Each year, there are approximately 20 million new cases of sexually transmitted infections (STIs) in the United States.1 Sexually transmitted infections are often asymptomatic or present with mild symptoms, making them difficult to diagnose and treat.2 The price of these infections is steep; sequelae include pelvic inflammatory disease, infertility, and cervical cancer.2 Health care costs related to STIs and their complications total nearly 16 billion per year.2 Rates of STI peak in young adulthood, with about one third of reported cases of gonorrhea and chlamydia occurring among those aged 20 to 24 years, and over half of infections occurring before the age of 25 years.1
A strong and consistent relationship has been found between childhood sexual abuse and sexual risk behavior including unprotected intercourse, early onset sexual activity, sex trade involvement, sex while using drugs or alcohol, multiple partnerships, and STI.3–7 Although some studies have found similar associations for childhood physical abuse and neglect, this research is far less conclusive.8–11 These studies are largely cross-sectional and fail to monitor trends over time. Few are nationally representative, but rather examine homogenous groupings by race, class, geography, or risk profile.12–16 Furthermore, men and women are often studied separately, and studies that include both do not necessarily consider gender as a moderator.17
A small body of research has linked exposure to direct (experienced or threatened) and indirect (witnessed) violence and sexual risk outcomes. Many of these studies use a single violence variable that does not distinguish between effects of direct and indirect exposures.18–20 Those that do study these effects separately often use geographically specific or high-risk population samples that are not generalizable.21–24
Two important studies of adverse childhood experiences (ACEs) examined 7 ACEs, 4 of which were indicators of household dysfunction, including having a battered mother and household members who abused substances, had mental illness, or were incarcerated.25,26 Hillis et al25 found that increasing number of ACEs had a graded relationship with self-reported STI among men and women, and with early first sexual intercourse, self-perceived acquired immune deficiency syndrome risk and number of sexual partners in a female-only study.26 Each of the 7 ACEs tended to be associated with the outcomes; however, models were not adjusted for other ACEs to estimate the independent associations.
The purpose of this study is to examine relationships between multiple forms of childhood maltreatment and STIs using data from The National Longitudinal Study of Adolescent to Adult Health (Add Health) including Wave I (1994–1995; adolescence, grades 7–12), Wave III (2001–2002; young adulthood, ages 18–26 years), and Wave IV (2008; adulthood, ages 24–32 years). We examined the associations of childhood neglect, abuse, household dysfunction and violence with multiple partnerships, sex trade involvement, and STI. We also explored gender as a modifier of these relationships. Because we know traumas often co-occur, we explored the total number of maltreatments as a predictor of STI-related outcomes.
MATERIALS AND METHODS
Study Population and Design
The institutional review boards at the University of Florida and NYU School of Medicine approved this study. Add Health is a nationally representative longitudinal study created to better understand factors underlying health outcomes. Participants were enrolled during adolescence and followed up into adulthood by means of 3 in-home interviews, the details of which are described on the Add Health website.27 During Wave I (1994–1995), 20,745 participants in grades 7–12 were asked about a range of topics including their sexual risk behavior. Of the original cohort, 15,197 were reinterviewed at Wave III (2001–2002, ages 18–26) and 15,701 were reinterviewed at Wave IV (2007–2008, ages 24–32). The Add Health study boasts excellent retention, in which 77.4% and 80.3% of eligible participants of the original cohort were reinterviewed at Waves III and IV, respectively.28 A total of 12,288 participants provided data for all 3 waves and had sample weights. Wave II was not included in this study due to its proximity to Wave I.
We used survey procedures in SAS 9.4 (SAS Institute Inc., Cary, NC) to account for the complex sample survey design. We estimated weighted prevalence of the individual traumas and trauma scores by STI outcomes and the sociodemographic covariates by self-reported STI for descriptive purposes. With logistic regression, we estimated unadjusted odds ratio (OR) and adjusted OR (AOR) and 95% confidence intervals (CI) for associations of covariates and outcomes for each of the 9 traumas. We judged the predictive importance of each trauma variable by the magnitude of the OR and AOR and the width of the CI. We used the same method to estimate ORs, AORs, and CIs for number of traumas experienced at each outcome and calculated P values for the linear trend test using orthogonal polynomial contrasts. To assess modification by gender, we included interaction terms for gender and trauma variables in multivariable models; only when the P value for an interaction term coefficient was less than 0.15 did we present the gender-specific estimates in tables.
