 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
ARTICLE |
a Departments of Psychiatry and Behavioral Sciences
b Preventive Medicine, Feinberg School of Medicine, Northwestern University, Chicago, Illinois
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
METHODS. At the baseline interview, HIV/sexually transmitted infection drug and sex risk behaviors were assessed in a stratified random sample of 800 juvenile detainees aged 10 to 18 years. Participants were reinterviewed approximately 3 years later. The final sample in these analyses (n = 724) included 316 females and 408 males; there were 393 African American participants, 198 Hispanic participants, 131 non-Hispanic white participants, and 2 participants who self-identified their race as "other."
RESULTS. More than 60% of youth had engaged in 
10 risk behaviors at their baseline interview, and nearly two thirds of them persisted in 10 risk behaviors at follow-up. Among youth living in the community, many behaviors were more prevalent at follow-up than at baseline. Among incarcerated youth, the opposite pattern prevailed. Compared with females, males had higher prevalence rates of many HIV/sexually transmitted infection risk behaviors and were more likely to persist in some behaviors and develop new ones. Yet, injection risk behaviors were more prevalent among females than males and were also more likely to develop and persist. Overall, there were few racial and ethnic differences in patterns of HIV/sexually transmitted infection risk behaviors; most involved the initiation and persistence of substance use among non-Hispanic whites and Hispanics.
CONCLUSIONS. Because detained youth have a median stay of only 2 weeks, HIV/sexually transmitted infection risk behaviors in delinquent youth are a community public health problem, not just a problem for the juvenile justice system. Improving the coordination among systems that provide HIV/sexually transmitted infection interventions to youth— primary care, education, mental health, and juvenile justice—can reduce the prevalence of risk behaviors and substantially reduce the spread of HIV/sexually transmitted infection in young people.
Key Words: adolescent sexual behavior • HIV/AIDS • juvenile delinquents • longitudinal study • incarceration • corrections • drug use
Abbreviations: STI—sexually transmitted infection • CCJTDC—Cook County Juvenile Temporary Detention Center • CFR—Code of Federal Regulations • NIDA—National Institute on Drug Abuse • RBA—Risk Behavior Assessment
Adolescents and young adults are disproportionately affected by HIV and other sexually transmitted infections (STIs). Youth aged 15 to 24 years represent approximately 25% of sexually active persons in the United States but accounted for nearly 50% of new STI cases (9.1 million) in 2000.1 Between 2001 and 2005, HIV/AIDS diagnoses increased >20% in persons aged 13 to 24 years.2 Advances for treating AIDS have slowed mortality.3,4 Still, among persons aged 25 to 34 years, HIV is the sixth leading cause of death among non-Hispanic whites and Hispanics, the third leading cause of death among African Americans, and the leading cause of death in African American women.5
HIV/AIDS and other STIs are increasingly diseases of racial/ethnic minorities and youth.3,6 The National Longitudinal Study of Adolescent Health, which sampled >13000 young adults, found that the rate of HIV infection in African Americans was 4.9 cases per 1000 persons, compared with 0.22 cases per 1000 in other racial/ethnic groups.7,8 The most recent statistics compiled by the Centers for Disease Control and Prevention indicate that more than three quarters of persons <25 years diagnosed with HIV/AIDS are African American or Hispanic.9 Young minority females are at particular risk. African American and Hispanic females account for approximately 80% of HIV/AIDS diagnoses in females aged 13 to 24 years.9 Minorities have greater exposure to risk factors than do other groups, including low socioeconomic status, urban living, substance abuse, and limited access to health care.10–14
Minorities are also overrepresented in the juvenile justice system, where HIV/STI risk behaviors are prevalent.15–20 Detained youth report more risk behaviors and initiate them at younger ages than do youth in the community.21 Detained youth are likely to be at continued risk for HIV infection as they age. Adults in prison have higher rates of HIV/STI risk behaviors22–26 and HIV infection (1.8%) than the general population (0.2%).27 Sound public policy and effective interventions require data on the developmental course of HIV/STI risk behaviors. Because youth are detained for an average of only 2 weeks,15 their behaviors place persons in the community at risk.
