Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Nov 1.
Published in final edited form as: Health Psychol. 2019 Aug 5;38(11):1001–1009. doi: 10.1037/hea0000780

Depressive Symptoms as a Longitudinal Predictor of Sexual Risk Behaviors among African American Adolescents

Jacklyn D Foley 1, Peter A Vanable 1, Larry K Brown 2,3, Michael P Carey 2, Ralph J DiClemente 4, Daniel Romer 5, Robert F Valois 6
PMCID: PMC6800787  NIHMSID: NIHMS1044242  PMID: 31380687

Abstract

Objective:

Understanding individual level factors associated with sexual risk behavior among African American adolescents remains an important public health priority. The current secondary data analysis examined the longitudinal association between a baseline assessment of depressive symptoms and sexual risk behaviors reported six-months later; the purpose was to determine whether the association of depressive symptoms to risky sex varies as a function of gender. A secondary aim was to examine self-efficacy for sex refusal and condom use assessed at a three-month follow-up as mediators of the depressive symptoms-sexual risk relationship.

Methods:

The sample consisted of 782 sexually active African American adolescents (M age = 15.3, SD = 1.08; 54% female) recruited to participate in a sexual health intervention trial. Data analyses focused on vaginal sex, and outcomes included: (a) sexual activity with two or more partners in the previous three-months; (b) the relative frequency of condom use in the previous three-months; (c) non-condom use at last occasion of sex; and (d) positive STI screening.

Results:

Depressive symptoms predicted sex with two more partners for female participants, but no other risk markers for the sample as a whole. However, there was a significant indirect effect of depressive symptoms on condomless sex via decreased condom use self-efficacy for both male and female adolescents.

Conclusions:

These findings have important implications for HIV/STI prevention, where behavioral interventions may benefit from modules that include a focus on the influence of mood on self-efficacy for safer sex practices.

Keywords: Depressive symptoms, sexual risk, self-efficacy, African American adolescents

African Americans account for nearly 60% of all new HIV infections occurring among adolescents and young adults (ages 13–24) living in the United States (Centers for Disease Control and Prevention, 2016). African American adolescents are also at elevated risk for other sexually transmitted infections (STIs); rates of chlamydia, gonorrhea and syphilis are 5–10 times more prevalent among African Americans compared to their Caucasian counterparts (Centers for Disease Control and Prevention, 2017). Sexual health disparities are attributed, in part, to less access and utilization of HIV/STI prevention and care services (Levy et al., 2004; Parrish & Kent, 2008). Nonetheless, approximately half of African American adolescents report sexual activity, exceeding the rate for the general population of youth in high school, while also reporting lower rates of condom protected sex (Kann et al., 2018). Thus, HIV/STI prevention intervention development tailored to African American adolescents remains a critical area of study.

Behavioral interventions based in Social Cognitive Theory (SCT) have been successful in improving condom use and reducing HIV/STI incidence (Protogerou & Johnson, 2014). According to SCT, an individual must have the behavioral skills (e.g., correct condom use), as well as a sense of self-efficacy (i.e., judgment of one’s capacity to perform the behavior successfully) to optimize behavioral health outcomes. While interventions have previously focused on improving self-efficacy through learning and mastery of the identified skill, psychological states are also critical. That is, improving emotional states may ensure that a person is emotionally stable and cognitively engaged prior to attempting a new behavior, which may enhance the likelihood of more successful outcomes (Bandura, 1986).

Consistent with SCT, depressive symptoms have the potential to undermine safer sex behaviors via decrements in self-efficacy among African American adolescents. Indeed, depressive symptoms may be more prevalent among African American adolescents and linked to experiences and perceptions of racial discrimination (English, Lambert, & Ialongo, 2014; Seaton Douglass, 2014) with the population also reporting less engagement in mental health treatment (Olfson, Druss, & Marcus, 2015). Further, there is evidence that depressive symptoms are positively associated with perceived difficulty of, and lower self-efficacy for, safer sex practices among African American adolescents (Seth, Raiji, DiClemente, Wingood, & Rose, 2009; Wade, Harper, & Bauermeister, 2018) and with sexual risk behaviors among adolescents (e.g., Brawner et al., 2017; Elkington, Bauermeister, & Zimmerman, 2010) and adults (e.g., Coyle et al., 2019; Millar, Starks, Grov, & Parsons, 2017; Miltz et al., 2017). Finally, there is empirical evidence of a positive association between measures of self-efficacy with sex refusal and consistent condom use behaviors in both domestic and international samples of Black/African American adolescents (Closson et al., 2018; Crosby, 2016; Voisin, Hotton, Tan, & Diclemente, 2013) and adults (Nehl, Klein, Sterk, & Elifson, 2016). However, the majority of studies focus on cross-sectional associations with adult populations.

Only five studies have used a prospective design to explore depressive symptoms as a predictor of sexual risk behaviors among African American adolescents, and most studies focus on comparing female adolescents with and without depressive symptoms. For instance, DiClemente et al. (2001) found that African American female adolescents reporting depressive symptoms were more likely to report condomless sex in the past month, fear of the negative consequences associated with condom negotiation, and more perceived barriers to condom-protected sex six-months later (N = 522). Similarly, Seth et al. (2011) reported that African American female adolescents with depressive symptoms at baseline reported greater incidence of non-condom use at last sex and multiple sexual partners at the six-month follow-up (N = 715). However, a subsequent report with the same participants found that baseline depressive symptoms were not predictive of any instances of non-condom use at the 12 month follow-up (Voisin et al. (2013). Finally, Jackson, Seth, DiClemente, and Lin (2015) reported that both depressive symptoms and substance use (alcohol, marijuana, or ecstasy) were independently predictive of non-condom use at last sex over the 36 month follow-up period (N = 701).

