|Home | About | Journals | Submit | Contact Us | Français|
This study examined the relationship between substance use, mental health problems, and HIV sexual risk behaviors among a sample of foster care adolescents. Data were collected through structured baseline interviews from 320 adolescents (age 15-18 years) who resided in foster care placements and participated in a larger evaluation study of an HIV prevention program. Final logistic regression models indicated that delinquent behavior and marijuana use were the most significant predictors of engaging in any one HIV risk behavior. Adolescents who reported delinquent behaviors, alcohol use, and marijuana use, and who were females were more likely than their counterparts to engage in vaginal sex without using a condom. Future research is needed to further identify risk and protective factors for substance use, mental health problems, and HIV sexual risk behaviors among adolescents in foster care. HIV prevention efforts for these vulnerable adolescents should target those with substance use and delinquent behaviors.
As many as 250,000 adolescents currently reside in the care and custody of the United States foster care system (U.S. Department of Health and Human Services, 2006). Adolescents enter into foster care placements because of neglect, and/or physical, emotional, and sexual abuse, and therefore all have histories of maltreatment. Examination of the empirical research indicates that these adolescents are at increased risk for using substances (Thompson & Auslander, 2007), developing mental health problems (Landsverk, 1999; Pilowsky, 1995; Trupin et al., 1993), and engaging in HIV sexual risk behaviors (Auslander et al., 2002; Becker & Barth, 2000; Elze et al., 2001). However, the extent to which the co-occurrence of mental health problems and substance use places adolescents in foster care at increased risk for engaging in HIV sexual risk behaviors has not been sufficiently examined.
Adolescents involved with the foster care system have consistently reported substance use at rates equal to or greater than those in the general population. Findings from earlier studies indicated that 20% to 50% of adolescent samples reported use of alcohol or other substances while in foster care, with substantial numbers reporting problematic use and continued use after exiting care (Taussig, Clyman, & Landsverk, 2001). Additionally, the 1998-1999 Washington State Adolescent Foster Care Survey (Kohlenberg et al., 2002), which estimated the rates of alcohol and other drug use among adolescents age 12 to 17 years who resided in family foster care, found that 34% of the 231 participants had used alcohol at least once in the past year, with 13% using alcohol in the past month. Twenty-three percent reported marijuana use in the past year, with 10% reporting use in the past month. An examination of the 2002 and 2003 National Survey on Drug Use and Health (NSDUH; Substance Abuse and Mental Health Services Administration [SAMHSA], 2005), found that 38% of the approximately 680,000 youths (age 12-17) who had ever been in foster care had used alcohol in the past year and 34% had used marijuana or other illicit drugs. Caucasian youth who had ever been in foster care were more likely than their African American counterparts to have used alcohol (41% vs. 30%) or illicit drugs (36% vs. 27%) in the prior year. No gender differences in alcohol or illicit drug use were found. Thompson and Auslander (2007) noted similar findings among adolescents currently in foster care, with Caucasian adolescents being more likely than African Americans to report current marijuana use and use of both alcohol and marijuana. Moreover, females were as likely to use alcohol and/or marijuana as their male counterparts.
A number of studies provide evidence of the prevalence of mental health problems among foster care adolescents. Mech and associates (1994) reported that over half of the foster care youth in their study had social-emotional adjustment problems. Trupin and colleagues (1993) found that 72% of the children in Washington State's child welfare system evidenced severe emotional disturbances indistinguishable from a criterion group of children in the state's most intensive mental health treatment programs. Pilowsky's (1995) review of studies published between 1974 and 1994 found that externalizing disorders in particular may be more prevalent than internalizing disorders in the foster care population, as did Landsverk's (1999) later review of nine additional studies. Most recently, McMillen and colleagues (2005) found that 61% of their sample of older foster care adolescents from Missouri reported having at least one psychiatric disorder during their lifetime (37% in the last year), with 62% having onset of their earliest disorder prior to entering the foster care system.