There are, however, few comprehensive studies of HIV/STI risk behaviors in delinquent youth,17,18,21,28–34 and, to our knowledge, no longitudinal studies. Even after expanding our literature review to include "high-risk" youth, such as inner city youth and other impoverished populations, we found only 4 epidemiological studies with follow-up periods >6 months.35–38 Only 1 of these studies collected comprehensive information on HIV/STI sex and drug risk behaviors.36 None of these studies investigated how the development and persistence of HIV/STI risk behaviors differ by gender, race/ethnicity, and age.35–38
To our knowledge, this is the first large-scale longitudinal study of HIV/STI risk behaviors in delinquent youth. Our study has 2 methodologic strengths: a stratified random sample, large enough (n = 724) to generate reliable rates of HIV/STI risk behaviors for key demographic subgroups (eg, females and Hispanics) and comprehensive measures of HIV/STI drug and sex risk behaviors.
In this article, we address 3 questions:
We examine differences according to incarceration status and demographic variables (gender, race/ethnicity, and age).
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
14 years), gender, race/ethnicity (African American, non-Hispanic white, and Hispanic), and legal status (processed as a juvenile or an adult). The CCJTDC is used for pretrial detention and for offenders sentenced for <30 days. Consistent with juvenile detainees nationwide,15 >80% of detainees at CCJTDC were male, and most were racial/ethnic minorities. Additional information on our methods has been published elsewhere.20,21,39
Procedures to Obtain Assent and Consent
This research was approved by the institutional review boards of Northwestern University, the Centers for Disease Control and Prevention, and the US Office of Protection from Research Risks. At the baseline and follow-up interviews, participants signed either an assent form (if they were <18 years) or a consent form (if they were 18 years). The Northwestern University Institutional Review Board and the Centers for Disease Control and Prevention Institutional Review Board waived parental consent, consistent with federal regulations regarding research with minimal risk (45 Code of Federal Regulations [CFR] 46.116[c], 45 CFR 46.116[d], and 45 CFR 46.408[c]).40 We nevertheless tried to contact parents to provide them information and offer an opportunity to decline participation. Despite repeated attempts to contact the parent or guardian, for 43.8% of the participants, none could be found. In lieu of parental consent, an independent participant advocate representing the interests of the participants oversaw youth assent. Federal regulations allow for a participant advocate if parental consent is not feasible (45 CFR 46.116[d]).40
Participants
Collection of the baseline HIV/STI data began when funding became available, from February 1997 through June 1998. Among the 1052 youth sampled during this period, 3.9% (n = 41) refused to participate.21,41 There were no significant differences in refusal rates according to gender, race/ethnicity, or age. Fourteen participants did not complete the HIV/STI questions because of the interviewer's error. One participant was released from detention before finishing the interview; 196 participants left the detention center while we were locating their caretakers to obtain consent or before we could schedule an interview. The final number of youth who received the HIV/STI interview was 800; of these, 769 (96.1%) were interviewed at follow-up; 12 (1.5%) died before the follow-up; 3 (0.4%) withdrew from the study; and 16 (2.0%) were lost to follow-up. Time to follow-up was between 2.9 and 7.9 years (mean [SD] follow-up: 3.3 [0.6] years; median follow-up: 3.1 years).
Forty-five of the 769 participants were excluded from our analyses: 5 (0.7%) did not receive the HIV/STI risk behavior assessment at follow-up (because of time constraints or the interviewer's error); and 40 (5.2%) received their follow-up interview >4.5 years after their baseline interview. We chose 4.5 years for the cutoff, because, in this high-risk and highly mobile sample, participants can be difficult to track; using a stricter cutoff would restrict the generalizability of the sample. To ensure that our cutoff did not bias the findings, we compared the demographic characteristics (gender, race/ethnicity, and age) of participants who were interviewed between 3.5 and 4.5 years (n = 81; 11% of the sample) after baseline with those interviewed within 3.5 years after baseline; there were no significant differences. In addition, we examined whether our findings were affected by including these participants. We repeated all of the analyses using only participants interviewed within 3.5 years; the findings were substantially the same.
The final sample in these analyses (n = 724) included 316 females and 408 males; there were 393 African American participants, 198 Hispanic participants, 131 non-Hispanic white participants, and 2 participants who self-identified their race as "other." At baseline, 113 youth were processed as adults, and 611 were processed as juveniles. The median length of stay at CCJTDC was 15 days (range: 1–686 days; mean [SD] days: 40.7 [75.3]). At baseline, participants were aged 10 to 18 years (mean [SD] age: 14.8 [1.4]; median age: 15). At follow-up, participants were aged 13 to 22 years (mean [SD] age: 18.1 [1.4]; median age: 18). Time to follow-up was 2.9 to 4.5 years (mean [SD] time to follow-up: 3.2 [0.3] years; median: 3.1 years).