Only one study to date has included both male and female adolescents to examine the association between depressive symptoms and sexual risk behaviors (Brown et al. (2006). At a baseline assessment, males reporting high levels of depressive symptoms were also more likely to report non-condom use at last sex compared to males with few depressive symptoms. Risk did not vary as a function of depressive symptoms among female participants. In contrast, longitudinal analyses confirmed that baseline depressive symptoms were associated with non-condom use at last sex at the six-month follow-up among both male and female participants (N = 415). The authors did not examine gender as a moderator of the longitudinal association of depression to sexual risk, despite the emergence of gender differences in the cross-sectional analysis of baseline data.

In summary, empirical evidence points to significant associations among depressive symptoms, self-efficacy, and sexual risk behaviors but the literature on African American adolescents has focused almost exclusively on females. Past research has provided little consideration for gender differences in the relationship between mood and sexual risk-taking. Further, few studies include other indices of sexual health risks apart from condom use, such as the occurrence of multiple and/or concurrent sexual partners (e.g., Aral, 2010). Finally, in the literature, there has been no formal testing of potential explanatory variables (i.e., mediators) that may help to clarify how depressive symptoms lead to sexual risk via decrements in self-efficacy among African American adolescents.

This secondary analysis of data from a randomized clinical trial (RCT) sought to clarify the longitudinal association of depressive symptoms to sexual risk behaviors and determine whether the association of depressive symptoms to risky sex varied as a function of gender in a large sample of African American adolescents. We hypothesized that baseline depressive symptoms would predict sexual risk behaviors (i.e., having two or more vaginal sex partners, less frequent condom use during vaginal sex, and non-condom use at the most recent episode of vaginal sex) and a positive STI screen six-months later. This study extends the literature by examining if self-efficacy (i.e., one’s confidence in his or her ability to master a specific skill) would mediate the depressive symptoms-sexual risk relationship. We hypothesized that sex refusal self-efficacy would mediate the longitudinal association between depressive symptoms and sex with two or more partners, while condom use self-efficacy would mediate its association with condomless sex and positive STI screening.

Method

Participants

Participants were African American adolescents recruited to participate in Project iMPPACS1, a study that examined the impact of a mass media and small group intervention for HIV/STI risk prevention (Hennessy et al., 2013; Romer et al., 2009). The study was conducted in the United States in two matched northeast cities (Providence, Rhode Island, and Syracuse, New York) and two matched southeast cities (Columbia, South Carolina, and Macon, Georgia). Youth were eligible if they identified as African American and were aged 14 to 17 years at the time of recruitment (sample M = 15, SD = 1.05). We oversampled adolescent girls, who were expected to have higher rates of STIs, which yielded 945 (60%) female respondents. The majority (94%) of participants reported that they were living in their family’s home or apartment, most often with their mother (85%) and less frequently with their father (23%) or both parents (20%). Most participants (71%) qualified for a free or reduced-price lunch at school; 75% reported that they were in high school. Of the entire sample of 1,616 adolescents, 1,580 (98%) completed the three-month follow-up and 1,558 (96%) completed the six-month follow-up. In the restricted sample of 782 sexually active adolescents (i.e., adolescents that endorsed engagement in vaginal sex in the past three-months), participant ages ranged from 14 to 17 years (M = 15.3, SD = 1.08), with 420 (54%) female respondents.

Procedures

Recruitment.

Institutional Review Boards of the participating universities (i.e., Brown University, Syracuse University, University of South Carolina, Emory University, and University of Pennsylvania) approved all procedures prior to starting recruitment. Participants were recruited (1) from community-based organizations, such as Boys and Girls Clubs and community centers that provide recreational, social, and educational services for young people; (2) using street outreach; and (3) by referrals from participants or adults in the community. Prior to enrolling in the study, prospective participants and their parents were informed that the study was being conducted to test different approaches to improving health among adolescents. All participants and their parents or guardians provided written, informed assent and consent. Only 25 (1.5%) of the eligible adolescents refused or were unable to participate in the study. The recruitment procedures in the four cities produced equivalent samples in terms of sexual risk behavior, gender, and age (Romer et al., 2009).

Data collection.

Two of the four cities, Syracuse and Macon, were randomly assigned to receive the mass media campaign. The media campaign ran continuously during the 15-month recruitment period in each media city using nine 30-second television ads and twelve 60-second radio ads, and included themes to emphasize that condoms can help increase pleasure by reducing stress during sex about STIs and/or unintended pregnancy, as well as how waiting to initiate sex shows maturity and respect for one’s partner (Romer et al., 2009).There was no formal viewing of the advertisements as part of the study procedures. Prior to enrolling in the study, each youth provided assent, and his or her parents or guardians provided informed consent. Subsequently, the adolescents completed a baseline assessment through an audio computer-assisted self- interview (ACASI). A test-retest study of the ACASI was undertaken with a sample of African American adolescents in preparation for the trial, and the measures were observed to be reliable (Vanable et al., 2009). As part of the baseline, participants also provided a urine sample to assess for gonorrhea, chlamydia, and trichomoniasis. Urine collection occurred in a private room. Specimens were stored in refrigerators until packed in approved biospecimen boxes and shipped by overnight courier to the Emory University Microbiology Laboratory for assay. Youth who tested positive for any STI were treated and counseled by a medical care provider.

