Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 May 5.
Published in final edited form as: J Sex Res. 2012 Apr 10;50(5):489–501. doi: 10.1080/00224499.2012.661101

HIV Testing among Heterosexual Young Adults: The Influence of Individual Attitudes and Behaviors, Partner’s Risk-Taking and Relationship Dynamics

Monica A Longmore 1, Wendi L Johnson 1, Wendy D Manning 1, Peggy C Giordano 1
PMCID: PMC4010152  NIHMSID: NIHMS414155  PMID: 22489753

Abstract

This study relies on survey (N=704) and in-depth qualitative (N = 100) interviews (Toledo Adolescent Relationship Study) to examine individual, partner, and relationship barriers and facilitators to HIV testing in a sample of young adults. Consistent with the public health goal of routine testing, nearly 40% of respondents had an HIV test within the context of their current sexual relationship, and women were significantly more likely to have tested within the current relationship than were men. For women, it was both their own risky behavior, and the partners’ characteristics and related relationship dynamics that distinguished testers from non-testers. In contrast, for men their own risky behavior was the most salient factor influencing their odds of being tested. These results showcase gender specific approaches to best promote sexual health, i.e., routine HIV testing among young adults.

Keywords: HIV testing, Sexual activity, Young adults, Sexual health

INTRODUCTION

The CDC (2008; 2010) recently reported that the HIV/AIDS epidemic is more extensive than previously assumed and has revised their prevention strategy to focus, more so than in past decades, on heterosexual transmission among adolescents and young adults. Teens and young adults are among the fastest growing subgroups diagnosed with HIV, with 50% of new infections in the U.S. occurring among 13–24 year olds (Futterman, 2005). Currently, heterosexual transmission accounts for over 70% of women’s HIV cases in the U.S. (CDC, 2006). Although men who have sex with men are the largest subgroup at risk, the CDC’s revised strategy reflects a changing feature of the HIV/AIDs epidemic: increasing numbers of individuals, especially women, are infected heterosexually through primary and secondary sexual partners.

Behaviors associated with preventing HIV/AIDS include abstinence, consistent use of condoms and engaging in monogamous sexual relationships with non-infected partners (Britton et al., 1998; Misovich, Fisher, & Fisher, 1997). Yet unmarried young adults have increased risks for HIV exposure due to increases in sexual activity, greater number of partners, and greater risk of an unintended pregnancy (Chandra, Mosher, Copen, & Sionean, 2011; Mosher & Jones, 2010). Moreover, many young people are unaware of their HIV status. Undiagnosed individuals may inadvertently transmit the infection to others (Campsmith, Rhodes, & Hall, 2009; Harris, Gordon-Larsen, Chantala, & Udry, 2006). To prevent new HIV cases it is critical for individuals to get tested routinely, yet our understanding of factors associated with young adults’ testing practices is limited.

Using survey data from young adults involved in current or recent heterosexual intimate relationships, the Toledo Adolescent Relationships Study (TARS) (n = 704), we examine factors that may influence recent HIV testing. Recognizing the complexity of decisions involving sexuality and health behaviors, we explore (1) an individual’s own attitudes (condom efficacy, sexual permissiveness) and risk behaviors (sexual non-exclusivity, lifetime sexual partners), (2) partner’s risk behaviors (sexual non-exclusivity, lifetime sexual partners), and (3) relationship dynamics (love, trust, power, sexual communication, sexual coercion, and pregnancy) as factors that may act as obstacles or facilitators to HIV testing with this specific partner. The analyses also include demographic characteristics shown in prior work to influence testing. To supplement these analyses, we examine narratives from a subset of 100 TARS participants whose survey responses reflected prior risk behaviors. Narrative excerpts illustrate individuals’ self-described motives for testing, illuminating aspects of decision-making that may be difficult to capture in survey research. Identifying individual, partner and relationship characteristics associated with testing is important to a comprehensive understanding of how young adults’ decisions involving sexual health unfold within the context of their intimate relationships. Our findings may be useful in designing enhanced HIV testing and prevention efforts targeted toward this age group.

BACKGROUND

HIV Testing and Gender

In past decades, a primary reason for not testing for HIV was the stigma associated with requesting a test and the anxiety provoked while waiting for results (Chesney & Smith, 1999; Exner, Hoffman, Parikh, Leu, & Ehrhardt, 2002; Fortenberry et al., 2002; Spielberg, Kurth, Gorbach & Goldbaum, 2001). With rapid testing now widely accessible, and results available in about twenty minutes, it is faster and less burdensome to get tested. Yet current estimates indicate that as many as one-fifth of the HIV-infected individuals in the U.S. are still unaware of their status, either because they have never been tested or have not been tested recently (Campsmith et al., 2009; Hall et al., 2008), underscoring the need to better understand testing practices among young people.

Much work on testing prevalence among non-clinical samples of young adults has used the Youth Risk Behavior Surveillance (YRBS) (e.g., CDC, 2006), the Longitudinal Study of Adolescent Health (Add Health) (e.g., Hahm, Song, Ozonoff, & Sassani, 2009; Nguyen et al., 2006) and the National Survey of Family Growth (NSFG) (e.g., Jeffries, 2010; Trepka & Kim, 2010). These data sets, however, assess whether individuals have ever been tested or tested in the past twelve months, but not whether individuals have been tested since being with their most recent sexual partner. Consistent with widely advocated prevention strategies (e.g., Britton et al., 1998; Campsmith et al., 2009; Hall et al., 2008; Harris et al., 2006; Misovich et al., 1997), our study moves beyond prior work by assessing whether young adults in their current or most recent relationship are getting tested for HIV. We examine individual characteristics, partner’s risk behaviors, and relationship dynamics that distinguish testing decisions among men and women.

Our work integrates health behavior and gender vulnerability approaches. The traditional health behavior approach focuses on the significance of attitudes, efficaciousness and behavioral choices, including partner choices, which put individuals at risk (e.g., DiClemente & Peterson, 1994; Hahm et al., 2009; Longmore, Manning, Giordano, & Rudolph, 2003; Rosenstock, Stretcher, & Becker, 1994). Yet, health behavior models often assume that women have as much personal control over sexual situations, and ability to put their beliefs into action, as do men (Higgins, Hoffman, & Dworkin, 2010). Additionally, such models do not typically distinguish the individual’s own risky attitudes and behavior from the choice of a partner whose behavior puts the individual at risk for HIV.

Because women have greater odds of infection by primary or secondary male sexual partners, and disproportionately report being tested for HIV (Arrington-Sanders & Ellen, 2008), some scholarship has conceptualized gendered power and other relational dynamics as an important backdrop to the heterosexual HIV/AIDS epidemic. This approach, which Higgins, Hoffman, and Dworkin (2010) termed the gender vulnerability hypothesis, has emphasized contextual factors, apart from individual behavior and partner’s risk behaviors, which lead to women’s enhanced vulnerability to HIV. This perspective has highlighted relational dynamics (e.g., Amaro & Raj, 2000; Corbett, Dickson-Gomez, Hilario, & Weeks, 2009; Higgins et al., 2010; Sobo, 1997) as essential elements contributing to women’s HIV risk. This scholarship reflects women’s greater vulnerability to heterosexually transmitted HIV, but has tended to overlook men’s perspectives.

Recognizing the complexity of sexual health we integrate these approaches and advocate for examining the attitudinal and behavioral risk factors that the individual brings to the relationship, partner’s risk behaviors, and features of the relationship itself. The research reviewed below focuses primarily on the broader literature on sexual risk taking as a useful conceptual starting point for the current investigation. Yet more research is needed to determine whether variables identified in this literature also influence testing decisions, and whether they appear to have a similar effect on men and women’s decisions to be tested.

Individual Characteristics: Efficacy, Permissive Attitudes, and Risky Behaviors

If used consistently and correctly, condoms are effective in reducing the risk of HIV infection (CDC, 2010). Thus, given its relevance for HIV prevention, we examine efficaciousness in the negotiation and use of condoms on subsequent testing practices. Although studies have examined condom efficacy as a predictor of actual condom use (e.g., Longmore et al., 2003), and HIV coping efficacy on attitudes toward routine HIV testing (Arrington-Sanders, Ellen, & Leonard, 2009), condom efficacy has not been examined in large surveys of correlates of testing. Condom efficacy is likely, on average, higher for men; nevertheless, it likely is an important correlate of testing for women. Kyomugisha (2006) argued that men are more likely than women to control sexual encounters, and to control condom use, which may limit women’s ability to protect themselves. In a similar vein, in a clinic sample of HIV positive and negative girls, aged 12–16, the majority believed either both partners or only men were responsible for condom use because men wear the condom (Marhefka & Demetriou, 2009). If individuals believe they are efficacious in negotiating condoms, they may take the necessary steps to be tested. Conversely, it is possible that efficacious individuals may feel protected from HIV exposure and not get tested. We explore the influence of condom efficacy on HIV testing, but do not specify the relationship direction.