Independent Childhood Traumas Variables
We created 9 dichotomous measures of childhood trauma defined as: neglect (left alone when adult should have been present and/or basic needs unmet ≥ 6 times), emotional abuse (caregiver said hurtful things or made child feel unloved ≥ 6 times)< physical abuse (slapped, hit, kicked, or thrown by caregiver ≥ 6 times), sexual abuse (caregiver touched child or forced the child to touch him/her in sexual way), parental incarceration (parent/parent figure spent time in jail or prison), parental binge drinking (≥5 drinks on 1 occasion in past month), witnessed violence (saw someone shot or stabbed), threatened with violence (knife or gun pulled on child), and experienced violence (child shot or cut/stabbed).
We adjusted models for all other traumas to get the independent effect of each. We also created a predictor representing cumulative trauma score that ranged from 1 to 4+ traumas.
Independent Sociodemographic Correlate Variables
We included the following sociodemographic variables in adjusted models: age in years at Wave IV (24–27, 28–29, ≥30 years (referent); gender (female (referent), female), and race/ethnicity (non-Hispanic white (referent), non-Hispanic black, Hispanic, other) at Wave I; concern about paying bills, a measure of functional poverty, at Wave I (parent-reported) and Wave III (referent = no); and education at Wave IV (less than high school [referent], completed high school, greater than high school).
Dependent Outcome Variables
Multiple Sexual Partnerships
For Wave I, multiple partnerships was defined as having 2 or more lifetime partners. For Waves III and IV, these were defined as 2 or more partners in the past year.
Participants were considered to have been involved in sex trade if they answered yes to either paying someone to have sex or being paid to have sex. The survey measured lifetime occurrence at Waves I and III and past year occurrence at Wave IV.
During Wave I, participants were asked if they had ever been told by a health care provider that they had chlamydia, gonorrhea or trichomoniasis. A yes response to at least 1 infection was considered a positive self-reported STI. At Wave IV, self-reported STI was defined as having been told by a health care provider in the past year they had any of the 3 STIs.
At Wave III, a positive urine test for chlamydia, gonorrhea, or trichomoniasis was considered a positive test-identified STI.
Study Population Characteristics
In total, 12,288 participants were interviewed at all 3 waves. This sample was evenly distributed by gender. Most (66%) were white, followed by African Americans (16%) and Hispanics (12%). Nearly three quarters were educated beyond high school, whereas 8% did not complete high school. Self-reported functional poverty was around 15%. Emotional abuse was the most commonly reported trauma (16.1%), followed by neglect (12.4%), threatened violence (12.1%), physical abuse (11.6%), parental binge drinking (11.5%), witnessed violence (10.9%), and parental incarceration (10.2%). Just under 8% of participants reported sexual abuse, whereas only 5% had been shot or stabbed.
We chose to look at characteristics of participants reporting an STI at Wave IV because it most closely reflects the population at greatest risk at their current age (Table 1). Younger participants (aged 24–27 years) were the most likely to report an STI within the last year (4.3%), whereas their older counterparts (>30) were the least likely (2.7%). Women had more than 2 times the odds of self-reported STI compared with men, as did those with less than a high school education compared with high school graduates. Race was also a strong correlate of STI, with blacks having an OR of 4.51 compared with whites (95% CI, 3.35–6.05). Concern about bills during Waves I and III had only modest associations with self-reported STI at Wave IV.
Associations Between Traumas and Multiple Partnerships
At Wave I, 37% of participants reported ever having sexual intercourse, whereas 19.5% of participants reported having 2 or more partners in their lifetime (Table 2). In unadjusted models, all traumas were significantly associated with this outcome. In fully adjusted models, the strongest associations were with witnessed violence (AOR, 2.20; 95% CI, 1.79–2.71) and threatened violence (AOR, 2.19; 95% CI, 1.72–2.79). Neglect, sexual abuse, parental incarceration, and parental binge drinking also remained significant predictors, though with only modest associations. Emotional abuse, although not significantly associated in the total population, was weakly predictive for women (AOR, 1.41; 95% CI, 1.05–1.90) but the association was null for men. Alternatively, men who witnessed violence had an almost 3-fold increase in odds (AOR, 2.97; 95% CI, 2.19–4.02), whereas the association was null for females. The odds of multiple partnerships demonstrated a dose-response relationship to the number of traumas experienced (p < 0.0001), with the greatest odds among those who experienced 4+ traumas (AOR, 4.82; 95% CI, 3.53–6.56).