After completing the baseline assessment, participants in all cities were randomized to one of two small group interventions focusing either on safer sex (Stanton et al., 1995) or on general health promotion (“Promoting Health Among Teens;” Jemmott, Jemmott, Braverman, & Fong, 2005). The small group interventions included two intensive sessions. Finally, participants completed the measures (same battery as they completed at baseline) during follow-up assessments at 3-, 6-, 12-, 18- and 36-months. Urine specimen collected occurred at baseline, 6-, 12- and 18-month follow-ups. Youth received $30 for completing the initial assessment with gradual monetary increases (maximum $60) for subsequent assessments (see Hennessy et al., 2013; Romer et al., 2009). The current analyses used data from the baseline, and the three and six-month assessments.

Measures

Sexual activity with two or more partners.

At the six-month assessment, sex with two or more partners was coded based on responses to the question “With how many people have you had vaginal sex (penis in vagina) in the past three-months?” (Schroder, Carey, & Vanable, 2003). Participants were dichotomously categorized based on sexual activity with less than two partners (0) versus sexual activity with two or more partners (1) for the previous three-months.

Relative frequency of condom use.

At the six-month assessment, among sexually active participants, a single item assessed the relative frequency of condom use at the six-month assessment (Schroder et al., 2003). Using a six-point scale ranging from zero, “Never,” to five, “Every time,” participants were asked: “In the last three-months, how often would you say that you and your partner or partners used a condom from start to finish when you had vaginal sex?”

Condom use at last sex.

At the six-month assessment, sexually active participants were also asked to recall their most recent episode of sexual activity with the prompt, “Now all of the following questions ask about the LAST TIME you had vaginal sex.” Included with this assessment, participants were asked whether they engaged in vaginal sex, and if they responded affirmatively, they were asked to indicate whether they used a condom (Schroder et al., 2003). Participants were dichotomously categorized based on condom use (0) versus non-condom use (1) at the last occasion of vaginal sex.

Depressive symptoms.

At baseline, depressive symptoms were assessed using an eight-item version of the Center for Epidemiological Studies Depression Scale (CES-D; Sales, Lang, Hardin, Diclemente, & Wingood, 2010). For each item, participants indicated how often they felt the emotion in the last week using a four-point scale ranging from one, “less than 1 day,” to four, “5–7 days.” Sample items included, “I felt that I could not shake off the blues even with help from my family and friends” and “I thought my life had been a failure.” A summary score was derived by adding the scores for each of the eight items with a score ≥ 15 indicative of above-threshold depressive symptoms (Cronbach’s α = .88).

Sex refusal self-efficacy.

At the three-month assessment, participants rated their confidence level to refuse sex during the next month by responding to the following prompt, “How sure are you that you would be able to say no to having vaginal sex with someone” using a four-point scale ranging from one “definitely could not say no” to four “definitely could say no” (Zimmerman, Sprecher, Langer, & Holloway, 1995). A summary score was derived by averaging the scores on all five items (Cronbach’s α = .85).

Condom use self-efficacy.

At the three-month assessment, participants also indicated their self-efficacy to use a condom by responding to the following prompt, “How sure are you that you would use a condom when you have vaginal sex” using a four-point scale ranging from one “Not at all sure,” to four “Extremely sure” (Galavotti et al., 1995). A summary score was derived by averaging the scores on all five items (Cronbach’s α = .81).

Data Analyses

Descriptive analyses characterized the prevalence of depressive symptoms at the baseline assessment, as well as self-efficacy measures and sexual risk behaviors at the three- and six-month follow-up assessments, respectively. Subsequently, the bivariate (i.e., unadjusted) association between predictor, outcome, and potential covariate variables were examined. Potential covariates included, age, group intervention condition, and mass media intervention condition. Age at baseline was controlled for in the analyses because sexual activity increases as adolescents age (Eaton et al., 2010). Group intervention condition (HIV prevention vs. general health) and mass media intervention condition (no campaign vs. HIV prevention campaign) were included to control for intervention effects. Finally, hierarchical regression models were used to determine gender differences in the association between depressive symptoms, self-efficacy measures and sexual risk indices to inform direct and indirect effect models.

Mediation path models were conducted using the SPSS macro “PROCESS” (Hayes, 2013; Preacher & Hayes, 2008). PROCESS was designed to examine direct and indirect effect path models using bootstrapping statistical methods. The macro provided estimates of the unstandardized path coefficients in the mediation model and a bootstrap 95% confidence interval (CI) based on 10,000 resamples for the indirect/mediated effect. The 95% CI that did not include zero indicated a significant mediation effect.

Results

Descriptive Analyses

Summary statistics for depressive symptoms, sexual risk behaviors, and other key study variables are shown in Table 1. At baseline, 24% of participants reported above-threshold levels of depressive symptoms (defined as a CES-D score ≥15; M = 12.56, SD = 5.26). At the six-month assessment, 44% of participants (N = 346) reported vaginal sexual activity with two or more partners in the previous three-months. Participants indicated that they used condoms “most of the time” when engaging in vaginal sex in the previous three-months (M = 4.18, SD = 1.95 on a 1–6 scale). Nevertheless, 36% (N = 279) reported non-condom use at the last occasion of vaginal sex. Lastly, 7% (N = 58) participants screened positive for at least one STI at the six-month follow-up.

Table 1.

