that trusted partners are safe partners"--is a major barrier to HIV prevention in intimate relationships. As trust and intimacy develop over time between sexual partners, trust may become a substitute for safer sex behavior. In addition, women in long-term relationships may be more likely than women in casual or newer relationships to risk pregnancy because they believe that the relationship will continue or because they are more likely to want to have a child with their current partner.

Limitations

This study has several limitations that need to be considered when interpreting and applying the results. First, because the findings were from cross-sectional interview data, the associations between self-reported decision making and condom and contraceptive use cannot be construed as causal. Prospective data are needed. Second, the sample size was small and the selection of participants was not random; thus, findings should be interpreted with caution. In particular, because this sample consisted of primarily White, HIV-negative IDUs in Los Angeles, the findings may not be generalizable to young female IDUs and female partners of IDUs in other geographic locations. The small sample size limited our ability to adjust for all potentially confounding variables. Moreover, the use of single-item measures of sexual decision making may have limited reliability.

Additionally, as with most research on sexual behavior, the data were based on participants' self-reports, and we were unable to verify if what they reported was accurate. However, interviewers attempted to minimize underreporting of behaviors that tend to be stigmatized. Specifically, our interviewers had a long-standing presence in the community, had served as volunteers at the needle exchange, and were well accepted and trusted by the target population. Questions about stigmatized behaviors were strategically placed in the interview, allowing interviewers to develop a rapport with respondents before broaching potentially uncomfortable topics. At the outset of these sections, respondents were reminded that their answers were confidential and that there were no right or wrong answers. The study interviewers believed that the vast majority of respondents were candid and open with them and were comfortable discussing these behaviors. In addition, short recall time frames (i.e., the last 30 days) were used in an attempt to reduce reporting error due to recall issues.

In correlational studies based on self-report, it is always difficult to identify socially desirable responding. For example, it could be argued that respondents who did not use condoms might have wished to justify their behavior by saying it was not their decision. Conversely, those using condoms may have been eager to take credit for their socially desirable decision. The order of the questions in the survey instrument argues against this chain of events, however. Questions regarding sexual decision-making were asked prior to questions about specific sexual behaviors, including condom use. Respondents had no way to know that the decision-making questions would be followed by questions about condom use. It seems unlikely that respondents would consistently report their condom use based on their answers to the sexual decision-making questions. Thus, although we cannot rule out the possibility entirely, we do not believe the association between sexual decision-making and condom use is a result of socially desirable responses.

Finally, although this study sheds light on women's perceived roles in sexual decision making and their impact on safer sex behaviors, it does not tell us why some women believe that they are more involved in sexual decisions than other women. Future research should explore what types of relationship characteristics and intrapersonal characteristics (e.g., self-esteem, social skills, assertiveness skills) might be related to perceived power in sexual decision making.

Implications for Interventions

Despite these limitations, the study results offer important insights for the design of interventions directed at reducing unprotected intercourse among young female IDUs and female partners of IDUs. First, the finding that perceiving that one is actively participating in specific sexual decisions is positively associated with self-reported condom and contraceptive use suggests that interventions need to provide skills training to encourage and enable young women to make decisions about sexual behaviors and to have young women role-play this behavior. Programs that assist young women in becoming more empowered sexually could help them reduce their risk of HIV and other STDs. In addition, because the majority of women in our study reported joint decision making with their partners and because condom use is a dyadic behavior, prevention programs may benefit from targeting both men and women in individual-level interventions or including both members of a dyad in couple-level interventions.

Conclusion

Research on the influence of relationship factors, including power and sexual decision making, on women's safer sex behavior is in its infancy. This study examined the associations among partner-specific relationship characteristics, domain-specific sexual decision making, and condom and contraceptive use among a sample of female IDUs and female partners of IDUs. Our findings support an association between women's participation in sexual decision making and safer sex behavior and suggest that further research on this topic is warranted. The need to focus on gender dynamics when investigating sexual and HIV-prevention decision making and behavior has become increasingly clear. Much of the sexual behavior that puts women at risk for negative health outcomes is dyadic in nature and influenced by the male partner. Our findings indicate, therefore, that further HIV-prevention research should be guided by theoretical frameworks that include the social and cultural context of sexual behavior and add to our understanding of how relationship factors and gender dynamics influence HIV risk for women.