Procedures for Data Collection
At the baseline interview, face-to-face, structured interviews were conducted at the detention center in a private area; most interviews took place within 2 days of intake.20,21,39 At the follow-up, the same participants were interviewed, irrespective of where they lived. Participants were interviewed in the community (66.2%), at correctional facilities (26.2%), at residential placement facilities (2.5%), or by telephone if they lived in a community >2 hours away (5.1%). Baseline and follow-up interviews took 2 to 4 hours to complete.20,21,39 We used both male and female interviewers; female participants were interviewed only by female interviewers. Most interviewers had advanced degrees in psychology or an associated field and had experience interviewing at-risk youth. All of the interviewers were trained for 1 month by 1 of the authors (Dr Abram) and other supervisory staff. One third of the interviewers were fluent in Spanish.20,21,39
Measures
We examined behaviors associated with increased risk for HIV/STI, including sex risk behaviors and injection risk behaviors (sharing needles or "works" for drug injection, piercings, or tattoos).42–46 We also examined antecedents to HIV/STI risk behaviors, such as alcohol and other drug use, because they may indirectly lead to HIV/STIs by increasing high-risk sexual behaviors.
HIV/STI risk behaviors were assessed using the National Institute on Drug Abuse (NIDA) Risk Behavior Assessment (RBA).47 Although designed for adults, we chose the RBA because instruments designed for adolescents and young adults did not assess the breadth, frequency, and severity of HIV/AIDS risk behaviors common in our sample. A report issued by the Substance Abuse and Mental Health Services Administration recommends the RBA for the comprehensive assessment of HIV/AIDS risk behaviors among drug-using adolescents.48 The RBA is a reliable and valid measure of drug and sex risk behaviors.47,49,50 We supplemented the RBA with items from the Adolescent Health Survey from NIDA's Study of Street Youth at Risk for AIDS51 and Yale's AIDS Risk Inventory.52 Experts reviewed our measure at baseline, and we pilot-tested 58 participants. At baseline, lifetime drug use was assessed using screen items (1 for each substance) from the Diagnostic Interview Schedule 2.3.53 At baseline and follow-up, recency and frequency of drug use were assessed using NIDA's RBA.
Missing Data
Missing Cases
To assess the effect of attrition on generalizability, we compared participants who provided follow-up data with those who did not on the following variables: demographic characteristics (gender, race/ethnicity, and age) and HIV/STI risk behaviors reported at baseline. There were no significant differences except those who died were more likely to be male (P < .05), and those lost to follow-up were more likely to be non-Hispanic white or Hispanic (P < .05) and were less likely to have had sex with >1 partner (P < .05). Potential bias from demographic differences in attrition was adjusted by weighting the statistical analyses by sampling strata (see "Statistical Analysis" in "Methods").
Missing Data From Interviews Conducted by Telephone
Because telephone interviews needed to be shorter than face-to-face interviews, they are missing the following variables at follow-up (n = 37; 5.1%): use of specific drugs, types of sex with a high-risk partner, sex and unprotected sex while drunk or high, and trading sex and drugs. Comparing participants interviewed by telephone with those interviewed face-to-face revealed the following: (1) no significant demographic differences (gender, race/ethnicity, or age); (2) no significant differences in the prevalence of HIV/STI risk behaviors reported at baseline; and (3) no significant differences in the prevalence of other HIV/STI risk behaviors reported at follow-up.
Independent Variables
We compared HIV/STI risk behaviors by gender, race/ethnicity, and age. We also examined incarceration status since baseline. For behaviors assessed "since the last interview," participants were considered incarcerated if they self-reported that they had been "mostly in correctional facilities" since the baseline interview (21.4% of sample; 126 males and 29 females). For behaviors assessed "in the past 3 months" or less, participants were considered incarcerated if they self-reported that they had been "mostly in correctional facilities in the past 3 months" (23.1% of sample; 138 males and 29 females).
Statistical Analysis
All of the data were weighted to reflect the population at the CCJTDC. Because selected strata were oversampled, we used sample weights, based on CCJTDC's population, to estimate descriptive statistics and model parameters that reflect CCJTDC's population. Taylor series linearization was used to estimate SEs.54,55 Only statistically significant findings with P < .05 are noted in the text.