Participant Demographics and Descriptive Statistics for all Study Variables


Time 1		 %
Time 2
Time 3
Mean age (SD) 15.31 (1.08) -- --
Gender
  Male 46 -- --
  Female 54
Free/Reduced Lunch
  No 16 -- --
  Yes 76
  Don’t know 8
Mean CES-D score 12.56 (5.26) 12.57 (5.15) 12.35 (4.92)
Mean sex refusal self-efficacy score 3.19 (.89) 3.17 (.91) 3.06 (.83)
Mean condom use self-efficacy score 3.08 (.83) 3.06 (.85) 3.14 (.93)
Number of sexual partners
  Less than two 67 70 56
  Two or more 33 30 44
Relative frequency of condom use 4.17 (1.94) 4.09 (1.96) 4.18 (1.95)
Non-condom use, last occasion of sex
  No 71 69 64
  Yes 29 31 36
Positive STI screen
  No 89 -- 93
  Yes 11 7
Open in a new tab

Note. N = 782 sexually active participants with complete data at the six-month follow-up; Time 1 = baseline assessment; Time 2 = three-month follow-up; Time 3 = six-month follow-up; CES-D = Center for Epidemiological Studies Depression Scale; Relative frequency of condom use is assessed using a scale from 0 “never” to 5 “every time;” STI = sexually transmitted infection

Bivariate Analyses

Age was not a significant predictor of depressive symptoms (b = .10, SE = .17, p = .54), sex refusal self-efficacy (b = .02, SE = .03, p = .60), condom use self-efficacy (b = −.03, SE = .03, p = .38), sexual activity with two or more partners (b = −.07, SE = .07, p = .27), or a positive STI screen (b = −.11, SE = .13, p = .40). However, age was significantly predictive of the relative frequency of condom use for the previous three-months (b = −.29, SE = .06, p < .001), and non-condom use at last occasion of sex (b = .27, SE = .07, p < .001). Overall, older adolescents were more likely to endorse condomless sex.

Gender was a significant predictor of depressive symptoms (b = 2.36, SE = .35, p < .001), sex refusal self-efficacy (b = .78, SE = .06, p < .001) and condom use self-efficacy (b = .14, SE = .06, p = .02), such that female adolescents reported greater depressive symptoms, yet still endorsed higher confidence in their ability to refuse sex and use condoms. Gender was also a significant predictor of sex with two or more partners (b = −1.45, SE = .15, p < .001), the relative frequency of condom use (b = −.73, SE = .14, p < .001), non-condom use at the last occasion of sex (b = .78, SE = .16, p < .001) and the likelihood of a positive STI screen (b = 1.28, SE = .33, p < .001). With regards to sexual risk behavior, male adolescents were more likely to report sex with two or more partners, less consistent condom use overall for the previous three-months, non-condom use at last sex, and to test positive for a STI at the six-month follow-up.

In unadjusted regression models, depressive symptoms were not a significant predictor of sex refusal self-efficacy (b = .008, SE = .006, p = .19), sexual activity with two or more partners (b = .02, SE = .01, p = .20), or non-condom use at the last occasion of sex (b = .02, SE = .02, p = .25). Conversely, depressive symptoms predicted less condom use self-efficacy (b = −.02, SE = .006, p = .001) and less frequent condom use (b = −.04, SE = .01, p = .01). Depressive symptoms were also marginally significant for predicting a positive STI screen (b = .05, SE = .02, p = .06). Sex refusal self-efficacy was a significant predictor of sex with two or more partners (b = −.56, SE = .09, p < .001), such that those high in sex refusal self-efficacy were less likely to have two or more partners. Similarly, condom use self-efficacy was a significant predictor of more frequent condom use (b = .48, SE = .08, p < .001) and non-condom use at last sex (b = −.34, SE = .09, p < .001). Neither condom use (b = .15, SE = .17, p = .39) or sex refusal (b = .22, SE = .17, p = .19) self-efficacy were significant predictors of a positive STI screen.

Moderation Analyses

In hierarchical regression models, the interaction between depressive symptoms and participant gender was a significant predictor of sex refusal self-efficacy (b = −.03, SE = .01, p = .01). When stratified by gender, depressive symptoms predicted less sex refusal self-efficacy for females (b = −.02, SE = .006, p = .001), but not males (b = .01, SE = .01, p = .31). The interaction between depressive symptoms and gender was non-significant for sex with two or more partners (b = .05, SE = .03, p = .16). Depressive symptoms predicted a higher likelihood of sex with two or more partners for female (b = .02, SE = .004, p < .001), but not male adolescents (b = .007, SE .006, p = .26). Therefore, direct and indirect effect models were restricted to female adolescents to examine the associations between depressive symptoms, sex refusal self-efficacy, and sex with two or more partners.

The interaction between baseline depressive symptoms and participant gender was nonsignificant for condom use self-efficacy (b = .005, SE = .01, p = .69). The depressive symptoms by gender interaction was also not a significant predictor of the relative frequency of condom use (b = −.02, SE = .03, p = .59), non-condom use at last sex (b = .03, SE = .04, p = .47) or a positive STI screen (b = −.06, SE = .07, p = .37). Therefore, direct and indirect effect models included both male and female adolescents to examine the associations between depressive symptoms, condom use self-efficacy with condom use behaviors and STI screening data.

Direct Effect Models

All PROCESS models controlled for age, group intervention condition, mass media intervention condition, and the baseline assessment of sexual risk variables. Among female participants, depressive symptoms were a significant predictor of sexual activity with two or more partners (b = .06, SE = .02, p < .001; see Figure 1A). However, for the entire sample, depressive symptoms did not significantly predict the relative frequency of condom use for the last three-months (b = .02, SE = .02, p = .31; see Figure 1B), non-condom use at the last occasion of sex (b = −.02, SE = .02, p = .17; see Figure 1C) or a positive STI screen (b = .03, SE = .03, p = .28).

Figure 1.

Figure 1.

Open in a new tab

Path model for the association of depressive symptoms to sexual risk variables with the indirect association of self-efficacy modeled. Covariates were not included in the figure for simplicity.