Permissive attitudes are associated with risky sexual behaviors, such as multiple and concurrent sexual partners and casual sexual partnerships (e.g., Basen-Engquist, & Parcel, 1992; Jemmott & Jemmott, 1990; Lyons, Giordano, Manning, & Longmore, 2010). We are not certain, however, whether such attitudes will operate similarly when we focus on HIV testing rather than risky behaviors themselves. Associations between permissive sexual attitudes and reluctance to get tested for HIV have been found in some African societies, especially among men (e.g., Steinberg, 2008), yet whether this would be the case in the U.S. is unclear. For example, it is possible that individuals who hold permissive attitudes may experience less reluctance to seek information about their health status. We explore whether permissiveness is associated with testing, but do not specify, a priori, an expected direction.

Although efficacy beliefs and permissive sexual attitudes are expected to be influential, individuals’ life experiences, including prior risky sexual experiences, likely have greater implications for behavior (e.g., Bucx, Raaijmakers, & van Wel, 2010). Studies focusing on high-risk populations have found positive associations between behavioral risks and HIV testing (e. g., Anderson, Carey & Taveras, 2000; Choi & Catania, 1996; Denison, Lungu, Dunnett-Dagg, McCauley, & Sweat, 2006; Stein & Nyamathi, 2000; Setia, Quesnel-Vallee, Curtis, & Lynch, 2009; Straub et al. 2011; Tolou-Shams et al., 2007). We expect that those who get tested will be responding to their greater sexually risky behavior (e.g., sexual non-exclusivity and a higher average number of lifetime sex partners). We also expect these factors to be the strongest predictors of testing.

Partner Risk Behaviors: Prior Number of Sex Partners and Sexual Non-Exclusivity

Individuals who get tested likely believe that their sexual partner’s behavior is risky in terms of prior number of sexual partners and/or lack of sexual exclusivity. In a qualitative study of poor, urban, Black women, Sobo (1997) found that women in sexual relationships relied on perceptions of their partner’s behaviors, and not their own risk behaviors, to decide whether to get tested. Yet, Tolou-Shams et al. (2007) prospectively examining the likelihood of getting tested for HIV at a three month follow-up found no significant differences between individuals reporting sex with a perceived risky partner and those who did not report a high risk partner.

An important goal of the current investigation, then, is assessing whether the individual’s own risk portfolio or that of the partner is more salient regarding the decision to get tested. In addition, the analyses allow us to assess the impact of these two sets of concerns for men and women. The current study moves beyond prior work by investigating more systematically the role of the individual’s own attitudes and sexual risk behaviors, and the partner’s risk behaviors within the context of a large heterogeneous sample that includes male and female respondents.

Relationship Dynamics: Love, Trust, Power, Communication, Coercion, and Pregnancy

As the gender vulnerability hypothesis has highlighted, the dynamics of women’s intimate relationships may affect whether women get tested for HIV (e.g., Graffigna & Olson, 2009; Trieu, Modeste, Marshak, Males, & Bratton, 2010). Scholars have theorized that due to gendered socialization, women place a higher value on romantic love (e.g., Amaro, 1995; Amaro, Raj, & Reed, 2001; Blanc, 2001; Kyomugisha, 2006; Logan, Cole, & Leukefeld, 2002; Sobo, 1995; Worth, 1989). For example, in HIV/AIDS education focus groups composed of same-sex teens, girls resonated with emotional and boys with physical aspects of sex (Hoppe et al., 2004). It has been argued that romantic love is a barrier to safe sex and HIV testing for women because such feelings work against objective judgments of sexual risk (e.g., Holland & Eisenhart, 1990; Warr, 2001, p. 241). Recognizing the emphasis on romantic love, and the possible risks posed, we expect that both women and men who report higher levels of romantic love will be less likely to get tested for HIV.

Feelings of mistrust have a straightforward connection with decisions to secure information about one’s HIV status, but may also involve other considerations. Women who characterize their intimate relationship as involving trust may be reluctant to ‘rock the boat’ or to introduce what may be perceived as an indication of mistrust (Holland & Eisenhart, 1990). As romantic relationships evolve, individuals may become more motivated to protect the relationship from conflict that occurs as a result of bringing up the subject of HIV testing, even at the expense of self-protection (Misovich et al., 1997). We expect that both men and women in relationships characterized by trust will be less likely to get tested for HIV.

Power in one’s relationship may influence not only sexual risk-taking, but decisions about testing. For women, power imbalance is associated with reduced sexual autonomy (Blanc, 2001; Pulerwitz, Amaro, De Jong, Gortmaker, & Rudd, 2002). Yet, prior research based on the TARS sample of adolescents and young adults has documented that female respondents report greater power in their intimate relationships (Giordano, Longmore, & Manning, 2006; Giordano, Manning, Longmore, & Flanigan, 2010). Even when unfavorable power imbalances exist, Kyomugisha (2006) theorized that women tend to be socialized to tolerate these dynamics favoring male power, and often perceive themselves as powerless to negotiate or insist on safer sexual practices. However, prior research has not established whether power influences decisions to be tested. Individuals with more power may take care of themselves by getting tested; conversely, individuals with less power may get tested because they are not sure if partners are sexually exclusive. We explore the influence of power imbalances on HIV testing.

A comprehensive understanding of relationship dynamics also necessitates attention to sexual communication. Although sexual communication is critical for relationship quality (e.g., Lefkowitz, Boone, & Shearer, 2004; Longmore, Eng, Giordano, & Manning, 2009; Tolou-Sham et al., 2007), its influence on HIV testing is not clear. Individuals who are uncomfortable or experience difficulty communicating about sex or condom use may be reluctant to take the steps necessary to be tested (Tolou-Sham et al., 2007). Conversely, it could be that those who communicate easily about sexual matters may be more likely to get tested. We explore the relationship between sexual communication and HIV testing for men and women.

Coercive sex may be associated not only with greater sexual risk for women (Blanc, 2001; Kyomugisha, 2006, p. 4; Pulerwitz, Amaro, De Jong, Gortmaker, & Rudd, 2002), but with reluctance to get tested. Consistent with prior research and theorizing about sexual risk-taking, we expect that coercive sex will be associated with women’s lower odds of getting tested for HIV with a specific, recent partner. However, unlike prior studies we consider this relationship for male respondents as well.

Another important development within the relationship is the couple’s experience of a pregnancy. The consequences of unprotected sexual behavior are often experienced more intimately by women (i.e., through pregnancy and childbearing), yet these experiences are likely to provide better access to healthcare, particularly in the context of prenatal care (Bond, Lauby, & Batson, 2005: Hahm et al., 2009). We expect that experiencing a pregnancy with the current or most recent partner will be associated with greater odds of HIV testing for women. We also examine whether partner’s pregnancy is associated with HIV testing for men.

Demographic Characteristics

Studies examining U.S. demographic patterns have provided much needed baseline information regarding the influence of age, race, and socioeconomic status on testing (e.g., Hahm et al., 2009; Nguyen et al., 2006). Consistent with the age gradient associated with most health behaviors (Evans, Barer, & Marmor, 1994), men and women in their twenties are more likely than teens to ever be tested for HIV or to be tested in the past 12 months (Straub et al., 2011). We expect that age will be associated with HIV testing with a specific, most recent or current intimate partner. HIV risk and testing is also more common among economically disadvantaged groups (CDC, 2010; Karon, Fleming, Steketee, & DeCock, 2001; Santelli, Lowry, Brener, & Robin, 2000). Parents’ education, a measure of social advantage, is expected to be negatively associated with HIV testing. Testing is more prevalent among non-white individuals regardless of socioeconomic status (CDC, 2008; Duran et al., 2010; Hahm et al., 2009). We expect that non-White relative to White respondents will have higher odds of testing.

CURRENT INVESTIGATION

Our goal in the current investigation is to build on previous studies by assessing the association between individual risk characteristics, partner’s risk behaviors, and relationship dynamics on HIV testing among a sample of young women and men. Is HIV testing linked more closely to the individual’s own risk portfolio, that of the partner, or the dynamics within the relationship? We assess whether these correlates are differentially associated with women and men’s decisions to be tested within the context of their current or most recent relationship.