In young adulthood, 28.4% reported multiple partnerships in the past year, the highest prevalence of multiple partnerships of any wave (Tables 3 and 4). By this wave, 86% of participants reported sexual intercourse in their lifetime. Physical abuse, sexual abuse, and witnessed violence each showed about a 30% increase in odds in our adjusted analyses. Among those who had witnessed violence, only the association for males was significant (AOR, 1.67; 95% CI, 1.25–2.24). In addition, among men, but not women, who were threatened with violence, there was a significant association with multiple partnerships (AOR, 1.30; 95% CI, 1.00–1.68). Multiple partnerships in young adulthood is associated with an increase in number of traumas experienced, though these associations are weaker than in adolescence, and the trend is not statistically significant.
By Wave IV, nearly all participants had sexual intercourse (93%). The prevalence of multiple partnerships in the past year was 25%. In the unadjusted models, all traumas besides neglect were significantly associated with multiple partnerships. After adjustment, emotional abuse and sexual abuse remained significantly associated, though for sexual abuse, only the association for men was significant (AOR, 1.75; 95% CI, 1.18–2.60). Results based on number of traumas experienced were very similar to those for young adulthood, with odds ranging from 1.32 for 1 trauma to 1.60 for 4+ traumas.
Associations Between Traumas and Sex Trade
Lifetime sex trade during adolescence was uncommon (1.2%), but had notable associations with childhood traumas. In unadjusted analyses, being threatened with violence and witnessing violence were associated with about two and a half times the odds of sex trade and experiencing violence by odds of about three and a half. No other trauma had a significant association. In the adjusted analyses, only experiencing violence was significant in the total population (AOR, 2.21; 95% CI, 1.02–4.76), although strong associations were seen among women who reported neglect (AOR, 3.46; 95% CI, 1.18–10.13) and women threatened with violence (AOR, 5.78; 95% CI, 1.67–20.03). Sex trade had strong associations with number of traumas reported, with AORs as high as 7.28 for those experiencing 4+ traumas.
During young adulthood, the prevalence of lifetime sex trade rose to 4.6%. In unadjusted models, all traumas were significantly associated with sex trade, with the exception of emotional abuse. In fully adjusted analyses, we see an over 2-fold increase in odds of sex trade for those reporting sexual abuse (AOR, 2.17; 95% CI, 1.43–3.29) and witnessed violence (AOR, 2.28; 95% CI, 1.54–3.38), in addition to women reporting neglect (AOR, 2.03; 95% CI, 1.04,3.97) and women reporting physical abuse (AOR, 2.02; 95% CI, 1.14–3.57) (these associations were null for men). Also of note is the prevalence of sex trade during this period. Specifically, 10% of those who were sexually abused reported sex trade involvement in young adulthood. The prevalence was even higher for those who had been exposed to violence, with rates of 12.1%, 11.3% and 10.6% for witnessed, threatened, and experienced violence, respectively. Number of traumas is also associated with sex trade, with odds increasing in a stepwise pattern with each additional trauma (p < 0.0001).
In adulthood, 1.9% reported sex trade involvement in the past year. Just as in young adulthood, significant associations in unadjusted models were observed for most traumas, with the exception of emotional abuse and parental binge drinking. In the fully adjusted models, only witnessed violence remained significant (AOR, 2.13; 95% CI, 1.27–3.58). We also found a significant association with men reporting physical abuse (AOR, 1.97; 95% CI, 1.05–3.69) and parental incarceration (AOR, 2.10; 95% CI, 1.15–3.83). Association with number of traumas is similar to that in young adulthood (p = 0.0001).
