What you don’t know can hurt you: Perceptions of sex-partner concurrency and partner-reported behavior
Article Outline
Abstract
Background
The objectives of this study were as follows: (1) to determine the extent of agreement between adolescents’ perceptions of sex-partner concurrency (having a partner who has other overlapping sexual partnerships) and their partners’ self-reported concurrency, and (2) to identify the relationship-level factors associated with agreement.
Methods
Adolescents ages 14 to 19 years, along with their main sex partners, were recruited from a primary care clinic and a public sexually transmitted disease clinic and interviewed separately about their own concurrency status and their perception of their partner’s concurrency status. Information from both participants and their partners were included in this analysis (N = 90 couples).
Results
This sample consisted of African-American, sexually experienced heterosexual couples. For males and females who perceived their partner not to have other partners, 16% and 37% of their sex partners reported having other partners, respectively. Of males and females who perceived their partner to have other partners, 80% and 39% of their sex partners reported not having other partners, respectively. Multivariate logistic regression revealed that adolescents who had been with their partners for more than 6 months and considered themselves more emotionally close were nearly twice as likely to agree on concurrency.
Conclusions
Among adolescent couples, agreement between perceptions of sex-partner concurrency and partner-reported behavior was low. To the extent that partner self-reports are accurate, individuals who presume that they are in a mutually monogamous relationship often underestimate their own sexually transmitted disease risk. To appropriately tailor risk reduction messages, prevention efforts need to consider adolescents’ perceptions of concurrent sex partners.
Keywords: Concurrency , Sexual partnerships , Interpartner agreement , Risk perception , Adolescents , Sexual risk
Research has shown that concurrent partnerships (i.e., sexual partnerships that overlap in time) have a dramatic effect on the transmission patterns of sexually transmitted diseases (STDs) and human immunodeficiency virus (HIV) [1]. Concurrency is relatively common among adolescents, ranging from 12% to more than 60% in some populations [2], [3], [4], [5]. Adolescents with sex partners involved in concurrent partnerships are at increased risk for STDs because of a greater likelihood that their partner could acquire and transmit an infection.
Adolescents who are unaware of a partner’s concurrent relationships may not see the need to take preventive action. Empiric studies show that an individual’s perception of a sex-partner’s risk behavior influences his or her own perceived personal risk and subsequent condom use [6], [7], [8], [9]. In addition, adolescents consider a partner’s background and other relationship factors when assessing STD risk [8], [10]. Kershaw et al [10] found that female adolescents assess their STD risk in a sexual relationship based on contextual factors, such as the length of the relationship. The sum of this research suggests that adolescents tend to equate the length of a relationship and emotional safety with a decreased risk for infection [10], [11].
Although data on individual perceptions are important, the extent to which adolescents’ perceptions match partner behavior may be an indication of actual STD risk. Research cannot measure directly the sexual behaviors of individuals and therefore the extent to which couples agree on behaviors is our best assessment of what are likely to be accurate risk perceptions. Partnerships in which individual perceptions match partner-reported behavior may afford the opportunity to determine partner-specific risks better.
The existing data on interpartner agreement shows variable findings related to the population and behaviors under investigation [6], [12], [13], [14], [15], [16], [17], [18], [19], [20]. Studies examining behaviors within the partnership (frequency of intercourse, condom use) for the purpose of examining the reliability of individual reports show good to excellent agreement among both homosexual and heterosexual couples [14], [15], [16], [17], [18], [19], [20]. The few studies considering risk behaviors occurring outside of the partnership show low and moderate levels of agreement [6], [12], [21].
There is a dearth of research investigating partner agreement among adolescent populations. Seal [21] examined interpartner agreement among college students. Within this sample, 35% of males and 53% of females were not aware of their partner’s reported concurrency. Although these results may be generalizable to similar college populations who often are at low risk for STD or HIV infection, we remain uninformed of interpartner agreement among high-risk, inner-city youth.
The objectives of this study were to determine the extent of agreement between adolescents’ perceptions of sex-partner concurrency and their partners’ self-reported concurrency. We also sought to identify relationship variables associated with agreement. We hypothesized that individuals who perceived themselves to be close with their main sex partner have been with their partner longer, and those who have lived with their partner would be more likely to agree with their main sex partners about concurrency. We chose to examine relationship factors because characteristics of the partnership and feelings about the partner and the relationship have been shown to influence perceptions of risk [11].