Unstandardized path coefficients are shown; Time 1 = baseline assessment; Time 2 = three-month follow-up; Time 3 = six-month follow-up. Results for the likelihood of screening positive for an STI are not shown; all associations were nonsignificant.

*p < .05; ** p < .01; *** p <.001.

◊ Sample was restricted to female participants

Indirect Effect Models

Sex Refusal Self-Efficacy.

When the analysis was restricted to female participants, depressive symptoms were a significant predictor of sex refusal self-efficacy (b = −.02, SE = .006, p < .001). The longitudinal association between sex refusal self-efficacy and sex with two or more partners was also significant (b = −.30, SE = .15, p = .05). However, based on 10,000 bootstrap resamples, a test of the indirect effect of depressive symptoms on sex with two or more partners via sex refusal self-efficacy was non-significant (b = .007; bootstrap 95% CI: −.0002, .02; see Figure 1A). The longitudinal association between sex refusal self-efficacy and a positive STI screening was nonsignificant (b = .12, SE = .17, p = .48). Based on 10,000 bootstrap resamples, a test of the indirect effect of depressive symptoms on the likelihood of a positive STI screen was also nonsignificant (b = −.003; bootstrap 95% CI: −.01, .005).

Condom Use Self-Efficacy.

For the entire sample, depressive symptoms were a significant predictor of less condom use self-efficacy (b = −.02, SE = .007, p = .02). The positive longitudinal association between condom use self-efficacy and the relative frequency of condom use was also significant (b = .38, SE = .09, p < .001). A test of the indirect association between depressive symptoms and condom use via condom use self-efficacy was significant (b = −.006; bootstrap 95% CI: −.01, −.0005; see Figure 1B). Once more, the negative longitudinal association between condom use self-efficacy and non-condom use at the last occasion of sex was significant (b = −.33, SE = .10, p < .001). A test of the indirect effect of depressive symptoms on non-condom use at the last occasion of sex via condom use self-efficacy was also significant (b = .007; bootstrap 95% CI: .002, .01; see Figure 1C). However, the longitudinal association between condom use self-efficacy and a positive STI screening was nonsignificant (b = .12, SE = .17, p = .48). A test of the indirect effect of depressive symptoms on the likelihood of a positive STI screen was also nonsignificant (b = −.003; bootstrap 95% CI: −.01, .005).

Discussion

The current study is the first to examine the role of self-efficacy as a mediator of the longitudinal association between depressive symptoms and sexual risk behaviors in a sample of African American adolescents. Depressive symptoms were highly prevalent in the sample of adolescents who reported recent (i.e., previous three-months) vaginal sex. Baseline depressive symptoms predicted vaginal sex with two or more partners assessed at the six-month assessment for female participants only, suggesting that mood is differentially associated with sexual activity for male and females adolescents. Depressive symptoms did not directly predict the relative frequency of condom use, non-condom use at the last occasion of vaginal sex or likelihood of a positive STI screen for the sample as a whole when reported six-months later. However, baseline depressive symptoms predicted lower self-efficacy for condom use at the three-month follow-up which predicted inconsistent condom use and condom nonuse at the six-month follow-up for the entire sample. Therefore, this study provides the first empirical evidence that condom use self-efficacy may mediate the longitudinal association between depressive symptoms and condomless sex in a sample of African American male and female adolescents.

Although depressive symptoms predicted sexual activity with two or more sexual partners for females, the indirect effect of depressive symptoms via sex-refusal self-efficacy was not statistically significant. The literature suggests other hypotheses for why female adolescents with depressive symptoms engage in sexual activity. For example, poor self-esteem may explain the association of depressive symptom to engagement in sex (Lopez et al., 2011). Indeed, when African American adolescents were interviewed about their sexual behaviors, participants reported that they were “looking for love,” trying to “get attention” and seeking someone to “comfort them” when having sex (Brawner, Davis, Fannin, & Alexander, 2012). Therefore, female adolescents may compensate for their negative self-evaluations with external validation, including sexual contact, with less concern for the potential adverse consequences to their health.

More broadly, our findings also help to clarify equivocal findings for the association of depressive symptoms to sexual risk behavior among African American adolescents. Specifically, gender may moderate the influence of depressed mood as a risk factor for some, but not all behavioral indices of risk among adolescents. Therefore, a lack of consideration for gender differences in the depressive symptoms—sexual risk relationship may account for previous nonsignificant findings. Further, the lack of a direct association between depressive symptoms and condom use may suggest that the longer time duration between the assessment of depressive symptoms and sexual behavior could also be contributing to non-significant direct associations (Voisin et al., 2013); that is, one hypothesis to explain the discrepant findings is that depressive symptoms fluctuate over time, making them a poor predictor of long-term sexual behavior. It is also important to note that the sexual risk behaviors that made up the primary focus of this paper only increase one’s susceptibility to negative health outcomes. Indeed, the associations between depressive symptoms and self-efficacy with predicting a positive STI screen were nonsignificant, and likely the result of low power with only 7% of the sample testing positive for a STI at the six-month follow-up.

Study Limitations and Future Directions

The current study findings should be considered in the context of its limitations. First, the CES-D scale is a screening tool, and not appropriate to diagnose clinical depression or other mood disorders. Nonetheless, cross-sectional studies that have sampled clinically depressed African American adolescents confirm that depression is an important correlate of inconsistent condom use (Brawner et al., 2012). The sole longitudinal study with a clinical sample also found that clinical depression predicted condomless sex among African American female adolescents (Starr, Donenberg, & Emerson, 2012). Nonetheless, there is need for further research to establish depressive symptoms as a risk factor for sexual risk-taking among clinically depressed male and female adolescents.