Associations Between Traumas and Self-reported STI
In adolescence, the prevalence of self-reported STIs (chlamydia, gonorrhea, or trichomoniasis) was 1.7%. The traumas most strongly associated with STI in adjusted models were parental binge drinking (AOR, 2.22; 95% CI, 1.31–3.77), witnessed violence (AOR, 2.41; 95% CI, 1.28–4.53), and threatened violence (AOR, 1.96; 95% CI, 1.12–3.43). Women who reported parental incarceration had a 1.64 AOR (95% CI, 0.90–3.00), significantly higher than that for men. Those exposed to multiple traumas had an increase in odds that ranged from 1.83 (2 traumas) to 3.11 (4+ traumas).
In adulthood, 3.5% reported an STI in the past year. Unadjusted models showed significant associations for sexual abuse, parental incarceration, and witnessed violence. After adjustment, only parental incarceration was significantly associated (AOR, 1.70; 95% CI, 1.11–2.58). No significant gender differences were noted. Increased odds of STI were associated with having experienced 1 trauma (AOR, 1.92; 95% CI, 1.27–2.89), 2 traumas (AOR, 1.61; 95% CI, 1.00–2.59), and 4+ traumas (AOR, 2.07; 95% CI, 1.27–3.38).
Associations Between Traumas and Test-Identified STI
Urine samples were taken during young adulthood and 6.6% of participants tested positive for chlamydia, gonorrhea, or trichomoniasis. In general, traumas were more weakly associated with this outcome, with a notable exception in the unadjusted analyses being witnessed violence, which was associated with 2.12 times the odds of STI (95% CI, 1.57–2.86). After adjustment, only parental binge drinking was significantly associated (AOR, 1.46; 95% CI, 1.01–2.11). No significant gender differences were observed. Unlike for other outcomes, when looking at number of traumas experienced, only reporting 4+ traumas was significantly associated with STI in our adjusted models, though a dose-response relationship was observed (P < 0.03).
Associations were found between trauma score and sexual risk outcomes at all waves, though the highest ORs for each outcome occurred in adolescence. We also observed a dose-response relationship between trauma score and each outcome during at least 1 wave of the study. These findings contribute to literature suggesting that increasing trauma load during the stress-sensitive early years has negative effects on physical and mental well-being25,26,29 and that these effects may extend into adulthood. In addition, our findings are consistent with data from the Centers for Disease Control and Prevention,1 demonstrating that STI risk behaviors are highest in young adulthood, and further suggests that this risk is magnified by childhood maltreatment.
Violence exposures were strong independent correlates of adolescent sexual risk outcomes. Witnessing violence, in particular, was correlated with the outcomes to varying degrees of strength at all life stages. Notably, 12% of those who reported witnessing violence tested positive for an STI in young adulthood, which equated to over 2-fold increase in the odds. A number of studies have postulated the mechanism underlying this connection, finding that exposure to violence increased risk of substance abuse, intercourse while using substances, suicidal ideation and inconsistent condom use, all of which may increase chance of STI acquisition.21,23
A novel contribution of this analysis was the inclusion of indicators of household dysfunction. Parental binge drinking was the only trauma associated with biologically confirmed infection in young adulthood, whereas parental incarceration was the trauma most strongly associated with self-reported STI in adulthood. There is a paucity of research on the effects of these traumas, though measures reflecting household members who abused substances or were incarcerated were included in Hillis et al's25 ACE and STI study. Children who grow up with parent-related adversity may experience feelings of chaos, fear, helplessness and loneliness, and struggle with regulation of affective states and maintaining stable relationships. It is possible that they engage in risky sexual behaviors as they seek to form relationships outside their family, and these early-life patterns extend into adulthood.
Sexual abuse had only modest associations with most of our outcomes, despite being the most studied predictor of STI risk behavior.3–5,7 This is not the first study to cite this inconsistency, and some have speculated that sexual avoidance, common in victims of sexual abuse, may weaken associations.24
One notable exception is the strong association between sexual abuse and sex trade. Ten percent of those reporting sexual abuse also reported engaging in sex trade during young adulthood, a two-fold increase compared to those who did not report abuse. The National Institute of Mental Health Multisite Human Immunodeficiency Virus Prevention Trial, a large study of high-risk women recruited from STD clinics, found a similarly strong association between sex abuse and sex trade. In their study sample, 17.8% of participants who reported sexual abuse in childhood also reported engaging in sex for drugs or money in the past 90 days (OR, 2.28; 95% CI, 1.83–2.83). Our study results suggest that this strong link between abuse and sex trade exists in the general population as well.