Table 1. Distribution of Respondents by Number and Type
of Male Sex Partners (N = 94)

                                                       n     %

One main male partner only                             61   64.9
One non-main male partner only                         14   14.9
Two or three main male partners only                    3    3.2
Two or three non-main male partners only                 5    5.3
One main male partner and one non-main male partner     8    8.5
One main male partner and two non-main male partners    2    2.1
Two main male partners and one non-main male partner    1    1.1

Table 2. Self-Reported Respondent Characteristics (N = 94 (a))

Age in years
  14 to 17                                                       18.1
  18 to 20                                                       48.9
  21 to 30                                                       33.0
Ethnicity
  African American                                                1.1
  Asian / Pacific Islander                                        2.1
  Hispanic / Latino                                               7.4
  White                                                          79.8
  Biracial / Multiracial                                          9.6
Sexual orientation
  Heterosexual                                                   52.1
  Bisexual                                                       46.8
  Undecided                                                       1.1
Highest education completed
  Elementary school                                               2.1
  Middle school /junior high                                     22.3
  Some high school                                               37.2
  High school diploma                                            16.0
  Some college                                                   21.3
  Finished college                                                1.1
Currently homeless                                               68.1
Homelessness history
  Never                                                          12.8
  Homeless < 1 year                                              26.9
  1-3 years                                                      26.9
  4-5 years                                                      19.4
  > 5 years                                                      14.0
Childhood physical abuse, n = 92                                 39.1
Childhood sexual abuse, n = 90                                   27.8
Prostitution in last 3 months                                    10.6
History of foster/group home placement                           33.0
Ever told that they had a mental or psychological problem        59.6
Ever been tested for HIV, n = 93                                 91.4
Ever tested positive for HIV, n = 84                              0.0
Currently using any IV drugs                                     73.4
IV drug currently using
  Heroin, n = 93                                                 55.9
  Cocaine                                                        26.6
  Speed, n = 93                                                  40.9
  Speedball                                                      16.0
Currently using non-IV drugs                                     84.0
Non-IV drug currently using
  Marijuana                                                      67.0
  Cocaine/crack                                                  38.3
  Speed                                                          39.4
  Heroin                                                         16.0
  Speedball, n = 93                                               2.2
Confidence asking sex partner to use a condom
  Not confident at all                                            1.1
  Not very confident                                              1.1
  Somewhat confident                                              7.4
  Very confident                                                 90.4
Confidence asking longtime sex partner to use a condom
  Not confident at all                                            8.5
  Not very confident                                              7.4
  Somewhat confident                                             13.8
  Very confident                                                 70.2
Confidence refusing sex if partner did not want
to use a condom
  Not confident at all                                            9.6
  Not very confident                                              3.2
  Somewhat confident                                             11.7
  Very confident                                                 75.5

(a) Some variables have missing data; the ns for those variables
are indicated.

Table 3. Partner-Specific Relationship Characteristics,
Perceptions, and Sexual Behavior (N = 94 (a))

                                                                %
Relationship type
  Main partner                                                 71.3
  Non-main partner                                             28.7
Length of sexual relationship
  1 day                                                        10.6
  2-7 days                                                      8.5
  8-30 days                                                    17.0
  > 30 days-2 months                                            7.4
  > 2 months-6 months                                          22.3
  > 6 months-1 year                                            12.8
  > 1 year                                                     21.3
How likely partner would get upset if asked to use condoms,
n = 93
  Not likely at all                                            80.6
  Not very likely                                               8.6
  Somewhat likely                                               9.7
  Very likely                                                   1.1
How likely partner would get violent if asked to use condoms
  Not likely at all                                            96.8
  Not very likely                                               0.0
  Somewhat likely                                               2.1
  Very likely                                                   1.1
How likely partner would refuse sex if asked to use condoms
  Not likely at all                                            95.7
  Not very likely                                               2.1
  Somewhat likely                                               2.1
  Very likely                                                   0.0
Had vaginal sex in past month                                  96.8
Had anal sex in past month, n = 93 (b)                          7.5
Condom use for vaginal sex in past month, n = 91
  None of the time                                             52.7
  Less than half the time                                       7.7
  Half the time                                                 1.1
  More than half the time                                       6.6
  Every time                                                   31.9
Contraceptive use in past month, n = 91
  None of the time                                             31.9
  Less than half the time                                       6.6
  Half the time                                                 3.3
  More than half the time                                       7.7
  Every time                                                   50.5
Reported decision maker about condom use, n = 91
  I                                                            25.3
  We                                                           48.4
  He                                                            7.7
  Neither                                                      18.7
Reported decision maker about contraception, n = 92
  I                                                            46.7
  We                                                           31.5
  He                                                            0.0
  Neither                                                      21.7
Reported decision maker about when to have sex, n = 93
  I                                                            18.3
  We                                                           66.7
  He                                                           12.9
  Neither                                                       2.2