Relatively few sexual risk studies have focused solely on adolescents in foster care. Thus, studies with troubled adolescents from environmental contexts similar to those of adolescents in foster care (e.g., in detention, runaway, homeless) have provided much of what is known on the sexual risk of adolescents in foster care. Examination of such research indicates that many troubled adolescents engage in unsafe sexual activities, have sex while using alcohol and other substances, hold negative attitudes about condom use, and use condoms about one-third of the times that they have sex (Gillmore et al., 1994). Studies of runaway and homeless adolescents, as many as 40% of whom had previously lived in foster care, have reported high rates of mental health problems and high rates of sexual risk behaviors (Robertson, 1989). They often have a heightened awareness of sexual activity, initiate sexual activity earlier, engage in prostitution, and are estimated to be 6-12 times more likely to become infected with HIV than other adolescents (Rotheram-Borus & Koopman, 1991).
In one of the few studies of HIV risk to be specifically conducted among adolescents in foster care, Auslander and colleagues (2002) examined the relationship among mental health problems, sexual abuse, and HIV risk behaviors and intentions and found that externalizing behaviors, after controlling for demographics and behavioral intentions, were most strongly associated with HIV risk behaviors. Moreover, a significant race by gender interaction was identified as a determinant of both current HIV risk behaviors and behavioral intentions, with Caucasian females reporting higher risky behaviors and intentions on average than ethnic minority females and males. Additional analyses by Elze and associates (2001) indicated a significant relationship between the severity of sexual abuse and HIV risk behaviors among adolescents in foster care, after accounting for the contribution of other childhood trauma and emotional and behavioral problems. The current study furthers our understanding of HIV risk among adolescents in foster care by examining the associations of substance use, as well as mental health problems, with sexual risk behaviors.
The purpose of this study was to determine the extent to which substance use and specific mental health problems increased the likelihood of the adolescents engaging in HIV sexual risk behaviors and to examine the extent to which these patterns vary according to race, gender, and age. Problem Behavior Theory (PBT), a framework that has been utilized to explain and predict substance use, behavioral problems, and the onset of sexual activity in the general adolescent population (Jessor & Jessor, 1977), guided the study. PBT is based on the premise that problem behaviors are part of normal adolescent development and play a major role in the transition to adulthood. The theory hypothesizes that many adolescent risk behaviors are interrelated manifestations of a common underlying syndrome of problem behavior. Given that delinquency and substance use generally precede sexual activity (Stiffman et al., 1992), substance use and mental health problems could be viewed as components of a risk behavior syndrome that may predict sexual involvement and, by extension, sexual risk-taking. Due to the prevalence of mental health problems among adolescents in foster care, and the strong association of mental health problems and substance use among the general adolescent population (Zeitlin, 1999), it was hypothesized that the use of substances and endorsement of mental health problems would increase their likelihood of engaging in HIV sexual risk behaviors.
The present study utilizes baseline data collected from a sample of 320 older adolescents who had been placed in foster or other out-of-home care by the child welfare services of a Midwestern metropolitan area and had participated in a life-skills program designed to assist with their transition to independent living. Adolescents were eligible for the study if they were 15 -18 years of age prior to starting the program and placed in out-of-home care through child welfare services. Eligibility was assessed by social workers prior to participation in the study via a brief informal group meeting that assessed adolescent interest in the program and ability to interact appropriately with others in a group setting. Adolescents were excluded from the study if they displayed severe learning problems (i.e., unable to read or write) or severe behavior problems (e.g., violent behavior). Four adolescents were excluded from the study for severe behavioral problems, as they were seen as being incapable of participating without seriously disrupting the group process. For the present study, 31 cases from the original recruited sample of 351 were excluded because race was designated as “other.” This strategy was utilized to avoid ethnic lumping (Fontes, 1995), leaving a final sample comprised of 320 foster care adolescents aged 15 to 18 years (M = 16.3, SD = .84). Over half (53.8%, n = 172) were female, 46.2% (n = 148) were male, 66.3% (n = 212) self-identified as African American, and 33.7% (n = 108) self-identified as Caucasian.
Child welfare social workers, group home workers, foster or biological parents, and the adolescents themselves made referrals to the study. The institutional Review Board at Washington University in St. Louis approved all procedures. Informed consent was obtained from legal guardians and written assent was provided by adolescents before baseline interviews were completed. Structured baseline interviews were then conducted with the adolescents by trained MSW students prior to participating in the HIV prevention program. Each adolescent was paid five dollars for participating in the baseline interview.
