A systematic search for articles involving substance use and HIV risk behaviors among justice-involved youth published from January 1, 2013 to March 2, 2018 was conducted in PubMed, PsycINFO, Embase, Web of Science, Sociological Abstracts, Social Services Abstracts, and Google Scholar. The search combined four concepts: incarceration, youth, substance use, and HIV risk behaviors. A search strategy was developed in collaboration with a clinical librarian using an iterative process that involved testing search terms, keywords, and controlled vocabulary, including MeSH and Emtree terms, for each of the search concepts and examining the relevance of corresponding search results. The search strategy was peer reviewed by a second librarian using Peer Review of Electronic Search Strategy guidelines. Detailed search strategies for each database can be found in Appendix Table 1. The literature search yielded 858 articles. After excluding duplicates, 532 articles were screened for inclusion based on title and abstract and 478 were eliminated because of their irrelevance to the topic. Additionally, eight publications from the search results that addressed the intersection of HIV, substance use, and juvenile justice were excluded because of the study design; this included one systematic review of HIV/STI -prevention interventions for detained and delinquent youth, three studies examining intersecting risks of HIV, substance use, and justice involvement among HIV-positive youth, two studies of large existing adolescent health databases,greenhouse tables and two retrospective studies of at-risk populations examining how a history of juvenile justice involvement and/or incarceration may increase HIV risk. A total of 46 articles were therefore included in the final review. Inclusion criteria were as follows: was a peer-reviewed article; included a US-only population; youth were in 10–-18 -year -old age range at time of study enrollment ; participants were actively justice-involved at time of study enrollment ; data were collected on HIV/STI and substance use, which included any illicit substances for minors, including nicotine.
A team of three reviewers assessed and summarized findings of the final 46 articles. One reviewer selected a random sample of articles to review for quality assurance. Any disagreements in the abstract review were resolved after a consultation and detailed examination of the study. Across the 26 cross-sectional studies reviewed, 38% included community supervised samples. Findings suggest that, among any justice-involved youth, substance use confers enhanced risk for engaging in HIV- risk -related behaviors. Many studies used global measures of substance use, collapsing drug use into one broad category. Studies have found that drug use is associated with unprotected sex , number of sexual partners, drug use during sexual activity, and increased rates of STIs.For example, one study found that attention-deficit- hyperactivity disorder was associated with sexual risk behavior for youth with conduct problems, but that cannabis use completely accounted for this association. A cross- sectional brain fMRI imaging study focusing on alcohol detected activation of brain regions associated with risky decision-making, riskier peer norms, and number of days endorsing sex while using alcohol among a sample of community-supervised justice-involved youth. Studies examining multiple substances find differential effects of these substances on HIV risk. One cross-sectional study by Gillman and colleagues found that detained youth who reported frequently used only cannabis were less likely to engage in risky sexual behavior and reported greater intention to use condoms compared with those who frequently used alcohol and cannabis or just alcohol alone. In this study, alcohol use alone appears to increase the likelihood of risky sexual behavior among juvenile detainees than either alcohol and cannabis or just cannabis alone. Several cross-sectional studies identified critical social ecological factors associated with enhanced risk for substance use and HIV, such as gang violence, dating violence victimization, and a history of child maltreatment. Several studies also included psychiatric risk factors, such as depression. The majority of cross-sectional studies focused on individual-level factors associated with HIV and substance use risk and, as stated above, several studies included an examination of larger socioecological factors.
Notably, few studies incorporated interpersonal and or parent or family-level factors associations with risk.Of the ten studies published, 80% centered on intersecting risk of HIV and substance use over time among community-supervised justice -involved youth, such as truant court-involved youth, youth presenting to a court intake center youth on probation, and youth in the juvenile drug court. These studies assessed and supported a wide variety of individual-level factors associated with substance use and HIV risk, including callous- unemotional traits, incidence of pregnancy, clustering of problem behaviors, and history of sexual coercion. Several longitudinal studies used global measures of any substance use and others included measurement of single substances, such as alcohol, marijuana and cigarettes. The three studies that focused on following detained youth highlighted contextual risk factors for substance use and HIV such as racial disparities, community trauma, and child welfare involvement. The Northwestern Juvenile Project was a seminal study that followed detained youth starting in 1998 for 14 years . Abram and colleagues found that multiple sexual partners and unprotected vaginal sex remain prevalent at the 14- year follow-up but that risk varied by sex and race, with African American and Hispanic males most at -risk. Among females, non-Hispanic white youth were at greatest sexual risk; the authors attribute this to their higher rates of substance use disorder relative to African American and Hispanic women. Ramaswamy and colleagues examined cigarette smoking among a 16–-18- year -old re-entry population of Black and Latinx men to ascertain risk factors associated with nicotine use prior to detention and 1 year post re-entry; high rates of smoking were reported at follow-up and associated with foster care history and number of prior arrests; only use prior to detention was also related to a greater number of sexual partners. Lastly, Puja Seth and colleagues followed 188 African American female detained adolescents over 6 months and found that community trauma at baseline was a significant risk factor for future unprotected sexual activity and marijuana use as well as post traumatic stress disorder symptoms.Of the ten published intervention studies, with the exception of one within-subject design, all were RCTs.