The TARS provides a unique opportunity to examine our research question, and contributes to prior work in six key ways. First, the data include a range of attitudes, behaviors, and perceptions of partner’s behaviors, as well as relationship qualities, which permit a more comprehensive assessment of HIV testing correlates. Second, the data are from a more heterogeneous group of young women and men as opposed to those who are already seeking treatment (e.g., Hall, Darville, Barral, & Sucato, 2011) or for whom the extent of representativeness is unknown (e.g., Djokic et al., 2009). Third, our measures move beyond those found in many other studies by examining HIV testing with specific sexual partners whereas many social surveys emphasize having ever been tested or having been tested in the past 12 months. Fourth, we consider gender specific models recognizing that men and women may differ in behavior and the correlates of their behavior may vary. Fifth, a multi-method approach using both survey and narrative data will likely provide greater insights relative to relying solely on a single method approach (Deren et al., 2003). The qualitative data provide more nuanced understandings of the motivations underlying testing decisions. Finally, our study focuses on unmarried young adults during the life stages of late adolescence and early adulthood whose risks for HIV exposure is likely increased due to increases in sexual activity, greater number of partners, and decreases in consistent condom use relative to early adolescence.

DATA and METHOD

Data

The data are from the TARS, a longitudinal study based on a stratified random sample of the year 2000 enrollment records of all youths registered for the 7th, 9th, and 11th grades in Lucas County, Ohio, a largely urban metropolitan area that includes the city of Toledo. The sample is drawn from student rosters from 62 schools across seven school districts, although respondents did not have to attend class to be in the sample. Rosters were made available through Ohio’s Freedom of Information Act. The sample, devised by the National Opinion Research Center, includes oversamples of Black and Hispanic adolescents. In the first interview (W1) conducted in 2001, 1,321 adolescents participated in the study. Our study relies on data collected during the fourth interview (W4) conducted in 2006–07. In W4, 83% of the original sample (n = 1,088) were re-interviewed. Interviews were conducted in the respondent’s home using preloaded laptops to maintain privacy. Primary parents or guardians were interviewed at W1, and parent’s education is from this wave. Our analytic sample includes unmarried respondents who are or were sexually active within their current or most recent relationship (previous 24 months), resulting in a final sample size of N = 704 (394 = female and 310 = male young adults).

We also draw on excerpts from in-depth interviews conducted with a subset (n = 100) of the respondents who participated in the wave 4 structured interviews. Respondents were chosen due to their high risk profiles on prior surveys. The in-depth interviews were scheduled separately from the survey interviews. Topics, in general, parallel the survey protocol, but allow more detailed consideration of respondents’ romantic and sexual histories. Narrative excerpts are included to explore individuals’ self-described reasons for getting/ not getting tested, illuminating aspects of testing decisions that are more difficult to capture via surveys. While the structured questions tap individuals’ attitudinal and sexual risk behaviors, perceptions of partners’ behavioral risks, and relationship dynamics associated with testing, the unstructured interviews highlight how young adults themselves give meaning to getting or not getting tested.

Dependent variable

Having been tested for HIV since being with current or most recent sexual partner is measured by asking: “Since the two of you have been together, have you been tested for HIV/AIDS?” We ask respondents answering with respect to their most recent relationship: “While you were with [name], were you ever tested for HIV/AIDS?” Responses are coded no (0) and yes (1), with 40% having been tested in this relationship.

Respondents’ Attitudes and Sexual Risk Behaviors

Permissive attitudes are measured by asking the degree to which respondents agree with the following eight items: (1) “It's okay to sometimes date more than one person at a time;” (2) “Sometimes I like to date a girl [guy] just for the fun of it;” (3) “A person should only have sex with someone they love;” (4) “A person should only have sex if they are married;” (5) “I would have to be committed to a girl [guy] in order to have sex with her [him];” (6) “It would be okay to have sex with someone I wasn't dating;” (7) “I would feel comfortable having sex with someone I was attracted to, but did not know very well;” and (8) “It is okay to have sex with an old girlfriend [boyfriend].” Responses range from strongly disagree (1) to strongly agree (5) and coding reflects greater permissiveness. The scale score is calculated as the mean of the items. The scale mean and standard deviation are 2.8 and .73, respectively. Cronbach’s alpha is .82.

Condom efficacy is measured by asking whether respondents are sure that they could: (1) “plan ahead to have a condom available;” (2) “stop yourself in the heat of passion and use a condom;” and (3) “resist having sex if your partner didn't want to use a condom.” Responses range from very unsure (1) to very sure (5).1 The scale is calculated as the mean of the items answered. For respondents who were missing data on two of the three items (n = 15), missing values are imputed using the scale mean score. The mean and standard deviation for the sample are 3.9 and 1.2, respectively. Cronbach’s alpha for the scale is .82.

Not sexually exclusive is measured by asking: “How often have [did] you gotten [get] physically involved (“had sex”) with other girls [guys]? Response categories are never (1), hardly ever (2), sometimes (3), often (4) and very often (5), and for multivariate analyses are coded as (0) sexually exclusive and (1) not sexually exclusive. Twenty-one percent of the sample report that they are not sexually exclusive.

Lifetime number of sex partners is measured by asking: “In your lifetime, about how many sex partners have you had?” Lifetime number of sex partners range from 1 to 18, with a mean and standard deviation of 6.3 and 5.4, respectively2.

Partner’s Sexual Risk Behaviors

Partner not sexually exclusive is measured by asking respondents: “How often do you think [name] has gotten [was] physically involved (“had sex”) with other girls [guys]?” For multivariate analyses responses are coded as partner is believed to be sexually exclusive (0), and partner is not believed to be sexually exclusive (1). Eighteen percent do not believe that their partner is sexually exclusive.

Partner’s lifetime number of sex partners is measured by asking respondents: “About how many partners do you think [name] had sex with before the two of you became involved?” Responses range from 0 to 12, with a mean response of 3.8 and a standard deviation of 3.8.3

Relationship Dynamics

Romantic love is measured using a modified version of Hatfield and Sprecher’s (1986) passionate love scale. Respondents are asked the extent to which they agree with the following statements: (1) “I am very attracted to [name];” (2) “The sight of [name] turns me on;” (3) “I would rather be with [name] than anyone else;” and (4) “[name] always seems to be on my mind.” Responses range from strongly disagree (1) to strongly agree (5). The scale score is calculated as the mean of the four items, with a mean and standard deviation of 4 and .8, respectively. Cronbach’s alpha is .83.

Trust is measured by asking the extent to which respondents agree with the statement: “There are times when [name] cannot be trusted.” Responses range from 1 to 5, with higher scores reflecting greater trust. The mean and standard deviation are 3.7 and 1.3, respectively.

Relationship power is measured by asking respondents: “If the two of you disagree[d] about something, which of you usually gets [got] their way?” Respondents answering that they usually get their way are coded as 1, while those responding either neutrally or that their partner usually gets his/her way are coded as 0. Thirty percent report greater relationship power.

Difficulty communicating about sex is measured by asking respondents the extent to which they agree with the statement: “Sometimes I find (found) it hard to talk about sexual matters with [name].” Responses range from strongly disagree (1) to strongly agree (5). The mean and standard deviation are 2 and 1, respectively.

Sexual coercion is assessed by asking: “How often has (did) [name] insist on or made you have sex with her [him] when you didn’t want to?” Responses are never (1), once (2), twice (3), three to five times (4), 6 to 10 times (5), 11 to 20 times (6), and more than 20 times (7). The mean and standard deviation for the sample are 1.5 and 1.1, respectively.

Pregnancy with partner is coded as 1 for those respondents who reported they (female) or their partner (male) were currently pregnant and for those who reported having a child(ren) with their current or most recent partner, and 0 otherwise. Sixteen percent report a pregnancy.

Respondent’s Demographic Characteristics

Age is calculated from the respondent’s date of birth and the date of the interview. Respondents’ mean age is 20.6 years, and the standard deviation is 1.7. Race/ethnicity is classified as White (individuals who identified as non-Hispanic White) (62%) and non-White (all other individuals) (38%), with White as the contrast category in the multivariate analyses. Parent’s education is measured from the parent’s questionnaire, which was completed primarily by mothers. We ask: “How far did you go in school?” If the father answered the questionnaire and was married or cohabiting, we ask: “How far did your partner go in school?” Responses are coded as 0 for 12 years of education or less (45%) and 1 for more than 12 years (55%).

Analytic Strategy

We first conduct a descriptive analysis that compares those who have gotten tested and those who have not within the context of the current or most recent sexual relationship, for women and men separately. Next, in multivariate analyses, we use logistic regression to estimate the odds of testing for HIV since being in a sexual relationship with this current or most recent partner, for women and men separately. We provide a baseline model that includes the demographic variables, and then estimate separate models for the individual characteristics, partner’s risk behaviors and relational dynamics respectively. Our final model includes the full set of covariates. Lastly, we provide an analysis of narrative data in which respondents describe their reasons for getting or not getting tested. The interviews we elicited from a subset (n = 100) of the respondents accord with the survey questions, but provide additional explanation for testing decisions. We describe respondents using pseudonyms, age, and testing status.