The Youth Self-Report (YSR) of the Child Behavior Checklist (CBCL) was utilized to assess mental health problems (Achenbach, 1991). The YSR is designed to assess the emotional and behavioral problems of adolescents (11-18 years) in a standardized format and is comprised of 118 items concerning the psychological functioning of the adolescent over the preceding 6 month period. The following 8 YSR mental health subscales were computed and clinical-borderline cut-off scores were utilized to determine the following problems: Internalizing Problems (withdrawn; somatic complaints; anxious/depressed; social problems; thought problems; attention problems); and Externalizing Problems (delinquent behavior; aggressive behavior.) The subscales correspond to well-established clinical symptoms, and demonstrate significant associations with DSM diagnostic categories (Achenbach, 1991; Weinstein et al., 1990).
Selected items from the alcohol and other drug use sections of the Diagnostic Interview Schedule for Children-Revised Version (DISC-R), initially developed by Costello and colleagues (1984), were used to measure substance use in the 6 months prior to assessment. For this study, the use of the following substances was assessed: alcohol; marijuana; amphetamines; barbiturates; other tranquilizers; heroin; cocaine; hallucinogens; and inhalants. Inter-rater reliability and validity for the DISC-R have been reported to be high in previous research (Shaffer et al., 1996).
To assess HIV sexual risk behaviors, respondents reported on the frequency of the following sexual behaviors performed without a condom in the last 2 months: vaginal intercourse; anal intercourse; oral sex; vaginal, anal, or oral sex while under the influence of alcohol or other drugs; and trading vaginal, anal or oral sex for money, drugs, or shelter. Individual risk items were coded “1” if a behavior was endorsed and “0” if not.
Descriptive statistics determined the frequencies of substance use, mental health problems, and HIV sexual risk behaviors among the adolescents. Spearman's Rho correlations were calculated to examine multicollinearity among predictor variables. Bivariate relationships between each demographic, substance use, and mental health variable and HIV sexual risk behaviors were examined using logistic regression. Variables significant at the bivariate level (using odds ratios [OR] and 95% confidence intervals [CI]) were then entered into final multiple logistic regression models to identify which placed foster care adolescents at increased likelihood of engaging in HIV sexual risk behaviors (e.g., unprotected vaginal sex).
Table 1 presents the frequencies and percentages of reported substances use, mental health problems, and HIV sexual risk behaviors among the adolescents. Alcohol (39.4%, n = 126) and marijuana (35.6%, n = 114) were the substances most frequently used by the adolescents in last six months. Given the low reports of other substance use, only alcohol and marijuana use were included in logistic regression analyses. The most endorsed mental health problem was delinquent behavior, with over one fifth (21.9%, n = 70) of the adolescents meeting borderline-clinical level criteria. Unprotected vaginal sex was the most frequently endorsed HIV sexual risk behavior (17.8%, n = 57). Given the low endorsements of several HIV sexual risk behaviors, two sexual risk variables were used for further bivariate and multivariate analyses: 1) engagement in any HIV risk behavior; and 2) vaginal sex without a condom.
Bivariate logistic regression analyses (see Table 2) found that Caucasian adolescents were over one and a half times more likely (OR = 1.60; CI = 1.00, 2.55) to engage in any HIV risk behavior and almost two times more likely (OR = 1.84; CI = 1.03, 3.30) to engage in unprotected vaginal sex than African American adolescents. Females were slightly more likely (OR = 0.52; CI = 0.29, 0.95) than males to engage in unprotected vaginal sex. No differences in HIV risk behaviors were found based on age.
Bivariate logistic regression analyses determined that adolescents who used alcohol (last 6 months) were two times more likely (OR = 2.19; CI = 1.39, 3.45) than those who did not use alcohol to engage in any HIV risk behavior and almost four times more likely (OR = 3.89; CI = 2.12, 7.14) to have vaginal sex without a condom (see Table 2). Those who used marijuana were over two and a half times more likely (OR = 2.65; CI = 1.65, 4.24) than their counterparts to engage in any HIV risk behavior and nearly five times more likely (OR = 4.82; CI = 2.61, 8.88) to have vaginal sex without a condom.