Methods
Study population
The couples used in this analysis were part of the Perceived Risk of Sexually Transmitted Disease study, a 5-year longitudinal study of STD risk perceptions among adolescents 14 to 19 years of age. Between August 2000 and June 2002, trained research assistants attempted to recruit all sexually active (defined as vaginal intercourse in the past 6 months), 14- to 19-year-olds presenting for reproductive care at the Johns Hopkins University teenage clinic and 1 of 2 public STD clinics in Baltimore City, MD. These clinics primarily serve African-American, lower-income, inner-city residents. Written informed consent was obtained from each adolescent. Parental/guardian consent was not required for this study because minors are deemed fit to seek reproductive care without parental consent. We obtained approval from the Johns Hopkins University Joint Committee on Clinical Investigations for this study.
We collected data from face-to-face interviews conducted at the baseline phase of the longitudinal study. Each interview lasted approximately 1 to 1.5 hours. Questions included characteristics of the respondent’s sociodemographics, sexual partnerships, and sexual behaviors. Participants also provided partner-specific data. They were asked to list their sexual partners from the past 3 months and identify the type of sex partner as main or casual. A main partner was defined as someone you have sex with and you consider this person to be the person you are serious about whereas a casual partner was defined as someone you have sex with but you do not consider this person to be a main partner. For the most recent main partnership, perceptions about the sex-partner’s demographics and risk behaviors also were obtained, including whether the sex partner had concurrent sex partners over the course of their sexual relationship. Participants provided the addresses and telephone numbers for all listed partners. Partners were contacted about their eligibility to participate in the study and were instructed to contact study staff for participation. In addition to the baseline interview, urine samples were collected and tested for chlamydia and gonorrhea. We offered participants and partners $25 as compensation for the time and effort required for this study.
We approached, and screened for eligibility, 669 clinic attendees. A total of 542 (81%) individuals were eligible and 385 (71%) agreed to participate. We were able to enroll 1 sex partner for 132 of the original participants (index participant). However, for the purposes of this study, we restricted our analyses to couples in which the following requirements were met: (1) the index reported that the enrolled sex partner was his/her most recent main partner, (2) the enrolled main partner also identified the listed index participant as the most recent main partner. By using this protocol, the sample of index-partner combinations decreased from 132 to 90 couples, representing 180 individuals (Figure 1).
Figure 1.
Validation of adolescent partnerships.
Measures
Respondents were asked whether they had concurrent partnerships during their most recent main sex-partner relationship and whether their most recent main sex partner had concurrent partnerships during that relationship. Agreement was defined as an instance in which an individual perceived their partner to be in a concurrent relationship and their partner reported having other sex partners, or, alternatively, an instance in which an individual perceived their partner not to be in a concurrent relationship, and the partner reported no other sex partners. Demographic measures included gender, age, ethnicity, and education level of both the index participant and their partner. Partnership measures included duration of the relationship (continuous response in days or months), cohabitation (yes, no), closeness (very, somewhat, not very, not at all), trust (completely, somewhat, not very much, not at all), love (totally, somewhat, not very much, not at all).
Data analysis
Statistical analyses were conducted using STATA software (Intercooled STATA 7, College Station, TX). Given the nonindependence of male and female data, agreement analyses were conducted separately. The first analyses described the subject’s perceptions (no, yes, do not know) of their partner’s concurrency and the partner’s reported behavior (no, yes). We computed the percentage of individuals who perceived their partner to have concurrent relationships, have no concurrent relationships, or did not know for those partners reporting concurrency. Likewise, we computed the percentage of individuals who perceived their partners to have concurrent relationships, have no concurrent relationships, or did not know for those partners not reporting concurrency. The κ values then were calculated for males and females to measure the amount of agreement between the respondent’s perceptions of their partner’s concurrency (no, yes) and their partner’s reported concurrency (no, yes). Respondents reporting that they did not know whether their partner had other sex partners were excluded from the κ calculations. Conventionally, κ values of .40 or less represent fair to poor agreement, .41 to .60 represent moderate agreement, .61 to .80 represent substantial agreement, and .81 or more represent almost perfect agreement [22]. Finally, multiple logistic regression analyses were performed to assess the relationship of demographic and partnership variables on partner agreement for concurrency. Partnership duration was recategorized as less than 6 months or greater than 6 months. An emotional closeness variable was created representing a summed scale of closeness, trust, and love, and analyzed as continuous (α = .69). An age-discordance variable was created and defined as a partnership in which the male partner was 2 or more years older than the female partner (yes). Non–age-discordant partnerships were defined as partnerships in which the female was older, or both partners were of similar age (<2-year difference). Variables included in the multiple regression model achieved a p value of .10 in the bivariate analysis. Generalized estimating equations were used to adjust the logistic regression for the nonindependence of the male and female participants.