Finally, we explored modification of all associations by gender. Females with an incarcerated parent had significantly higher odds of testing positive for STI in young adulthood than males. Childhood neglect was far more correlated to sexual risk behavior in females, with significant gender differences in 4 of the 9 studied outcomes. On the other hand, the association between witnessed violence and multiple partnerships was much stronger for men. Little is understood about the role of gender in moderating sexual risk, though there is evidence that men tend to respond to trauma with externalized stress symptoms (ie, aggression), whereas women tend to show more internalized stress symptoms (ie, depression), which may impact sensitivity to certain traumas.10,30
These nationally representative data yielded results that are generalizable to the US population, and the study’s large sample size allowed us to examine associations for individual effects, including many that have not been studied in the context of STI. The longitudinal nature of Add Health data and its prospective design allowed us to observe relationships over time. Limitations of this analysis include the use of self-reported data, except for test-identified STIs during young adulthood, which may have introduced bias, likely from underreporting. Also, sexual risk outcomes during adolescence were measured as lifetime values, thus we could not ascertain whether they occurred before or after reported trauma. Additionally, there was some loss to follow up at each wave. Higher response rates were noted among participants who were women, white, and native-born, as well as by those with higher parental education and income levels at Wave I.28 The loss of participants who are more likely to have experienced trauma and STI outcomes would bias our results towards the null. However, Add Health investigated the potential bias due to attrition and found the effect to be negligible.28
Sexually transmitted infections are prevalent in United States, impacting every race, gender, and socioeconomic group. They carry the potential to cause life-threatening disease and permanent infertility. Sexual behavior is complex, and as such, STI prevention efforts should be as comprehensive as possible. A broad range of traumas were found to be independent correlates of sexual risk behavior, with increasing trauma score generally correlated with increasing odds of unfavorable outcomes. These findings underscore the need to consider trauma history in STI screening and prevention strategies. Reproductive health care providers should incorporate trauma screening and provide STI testing and follow up as appropriate. Approaches to care such as those espoused by The Sanctuary Model, which recognizes that certain types of interactions between patients and health professionals can compound the negative effects of past traumas,31 must also inform treatment for vulnerable populations. Moreover, adolescents exposed to trauma should be identified so that preventive strategies can be implemented before young adulthood when risk for infection is highest.
1. Center for Disease Control. CDC Fact Sheet: Reported STDs in the United States 2015; https://www.cdc.gov/nchhstp/newsroom/docs/factsheets/std-trends-508.pdf
. Accessed March 15, 2017.
2. World Health Organization. Sexually Transmitted Infections (STIs). 2015; http://www.who.int/mediacentre/factsheets/fs110/en/
. Accessed July 21, 2016.
3. Upchurch DM, Kusunoki Y. Associations between forced sex, sexual and protective practices, and sexually transmitted diseases among a national sample of adolescent girls. Womens Health Issues 2004; 14:75–84.
4. Howard DE, Wang MQ. Psychosocial correlates of U.S. adolescents who report a history of forced sexual intercourse. J Adolesc Health 2005; 36:372–379.
5. van Roode T, Dickson N, Herbison P, et al. Child sexual abuse and persistence of risky sexual behaviors and negative sexual outcomes over adulthood: findings from a birth cohort. Child Abuse Negl 2009; 33:161–172.
6. NIMH Multisite HIV Prevention Trial Group. A test of factors mediating the relationship between unwanted sexual activity during childhood and risky sexual practices among women enrolled in the NIMH Multisite HIV Prevention Trial. Women Health 2001; 33:163–180.
7. Sumner SA, Mercy JA, Buluma R, et al. Childhood sexual violence against boys: a study in 3 countries. Pediatrics 2016; 137.
8. Wilson HW, Widom CS. Sexually transmitted diseases among adults who had been abused and neglected as children: a 30-year prospective study. Am J Public Health 2009; 99(Suppl 1):S197–203.
9. Wilson HW, Widom CS. Pathways from childhood abuse and neglect to HIV-risk sexual behavior in middle adulthood. J Consult Clin Psychol 2011; 79(2):236–246.