Adolescents who met the criteria for delinquent behavior were almost three times more likely (OR = 2.80; CI = 1.60, 4.90) than those who did not to engage in any HIV risk behavior (see Table 2), over four and a half times more likely (OR = 4.68; CI = 2.52, 8.68) to engage in unprotected vaginal sex. No other mental health problem was found to be associated with engaging in any risk behavior or having unprotected vaginal sex.
Demographic, substance use, and mental health variables significantly associated with HIV risk behaviors at the bivariate level were tested for multicollinearity via Spearman Rho correlational analyses. Coefficients ranged from .014 to .482, indicating mild intercorrelations. Predictors significant at the bivariate level were then entered into multivariate models (Table 3) to identify which remained most predictive of engaging in (last 2 month) any HIV risk behavior and vaginal sex without a condom.
Marijuana use and delinquent behavior were found to be most predictive of engaging in any HIV risk behavior, with adolescents who smoked marijuana being almost two times more likely (OR = 1.87; CI = 1.08, 3.22) and those who reported delinquent behavior being over two times more likely (OR = 2.21; CI = 1.23, 3.96) than their counterparts to engage in any HIV risk behavior.
Gender, alcohol use, marijuana use, and delinquent behavior remained most predictive of having vaginal sex without a condom. Female adolescents were 50% more likely than males to engage in vaginal sex without a condom. Adolescents who used alcohol (OR = 2.33, CI = 1.12, 4.82) and marijuana (OR = 2.66, CI = 1.28, 5.51) were each over two times more likely than their counterparts to engage in vaginal sex without a condom. Those who met criteria for delinquent behavior were over three times more likely (OR = 3.26, CI = 1.67, 6.37) than those who did not to engage in vaginal sex without a condom.
This study confirmed that many of the foster care adolescents interviewed used substances, reported mental health problems, and engaged in sexual activities without the protection of condoms. Among the substances endorsed, alcohol and marijuana, as with the general adolescent population, were the most commonly used substances by the adolescents. This is consistent with multiple studies who have found strong relationships between childhood maltreatment and substance use among adolescents (e.g., Dube et al., 2003; Anda et al., 1999).
Most of the over 500,000 children and adolescents currently placed in the U.S. foster care system were removed from their homes as a result of neglect, caretaker absence, physical abuse, and/or sexual abuse (Children's Bureau, 2005; Garland, Landsverk, Hough, & Ellis-MacLeod, 1996). The long-term consequences of childhood maltreatment in terms of risk for later psychopathology have been documented extensively (e.g., Kendler et al., 2000; Johnson et al., 2002ab; Nelson et al., 2006). In addition, removal from the home and separation from parents, siblings, relatives, and friends can have devastating and long lasting emotional and psychological effects, which have also been shown to increase risk for substance abuse and mental health problems (e.g., Beitchman, Zucker, Hood, DaCosta & Akman, 1991; Beitchman, Zucker, Hood, DaCosta, Akman, & Cassavia, 1992).
Caviola and Schiff (1989) describe drug use as the first mode of defense to deal with the trauma of abuse. Other researchers characterize substance use among abused adolescents as a means to self-medicate for abuse-related mental health problems such as depression, anxiety, or posttraumatic distress; dissociate themselves from painful emotions; cope with unresolved psychological distress from their abuse experiences; and blur upsetting memories (Briere & Elliot, 1994; Harrison, Fulkerson, & Beebe, 1997). Substance use as a maladaptive coping strategy may generalize to other stressful situations. In clinical settings, abused adolescents report high rates of substance abuse (McClellan et al., 1995), and substance-abusing adolescents report high rates of childhood abuse (Caviola & Schiff; 1988). When compared to their nonabused peers, abused adolescents in both clinical and community settings report more illicit drug use (e.g., Cohen, Berliner, & Mannarino, 2000), more alcohol use (Brown et al., 1997), more poly drug use, including alcohol and marijuana (Kilpatrick et al., 2000), earlier initiation of substance use (Garnefski & Diekstra, 1997), more frequent attempts to self-medicate painful emotions, and more diagnosed substance use disorders (Singer, Petchers, & Hussey, 1989). These associations continue into adulthood, with significantly greater substance abuse found not only among adult survivors of childhood sexual abuse (Briere, 1988), but also among survivors of other forms of adverse childhood experiences (e.g., Felitti et al., 1998). Moreover, high rates of comorbid mental health problems and substance abuse have been documented in survivors of both physical and sexual forms of interpersonal violence (Kilpatrick, Saunders, & Smith, 2001).