Results
The sample included 90 sexually experienced adolescent couples, the majority of whom were African-American (Table 1). Because of differential reporting and missing data on concurrency perceptions and reported behaviors, 4 males were excluded from the present analysis. Thus, our final sample consisted of 176 individuals (90 females, 86 males). The mean age was 18.9 (SD = 2.5) years for males and 17.3 (SD = 1.5) years for females (p < .001). Over one third of males currently were enrolled in school, compared with 62.2% of females (p < .001). Males had a younger age of first intercourse relative to females (13.2 vs. 14.6; p < .001). Males were significantly more likely to report concurrency compared with females (38.4% vs. 13.3%; p < .001). Although the 88 individuals excluded from this analysis (as described earlier) were more likely to be older (p = .04), no other significant differences were found.
Table 1. The prevalence of selected demographic variables and risk behaviors among male and female subjects
| Characteristic | Males (N = 86) | Females (N = 90) |
|---|---|---|
| Age | 18.9 (SD=2.5) | 17.3(SD=1.5) |
| Race⁎ | ||
| African-American | 84(97.7) | 88(97.8) |
| White/other | 1(1.2) | 1(1.1) |
| School status† | ||
| Yes | 28(32.6) | 56(62.2) |
| No | 56(65.2) | 34(37.8) |
| Gonorrhea history | ||
| Yes | 27(31.4) | 19(21.1) |
| No | 59(68.6) | 71(78.9) |
| Condom use during last sexual contact‡ | ||
| Yes | 33(38.4) | 39(43.3) |
| No | 51(59.3) | 49(54.4) |
| Age at first intercourse§ | 13.2(SD=2.0) | 14.6(SD=1.4) |
| Concurrency | 33(38.4) | 12(13.3) |
⁎ One missing value for males and females. |
† Two missing values for males. |
‡ Two missing values for males and females. |
§ One missing value for males and 2 missing values for females. |
Measures of agreement between the respondent’s perceptions and the partner-reported concurrent relationships are shown in Table 2. Among male participants who perceived their partner as being in concurrent relationships, only 20.0% (n = 3) of their female partners reported having other partners, and the remaining majority of their partners reported no concurrency (80.0%; n = 12). Additional data on these 12 males revealed that 41.6% (n = 5) were involved in concurrent relationships themselves (data not shown). For those males who did not perceive their partners to have a concurrent relationship, 16.0% (n = 8) of their partners reported being in a concurrent relationship. Half of the males who underestimated their partner’s concurrency reported having no concurrent relationship (data not shown).
Table 2. Agreement between respondent’s perception of sex partner concurrency and partner’s reported behavior
| Partner’s self-reported concurrent relationship | |||
|---|---|---|---|
| Yes | No | Totals | |
| Male perception of partners’ concurrent relationship | |||
| No | 8(16.0) | 42(84.0) | 50(100%) |
| Yes | 3(20.0) | 12(80.0) | 15(100%) |
| Do not know | 2(9.5) | 19(90.5) | 21(100%) |
| κ coefficient | .04 | ||
| Female perception of partners’ concurrent relationship | |||
| No | 20(37.0) | 34(63.0) | 54(100%) |
| Yes | 16(61.5) | 10(38.5) | 26(100%) |
| Do not know | 2(20.0) | 8(80.0) | 10(100%) |
| κ coefficient | .22 | ||
Among females who perceived their partners as being in a concurrent relationship, 61.5% (n = 16) of their male partners reported having other partners, whereas only 38.5% (n = 10) reported no concurrent relationships. Sixty percent of females (n = 54) felt their partners were not in concurrent partnerships; however, 37.0% (n = 20) of their partners responded that they did have other partners. This was more than twice the proportion of males who underestimated sex-partner concurrency. In addition, similar proportions of females both overestimated and underestimated the concurrency of their male partners. Among these females, 80% and 90% reported no concurrent relationships, respectively. Overall, males’ and females’ perceptions of sex-partner concurrency agreed poorly with their partner’s reported concurrency. The κ scores were .04 and .22 for males and females, respectively.