RESULTS

Descriptive Analyses

Table 1 presents the weighted means and percentages for the dependent and independent variables. The TARS data are drawn from a stratified, random sample; thus, each respondent has a unique probability of inclusion. Survey weights are then calculated based upon the probabilities. This allows us to transform point estimates into values that are more representative of a national sample.

Table 1.

Means and Percentages for All Unmarried Respondents by Gender (N = 704)

FEMALES (n=394) MALES (n = 310)
HIV Test Since w/Partner HIV Test Since w/Partner
Yes (48%) No (52%) Yes (30%) No (70%)
Individual Characteristics
   Permissive attitudes 2.6 2.6 3.3* 3.1
   Condom efficacy 4.0† 4.2 3.8 3.8
   Not sexually exclusive 16.9% 13.2% 34.3%† 24.7%
   Lifetime number of sex partners 6.0† 4.8 9.5*** 6.6
Partner Risk Behaviors
   Partner's number of sex partners 5.5** 4.1 2.7 2.6
   Partner not sexually exclusive 25.4%** 12.6% 14.5% 18.6%
Relationship Dynamics
   Romantic love 4.1 4.1 3.9† 3.8
   Trust 3.5** 3.9 3.9 3.8
   More power 44.2%* 34.1% 17.7% 21.6%
   Difficulty communicating 2.0* 1.8 1.9† 2.2
   Sexual coercion 1.4 1.4 1.6 1.5
   Pregnancy w/partner 36.1%*** 6.1% 20.7%** 7.8%
Demographic characteristics
   Age 20.5 20.3 21.1** 20.4
   Race
      White non-Hispanic 36.7%*** 63.3% 20.7%*** 79.3%
      Non-White/Hispanic 66.8%*** 33.2% 40.3%*** 59.7%
   Parent's Education
      High school graduate or less 54.7%** 45.3% 38.1%*** 61.9%
     Some college/college degree 41.7%** 58.3% 21.2%*** 78.8%
Open in a new tab

Source: Toledo Adolescent Relationships Study

Note: Significance levels based on zero-order logistic regression

Forty-eight percent of the women in our sample have been tested for HIV since being sexually intimate with their current or most recent partner. Women who report being tested for HIV have similar attitudes, but score slightly lower on condom efficacy, relative to non-testers (4.0 versus 4.2, respectively). Women who have tested for HIV, on average, report a higher number of lifetime sexual partners compared with non-testers (6 partners versus 5, respectively). A higher percentage of testers (25%), relative to non-testers (13%), also report that their partners are not sexually exclusive, and that their partners, on average, have a higher number of lifetime sexual partners compared with non-testers (6 partners versus 4, respectively). Women who have tested for HIV, relative to non-testers, report similar scores on the passionate love scale, but report lower scores when asked whether their partners can be trusted (3.5 versus 3.9, respectively). Forty-four percent of testers, relative to 34% of non-testers, report a power differential that favors them. Testers, on average, score higher when asked whether it is hard to talk about sexual matters with their partners (2.0 versus 1.8, respectively), and a higher percentage (36%) of testers relative to non-testers (6%) have had a pregnancy with this partner.

We find 30% of men have been tested for HIV while in their current or most recent relationship. Consistent with prior studies fewer men than women have been tested for HIV. Men who have tested for HIV within the context of their current or most recent relationship, on average, report more permissive attitudes than non-testers (3.3 versus 3.1, respectively), but have similar scores on the condom efficacy scale. A higher percentage of testers (34%), compared with non-testers (25%), report that they are not sexually exclusive, and on average, report a greater number of lifetime sexual partners (10 versus 7 partners, respectively). Male testers and non-testers have partners with similar numbers of prior sexual partners and are similar in reporting that their partners are not sexually exclusive. Testers report, on average, higher levels of romantic love than non-testers (3.9 versus 3.8, p < .10), but score similarly in terms of trust. Testers report, on average, less difficulty communicating about sexual topics relative to men who are not tested (1.9 versus 2.2, respectively). A higher percentage of men who are tested (21% versus 8%) report a pregnancy with this partner.

Taken together, findings at the bivariate level lead us to conclude that young women’s own risk behaviors as well as their partners’ risk behaviors are associated with HIV testing. In contrast, young men account for testing based on their own risk behavior. The relationship dynamics appear to operate differently regarding testing among men and women.

Multivariate Analyses: Women and HIV Testing

Table 2 displays the multivariate models of women’s odds of HIV testing while in their current or most recent relationship. Model 1 includes the demographic indicators and shows that age has a modest positive effect on the odds of being tested for HIV. Additionally, non-White compared with White women are three times more likely to be tested, and women whose parents are more highly educated are less likely to be tested. Model 2 shows that when women’s risk characteristics and demographic covariates are examined, the former are not associated with testing. Thus including women’s risk characteristics does little to reduce the effects of race and parent’s education on the odds of being tested for HIV. Thus, with regard to women, we did not find support for our expectation that the individual’s behavioral risks would significantly correlate with HIV testing.

Table 2.

Odds Ratios for the Logistic Regression of Tested for HIV Since with Partner among Unmarried Women (N = 394)

Model 1 Model 2 Model 3 Model 4 Model 5
eb SE eb SE eb SE eb SE eb SE
Intercept 0.08† 1.29 0.26 1.47 0.08† 1.31 0.06† 1.55 0.15 1.82
Individual Characteristics
   Permissive attitudes 0.82 0.18 0.73 0.21
   Condom efficacy 0.91 0.10 0.98 0.11
   Not sexually exclusive 0.89 0.31 0.66 0.41
   Lifetime number of sex partners 1.04 0.03 1.04 0.03
Partner Risk Behaviors
   Partner’s number of sex partners 1.05* 0.03 1.04 0.03
   Partner not sexually exclusive 1.82* 0.29 2.29† 0.45
Relationship Dynamics
   Romantic love 1.30 0.17 1.21 0.18
   Trust 0.81* 0.10 0.92 0.12
   Power
     (Same/less power)
     More power 1.33 0.24 1.34 0.24
   Difficulty communicating
about sex		 1.28† 0.13 1.23 0.13
   Sexual coercion 0.75* 0.13 0.70** 0.14
   Pregnancy w/partner 7.04*** 0.35 7.10*** 0.36
Demographic characteristics
   Age 1.12† 0.06 1.09 0.06 1.10 0.06 1.10 0.07 1.07 0.07
   Race/Ethnicity
     (White non-Hispanic)
     Non-White/Hispanic 3.02*** 0.22 2.97*** 0.22 2.69*** 0.22 2.09** 0.24 1.97** 0.25
   Parent Education
     (12 years or less)
     More than 12 years 0.57** 0.21 0.58* 0.22 0.58* 0.22 0.67† 0.23 0.66† 0.24
Likelihood Ratio χ2 39.84*** 43.68*** 49.31*** 94.25*** 103.30***
Open in a new tab

Note: The omitted category is in parentheses. Source: Toledo Adolescent Relationship Study.

†

p < .10.

*

p < .05.

**

p < .01.

***

p < .001.

Model 3 presents the associations between partner’s risk behaviors, demographic measures, and HIV testing. Consistent with our expectations, and the bivariate results, partner’s sexual non-exclusivity and number of lifetime sex partners are positively associated with women’s HIV testing net of demographic background. Race and parent’s education remain significantly related to testing.

Relationship dynamics and demographic characteristics are included in Model 4. Consistent with bivariate results, mistrust, and difficulty communicating about sex are significantly related to testing; however, the effect of greater power in the relationship is no longer significantly related to testing. Examining the influence of the covariates reveals that including race in the model reduces the effect of greater power on HIV testing. This is likely due to the positive relationship between being non-White or Hispanic and reporting a relatively favorable power balance (r = 0.15; p < .01, results not shown). Sexual coercion, while not significantly related to testing at the bivariate level, is associated with significantly lower odds of being tested for HIV, net of demographic background. Further examination of the covariates reveals that when race is included in the model, the effect of sexual coercion on HIV testing becomes statistically significant. While non-White women report higher levels of sexual coercion in their relationships as reflected in the correlation of these variables (r = 0.12, p < .01, results not shown), the inclusion of an interaction term indicates that the impact of sexual coercion on HIV testing does not differ significantly for White and non-White women (results not shown). Additionally, women who experienced a pregnancy with this partner are seven times more likely to be tested for HIV. While the effect of race on the odds of being tested is reduced, it remains statistically significant, and parent’s education continues to exert a modest negative effect on the odds of being tested.