Likewise, findings from this study indicated that the mental health problems reported among the sample are consistent with those found in prior studies of foster care adolescents (e.g., Auslander et al., 2002; Landsverk, 1999; Pilowsky, 1995; Trupin et al., 1993), with externalizing problems being more prevalent than internalizing disorders. Delinquent and aggressive behaviors are collectively referred to as externalizing behavior problems or conduct disorders and are associated with lack of impulse control (Achenbach, 1991). Delinquent behaviors were found to be present in 21.9% of the sample and aggressive behaviors in 10.7%. Risk taking behaviors, whether they are related to sex or substances, often co-occur and are more common among adolescents with behavior problems such as delinquent and aggressive behaviors (Jessor, 1998).
An important contribution of this study is to identify the substances and mental health problems that are most associated with HIV risk behaviors in this population. Findings indicate that marijuana use and delinquent behaviors were most predictive of both engaging in any HIV risk behavior and vaginal sex without a condom. Additionally, gender and alcohol use, predicted engaging in vaginal sex without a condom. Interestingly, delinquent behavior was the only mental health problems behavior that uniquely predicted HIV risk behaviors, thus suggesting that engaging in HIV risk behaviors is not necessarily related to all mental health problems, but is perhaps part of an externalizing syndrome among adolescents in foster care, among which, substance use is a component.
Child maltreatment traumatizes children through different mechanisms (i.e., traumatic sexualization, betrayal, stigmatization, and powerlessness) that lead to cognitive distortions, impaired decision making and problem solving, and sexual preoccupation and re-victimization. When left untreated, delinquent behaviors and substance use become common maladaptive methods of coping, along with sexual risk taking. Although this study included cross-sectional data, the results suggest a possible time-order, whereby maltreatment leads to behavioral problems such as delinquency and substance use, which in turn, increased the likelihood of engaging sexual risk behaviors.
Another important finding is the higher risk of females engaging in unprotected vaginal sex compared to their male counterparts. One explanation for this finding is that girls experience higher rates of child sexual abuse than boys and that the subsequent use of alcohol and marijuana to cope with the trauma, coupled with untreated mental health problems, makes them more at risk than boys. Moreover, the powerlessness associated with sexual abuse leads to a greater likelihood of sexual re-victimization and sexual risk taking. Future research with female adolescents should adapt evidence-based trauma-focused cognitive behavioral therapy for sexually abused girls in foster care as a potential strategy to reduce mental health problems, and the subsequent risk taking behaviors, such as substance use and unsafe sex.
This study is not without its limitations. Because this study used a cross-sectional design, causal sequences could not be tested between substance use, mental health problems, and HIV sexual risk behaviors. However, this design was sufficient for identifying the likelihood engaging in specific HIV sexual risk behaviors given specific substance use and mental health problems. Substance abuse and psychiatric disorders were not measured in this study, which limits the comparability of findings to other studies that employed diagnostic measures. All data were collected through face-to-face self-report interviews and participant responses may have been biased by social desirability. Many studies of older youth in the foster care system, however, have been conducted with case records or administrative data only, so this limitation is also a comparative strength of this study. Additionally, self-report questionnaires may be problematic for youths who have poor literacy skills. To reduce social desirability bias among lower literacy subjects, future research may benefit from using computer-assisted self-administered interviews with an audio component for sensitive questions about sexual behaviors and alcohol and drug use. A last limitation of the study is that the sample was not random, but rather a convenience sample of adolescents in foster care. Although the sample in the present study appears representative of the national foster care population (e.g., overrepresentation of African Americans, high rates of substance abuse and mental health problems), it may not be representative of all adolescents in foster care.
HIV prevention interventions are going to have to move beyond the current educational standards by assessing substance use and mental health problems among adolescents in foster care and targeting interventions to those with such problems. Additional aspects of the adolescents' lives that may impact their motivation and ability to act on information or change behavior must be incorporated into prevention methods. The social context of sexual risk behaviors, issues of social marginalization, ethnic and cultural factors, sexual orientation issues, and the cognitive and emotional aspects of adolescence should not be ignored.