Changes in the prevalence of behaviors between the baseline and follow-up interviews were assessed using paired differences with an adjusted Wald F statistic.56 Logistic regression was used to assess demographic differences in the prevalence (Tables 1 and 2), development (Tables 3 and 4), and persistence (Tables 5 and 6) of individual risk behaviors.57 The independent variables in the regression models were incarceration status only (Tables 1 and 2), incarceration status and gender (Tables 2, 3, and 5), incarceration status and race/ethnicity (Tables 4 and 6), and incarceration status and age. We tested for differences between specific groups (eg, African American versus Hispanic) only when the overall model was significant at the P < .05 level. We controlled for incarceration status in all of the analyses by either computing separate prevalence rates for those incarcerated and those in the community or including incarceration status in logistic regression models.
|
|
|
|
|
|
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Table 1 also shows differences according to incarceration status. Among males in the community, most behaviors were more prevalent at follow-up. Only 2 behaviors, recent and frequent use of marijuana, were significantly less prevalent. Among incarcerated males, the opposite pattern prevailed; many behaviors were less prevalent at follow-up. Only 1 behavior, oral sex with a high-risk partner, was significantly more prevalent.
Females
Table 2 shows that the prevalence of the following behaviors increased at follow-up: sexual activity, vaginal sex, recent unprotected vaginal sex, oral sex, recent unprotected oral sex, unprotected sex while drunk or high, and trading sex and drugs. In contrast, multiple sex partners (>1 in the past 3 months), use of alcohol, use of marijuana, and recent use of marijuana were less prevalent at follow-up.
Table 2 also reports differences according to incarceration status. Among females in the community, many behaviors were more prevalent at follow-up. Only 4 behaviors, multiple sex partners (>1 and >3 in the past 3 months), use of marijuana, and recent use of marijuana, were significantly less prevalent. Among incarcerated females, the opposite pattern prevailed; only 1 behavior, oral sex, was significantly more prevalent.
Prevalence at Follow-up
Males
Table 1 shows that at follow-up, nearly all of the males were sexually active, had vaginal sex, and used alcohol and marijuana. The prevalence of unprotected sex in the past 3 months was also notable: nearly half had recent unprotected vaginal sex, more than one third had recent unprotected oral sex, 7% had recent unprotected anal sex, and more than half had unprotected sex while drunk or high.
Table 1 also shows that, at follow-up, more males in the community than incarcerated males reported the following behaviors: sexually active, multiple sex partners (>1 and >3 in the past 3 months), vaginal sex, recent unprotected vaginal sex, recent unprotected oral sex, recent unprotected anal sex, traded sex and drugs, recent use of alcohol, frequent use of alcohol, use of marijuana, recent use of marijuana, and frequent use of marijuana. In contrast, more incarcerated males engaged in anal sex with a high-risk partner than those in the community.
Females
Table 2 shows that, at follow-up, nearly all of the females were sexually active and had vaginal sex; more than four fifths used alcohol and marijuana. The prevalence of unprotected sex in the past 3 months was also notable: nearly two thirds had recent unprotected vaginal sex, more than one third had recent unprotected oral sex, nearly 4% had recent unprotected anal sex, and nearly half had unprotected sex while drunk or high.
Table 2 also shows that, at follow-up, more females in the community than incarcerated females reported the following behaviors: sexually active, vaginal sex, recent unprotected vaginal sex, recent use of alcohol, frequent use of alcohol, recent use of marijuana, and frequent use of marijuana.
Gender Differences
There were many gender differences, which are reported in Table 2. More males than females reported the following behaviors: multiple sex partners (>1 and >3 in the past 3 months), vaginal sex, oral sex with a high-risk partner, anal sex (receptive and/or insertive), anal sex with a high-risk partner, sex while drunk or high, frequent use of alcohol, and use of marijuana. In contrast, more females than males reported the following behaviors: receptive anal sex and injection drugs.
Age Differences (Data Not Shown)
More youth 18 years (n = 502) than youth <18 years (n = 222) reported the following behaviors: recent unprotected anal sex (prevalence rate: 2.0% vs 7.9%; P < .05), recent use of alcohol (prevalence rate: 28.9% vs 52.9%; P < .05), and tattooing (prevalence rate: 19.8% vs 50.1%; P < .001).