(a) Some variables have missing data or were not applicable
(if respondents had not had vaginal sex with that partner
in the last month); the ns for those variables are indicated,
(b) All respondents who had anal sex in the past month
reported that they used condoms none of the time for
anal sex in the past month.

Table 4. Association Between Relationship Characteristics and
Participation in Sexual Decision Making

                                               Relationship type

                                               Main      Non-main
                                              partner    partner
                                             (n = 67)    (n = 27)
                                               % (n)       % (n)

Who makes decisions about condom use
  I / we                                     75.8 (50)   68.0 (17)
  He / neither                               24.2 (16)   32.0 (8)

Who makes decisions about birth control
  I / we                                     81.8 (54)   69.2 (18)
  He / neither                               18.2 (12)   30.8 (8)

Who makes decisions about when to have sex
  I / we                                     89.6 (60)   73.1 (19)
  He / neither                               10.4 (7)    26.9 (7)

                                             Relationship
                                              length (a)
                                             [Mean (SD)]

Who makes decisions about condom use
  I / we                                      4.6 (1.9)
  He / neither                                4.2 (2.2)

Who makes decisions about birth control
  I / we                                      4.8 (1.9) *
  He / neither                                3.7 (2.1)

Who makes decisions about when to have sex
  I / we                                      4.7 (2.0) *
  He / neither                                3.4 (1.7)

(a) Length of sexual relationship was measured on a 7-point
scale ranging from one day to more than one year. T tests
compared the relationship lengths of women reporting that
I/we made these decisions and women reporting that he/neither
made these decisions.

* p < .05.

Table 5. Adjusted Odds Ratios (and 95 % Confidence
Intervals) From Multiple Logistic Regression
Analysis of Condom Use With Partner in Past
Month (n = 87)

                                           Odds ratio      (95% CI)

Age                                           1.00       (0.78, 1.27)
Education (a)                                 0.63       (0.38, 1.07)
Number of male partners reported
    One (ref)                                 1.00
    Two or more                               0.37       (0.07, 2.02)
Condom self-efficacy
    Very confident on all three items         0.97       (0.33, 2.85)
    Somewhat, not very, or not confident
      on one or more items (ref)              1.00
Relationship type
    Main partner (ref)                        1.00
    Non-main partner                          2.44       (0.48, 12.36)
Length of sexual relationship (b)             0.60 *     (0.41, 0.88)
Who makes decisions about condom use
    I/we                                      7.28 *     (1.66, 31.90)
    He/neither (ref)                          1.00
Who makes decisions about birth control
    I/we                                      1.19       (0.24, 6.08)
    He/neither (ref)                          1.00
Who makes decisions about when
  to have sex
    I/we                                     18.81 *     (2.40, 147.23)
    He/neither (ref)                          1.00

Note. ref = reference group.

(a) Education was treated as a continuous variable, due to its
close-to-normal distribution and large number of categories,
(b) Length of sexual relationship was treated as a continuous
variable, due to its close-to-normal distribution and large
number of categories.

* p < .01.

Table 6. Adjusted Odds Ratios (and 95 % Confidence
Intervals) From Multiple Logistic Regression
Analysis of Birth Control Use With Partner in
Past Month (n = 87)
                                           Odds ratio     (95% CI)

Age                                           0.82      (0.63, 1.07)
Education (a)                                 0.92      (0.52, 1.61)
Number of male partners
  One (ref)                                   1.00
  Two or more                                 0.05 *    (0.01, 0.37)
Condom self-efficacy
  Very confident on all three items           0.28      (0.07, 1.19)
  Somewhat, not very, or not confident
    on one or more items (ref)                1.00
Relationship type
  Main partner (ref)                          1.00
  Non-main partner                            0.90      (0.14, 5.68)
Length of sexual relationship (b)             0.43 *    (0.26, 0.73)
Who makes decisions about condom use
  I/we                                        1.47      (0.31, 6.93)
  He/neither (ref)                            1.00
Who makes decisions about birth control
  I/we                                       19.55 *    (2.88, 132.83)
  He/neither (ref)                            1.00
Who makes decisions about when to have sex
  I/we                                        6.00      (0.82, 43.91)
  He/neither (ref)                            1.00

Note. ref = reference group.