Brief motivational interventions would likely be feasible and effective for adolescents in foster care. Ample evidence from randomized clinical trials supports the effectiveness of brief interventions to reduce substance use across a variety of populations, including primary care, medical, child-welfare, substance abuse, and mental health settings (e.g., Neighbors et al., 2006; Toumbourou et al., 2007). Brief alcohol interventions reduced drinking among young adults about to enter intensive substance abuse treatment (Werch, 2000), binge and heavy drinking college students (Marlatt, 1998; Carey et al., 2006), and homeless adolescents (Peterson et al., 2006; Baer et al., 2007).
Brief motivational interviewing integrates the relationship-building principles of humanistic therapy with more active cognitive behavioral strategies targeted to the client's stage of readiness to change (Prochaska, DiClemente, & Norcross, 1992). It is a client-centered method for enhancing intrinsic motivation to change by exploring and resolving ambivalence (Miller & Rollnick, 2002). Motivational interviewing strategies have been shown to reduce alcohol, marijuana, and tobacco use among adolescents and young adults (McCambridge & Strang, 2004;) and augment the effectiveness of skills-based HIV risk reduction interventions (Carey et al., 2006; Belcher et al., 1998). Masterman and Kelly (2003) argue that motivational interviewing within a harm reduction framework is well suited for adolescent and young adult populations who may be less likely to be interested in changing their behaviors, especially those who have already initiated some risk behaviors, because these approaches do not require a lifetime commitment to change and are based on harm reduction principles (Baer & Peterson, 2002; Monti, et al., 2001). They also mesh well with developmental issues, such as autonomy, individuation, and reactivity to being told what to do. Brief, harm-reduction interventions therefore present relatively few logistical and psychological barriers for at-risk individuals who may be ambivalent about risk reduction (Rutledge et al., 2001; Stern et al., 2007).
Subsequently, brief motivational interviewing interventions for delinquent adolescents in foster care with impulse control or resistance issues should incorporate the following rationale and basic principles: the interviewer expresses empathy; discrepancies between current and desired behavior/self-image are developed; ambivalence is presented as a natural and normal part of behavior change; alternative behaviors/options are identified; argumentation is avoided; interviewer rolls with client resistance; stage of readiness to change behavior is acknowledged and respected; self-efficacy, responsibility, and optimism are facilitated; personalized feedback and behavior plans are developed; and behavioral skills training is provided. The approach should emphasizes personal choice and responsibility, view client resistance as normative, build upon client strengths and capacities, and be tailored to the client's individual culture, beliefs, motivations, and behaviors. The preventive information given, motivational techniques employed, and behavioral skills training provided (or not) should vary by stage of readiness to change. As it is likely that mental health problems, and associated functional impairment, will impede the learning of prevention information and adoption of safer behavioral practices. Future research should examine the influence of mental health problems, particularly delinquent behavior, on substance use and HIV intervention outcomes.
Given the disproportionate HIV risk among female foster care adolescents, interventions should be developed that target their special prevention needs. Findings of studies that have examined the differential effects of HIV risk reduction programs on male and female college students (O'Leary et al., 1996) suggest that low-intensity risk reduction programs do not reach females. Given this, interventions for female adolescents in foster care will likely require adequate intensity to address their perceived invulnerability to HIV and motivation regarding behavior change. Group interventions led by peers of the adolescents should be particularly effective, as they may combat the greater societal norms related to sex roles and behaviors. Additionally, females in foster care will require individualized skills training and practice that are not appropriate for group settings (e.g., male and female condom training). A small number of studies have reported success in reducing female HIV risk using more intensive, individualized interventions (e.g., DiClemente & Wingwood, 1995).
Approximately 20,000 adolescents aged 16 and older exit the foster care system. Studies of these older adolescents have shown that they are more likely than those in the general population to drop out of high school, be unemployed, and be dependent on public assistance. Many find themselves homeless, without health care, in prison, or parents at an early age (e.g., Courtney et al., 2001). Substance abuse has also been shown to be a significant barrier in the development of independent living skills needed to successfully transition from foster care to self-sufficiency. Because of these dire outcomes, all of which are associated with increased HIV risk, it is imperative that forthcoming research identify risk and protective factors for substance use, mental health problems, and HIV sexual risk behaviors and develop effective prevention efforts that target these factors among adolescents in foster care.