Model 5 presents the results of the full model, which includes all three sets of variables and the demographic covariates. For women, partner’s sexual non-exclusivity, sexual coercion and pregnancy with partner remain significantly associated with HIV testing. The partner’s lifetime number of sex partners, mistrust, and sexual communication difficulty are no longer significantly related to testing once these other factors are taken into account. Further examination of the covariates reveals that controlling for trust reduces the effect of partner’s prior number of sex partners on HIV testing. A Sobol test (z = 2.48; p < .05) supports the notion that women who believe that their partner has a high number of prior sexual partners demonstrate a lower degree of trust, and thus, are more likely to get tested. Yet trust is not significantly related to testing in the full model. This is largely due to inclusion of the partner’s sexual non-exclusivity, which demonstrates a strong negative correlation with trust (r = - 0.489, p < .001). Consequently, the positive effect of having a partner who is not sexually exclusive, overrides the negative effect of trust on being tested. Examination of the covariates reveals that neither individual characteristics, nor partner risk behaviors alone, explain the lack of a statistically significant effect of sexual communication on HIV testing. Finally, non-White women continue to be significantly more likely to be tested for HIV, while those whose parents have higher education are less likely to be tested, net of respondent, partner and relationship characteristics.

These results highlight that among women, partner risk behaviors and relationship dynamics are of greater importance than individual risk characteristics in influencing the odds of HIV testing. We confirm this by conducting nested χ2 tests comparing Models 2, 3, and 4 to Model 1. For Model 2 the nested χ2 test is not significant indicating that including women’s individual risk characteristics do not add to model fit. However, comparisons of Models 3 and 4 to Model 1 yield significant results (χ2 = 9.47, 2 df; p < .01 and χ2 = 54.41, 6 df; p < .001 respectively) showcasing that partner’s risk behaviors and relationship dynamics add to the fit of the model.

Men and HIV Testing

Model 1 in Table 2 shows how the demographic indicators are associated with the odds of men testing for HIV within the context of their current or most recent relationship. Similar to women, the demographic characteristics are significantly related to HIV testing. However, among men, age emerges as a stronger predictor compared to women. Model 2 includes men’s individual risk and demographic characteristics. The relationships between permissive attitudes, sexual non-exclusivity and HIV testing shown in the bivariate analyses are no longer statistically significant once the individual risk and demographic characteristics are included. Men’s lifetime number of sex partners continues to be significantly associated with increased odds of being tested for HIV, as do the demographic variables. Consistent with the bivariate results, Model 3 shows that partner’s lifetime number of sex partners and partner sexual non-exclusivity are not positively related to the odds men are tested for HIV. In Model 3 age, race and parent’s education remain significantly associated with being tested for HIV.

Model 4 includes the relationship and demographic variables. The bivariate associations between men’s feelings of romantic love, and lack of difficulty communicating about sex are no longer significantly related to testing. An examination of the covariates reveals that including the relationship variables as a block accounts for the diminishment of the effects of these variables on the odds of getting tested. Pregnancy with partner is significantly associated with increased odds of being tested, but is a weaker influence among men than women. Demographic characteristics yield results similar to those in the previous models.

Model 5 includes men’s individual characteristics, partner’s risk behaviors, relationship dynamics and demographic controls. The findings in Model 5 are similar to those reported in the earlier models. Men who report higher numbers of lifetime sex partners have higher odds of testing for HIV. Pregnancy remains related to HIV testing. Age, race and parent education are significantly associated with the odds of being tested for HIV.

These results show that men are less reliant on the relationship context to inform them of their need to be tested for HIV. Rather it appears men rely on their personal sexual history to inform them on the need to seek HIV testing. Again, this is confirmed by conducting nested χ2 tests, which reveal a significant result for only Model 2 (χ2 = 9.57, 4 df; p < .05). Thus, individual risk adds to the model fit. Among men, their partner’s characteristics and most relationship dynamics do not significantly contribute to model fit.

A Qualitative Lens on Intimate Relationships and HIV Testing

Consistent with the quantitative results, the women who participated in the in-depth interviews were more likely to be tested than their male counterparts. In addition, and consistent with the quantitative results, women we interviewed were not averse to mentioning a partner’s extensive sexual history, and sometimes did appear to connect this directly to their testing decisions. As Ana, a 21 year old , suggested: “David, he’s been with a lot, I know that. And that was like a lot of the reason where um, I did get testing done...”

While women’s narratives emphasized their partners’ past experiences, men’s narratives often focused on their being good judges of character, which, they believed, worked to diminish their risk of exposure to HIV. For example, Jordan, 21 years old, described his girlfriend Andrea, and the reasons he did not need to be tested:

I know she didn’t even have sex until last year. I mean, it’s still not a sure thing but, like her parents are so strict, yeah, like they’re the “you’re not having sex until you get married.” Like real strict, like there are no thongs going through the washer in that place, put it that way. It still doesn’t mean anything, but, I don’t know, she had one boyfriend. I just knew that I was clean.

Another gendered pattern is that some men appeared to rely on testing-by-proxy, recognizing that women are likely to have more routine access to health care; and, consistent with the quantitative results (i.e., positive effect of pregnancy status) testing may occur in connection with women’s sexual health and prenatal care. Daniel, age 23, references his partner’s visits to obtain birth control in connection with his own decision not to get tested:

“But I mean she did (get tested), because she had to get put on birth control so… I didn’t… I guess I take it for granted that I’m fine. I’m pretty sure I am…”

Such references may suggest that the female partner’s test results provide a convenient way for some men to assess their own HIV status, but this strategy does, nevertheless, encompass assessments of the character of the partner with whom the individual is involved:

[Interviewer: Did you ever get tested?] No. She does. [Interviewer: So you rely on her test. So she doesn’t come up positive then you know…?] I am not (positive). Exactly. [Interviewer: That would be kind of after the fact wouldn’t it?] Yeah, I guess [but] I am not sleeping with whores.

Thus, in the above example, the respondent’s willingness to rely on second hand reports that his partner provides involves a characterization about her trustworthiness and feelings of confidence that she has not had recent unprotected sex with other partners.

One problem with relying on these routine testing situations for women and their partners is that it is not always clear when respondents are referencing testing for STDs in general and HIV specifically (e.g., Anderson & Blake, 2011). It is possible that respondents who are going in for routine examinations and having blood work performed are assuming they are being tested for HIV when they are not (Sobo, 1994). When asked whether they have been tested for HIV, some female respondents will simply state: “I get regular exams.”

The vagueness of these details of testing may reflect tensions that arise around the prospect of being tested, either from a fear of receiving positive test results (Spielberg, Kurth, Gorbach, & Goldbaum, 2001), or fear of a partner’s reaction. As suggested above, gendered power dynamics that compromise women’s ability to successfully negotiate condom use (Rosenthal & Levy, 2010), may also constrain conversations revolving around the issue of HIV testing. As romantic relationships evolve, individuals may become more motivated to protect the relationship from conflict that occurs as a result of bringing up the subject of HIV testing, even at the expense of self-protection (Misovich et al., 1997). The bivariate results showed that women who reported a more favorable power balance were significantly more likely to be tested, although these results were not significant in the full model. Reports about coercive sex were not significantly related to testing at the bivariate level, but this variable was significant in the full model (female results only). Although these results are thus far from definitive, such findings combined with comments within the longer narratives highlight the need for more research on conflicts and power dynamics and their association with testing behaviors within the context of a given intimate relationship. For example, Susie, a 22 year old respondent who had not been tested, noted that this topic was indeed associated with prior conflict with her partner:

It was probably in the back of mind if he had been tested and if he had been with girls who had something… I started to get into a conversation with him and he absolutely took it the wrong way…

One contribution of combining qualitative and quantitative results is that they document correlates of testing behavior within the context of a specific relationship, rather than the more typical strategy of assessing only whether an individual has ever been tested. However, this focus on a current or most recent relationship may overstate the notion that male respondents are not as likely to take partner characteristics or relationship features into account. Donnie, a 19 year old respondent, got tested because of his former girlfriend’s behavior:

And the HIV and STD test I get done. And, just to make sure. Because my ex-girlfriend, I had found out after we split up, her sex history, and it was not good. She had been around the block more than once. And that made me feel a little bit funny. So I just got tested, just to make sure.

The longer narratives are thus also useful because they highlight that experiences involving past partners may be associated with testing behaviors, and are thus part of the package of concerns that individuals bring with them as they forge the next relationship.

DISCUSSION

Although the benefits of detecting and treating HIV early are widely recognized, knowledge of facilitators that encourage and obstacles that deter HIV testing among young adult heterosexual men and women in the U.S. is limited. Overall, 40% of respondents were tested for HIV since being with their current or most recent partner; and women were significantly more likely than men to get tested (48% versus 30%, respectively). Our estimates are consistent with other studies (e.g., Arrington-Sanders & Ellen, 2008), which find that a higher percentage of women than men get tested for HIV. Our findings are also consistent with prevalence rates for women in the NSFG (Trepka & Kim, 2010), and are consistent with but higher than prevalence rates based on wave III of the Add Health, which show that about 23% of sexually experienced women aged 18 - 27 reported HIV testing in the past year (Hahm et al., 2009). Differences with the Add Health are likely due to the distinction between getting tested in the past year versus getting tested within the context of one’s most current or most recent intimate relationship, which can extend to the prior 24 months. Our study showcases the value of querying about testing within a specific relationship.