Second, the sample included only African American adolescents; thus, the results may not generalize beyond this population. Existing research supports the longitudinal association of depression to inconsistent condom use among ethnically diverse samples of adolescents (e.g., Lehrer, Shrier, Gortmaker, & Buka, 2006), but further research is needed to replicate the mediational effects of condom use self-efficacy. Among studies that have reported on a negative association between depressive symptoms and condom use among adolescents, only one study examined potential mediators, with alcohol consumption and marijuana use emerging as explanatory variables (Shrier, Harris, Sternberg, & Beardslee, 2001).

Third, we used self-report to assess sexual risk behaviors. Self-report is imperfect and may be influenced by recall biases (Schroder et al., 2003). Thus, a next step will be to confirm the association of depressive symptoms to sexual risk behaviors using ecological momentary assessment (EMA) that reduces self-report biases by allowing the participants to report on their behavior in real time. Indeed, research has shown that EMA is well suited for studying episodic and transient experiences, such as mood and substance use (Shiffman, 2009). Consequently, future research should focus on intensive longitudinal designs to establish depressive symptoms as a risk factor for HIV/STI transmission among adolescents.

Conclusions

We found a direct longitudinal association between depressive symptoms and sexual activity with two or more partners for female adolescents, and an indirect association between depressive symptoms and condom use via decrements in condom use self-efficacy for both males and females. These findings suggest that sexual risk reduction interventions for adolescents may be improved by including modules that address the influence of mood on sexual behaviors. Specifically, alleviating depressive symptoms may facilitate safer sex practices, in part, by increasing self-efficacy, to reduce the likelihood of HIV/STI transmission. Indeed, among African American adults participating in a safer sex behavioral intervention, decreases in sexual risk-taking were optimized when the intervention lead to reductions in depressed mood (Lennon, Huedo-Medina, Gerwien, & Johnson, 2012). Nonetheless, hypotheses have yet to be tested among adolescent samples for whom depression is not uncommon (Costello, Erkanli, & Angold, 2006; Costello, Mustillo, Erkanli, Keeler, & Angold, 2003). Therefore, there is need for a greater emphasis on depressive symptoms to promote sexual health and reduce HIV/STI incidence among African American adolescents.

Acknowledgment:

This study was conducted through the iMPPACS network supported by the National Institutes of Mental Health (3U01MH066809-05S1; Pim Brouwers, Project Officer) at the following sites and local contributors: Columbia, SC (MH66802; Robert Valois [PI], Naomi Farber, Andre Walker); Macon, GA (MH66807; Ralph DiClemente [PI], Gina Wingood, Laura Salazar, Rachel Joseph, Delia Lang); Philadelphia, PA (MH66809; Daniel Romer [PI], Sharon Sznitman, Bonita Stanton, Michael Hennessy, Susan Lee, Eian More, Ivan Juzang, and Thierry Fortune); Providence, RI (MH66785; Larry Brown [PI], Christie Rizzo, Nanetta Payne); and Syracuse, NY (MH66794; Peter Vanable [PI], Michael Carey, Rebecca Bostwick).

Footnotes

Conflicts of Interest: There are no conflicts of interest to report.

Trial Registration: ClinicalTrials.gov

1

An acronym developed for “in Macon, Philadelphia, Providence, Atlanta, Columbia and Syracuse”