The results of bivariate and multivariate associations are shown in Table 3. Only emotional closeness was associated significantly with agreement on sex-partner concurrency in the bivariate model. Although partnership duration was slightly greater than our value for inclusion into the multivariate model, we retained this variable because of its theoretical relevance. Thus, partnership duration, emotional closeness, and age discordance were entered into the multivariate model controlling for age and gender. In this model, those individuals emotionally closer to their partners were 1.3 times as likely to have agreement between their concurrency perceptions and partner-reported behavior (odds ratio = 1.33; 95% confidence interval = 1.06–1.67). However, individuals who were in the partnership for more than 6 months were less likely to have interpartner agreement on concurrency (odds ratio = .43; 95% confidence interval = .20–.94).
Table 3. Bivariate and multivariate models predicting interpartner agreement on sex partner concurrency
| Variable | Bivariate model | Multivariate model⁎ | ||||
|---|---|---|---|---|---|---|
| Odds ratio | 95% confidence interval | p | Odds ratio | 95% confidence interval | p | |
| Gender | ||||||
| Male | ||||||
| Female | 1.22 | .59–2.50 | .59 | 1.12 | .56–2.29 | .73 |
| Age | 1.04 | .90–1.20 | .63 | 1.01 | .87–1.17 | .87 |
| Duration of relationship | ||||||
| <6 mo | ||||||
| >6 mo | .59 | .30–1.15 | .12 | .43 | .20–.94 | .03 |
| Cohabitation | ||||||
| No | ||||||
| Yes | 1.10 | .58–2.05 | .78 | |||
| Emotional closeness | 1.26 | 1.02–1.55 | .03 | 1.33 | 1.06–1.67 | .02 |
| Age discordance | ||||||
| Similar age | ||||||
| Male older | 1.69 | .92–3.13 | .09 | 1.56 | .78–3.12 | .21 |
⁎ Model included partnership duration, emotional closeness, and age discordance controlling for age and gender. |
Because this last finding was counterintuitive, the interactive effects of partnership duration and emotional closeness on agreement were examined. The data were stratified into 2 groups based on partnership duration. In these models, the association between emotional closeness and interpartner agreement was significant only for those who were in the partnership for longer than 6 months (Figure 2). For these couples, interpartner agreement was nearly 2 times more likely if they were emotionally close (odds ratio = 1.82; 95% confidence interval = 1.20–2.79). The effect of emotional closeness was not significant for those in shorter duration partnerships.
Figure 2.
Agreement as a function of emotional closeness.
Discussion
We conducted this study to determine the extent of agreement between adolescents’ perceptions of sex-partner concurrency and partner-reported concurrency, in addition to identifying predictors of interpartner agreement. Adolescent perceptions often disagreed with partner-reported behavior and overall interpartner agreement was low. Males tended to overestimate concurrency of their female partners. The greatest discordance occurred among the males who perceived their partners as having other partners, when in actuality the overwhelming majority of their female partners reported no concurrent partnerships. One possible explanation for this observation is that these males were involved in concurrent partnerships themselves. Nearly 42% of these males reported having concurrent partnerships, of which the majority of their female partners were unaware. It could be that the males were projecting their own concealed concurrency onto their partners. Furthermore, these males may rationalize their behavior by convincing themselves that their partners are involved in concurrent relationships as well.
More important is the observation that females tended to underestimate sex-partner concurrency compared with males. Among females who did not perceive their partners to be in concurrent relationships, nearly 40% of their partners reported such behavior, whereas among males who did not perceive their partners as having concurrent relationships, only 16% reported concurrency. In addition, females exhibited similar proportions of overestimation and underestimation of their male partner’s concurrent relationships. The observation that the majority of these females themselves reported no concurrent relationship leads to 2 plausible explanations. For those who overestimated, perhaps there is a normative expectation that male partners will have more than one sexual relationship. Qualitative research currently being conducted by the first author suggests that female adolescents believe their male partners are more susceptible to pressures to initiate concurrent relationships. In contrast, those females underestimating their partner’s concurrency may be operating under an assumption of monogamy. This particularly is true for those adolescents involved with a main partner. Sobo [23] found that women often fool themselves by creating “monogamy narratives” to justify unprotected sex with long-term partners. The inner-city African-American women of Sobo’s study created idealized, monogamous unions, oftentimes denying the possibility that their partners were involved in concurrent relationships [23].