The indicators associated with HIV testing differ among men and women. Our findings are in accord with prior work on health behaviors by indicating that permissive attitudes among men and efficacy among women are related to being tested. However, these factors are not the most important predictors of whether men and women get tested. While individual factors such as permissive attitudes and condom efficacy may play a role in condom use generally, which is an important sexual health behavior, in our sample of heterosexual couples these factors seem to have little effect on HIV testing within relationships. As expected, among both men and women pregnancy with their partner is associated with greater odds of testing which is most likely tied to increased interactions with reproductive health services. Notably, pregnancy in their current or most recent relationship is relatively rare (22% of women and 12% of men) indicating that only a subset of women and men experience the positive effects of pregnancy on testing. Even among those experiencing a pregnancy within their current or most recent relationship, 15% of women and almost half of men were not tested for HIV.

A primary factor that predicts HIV testing among men is their own number of lifetime sexual partners. These are not specific to the relationship and simply present an assessment of men’s own sexual risk portfolio. Consideration of the respondents’ full romantic histories provide a more nuanced perspective suggesting that men do take relationship features into account, albeit in a somewhat different manner. The qualitative evidence showcases that men rely on their own, sometimes inaccurate, judgments about their partner’s risk level. Additionally, men appeared to rely on women getting tested to assess their own HIV status. One problem with relying on these routine testing situations for women and their partners is that it is not always clear when respondents are referencing testing for STDs in general and HIV specifically (Anderson & Blake, 2011). Yet, individuals’ assessments of personal risk are often based on myths and misperceptions or what has been referred to as “folk constructions” (Essien, Meshack, & Ross, 2002). As noted by Essien and colleagues, a prime example of a folk construction is assuming that information provided by one’s partner with respect to previous sexual activities or simply observations that she is “clean” or a “good girl” is sufficient to correctly judge levels of risk.

These results seem to suggest that men are less reliant on the relationship context to inform them on the need to be tested for HIV. It appears from these analyses that men rely more on their own personal sexual histories to inform them on the need to seek HIV testing. In contrast, women’s testing decisions appear to rest on their assessments of their partner’s sexual exclusivity and partner’s sexual experience. The qualitative analysis confirms these findings and demonstrates through women’s narratives their emphasis on their partners’ past experiences. These findings are consistent with the gender vulnerability approach.

An important focus of this paper was the attention to relationship dynamics. While relationship dynamics are not the central factors associated with testing, we observed some gender specific associations, which may deserve additional research scrutiny. Women with greater power, higher mistrust scores, and greater difficulty communicating about sex report higher odds of testing. In contrast, men who report greater love and less difficulty communicating about sexual topics experience lower odds of testing. Thus, the associations between relationship factors and testing appear to be operating in almost opposite manners for men and women. These distinct relationship patterns suggest that further attention to couple-level analyses may help elucidate the meaning of these relationship dynamics among men and women.

Our study has several limitations. It is possible that HIV testing occurred prior to measurement of some of the independent variables. For example, if an HIV test was negative, perhaps this increased an individual’s sense of trust. In our qualitative data, this point was expressed by Jason, an 18 year old respondent, in the following manner:

Sarah, yea me and Sarah were the ones that went down and got tested together… Just I, I mean I knew who she was. I knew I mean I just, basically after me and (previous partner) you know after she kind of messed around on me like that, trust with women wasn’t exactly high at the time so I just made sure we both went down there…”

Our findings represent important associations and as with all cross-sectional research, statistical associations need to be interpreted carefully and need to give attention to other possible interpretations. Another limitation is that our study is not representative of individuals who are at high risk of HIV. An asset is that we present a representative sample of young adults but more refined analyses of young adults facing high HIV risk are warranted. Our research is also limited by using self-report data which may be under-estimating HIV testing. Research that triangulates self-report and clinic based data would help to assess potential biases in our findings. Finally, certain factors not assessed in our study may be important to consider for assessing barriers and facilitators to HIV testing. For example, our study did not assess whether individuals were aware of significant others who may have tested positive for HIV (Bond et al., 2005). These factors merit continued attention in future research.

The study is important for reporting findings on HIV testing practices from a representative sample of young adult men and women in current or recent intimate relationships. Although prior research has described demographic characteristics and the influence of individuals’ own risk behaviors, our work provides a key first step toward understanding the motivations to get tested that emanate from the influence of the partner’s risk behavior or that are associated with qualities of the relationship itself, such as power differences. Our work also highlights the importance of a gender specific lens on assessments of HIV testing.

Table 3.

Odds Ratios for the Logistic Regression of Tested for HIV Since with Partner Among Unmarried Men (N = 310)

Model 1 Model 2 Model 3 Model 4 Model 5
eb SE eb SE eb SE eb SE eb SE
Intercept 0.00*** 1.75 0.00** 1.96 0.00*** 1.77 0.00*** 1.99 0.00*** 2.35
Individual Characteristics
   Permissive attitudes 0.97 0.24 1.17 0.26
   Condom efficacy 1.04 0.12 1.02 0.12
   Not sexually exclusive 0.74 0.34 1.03 0.39
   Lifetime number of sex partners 1.08** 0.03 1.08** 0.03
Partner Risk Behaviors
   Partner’s number of sex
partners		 1.01 0.05 0.96 0.06
   Partner not sexually exclusive 0.68 0.38 0.69 0.46
Relationship Dynamics
   Romantic love 1.31 0.19 1.34 0.21
   Trust 1.12 0.13 1.15 0.14
   Power
   (Same/less power)
   More power 0.88 0.36 0.73 0.38
   Difficulty communicating about
sex		 0.78 0.16 0.76 0.17
   Sexual coercion 1.10 0.12 1.11 0.12
   Pregnancy w/partner 2.45* 0.39 2.34* 0.41
Demographic characteristics
   Age 1.32*** 0.08 1.28** 0.08 1.30** 0.08 1.31** 0.09 1.26** 0.09
   Race/Ethnicity
     (White non-Hispanic)
     Non-White/Hispanic 2.66*** 0.27 2.39** 0.29 2.71*** 0.27 2.56*** 0.28 2.09* 0.30
Parent Education
   (12 years or less)
   More than 12 years 0.47** 0.27 0.46** 0.27 0.46** 0.27 0.49* 0.28 0.49* 0.29
Likelihood Ratio χ2 38.03*** 47.60*** 39.09*** 50.09*** 62.14***
Open in a new tab

Note: The omitted category is in parentheses. Source: Toledo Adolescent Relationship Study.

†

p < .10.

*

p < .05.

**

p < .01.

***

p < .001.

Acknowledgments

This research was supported by grants from The Eunice Kennedy Shriver National Institute of Child Health and Human Development (HD036223 and HD044206), and by the Center for Family and Demographic Research, Bowling Green State University, which has core funding from The Eunice Kennedy Shriver National Institute of Child Health and Human Development (R24HD050959-01).

Footnotes

1

Two additional response categories were provided in the original questionnaire so that respondents could indicate that they 1) never use birth control and 2) would never become intimate with someone before marriage. Only a small number of respondents (n = 22) provided one of these answers on one or more of the items. We coded these few responses as missing on the individual items.

2

Responses for lifetime number of sex partners that exceeded 17 were capped at 18. These responses represent the 90th percentile of all responses. That is 90% of all responses fell below 18.

3

Responses for partner’s number of sex partners that exceeded 11 were capped at 12. These responses represent the 90th percentile of all responses. That is, 90% of all responses fell below 12.