(a) Education was treated as a continuous variable, due to its close
to normal distribution, and large number of categories, (b) Length
of sexual relationship was treated as a continuous variable, due to
its close-to-normal distribution, and large number of categories.

* p < .01.

This research was supported by the National Institute on Drug Abuse, Grant ROI-DA10661. Investigators for the project include Susanne D. Montgomery (PI), Michael Claus, Christine Johnson De Rosa, Susan Ennett, S. Marie Harvey, Justeen Hyde, Ellen Iverson, Michele D. Kipke, and Louise Ann Rohrbach. The authors appreciate the hard work of the interviewing team: Kimberly Brown, Cecilia Mejia, Cristina Polyzoides, Jason Collins, Eileen Argueta, Margaret Haines, Anthony Espinosa, Ana Quiran, Veronica Montenegro, and German Rodriguez. In addition, we give thanks to Anna Wilson for her assistance with data analysis.

(1) We also conducted analyses using dichotomous measures of consistent condom and contraceptive use (e.g., used condoms every time vs. used condoms more than half of the time or less frequently). Because the sexual decision making results showed a similar pattern when using those measures, we do not further describe those results in this article.

REFERENCES

Agnew, C. R. (1999). Power over interdependent behavior within the dyad: Who decides what a couple does? In L. J. Severy & W. Miller (Eds.), Advances in population: Psychosocial perspectives, Vol. 3 ( pp. 162-188). London: Jessica Kingsley Publishers.

Amaro, H. (1995). Love, sex, and power: Considering women's realities in HIV prevention. American Psychologist, 50, 437-443.

Amaro, H., & Gornemann, I. (1992). HIV/AIDS related knowledge, attitudes, beliefs and behaviors among Hispanics: Report of findings and recommendations. Boston: Boston University School of Public Health and Northeast Hispanic Consortium.

Amaro, H., & Raj, A. (2000). On the margin: Power and women's HIV risk reduction strategies. Sex Roles, 42, 723-749.

Azjen, I., & Fishbein, M. (1977). Attitude-behavior relations: A theoretical analysis and review of empirical research. Psychological Bulletin, 84, 888-918.

Bailey, S. L., Camlin, C. S., & Ennett, S. T. (1998). Substance use and risky sexual behavior among homeless and runaway youth. Journal of Adolescent Health, 23, 378-388.

Baker, A., Kochan, N., Dixon, J., Wodak, K., & Heather, N. (1994). Drug use and HIV risk-taking behavior among injection drug users not currently in treatment in Sydney, Australia. Drug and Alcohol Dependence, 34, 155-160.

Bandura, A. (1994). Social cognitive theory and exercise of control over HIV infection. In R. J. DiClemente and J. Peterson (Eds.), Preventing AIDS: Theories and methods of behavioral interventions (pp. 25-59). New York: Plenum Publishing Corp.

Battjes, R. J., Leukefeld, C. G., & Pickens, R. W. (1992). Age at first injection and HIV risk among intravenous drug users. American Journal of Drug and Alcohol Abuse, 18, 263-273.

Becker, M. H. (1974). The health belief model and personal health behavior. Health Education Monographs, 2, 324-508.

Booth, R. E. (1995). Gender differences in high risk sex behaviors among heterosexual drug injectors and crack smokers. American Journal of Drug and Alcohol Abuse, 18, 419-432.

Brown, S. S., & Eisenberg, L. (Eds.). (1995). The best intentions: Unintended pregnancy and the well-being of children and families. Washington, DC: National Academy Press.

Cabral, R. J., Pulley, L., Artz, L. M., Brill, I., & Macaluso, M. (1998). Women at risk of HIV/STD: The importance of male partners as barriers to condom use. AIDS and Behavior, 2, 75-85.