10. Haydon AA, Hussey JM, Halpern CT. Childhood abuse and neglect and the risk of STDs in early adulthood. Perspect Sex Reprod Health 2011; 43:16–22.
11. Bensley LS, Van Eenwyk J, Simmons KW. Self-reported childhood sexual and physical abuse and adult HIV-risk behaviors and heavy drinking. Am J Prev Med 2000; 18:151–158.
12. Medrano MA, Hatch JP. Childhood trauma, sexually transmitted diseases and the perceived risk of contracting HIV in a drug using population. Am J Drug Alcohol Abuse 2005; 31:403–416.
13. Kalichman SC, Gore-Felton C, Benotsch E, et al. Trauma symptoms, sexual behaviors, and substance abuse: correlates of childhood sexual abuse and HIV risks among men who have sex with men. J Child Sex Abus 2004; 13:1–15.
14. Senn TE, Carey MP, Vanable PA, et al. Childhood sexual abuse and sexual risk behavior among men and women attending a sexually transmitted disease clinic. J Consult Clin Psychol 2006; 74:720–731.
15. Cunningham RM, Stiffman AR, Dore P, et al. The association of physical and sexual abuse with HIV risk behaviors in adolescence and young adulthood: implications for public health. Child Abuse Negl 1994; 18:233–245.
16. Medrano MA, Desmond DP, Zule WA, et al. Histories of childhood trauma and the effects on risky HIV behaviors in a sample of women drug users. Am J Drug Alcohol Abuse 1999; 25:593–606.
17. Senn TE, Carey MP, Vanable PA. Childhood and adolescent sexual abuse and subsequent sexual risk behavior: evidence from controlled studies, methodological critique, and suggestions for research. Clin Psychol Rev 2008; 28:711–735.
18. Wilson HW, Donenberg GR, Emerson E. Childhood violence exposure and the development of sexual risk in low-income African American girls. J Behav Med 2014; 37:1091–1101.
19. Brady SS, Donenberg GR. Mechanisms linking violence exposure to health risk behavior in adolescence: motivation to cope and sensation seeking. J Am Acad Child Adolesc Psychiatry 2006; 45:673–680.
20. Voisin DR, Hotton A, Neilands T. Exposure to community violence and sexual behaviors among African American youth: testing multiple pathways. Behav Med 2016; 25:1–9.
21. Berenson AB, Wiemann CM, McCombs S. Exposure to violence and associated health-risk behaviors among adolescent girls. Arch Pediatr Adolesc Med 2001; 155:1238–1242.
22. Voisin DR. The relationship between violence exposure and HIV sexual risk behavior: does gender matter? Am J Orthopsychiatry 2005; 75:497–506.
23. Voisin DR, Salazar LF, Crosby R, et al. Witnessing community violence and health-risk behaviors among detained adolescents. Am J Orthopsychiatry 2007; 77:506–513.
24. Wilson HW, Woods BA, Emerson E, et al. Patterns of violence exposure and sexual risk in low-income, urban African American girls. Psychology of violence 2012; 2:194–207.
25. Hillis SD, Anda RF, Felitti VJ, et al. Adverse childhood experiences and sexually transmitted diseases in men and women: a retrospective study. Pediatrics 2000; 106:E11.
26. Hillis SD, Anda RF, Felitti VJ, et al. Adverse childhood experiences and sexual risk behaviors in women: a retrospective cohort study. Fam Plann Perspect 2001; 33:206–211.
27. The National Longitudinal Study of Adolescent to Adult Health. http://www.cpc.unc.edu/projects/addhealth
. Accessed November 25, 2016.
28. Harris KM. The Add Health Study: Design and Accomplishments 2013; http://www.cpc.unc.edu/projects/addhealth/documentation/guides/DesignPaperWIIV.pdf
. Accessed April 13, 2017.
29. Khoury L, Tang YL, Bradley B, et al. Substance use, childhood traumatic experience, and posttraumatic stress disorder in an urban civilian population. Depress Anxiety 2010; 27:1077–1086.
30. Thompson MP, Kingree JB, Desai S. Gender differences in long-term health consequences of physical abuse of children: data from a nationally representative survey. Am J Public Health 2004; 94:599–604.
31. The Sanctuary Model. http://sanctuaryweb.com/
. Accessed November 27, 2016.