REFERENCES

  1. Amado H. Love, sex, and power: Considering women’s realities in HIV prevention. American Psychologist. 1995;50:437–447. doi: 10.1037//0003-066x.50.6.437. [DOI] [PubMed] [Google Scholar]
  2. Amaro H, Raj A. On the margin: Power and women’s HIV risk reduction strategies. Sex Roles. 2000;42(7/8):723–749. [Google Scholar]
  3. Amaro H, Raj A, Reed E. Women’s sexual health: The need for feminist analyses in public health in the decade of behavior. Psychology of Women Quarterly. 2001;25(4):324–334. [Google Scholar]
  4. Anderson J, Blake D. Adolescent and young adult understanding of the differences between Chlamydia, HIV, AIDS. Journal of Adolescent Health. 2011;48(2):S67. doi: 10.1177/0009922813479162. [DOI] [PubMed] [Google Scholar]
  5. Anderson JE, Carey JW, Taveras S. HIV testing among the general U.S. population and persons at increased risk: Information from national surveys, 1987–1996. American Journal of Public Health. 2000;90(7):1089–1095. doi: 10.2105/ajph.90.7.1089. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Arrington-Sanders R, Ellen J. Prevalence of self-reported human immunodeficiency virus testing among a population-based sample of urban African-American adolescents. Journal of Adolescent Health. 2008;43(3):306–308. doi: 10.1016/j.jadohealth.2008.01.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Arrington-Sanders R, Ellen J, Leonard L. The impact of HIV self-efficacy on attitudes toward routine HIV testing among sexually active adolescents. Journal fo Adolescent Health. 2009;44(2):S26–S27. [Google Scholar]
  8. Basen-Engquist K, Parcel GS. Attitudes, Norms, and Self-efficacy: A model of adolescents’ HIV-related sexual risk behavior. Health Education and Behavior. 1992;19(2):263–277. doi: 10.1177/109019819201900209. [DOI] [PubMed] [Google Scholar]
  9. Blanc AK. The effect of power in sexual relationships on sexual and reproductive health: An examination of the evidence. Studies in Family Planning. 2001;32(3):189–213. doi: 10.1111/j.1728-4465.2001.00189.x. [DOI] [PubMed] [Google Scholar]
  10. Bond L, Lauby J, Batson H. HIV testing and the role of individual- and structural-level barriers and facilitators. AIDS Care. 2005;17(2):125–140. doi: 10.1080/09541020512331325653. [DOI] [PubMed] [Google Scholar]
  11. Britton PJ, Levine OH, Jackson AP, Hobfoll SE, Shepherd JB, Lavin JP. Ambiguity of monogamy as a safer-sex goal among single, pregnant, inner-city women: Monogamy by whose definition? Journal of Health Psychology. 1998;3(2):227–232. doi: 10.1177/135910539800300206. [DOI] [PubMed] [Google Scholar]
  12. Bucx F, Raaijmakers Q, Van Wel F. Life course stage in young adulthood and intergenerational congruence in family attitudes. Journal of Marriage and Family. 2010;72:117–134. [Google Scholar]
  13. Campsmith M, Rhodes P, Hall HI. Estimated prevalence of undiagnosed HIV infection in the United States at the End of 2006; Poster presented at the 16th conference on Retroviruses and Opportunistic Infections; Montreal, Canada. 2009. Feb, [Google Scholar]
  14. Center for Disease Control and Prevention. Revised recommendations for HIV testing of adults, adolescents and pregnant women in health care settings. Morbidity and Mortality Weekly Report. 2006;55(No. RR 14):1–17. [PubMed] [Google Scholar]
  15. Center for Disease Control and Prevention. Persons tested for HIV – United States, 2006. Morbidity and Mortality Weekly Report. 2008;57(36):845–849. [PubMed] [Google Scholar]
  16. Center for Disease Control and Prevention. New CDC analysis reveals strong link between poverty and HIV infection. 2010 Retrieved from http://www.cdc.gov/nchhstp/newsroom/povertyandhivpressrelease.html.
  17. Chandra A, Mosher WD, Copen C, Sionean C. Sexual behavior, sexual attraction, and sexual identity in the United States: Data from the 2006 – 2008 National Survey of Family Growth (Report No. 36) Hyattsville, MD: National Center for Health Statistics; 2011. [PubMed] [Google Scholar]
  18. Chesney MA, Smith AW. Critical delays in HIV testing and care: The potential role of stigma. American Behavioral Scientist. 1999;42(7):1162–1174. [Google Scholar]
  19. Choi K, Catania JA. Changes in multiple sexual partnerships, HIV testing, and condom use among U.S. heterosexuals 18 to 49 years of age, 1990 and 1992. American Journal of Public Health. 1996;86(4):554–556. doi: 10.2105/ajph.86.4.554. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Corbett AM, Dickson-Gomez J, Hilario H, Weeks MR. A little thing called love: Condom use in high-risk primary heterosexual relationships. Perspectives on Sexual and Reproductive Health. 2009;41(4):218–224. doi: 10.1363/4121809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Denison JA, Lungu N, Dunnett-Dagg WA, McCauley A, Sweat MD. Horizons Research Summary. Washington, DC: Population Council; 2006. Social relationships and adolescents’ HIV counseling and testing decisions in Zambia. [Google Scholar]
  22. Deren S, Oliver-Velez D, Finlinson A, Robles R, Andia J, Colón HM, Shedlin M. Integrating qualitative and quantitative methods: Comparing HIV-related risk behaviors among Puerto Rican drug users in Puerto Rico and New York. Substance Use & Misuse. 2003;38(1):1–24. doi: 10.1081/ja-120016563. [DOI] [PubMed] [Google Scholar]
  23. DiClemente RJ, Peterson JL. Preventing AIDS: Theories and methods of behavioral interventions. New York, NY: Plenum; 1994. [Google Scholar]
  24. Djokic D, Englund J, Daum R, Martin R, Dozier T, Potts S, Marcinak J. HIV knowledge and attitudes toward HIV testing of South Side Chicago Housing Authority residents. AIDS Patient Care and STDs. 2009;23:23–28. doi: 10.1089/apc.2008.0110. [DOI] [PubMed] [Google Scholar]
  25. Duran D, Usman HR, Beltrami J, Alvarez ME, Valleroy L, Lyles CM. HIV Counseling and Testing among Hispanics at CDC-Funded Sites in the United States, 2007. American Journal of Public Health. 2010;100:S152–S158. doi: 10.2105/AJPH.2009.166355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Essien EJ, Meshack AF, Ross MW. Misperceptions about HIV transmission among heterosexual African-American and Latino men and women. Journal of the National Medical Association. 2002;94(5):304–312. [PMC free article] [PubMed] [Google Scholar]
  27. Evans RG, Barer ML, Marmor TR. Why are some people healthy and others not? The determinants of health of populations. New York, NY: Aldine de Gruyter; 1994. [Google Scholar]
  28. Exner TM, Hoffman S, Parkih K, Leu C, Ehrhardt AA. HIV counseling and testing: Women’s experiences and the perceived role of testing as a prevention strategy. Perspectives on Sexual and Reproductive Health. 2002;34(2):76–83. [PubMed] [Google Scholar]
  29. Fortenberry JD, McFarlane M, Bleakley A, Bull S, Fishbein M, Grimley DM, Malotte CK, Stoner BP. Relationships of stigma and shame to gonorrhea and HIV screening. American Journal of Public Health. 2002;92(3):378–381. doi: 10.2105/ajph.92.3.378. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Futterman D. HIV in adolescents and young adults: Half of all new infections in the United States. Topics in HIV Medicine. 2005;13:101–105. [PubMed] [Google Scholar]
  31. Giordano PC, Longmore MA, Manning WD. Gender and the meanings of adolescent romantic relationships: A focus on boys. American Sociological Review. 2006;71:260–287. [Google Scholar]
  32. Giordano PC, Manning WD, Longmore MA, Flanigan CM. Developmental shifts in the character of romantic and sexual relationships from adolescence to young adulthood. University Park, PA: Paper presented at the Early Adulthood in a Family Context, Penn State’s annual symposium on family issues; 2010. Oct, [Google Scholar]
  33. Graffigna G, Olson K. The ineffable disease: Exploring young people’s discourses about HIV/AIDS in Alberta, Canada. Qualitative Health Research. 2009;19(6):790–801. doi: 10.1177/1049732309335393. [DOI] [PubMed] [Google Scholar]
  34. Hahm HC, Song IH, Ozonoff A, Sassani JC. HIV testing among sexually experience Asian and Pacific Islander young women: Association with routine gynecologic care. Women’s Health Issues. 2009;19:279–288. doi: 10.1016/j.whi.2009.05.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Hall C, Darville T, Barral R, Sucato G. HIV testing in high-risk adolescent females screened for other STIs. Journal of Adolescent Health. 2011;48(2):S67–S68. [Google Scholar]
  36. Hall HI, Ruiguang S, Gerstle JE, Lee LM. Assessing the completeness of reporting the Human Immunodeficiency Virus diagnoses in 2002–2003: Capture-recapture methods. American Journal of Epidemiology. 2006;164(4):391–397. doi: 10.1093/aje/kwj216. [DOI] [PubMed] [Google Scholar]