Development of HIV/STI Risk Behaviors
Table 3 shows that many participants who had not reported risk behaviors at baseline had developed them by follow-up. For example, among those who had not previously reported unprotected vaginal sex, nearly 40% of males and more than half of females reported such behavior at follow-up. More than 40% of males and nearly 40% of females began engaging in unprotected sex while drunk or high at follow-up.
Gender Differences
Table 3 shows that, at follow-up, more males than females had begun engaging in the following behaviors: multiple sex partners (>1 and >3 in the past 3 months), vaginal sex with a high-risk partner, anal sex (receptive and/or insertive), anal sex with a high-risk partner, use of alcohol, and frequent use of alcohol. In contrast, at follow-up more females than males had begun engaging in receptive anal sex, use of substances other than alcohol and marijuana, and injection drugs.
Racial/Ethnic Differences
Table 4 shows that there were few racial and ethnic differences in the development of risk behaviors; most involved the initiation of substance use among non-Hispanic whites and Hispanics. More non-Hispanic whites and Hispanics than African Americans had begun engaging in recent use of alcohol, use of substances other than alcohol or marijuana, recent use of substances other than alcohol or marijuana, and frequent use of substances other than alcohol or marijuana. More non-Hispanic whites than African Americans had begun having recent unprotected oral sex. More Hispanics than non-Hispanic whites had begun having recent unprotected anal sex.
Persistence of HIV/STI Risk Behaviors
Table 5 shows that persistence of sex and drug risk behaviors was common for both males and females. For example, among youth who had engaged in unprotected vaginal sex at baseline, more than half of males and nearly 70% of females persisted in this behavior at follow-up. More than three quarters of males and nearly 60% of females persisted in unprotected sex while drunk or high at follow-up. More than 70% of males and nearly 70% of females persisted in using substances other than alcohol and marijuana at follow-up.
Gender Differences
There were few gender differences in the tendency for risk behaviors to persist. At follow-up, more males than females had persisted in the following sex risk behaviors: multiple sex partners (>1 in the past 3 months), sex while drunk or high, and use of marijuana.
Racial/Ethnic Differences
Table 6 shows that there were few racial or ethnic differences in the tendency for behaviors to persist. More non-Hispanic whites and Hispanics than African Americans persisted in the following behaviors: oral sex and use of substances other than alcohol and marijuana. More Hispanics than African Americans persisted in recent use of alcohol and tattooing.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
1 unprotected sexual risk behavior at follow-up. More than 60% had engaged in 10 risk behaviors at their baseline interview,21 and nearly two thirds of them persisted in 10 risk behaviors at follow-up. Irrespective of gender, race/ethnicity, or age, sex risk behaviors were prevalent and likely to persist and develop. At follow-up, more than one third of males and one fourth of females reported engaging in vaginal sex with a high-risk partner. At baseline, more than one third of males and more than half of females reported engaging in recent unprotected vaginal sex. At follow-up, more than half of youth persisted in this behavior, and more than one third developed this behavior. These findings underscore the importance of providing early HIV/STI interventions, continued outreach, and long-term interventions that focus on sex risk behaviors.
Injection risk behaviors were uncommon at baseline and at follow-up. However, our findings on risk behaviors related to noninjection drug use are of great concern. One half of our participants had a substance use disorder at baseline,20 and >80% of youth reported using alcohol and marijuana at follow-up. At baseline, more than one third of participants engaged in unprotected sex while drunk or high. At follow-up, approximately three fifths of the youth persisted in this behavior, and two fifths of the youth developed this behavior. Substance abuse can lead to high-risk sexual behaviors by affecting decision-making, compromising judgment, decreasing the likelihood of condom use, and increasing the likelihood of sex-for-drug exchanges and injection drug use.58–72 Yet, research on noninjection drug use and HIV/STIs has lagged, considering its importance in the current HIV/STI epidemic.