Catania, J. A., Kegeles, S. M., & Coates, T. J. (1992). Towards an understanding of risk behavior: An AIDS risk reduction model (ARRM). Health Education Quarterly, 17, 53-72.

Centers for Disease Control and Prevention. (1999). HIV/AIDS surveillance report, 11(2).

Centers for Disease Control and Prevention. (2000). HIV/AIDS surveillance report, 12(2).

Clements, K., Gleghorn, A., Garcia, D., Katz, M., & Marx, R. (1997). A risk profile of street youth in Northern California: Implications for gender specific human immunodeficiency virus prevention. Journal of Adolescent Health, 20, 343-353.

Connell, R. W. (1987). Gender and power. Stanford, CA: Stanford University Press.

Davies, A. G., Dominy, N. J., Peters, A. D., & Richardson, A. M. (1996). Gender differences in HIV risk behaviours among regular amphetamine users. AIDS Care, 8, 517-527.

DeMatteo, D., Major, C., Block, B., Coates, R., Fearon, M., Goldberg, E., et al. (1999). Toronto street youth and HIV/AIDS: Prevalence, demographics, and risks. Journal of Adolescent Health, 23, 378-388.

De Rosa, C. J., Montgomery, S. B., Hyde, J., Iverson, E., & Kipke, M. D. (2001). HIV risk behavior and HIV testing: A comparison of rates and associated factors among homeless and runaway adolescents in two cities. AIDS Education & Prevention, 13, 131-148.

Eldridge, G. D., St. Lawrence, J. S., Little, C. E., Shelby, M. C., & Brasfield, T. L. (1995). Barriers to condom use and barrier method preferences among low-income African-American women. Women & Health, 23, 73-89.

Ennett, S. T., Federman, E. B., Bailey, S. L., Ringwalt, C. L., & Hubbart, M. L. (1999). HIV-risk behaviors associated with homeless characteristics in youth. Journal of Adolescent Health, 25, 344-353.

Ensign, J., & Santelli, J. (1998). Health status and service use: Comparison of adolescents at a school-based health clinic with homeless adolescents. Archives of Pediatric Adolescent Medicine, 152, 20-24.

Fullilove, M. T., Fullilove, R. E., Haynes, K., & Gross, S. (1990). Black women and AIDS prevention: A view towards understanding the gender rules. The Journal of Sex Research, 27, 47-64.

Gomez, C. A., & Marin, B. V. (1996). Gender, culture, and power: Barriers to HIV-prevention strategies for women. The Journal of Sex Research, 33, 355-362.

Green, J. M., Ennett, S. T., & Ringwalt, C. L. (1999). Prevalence and correlates of survival sex among runaway and homeless youth. American Journal of Public Health, 89, 1406-1409.

Green, J. M., & Ringwalt, C. L. (1998). Pregnancy among three national samples of runaway and homeless youth. Journal of Adolescent Health, 23, 370-377.

Harrison, J. S., Kay, K. L., Dixon, D., Peters, M., & Moore, J. (1996, September). Defining and measuring power: Researcher and study participant perspectives. Paper presented at the Conference on Psychosocial and Behavioral Factors in Women's Health, Washington, DC.

Harvey, S. M., Beckman, L. J., & Doty, M. (1999). Couple dynamics in sexual and reproductive decision making among Mexican immigrants. Advances in Population, 3, 251-279.

Harvey, S. M., Bird, S. T., Galavotti, C., Duncan, E. A., & Greenberg, D. (in press). Relationship power, sexual decision making and condom use among women at risk for HIV/STDs. Women & Health.

Henshaw, S. K. (1998). Unintended pregnancy in the United States. Family Planning Perspectives, 30, 24-29, 46.

Johnston, L. D., O'Malley, P. M., & Bachman, J. G. (2000). Monitoring the future: National survey results on adolescent drug use; Overview of key findings, 1999 (NIH Publication No. 00-4690). Bethesda, MD: National Institute on Drug Abuse.

Kipke, M. D., Montgomery, S. B., Simon, T. R., Unger, J. B., & Johnson, C. J. (1997). Homeless youth: Drug use patterns and HIV risk profiles according to peer group affiliation. AIDS and Behavior, 25, 344-353.

Kipke, M. D., O'Connor, S., Palmer, R., & MacKenzie, R. G. (1995). Street youth in Los Angeles: Profile of a group at high risk for human immunodeficiency virus infection. Archives of Pediatrics and Adolescent Medicine, 149, 513-519.