All interventions were conducted with community supervised justice -involved youth; however, two interventions began in detention and then continued with youth post-release. Two intervention trials published two sets of papers with the same intervention just at different stages. Interventions ranged with respect to format and intensity, ranging from brief to more intensive. For example, Bryan and colleagues recently demonstrated the efficacy of a single 3-h substance use and sexual risk -reduction intervention session for detained youth in reducing STI incidence at 12- month follow-up. In a pilot study of adolescents recruited from a juvenile drug court , participants were randomized to either a 6-month-long, sexual risk – reduction protocol with an emphasis on contingency management and family involvement or a treatment-as-usual arm; preliminary results demonstrated that adolescents randomized to the intervention exhibited slower increases in sexual behavior over the study period. Likewise, Donenberg and colleagues randomized youth on probation to a 6- month sexual risk -reduction program or a time-matched health promotion program ; in a moderator analysis, high-risk adolescents reported significantly lower risk behaviors than controls at the end of the trial. Five interventions incorporated parent training or family-based approaches. Efficacy of the interventions was variable. For instance, Perrino and colleagues randomized community-dwelling, Hispanic youth with a history of delinquency to either a multifaceted, family-based intervention or a community-practice control; youth in the intervention arm were significantly less likely to endorse internalizing symptoms at 6 to 12 months following initial assessment. This effect was especially pronounced among youth with worse caregiver-youth communication at baseline, which suggests that the internalizing-reducing effect of the intervention may have been mediated by improved communication. Another study that focused on drug court-involved youth, randomized to either a family-focused intervention for substance use disorders or usual services for 1 year, no clinically significant treatment-attributable effects were detected for substance use, sexual risk behaviors, and HIV- risk behaviors. The authors posited that treatment effects may have been obscured by interventions that were implemented to all study participants by the juvenile drug court. In a study from our group, juvenile drug court-involved adolescents and an involved caregiver were randomized to either a five-session,growing cannabis indoors family-based substance use and HIV/STI- risk -reduction arm or a psycho education-only arm. At 3 months, results suggested that the intervention was associated with increased motivation to change marijuana use, lead to a decrease in marijuana use, and decrease in risky sex over time. Dembo and colleagues tested a brief intervention for truant youth and families that found a two – session youth-only intervention to be superior to that which added a parent-only session in robustly decreasing recidivism at 12 and 24 months; increased rates of recidivism were noted among youth with more substance use and sexual- risk behaviors at baseline. Both interventions that began in detention and continued post-release included a first, single session while in detention and then delivered the remainder of the intervention while community-supervised. DiClemente and colleagues tested a gender responsive intervention uniquely tailored for African American detained girls and designed to reduce new STIs, increase safe-sex practices, and improve psychosocial markers. Participants were randomized to Imara, which included three individual and four phone-based sessions, or time-matched psychosocial intervention; 3 months post-intervention, participants in the intervention arm reported more frequent condom use self-efficacy, STI-HIV knowledge, and condom-use skills. However, groups did not differ on new STI cases, condom use, or number of sexual partners.
Similarly, Rowe and colleagues randomized drug-involved, detained youth in two sites to either Multidimensional Family Therapy or enhanced treatment as usual; both groups demonstrated reduced rates of unprotected sex acts and STI incidence from intake to 9 months. Of note, youth in both conditions received structured HIV/STI prevention but only those in MDFT received family-based HIV/STI prevention after release. Only in one of the two sites, the MDFT group demonstrated a lower frequency of sex acts and unprotected sex compared to the usual-care group at 9- month follow-up. Our review illustrates recent advances in our understanding of how substance use is interrelated with HIV risk behaviors for justice-involved youth. Over the past 5 years, the number and scope of studies examining and addressing intersecting risks among this population have increased. Necessarily, the field has expanded its focus to include juvenile justice populations outside of youth in detention settings. Specifically, there have been a proliferation of studies focused on one particular subset of this at-risk population: community-supervised justice-involved youth . By including community-based juvenile justice samples, the field is beginning to identify distinct risks, expanded juvenile justice settings, and new frameworks to develop, test, and implement critical substance use and HIV -prevention intervention programs. The field has observed an increase in the proportion of longitudinal studies examining risk in this population, yet the majority of studies continue to be cross-sectional. The lack of longitudinal studies limits the field’s ability to ascertain how risk develops over time; a critical component of HIV-prevention efforts among a subset of youth that are at great risk for future substance use and adult criminal justice involvement . There has also been a marked increase in the number of published interventions, but overall there is still a dearth of available evidence-based behavioral treatments. Programs that target substance use and HIV risk behaviors in tandem are sorely needed; as are interventions that tailor HIV -prevention content to substance -using youth in the justice system. In addition, intervention effects remain short-term and thus effects of interventions into emerging adulthood remain unknown. The Northwestern Juvenile Project and Leve and colleagues conducted studies following justice -involved youth from adolescence into critical adult HIV risk windows. Leve and colleagues’ study followed justice-involved girls post- intervention, but intervention effects were not the focus of the longer-term follow-up study. As such, we are very limited in understanding how interventions developed and delivered in adolescence impact HIV risk into adulthood. Evidence examined as part of this review suggests that substance use promotes HIV risk behaviors for justice-involved youth populations. With one exception, the cross-sectional literature underscores that substance use is associated with engaging in a variety of sexual risk behaviors including inconsistent condom use, number of sexual partners, and history of STIs.