Our research highlights the importance of relationship characteristics in interpartner agreement on concurrency. Specifically, duration of relationship and emotional closeness were associated significantly with interpartner agreement. An interaction between duration and emotional closeness revealed that adolescents in partnerships for longer than 6 months, who are close to their partners emotionally, were nearly twice as likely to have agreement on sex-partner concurrency compared with those less close emotionally. This agreement may be an indication of partner communication, an important component of STD and HIV prevention efforts. Partnerships in which both members agree about concurrency may reflect closer relationships with open communication. Research investigating partner notification indicates that those individuals with high self-efficacy and strong emotional ties to their partner are more likely to disclose risk information [24]. In a study of African-American adolescents, those who had better relationship quality (defined as emotional, affiliative, and supportive characteristics) were 1.17 times more likely to notify partners of positive STD results [24].
Similar to the research conducted by Stoner et al [12], our data do not address the validity of adolescent responses, but rather the degree to which both individuals in a partnership agree on the occurrence of concurrency. Although these findings suggest that adolescents in longer, more emotionally close partnerships may communicate about sexual risk behaviors occurring outside of the partnership, the validity of what is communicated and internalized is still in question. Among couples showing interpartner agreement, our results indicate the accuracy of adolescent perceptions only to the extent of the validity of their partner’s self reports. When couples agree, both may be giving accurate and valid reports, or they may be giving reliable but inaccurate reports [15]. Perceptions of a partner’s risk depend on that partner’s truthfulness and accurate disclosure of risk [12]. When conversations about partner-specific risk occur, many youth lie about their past sexual history [25]. Thus, the low levels of agreement observed in our study, specifically the underestimates of sex-partner concurrency, may be an indication of untruthfulness on the part of the concurrent partners. Alternatively, adolescents may be assuming a mutually monogamous relationship without engaging in conversation with their main partners. Research also has shown that young adults confuse the emotional safety of a regular partner with physical safety independent of any verified decrease in the risk for STD or HIV infection (i.e., mutual monogamy) [26]. Sixty-one percent of participants reported recalling or assuming monogamy in an emotionally safe regular-partner scenario, even though this scenario made no mention of mutual monogamy [26]. The large percentage of females underestimating their partner’s concurrent relationships observed in our study may be explained by these hypotheses.
Strengths
This study builds on previous research with adults and examines interpartner agreement for sex-partner concurrency and its associated predictors among high-risk adolescents. This study examined partnership variables associated with agreement. An additional strength of this work is that we explicitly looked at concurrency within a defined main partnership from the perspective of the adolescent. We also used sociometric data in the examination of risk behaviors that occur outside of the immediate partnership. The majority of studies investigating interpartner agreement focus on the reliability of behaviors occurring within the partnership (frequency of intercourse and condom use). Our aim was not to determine the reliability or validity of reports, but rather the extent to which adolescents in main partnerships agree on the occurrence of a particular risk behavior.
Limitations
Limitations of this research include the examination of an absolute measure of agreement. We did not distinguish between interpartner agreement on the presence versus absence of concurrency. As it relates to STD- and HIV-prevention efforts, adolescents who agree on the presence of concurrency and those who underestimate their partner’s concurrency are important targets for intervention. The data presented here do not provide information on that distinction because of the small cell size for those couples agreeing on the presence of concurrency. In addition, we could not determine the extent to which disagreement was related to inaccurate perceptions or invalid partner report [6]. Our study used self-reported data, and there is the possibility of differential reporting of concurrent relationships by gender. It is plausible that the overestimation of concurrency shown by males could be caused by underreporting by their female partners because of social desirability. Further, the underestimates of females could be because of overreporting of concurrent relationships by their male partners. However, in this case, as suggested by Ellen et al [6], there is a tendency to underreport high-risk behaviors, thus the high level of underestimation by females in all probability is indicative of misperceptions rather than overreporting of risks. Our study did not explicitly examine the relationship between partner communication and interpartner agreement on concurrency, although agreement would suggest some common knowledge of partner risk behaviors supporting communication. Finally, our sample included predominantly African-American adolescents recruited from reproductive health and STD clinics. The results observed may not be generalizable to other populations who face a lower risk for STD acquisition.