Kline, A., Kline, E., & Oken, E. (1992). Minority women and sexual choice in the age of AIDS. Social Science and Medicine, 34, 447-457.

Misovich, S. J., Fisher, J. D., & Fisher, W. A. (1997). Close relationships and elevated HIV risk behavior: Evidence and possible underlying psychological processes. Review of General Psychology, 1, 72-107.

Montgomery, S. B., Hyde, J., DeRosa, C. J., Rohrbach, L. A., Ennett, S., Harvey, S. M., et al. (in press). Gender differences in HIV risk behaviors among young injectors and their social network members. American Journal of Drug and Alcohol Abuse.

Murrill, C. S., Prevots, R., Miller, M. S., Linley, L. A., Royalty, J. E., & Gwinn, M. (2001). Incidence of HIV among injection drug users entering drug treatment programs in four US cities. Journal of Urban Health, 78, 152-161.

Nelson, K. E., Vlahov, D., Solomon, L., Cohn, S., & Munoz, A. (1995). Temporal trends in incident immunodeficiency virus infection in a cohort of injection drug users in Baltimore, MD. Archives of Internal Medicine, 155, 1305-1311.

Noell, J., Rhode, P., & Ochs, L. (1997, November). Childhood sexual abuse, adolescent sexual coercion, and pregnancy in homeless adolescents. Paper presented at the meeting of the American Public Health Association, Indianapolis, IN.

Prochaska, J. O., & DiClemente, C. C. (1983). Stages and processes of self-change of smoking: Toward an integrative model of change. Journal of Consulting and Clinical Psychology, 51, 390-395.

Prochaska, J. O., & DiClemente, C. C. (1984). The transtheoretical approach: Crossing the traditional boundaries of therapy. Homewood, IL: Dow-Jones/Irwin.

Pulerwitz, J., Gortmaker, S. L., & DeJong, W. (2000). Measuring sexual relationship power in HIV/STD research. Sex Roles, 42, 637-660.

Sly, D. F., Quadagno, D., Harrison, D. F., Eberstein, I., & Riehman, K. (1997). The association between substance use, condom use and sexual risk among low income women. Family Planning Perspectives, 29, 132-136.

Sobo, E. (1995). Choosing unsafe sex. Philadelphia: University of Pennsylvania Press.

Soet, J. E., Dudley, W. N., & DiIorio, C. (1999). The effects of ethnicity and perceived power on women's sexual behavior. Psychology of Women Quarterly, 23, 707-723.

Soler, H., Quadagno, D., Sly, D. F., Riehman, K. S., Eberstein, I. W., & Harrison, D. F. (2000). Relationship dynamics, ethnicity and condom use among low-income women. Family Planning Perspectives, 32, 82-101.

Wingood, G. M., & DiClemente, R. J. (1998). Gender-related correlates and predictors of consistent condom use among young adult African-American women: A prospective analysis. International Journal of STD & AIDS, 9, 139-145.

Wingood, G. M., & DiClemente, R. J. (2000). Application of the theory of gender and power to examine HIV-related exposures, risk factors, and effective interventions for women. Health Education & Behavior, 27, 539-565.

Wingood, G. M., Hunter-Gamble, D., & DiClemente, R. J. (1993). A pilot study of sexual communication and negotiation among young African American women: Implications for HIV prevention. Journal of Black Psychology, 19, 190-203.

Worth, D. (1989). Sexual decision making and AIDS: Why condom promotion among vulnerable women is likely to fail. Studies in Family Planning, 20, 297-307.

Wyatt, G. E., Carmona, J. V., Loeb, T. B., Guthrie, D., Chin, D., & Gordon, G. (2000). Factors affecting HIV contraceptive decision making among women. Sex Roles, 42, 495-521.

Manuscript accepted December 2, 2002

Address correspondence to S. Marie Harvey, Research Program on Women's Health, Center for the Study of Women in Society, 1201 University of Oregon, Eugene, OR 97403-1201; e-mail: [e-mail].

S. Marie Harvey
University of Oregon

Sheryl Thorburn Bird
Oregon State University

Christine Johnson De Rosa
Childrens Hospital Los Angeles

Susanne B. Montgomery
Loma Linda University

Louise Ann Rohrbach
University of Southern California