References

  1. Aral SO (2010). Partner Concurrency and the STD/HIV Epidemic. Curr Infect Dis Rep, 12(2), 134–139. doi: 10.1007/s11908-010-0087-2 [DOI] [PubMed] [Google Scholar]
  2. Bandura A (1986). Social foundations of thought and action Englewood Cliffs, NJ, 1986. [Google Scholar]
  3. Brawner BM, Davis ZM, Fannin EF, & Alexander KA (2012). Clinical depression and condom use attitudes and beliefs among African American adolescent females. J Assoc Nurses AIDS Care, 23(3), 184–194. doi: 10.1016/j.jana.2011.03.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brawner BM, Jemmott LS, Wingood GM, Reason J, Daly B, Brooks K, & Lanier Y (2017). Feelings Matter: Depression Severity and Emotion Regulation in HIV/STI Risk-Related Sexual Behaviors. Journal of Child and Family Studies, 26(6), 1635–1645. doi: 10.1007/s10826-017-0674-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Brown LK, Tolou-Shams M, Lescano C, Houck C, Zeidman J, Pugatch D, & Lourie KJ (2006). Depressive symptoms as a predictor of sexual risk among African American adolescents and young adults. J Adolesc Health, 39(3), 444.e441–448. doi: 10.1016/j.jadohealth.2006.01.015 [DOI] [PubMed] [Google Scholar]
  6. Centers for Disease Control and Prevention. (2016). HIV Infection Risk, Prevention, and Testing Behaviors among Men Who Have Sex With Men—National HIV Behavioral Surveillance, 20 U.S. Cities, 2014 Retrieved from http://www.cdc.gov/hiv/library/reports/surveillance/#panel2.
  7. Centers for Disease Control and Prevention. (2017). Sexually Transmitted Disease Surveillance 2016 Retrieved from Atlanta: U.S. Department of Health and Human Services: [Google Scholar]
  8. Closson K, Dietrich JJ, Lachowsky NJ, Nkala B, Palmer A, Cui Z, … Kaida A. (2018). Gender, Sexual Self-Efficacy and Consistent Condom Use Among Adolescents Living in the HIV Hyper-Endemic Setting of Soweto, South Africa. AIDS Behav, 22(2), 671–680. doi: 10.1007/s10461-017-1950-z [DOI] [PubMed] [Google Scholar]
  9. Costello EJ, Erkanli A, & Angold A (2006). Is there an epidemic of child or adolescent depression? Journal of Child Psychology and Psychiatry, 47(12), 1263–1271. [DOI] [PubMed] [Google Scholar]
  10. Costello EJ, Mustillo S, Erkanli A, Keeler G, & Angold A (2003). Prevalence and development of psychiatric disorders in childhood and adolescence. Arch Gen Psychiatry, 60(8), 837–844. doi: 10.1001/archpsyc.60.8.837 [DOI] [PubMed] [Google Scholar]
  11. Coyle RM, Lampe FC, Miltz AR, Sewell J, Anderson JF, Apea V, … Group AS. (2019). Associations of depression and anxiety symptoms with sexual behaviour in women and heterosexual men attending sexual health clinics: a cross-sectional study. Sex Transm Infect, sextrans-2018-053689 doi: 10.1136/sextrans-2018-053689 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Crosby RA (2016). Dealing with pre-exposure prophylaxis-associated condom migration: changing the paradigm for men who have sex with men. Sex Health doi: 10.1071/sh16128 [DOI] [PubMed] [Google Scholar]
  13. DiClemente RJ, Wingood GM, Crosby RA, Cobb BK, Harrington K, & Davies SL (2001). Parent-adolescent communication and sexual risk behaviors among African American adolescent females. The Journal of Pediatrics, 139(3), 407–412. [DOI] [PubMed] [Google Scholar]
  14. Eaton DK, Kann L, Kinchen S, Shanklin S, Ross J, Hawkins J, … Prevention. (2010). Youth risk behavior surveillance - United States, 2009. MMWR Surveill Summ, 59(5), 1–142. [PubMed] [Google Scholar]
  15. Elkington KS, Bauermeister JA, & Zimmerman MA (2010). Psychological distress, substance use, and HIV/STI risk behaviors among youth. J Youth Adolesc, 39(5), 514–527. doi: 10.1007/s10964-010-9524-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. English D, Lambert SF, & Ialongo NS (2014). Longitudinal Associations Between Experienced Racial Discrimination and Depressive Symptoms in African American Adolescents. Developmental Psychology, 50(4), 1190–1196. doi: 10.1037/a0034703 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Galavotti C, Cabral RJ, Lansky A, Grimley DM, Riley GE, & Prochaska JO (1995). Validation of measures of condom and other contraceptive use among women at high risk for HIV infection and unintended pregnancy. Health Psychol, 14(6), 570–578. [DOI] [PubMed] [Google Scholar]
  18. Hayes AF (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach New York, NY, US: Guilford Press. [Google Scholar]
  19. Hennessy M, Romer D, Valois RF, Vanable P, Carey MP, Stanton B, … Salazar LF. (2013). Safer sex media messages and adolescent sexual behavior: 3-year follow-up results from project iMPPACS. Am J Public Health, 103(1), 134–140. doi: 10.2105/AJPH.2012.300856 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Jackson JM, Seth P, DiClemente RJ, & Lin A (2015). Association of Depressive Symptoms and Substance Use With Risky Sexual Behavior and Sexually Transmitted Infections Among African American Female Adolescents Seeking Sexual Health Care. Am J Public Health, 105(10), 2137–2142. doi: 10.2105/AJPH.2014.302493 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Jemmott JB 3rd, Jemmott LS, Braverman PK, & Fong GT (2005). HIV/STD risk reduction interventions for African American and Latino adolescent girls at an adolescent medicine clinic: a randomized controlled trial. Arch Pediatr Adolesc Med, 159(5), 440–449. doi: 10.1001/archpedi.159.5.440 [DOI] [PubMed] [Google Scholar]
  22. Kann L, McManus T, Harris WA, Shanklin SL, Flint KH, Queen B, … Ethier KA. (2018). Youth Risk Behavior Surveillance — United States, 2017. MMWR Surveillance Summaries, 67(8), 1–114. doi: 10.15585/mmwr.ss6708a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lehrer JA, Shrier LA, Gortmaker S, & Buka S (2006). Depressive symptoms as a longitudinal predictor of sexual risk behaviors among US middle and high school students. Pediatrics, 118(1), 189–200. doi: 10.1542/peds.2005-1320 [DOI] [PubMed] [Google Scholar]
  24. Lennon CA, Huedo-Medina TB, Gerwien DP, & Johnson BT (2012). A role for depression in sexual risk reduction for women? A meta-analysis of HIV prevention trials with depression outcomes. Soc Sci Med, 75(4), 688–698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Levy RW, Rayner CR, Fairley CK, Kong DC, Mijch A, Costello K, … Melbourne Adherence G. (2004). Multidisciplinary HIV adherence intervention: a randomized study. AIDS Patient Care STDS, 18(12), 728–735. doi: 10.1089/apc.2004.18.728 [DOI] [PubMed] [Google Scholar]