Implications
This study supports the need for increased awareness of inaccurate risk perception within adolescent partnerships, particularly in regard to sexual concurrency [12]. Adolescents either are not asking about or are receiving false information about the risk behaviors of their partners [6]. Therefore, they are perceiving themselves inaccurately to be at low risk for STDs, including HIV, and may fail to take appropriate preventive action [6]. Interventions should promote enhanced awareness and risk disclosure by building partner communication skills [12]. Additional focus should be on partner-specific risk perception. The perception of lower levels of risk is associated with higher levels of commitment and longer relationships [11]. Efforts to increase partner-specific risk awareness among adolescents in long-term or regular partnerships are critical, particularly because these relationships are characterized by low levels of condom use. Future research should explore the relationship between accurate partner-specific risk assessments through interpartner agreement and subsequent preventive behavior.
Acknowledgments
Supported by a grant from the National Institute of Allergy and Infectious Diseases (5 RO1 AI 36986-08).
Presented and awarded the New Investigator Award at the Society for Adolescent Medicine Scientific Meeting, St. Louis, MO, March 24-27, 2004.
References
- . Concurrent partnerships and transmission dynamics in networks . Soc Networks . 1995;17:299–318
- The role of sequential and concurrent sexual relationships in the risk of sexually transmitted diseases among adolescents . J Adolesc Health . 2003;32:296–305
- . American adolescents (sexual mixing patterns, bridge partners and concurrency) . Sex Transm Dis . 2002;29:13–19
- Levels of sexual activity among adolescent males in the US . Fam Plann Perspect . 1991;23:162–167
- Concurrent sex partners and risk for sexually transmitted diseases among adolescents . Sex Transm Dis . 1999;26:208–212
- Individuals’ perceptions about their sex partners’ risk behaviors . J Sex Res . 1998;35:328–332
- . Increased attention to human sexuality can improve HIV-AIDS prevention efforts (key research issues and directions) . J Consult Clin Psychol . 1995;65:907–918
- . Sexual communication in the age of AIDS (the construction of risk and trust among young adults) . Soc Sci Med . 1995;41:1311–1323
- Determinants of self-perceived risk for AIDS . J Health Soc Behav . 1990;3:384–394
- Misperceived risk among female adolescents (social and psychological factors associated with sexual risk accuracy) . Health Psychol . 2003;22:523–532
- . Partner-specific risk perception (a new conceptualization of perceived vulnerability to STDS) . J Appl Soc Psychol . 1999;29:667–684
- Avoiding risky sex partners (perception of partners’ risks v partner’s self reported risks) . Sex Transm Infect . 2003;79:197–201
- . The accuracy of husbands’ and wives’ reports of the frequency of marital coitus . Popul Stud . 1964;18:165–173
- Validity of sexual histories in a prospective study of male sexual contacts of men with AIDS or an AIDS-related condition . Am J Epidemiol . 1988;128:719–728
- Reliability of partner reports of sexual history in a heterosexual population at a sexually transmitted disease clinic . Sex Transm Dis . 1996;23:446–452
- . Reliability of reports of sexual behavior (a study of married couples in rural West Africa) . Am J Epidemiol . 1995;141:1194–1200
- . Sexual histories of heterosexual couples with one HIV-infected partner . Am J Public Health . 1990;80:990–991
- Reliability of sexual histories in heterosexual couples . Sex Transm Dis . 1995;22:169–172
- Corroboration of sexual histories among male homosexual couples . Am J Epidemiol . 1992;135:79–84
- Interpartner reliability of reporting of recent sexual behaviors . Am J Epidemiol . 1991;134:1159–1166
- . Interpartner concordance of self-reported sexual behavior among college dating couples . J Sex Res . 1997;34:39–55
- . Statistical Methods for Rates and Proportions . New York: John Wiley & Sons; 1981;
- . Inner-city women and AIDS (the psycho-social benefits of unsafe sex) . Cult Med Psychiatry . 1993;17:455–485
- The role of self efficacy and relationship quality in partner notification by adolescents with sexually transmitted infections . Arch Pediatr Adolesc Med . 2002;156:113–137
- . Sex, lies, and HIV . N Engl J Med . 1990;322:774–775
- . Blurring emotional safety with physical safety in AIDS and STD risk estimations (the casual/regular partner distinction) . J Appl Soc Psychol . 2000;30:2467–2490
PII: S1054-139X(05)00054-6
doi:10.1016/j.jadohealth.2005.01.012
© 2006 Society for Adolescent Medicine. Published by Elsevier Inc. All rights reserved.