  37. Hall HI, Song R, Rhodes P, Prejean J, An Q, Lee LM, Janssen RS. Estimates of HIV incidence in the United States. Journal of the American Medical Association. 2008;300(5):520–529. doi: 10.1001/jama.300.5.520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Harris KM, Gordon-Larsen P, Chantala K, Udry JR. Longitudinal trends in race/ethnic disparities in leading health indicators from adolescence to young adulthood. Archives of Pediatric and Adolescent Medicine. 2006;160:74–81. doi: 10.1001/archpedi.160.1.74. [DOI] [PubMed] [Google Scholar]
  39. Hatfield E, Sprecher S. Measuring passionate love in intimate relationships. Journal of Adolescence. 1986;9:383–410. doi: 10.1016/s0140-1971(86)80043-4. [DOI] [PubMed] [Google Scholar]
  40. Higgins JA, Hoffman S, Dworkin S. Rethinking gender, heterosexual men, and women’s vulnerability to HIV/AIDS. American Journal of Public Health. 2010;100(3):435–445. doi: 10.2105/AJPH.2009.159723. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Holland DC, Eisenhart MA. Educated in romance: Women, achievement, and college culture. Chicago, IL: University of Chicago Press; 1990. [Google Scholar]
  42. Hoppe MJ, Graham L, Wilsdon A, Wells EA, Nahom D, Morrison DM. Teens speak out about HIV/AIDS: Focus group discussions about risk and decision-making. Journal of Adolescent Health. 2004;35(4):345.e27–345.e35. [PubMed] [Google Scholar]
  43. Jeffries WL. HIV testing among bisexual men in the United States. AIDS Education and Prevention. 2010;22(4):356–370. doi: 10.1521/aeap.2010.22.4.356. [DOI] [PubMed] [Google Scholar]
  44. Jemmott LS, Jemmott JB. Sexual knowledge, attitudes, and risky sexual behavior among inner-city black male adolescents. Journal of Adolescent Research. 1990;5(3):346–369. [Google Scholar]
  45. Karon JM, Fleming PL, Steketee RW, DeCock KM. HIV in the United States at the turn of the century: An epidemic in transition. American Journal of Public Health. 2001;91(7):1060–1068. doi: 10.2105/ajph.91.7.1060. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Kyomugisha F. HIV, African American women and high risk in heterosexual relationships. Journal of African American Studies. 2006;10(2):38–50. [Google Scholar]
  47. Lefkowitz ES, Boone TL, Shearer CL. Communication with best friends about sex-related topics during emerging adulthood. Journal of Youth and Adolescence. 2004;33(4):339–351. [Google Scholar]
  48. Logan TK, Cole J, Leukefeld C. Women, sex, HIV: Social and contextual factors, meta-analysis of published interventions, and implications for practice and research. Psychological Bulletin. 2002;128(6):851–885. doi: 10.1037/0033-2909.128.6.851. [DOI] [PubMed] [Google Scholar]
  49. Longmore MA, Manning WD, Giordano PC, Rudolph JL. Contraceptive self-efficacy: Does it influence adolescents’ contraceptive use? Journal of Health and Social Behavior. 2003;44:45–60. [PubMed] [Google Scholar]
  50. Longmore MA, Eng AL, Giordano PC, Manning WD. Parenting and adolescents’ sexual initiation. Journal of Marriage and the Family. 2009;71:969–982. doi: 10.1111/j.1741-3737.2009.00647.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Lyons H, Giordano PC, Manning WD, Longmore MA. Identity, peer relationships, and adolescent girls’ sexual behavior: An exploration of the contemporary double standard. Journal of Sex Research. 2010 doi: 10.1080/00224499.2010.506679. forthcoming. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Marhefka SL, Demetriou NK. Who is responsible for preventing HIV transmission? Attitudes of HIV+ and HIV− adolescent girls. Journal of Adolescent Health. 2009;44(2):S26. [Google Scholar]
  53. Misovich SJ, Fisher JD, Fisher WA. Close relationships and elevated HIV risk behavior: Evidence and possible underlying psychological processes. Review of General Psychology. 1997;1(1):72–107. [Google Scholar]
  54. Mosher WD, Jones J. Use of contraception in the United States: 1982 – 2008: Data from the National Survey of Family Growth (Series 23, No. 29) Hyattsville, MD: National Center for Health Statistics; 2010. [PubMed] [Google Scholar]
  55. Nguyen TQ, Ford CA, Kaufman JS, Leone PA, Suchindran C, Miller WC. HIV testing among young adults in the United States: Associations with financial resources and geography. American Journal of Public Health. 2006;96:1031–1034. doi: 10.2105/AJPH.2005.063248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Pulerwitz J, Amaro H, De Jong W, Gortmaker SL, Rudd R. Relationship power, condom use and HIV risk among women in the USA. AIDS Care: Psychological and Socio-medical Aspects of AIDS/HIV. 2002;14(6):789–800. doi: 10.1080/0954012021000031868. [DOI] [PubMed] [Google Scholar]
  57. Rosenstock IM, Stretcher VJ, Becker MH. The health belief model and HIV risk behavior change. In: DiClemente RJ, Peterson JL, editors. Preventing AIDS: Theories and methods of behavioral interventions. New York, NY: Plenum; 1994. pp. 5–24. [Google Scholar]
  58. Rosenthal L, Levy SR. Understanding women’s risk for HIV infection using social dominance theory and the four bases of gendered power. Psychology of Women Quarterly. 2010;34:21–35. [Google Scholar]
  59. Santelli JS, Lowry R, Brener ND, Robin L. The association of sexual behaviors with socioeconomic status, family structure, and race/ethnicity among U.S. adolescents. American Journal of Public Health. 2000;90(10):1582–1588. doi: 10.2105/ajph.90.10.1582. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Setia MS, Quesnel-Vallee A, Curtis S, Lynch J. Assessing the role of individual and neighbourhood characteristics in HIV testing: Evidence from a population based survey. The Open AIDS Journal. 2009;3:46–54. doi: 10.2174/1874613600903010046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Sobo EJ. Attitudes toward HIV testing among impoverished inner-city African-American women. Medical Anthropology. 1994;16:17–38. doi: 10.1080/01459740.1994.9966107. [DOI] [PubMed] [Google Scholar]
  62. Sobo EJ. Finance, romance, social support and condom use among impoverished inner-city women. Human Organization. 1995;54(2):115–128. [Google Scholar]
  63. Sobo EJ. Self-disclosure and self-construction among HIV-positive people: The rhetorical uses of stereotypes and Sex. Anthropology & Medicine. 1997;4(1):67–87. [Google Scholar]
  64. Spielberg F, Kurth A, Gorbach PM, Goldbaum G. Moving from apprehension to action: HIV counseling and testing preferences in three at-risk populations. AIDS Education and Prevention. 2001;13(6):524–540. doi: 10.1521/aeap.13.6.524.21436. [DOI] [PubMed] [Google Scholar]
  65. Stein JA, Nyamathi A. Gender differences in behavioural and psychosocial predictors of HIV testing and return for test results in a high-risk population. AIDS Care. 2000;12(3):343–356. doi: 10.1080/09540120050043007. [DOI] [PubMed] [Google Scholar]
  66. Steinberg J. Three Letter Plague: A Young Man’s Journey through a Great Epidemic. New York: Vintage; 2008. [Google Scholar]
  67. Straub DM, Arrington-Sanders RD, Harris R, Willard N, Kapogiannis B, Emmanuel P, Futterman D, Ellen JM. Correlates of HIV testing history among urban youth recruited through venue-based testing in 15 U.S. cities. Sexually Transmitted Diseases. 2011:1. doi: 10.1097/OLQ.0b013e318214bb70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Tolou-Shams M, Payne N, Houch C, Pugatch D, Beausoleil N, Brown LK Project SHIELD Study Group. HIV testing among at-risk adolescents and young adults: A prospective analysis of a community sample. Journal of Adolescent Health. 2007;41:586–593. doi: 10.1016/j.jadohealth.2007.06.004. [DOI] [PubMed] [Google Scholar]
  69. Trepka MJ, Kim S. Prevalence of human immunodeficiency virus testing and high-risk human immunodeficiency virus behavior among 18 to 22 year-old students and nonstudents: Results of the national survey of family growth. Sexually Transmitted Diseases. 2010;37(10):653–659. doi: 10.1097/OLQ.0b013e3181e1a766. [DOI] [PubMed] [Google Scholar]
  70. Trieu SL, Modeste NN, Marshak HH, Males MA, Bratton SI. Partner communication and HIV testing among U.S. Chinese college students. American Journal of Health Behavior. 2010;34(3):362–373. doi: 10.5993/ajhb.34.3.11. [DOI] [PubMed] [Google Scholar]
  71. Warr DJ. The importance of love and understanding: Speculation on romance in safe sex health promotion. Women’s Studies International Forum. 2001;24(2):241–252. [Google Scholar]
  72. Worth D. Sexual decision-making and AIDS: Why condom promotion among vulnerable women is likely to fail. Studies in Family Planning. 1989;20(6):297–307. [PubMed] [Google Scholar]

RESOURCES