  26. Lopez V, Kopak A, Robillard A, Gillmore MR, Holliday RC, & Braithwaite RL (2011). Pathways to sexual risk taking among female adolescent detainees. J Youth Adolesc, 40(8), 945–957. doi: 10.1007/s10964-010-9623-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Millar BM, Starks TJ, Grov C, & Parsons JT (2017). Sexual Risk-Taking in HIV-Negative Gay and Bisexual Men Increases with Depression: Results from a U.S. National Study. AIDS Behav, 21(6), 1665–1675. doi: 10.1007/s10461-016-1507-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Miltz AR, Rodger AJ, Sewell J, Speakman A, Phillips AN, Sherr L, … Group AS. (2017). Clinically significant depressive symptoms and sexual behaviour among men who have sex with men. BJPsych Open, 3(3), 127–137. doi: 10.1192/bjpo.bp.116.003574 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Nehl EJ, Klein H, Sterk CE, & Elifson KW (2016). Prediction of HIV Sexual Risk Behaviors Among Disadvantaged African American Adults Using a Syndemic Conceptual Framework. AIDS Behav, 20(2), 449–460. doi: 10.1007/s10461-015-1134-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Olfson M, Druss BG, & Marcus SC (2015). Trends in mental health care among children and adolescents. New England Journal of Medicine, 372(21), 2029–2038. [DOI] [PubMed] [Google Scholar]
  31. Parrish DD, & Kent CK (2008). Access to care issues for African American communities: implications for STD disparities. Sex Transm Dis, 35(12 Suppl), S19–22. doi: 10.1097/OLQ.0b013e31818f2ae1 [DOI] [PubMed] [Google Scholar]
  32. Preacher KJ, & Hayes AF (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods, 40(3), 879–891. [DOI] [PubMed] [Google Scholar]
  33. Protogerou C, & Johnson BT (2014). Factors underlying the success of behavioral HIV-prevention interventions for adolescents: a meta-review. AIDS Behav, 18(10), 1847–1863. doi: 10.1007/s10461-014-0807-y [DOI] [PubMed] [Google Scholar]
  34. Romer D, Sznitman S, DiClemente R, Salazar LF, Vanable PA, Carey MP, … Juzang I. (2009). Mass media as an HIV-prevention strategy: using culturally sensitive messages to reduce HIV-associated sexual behavior of at-risk African American youth. Am J Public Health, 99(12), 2150–2159. doi: 10.2105/ajph.2008.155036 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Sales JM, Lang DL, Hardin JW, Diclemente RJ, & Wingood GM (2010). Efficacy of an HIV prevention program among African American female adolescents reporting high depressive symptomatology. J Womens Health (Larchmt), 19(2), 219–227. doi: 10.1089/jwh.2008.1326 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Schroder KE, Carey MP, & Vanable PA (2003). Methodological challenges in research on sexual risk behavior: I. Item content, scaling, and data analytical options. Ann Behav Med, 26(2), 76–103. doi: 10.1207/s15324796abm2602_02 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Seaton EK, & Douglass S (2014). School diversity and racial discrimination among African-American adolescents. Cultur Divers Ethnic Minor Psychol, 20(2), 156–165. doi: 10.1037/a0035322 [DOI] [PubMed] [Google Scholar]
  38. Seth P, Patel SN, Sales JM, DiClemente RJ, Wingood GM, & Rose ES (2011). The impact of depressive symptomatology on risky sexual behavior and sexual communication among African American female adolescents. Psychol Health Med, 16(3), 346–356. doi: 10.1080/13548506.2011.554562 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Seth P, Raiji PT, DiClemente RJ, Wingood GM, & Rose E (2009). Psychological distress as a correlate of a biologically confirmed STI, risky sexual practices, self-efficacy and communication with male sex partners in African-American female adolescents. Psychol Health Med, 14(3), 291–300. doi: 10.1080/13548500902730119 [DOI] [PubMed] [Google Scholar]
  40. Shiffman S (2009). Ecological momentary assessment (EMA) in studies of substance use. Psychol Assess, 21(4), 486–497. doi: 10.1037/a0017074 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Shrier LA, Harris SK, Sternberg M, & Beardslee WR (2001). Associations of depression, self-esteem, and substance use with sexual risk among adolescents. Prev Med, 33(3), 179–189. doi: 10.1006/pmed.2001.0869 [DOI] [PubMed] [Google Scholar]
  42. Stanton B, Black M, Feigelman S, Ricardo I, Galbraith J, Li X, … Nesbitt R. (1995). Development of a culturally, theoretically and developmentally based survey instrument for assessing risk behaviors among African-American early adolescents living in urban low-income neighborhoods. AIDS Educ Prev, 7(2), 160–177. [PubMed] [Google Scholar]
  43. Starr LR, Donenberg GR, & Emerson E (2012). Bidirectional linkages between psychological symptoms and sexual activities among African American adolescent girls in psychiatric care. J Clin Child Adolesc Psychol, 41(6), 811–821. doi: 10.1080/15374416.2012.694607 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Vanable PA, Carey MP, Brown JL, DiClemente RJ, Salazar LF, Brown LK, … Stanton BF. (2009). Test-retest reliability of self-reported HIV/STD-related measures among African-American adolescents in four U.S. cities. J Adolesc Health, 44(3), 214–221. doi: 10.1016/j.jadohealth.2008.09.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Voisin DR, Hotton A, Tan K, & Diclemente R (2013). A Longitudinal Examination of Risk and Protective Factors Associated with Drug Use and Unsafe Sex among Young African American Females. Child Youth Serv Rev, 35(9), 1440–1446. doi: 10.1016/j.childyouth.2013.05.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Wade R, Harper GW, & Bauermeister JA (2018). Psychosocial Functioning and Decisional Balance to Use Condoms in a Racially/Ethnically Diverse Sample of Young Gay/Bisexual Men Who Have Sex with Men. Arch Sex Behav, 47(1), 195–204. doi: 10.1007/s10508-016-0912-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Zimmerman RS, Sprecher S, Langer LM, & Holloway CD (1995). Adolescents’ perceived ability to say “no” to unwanted sex. J Adolesc Res, 10(3), 383–399. [DOI] [PubMed] [Google Scholar]

RESOURCES