Taken together, these findings mirror the changing patterns of transmission of HIV/STIs in the general population. In the early stages of the HIV/STI epidemic, the most common patterns of transmission were injection drug use (approximately one quarter of AIDS cases) and male-to-male sex (two thirds of AIDS cases).73 Male-to-female sexual contact at that time accounted for only 4% of AIDS cases73; it now accounts for one third of reported HIV/AIDS cases.74
We found a number of gender differences, even after adjusting for incarceration status. Compared with females, males had higher prevalence rates of many HIV/STI risk behaviors and were more likely to persist in some behaviors and develop new ones. Yet, injection risk behaviors were more prevalent among females than males and were also more likely to develop and persist. Our findings emphasize the need to develop interventions tailored to specific patterns of risk and transmission.74 For example, nearly 80% of females contract HIV/AIDS from vaginal sex compared with only 16% of males.74 Gender-specific interventions are especially important now that females compose nearly 20% of juvenile detainees15 and 26% of AIDS cases2 (compared with 7% during the early years of the epidemic73).
Incarceration status was an important variable. Among youth in the community, many behaviors were more prevalent at follow-up than at baseline. Among incarcerated youth, the opposite pattern prevailed. Our findings add to the growing debate on the role of incarceration in the HIV/STI epidemic. One view is that correctional facilities are "breeding grounds"75 for HIV/AIDS. Others suggest that the disproportionately high prevalence of HIV/AIDS in correctional facilities occurs because behaviors that put persons at risk for HIV/AIDS (eg, drug use, prostitution) also put them at risk for incarceration.75
Although risk behaviors may be less common in correctional facilities than in the community, they may carry substantially greater risk. For example, to prevent HIV/STI transmission, prisoners may use plastic gloves and hand lotion instead of lubricated condoms.24,76 Similarly, to inject drugs, inmates may share needles or "works" or use dirty equipment if sterilization is unavailable.24,76–78 Moreover, the probability of infection is also higher, because more persons in prisons than in the community are infected with HIV.27
Overall, there were few racial and ethnic differences in patterns of HIV/STI risk behaviors; most involved the initiation and persistence of substance use among non-Hispanic whites and Hispanics. There were surprisingly few racial/ethnic differences in sex risk behaviors. Yet, because of the disproportionate numbers of African Americans who cycle through correctional facilities,79,80 the pediatrics community must focus on implementing culturally appropriate interventions for African American youth and young adults. More than any other racial/ethnic group, African Americans are disproportionately incarcerated and affected by HIV/AIDS.81 Although African Americans compose only 13% of the general population,82 and juvenile crime rates are relatively similar across race/ethnicity,83 African Americans compose about 40% of incarcerated youth and adults15,80,84 and 49% of new cases of HIV/AIDS (52% among men and 66% among women).2
The prevalence of HIV/STI risk behaviors in our sample is similar to that of other high-risk youth: those living on the street, drug users, and those living in the inner city.85–90 Primary risk reduction interventions may not reach these youth. Although most schools now provide HIV/STI education,91,92 youth who are frequently truant, such as delinquent and homeless youth, are unlikely to receive school-based interventions.93 Moreover, delinquent youth are overrepresented in groups that are uninsured (including the poor,94 youth living in central cities,95 and older adolescents95–98), reducing the likelihood that they will have a primary care physician from whom they could receive primary interventions.99–101 Public clinics and emergency services are often the primary source of health care for high-risk youth.102,103 As recently recommended by the American Academy of Pediatrics,104 public clinics should integrate HIV prevention, especially sex education and substance abuse treatment, into primary medical care. HIV/STI interventions should also be provided in detention centers and in juvenile courts that, based on recent statistics, could reach as many as 1.6 million youth annually.15
Limitations
It was not possible to assess actual HIV/STI risk behaviors, such as having unprotected sex with an infected partner and sharing injection/piercing equipment with an infected partner. Moreover, it was not feasible to obtain biological outcome measures, such as HIV or STI tests. Thus, our measures of HIV/STI risk are proximal. Findings might have been slightly different had follow-up data been available for participants who died, withdrew from the study, or were lost to follow-up. We examined HIV/STI risk behaviors during 2 periods of our subjects' lives. Our analyses do not address causal mechanisms underlying HIV/STI risk. Our findings, drawn from 1 site, may pertain only to youth who were detained during adolescence in urban detention centers of similar demographic composition. Our sample (though larger than most previous investigations) limited our analyses of demographic subgroups that are less common in detention centers, such as young, non-Hispanic white females. Finally, the data are subject to the limitations of self-reporting. Participants may have underreported some behaviors and exaggerated others.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Many more people than the authors contributed to this project. This study could not have been accomplished without the advice of Ann Hohmann, PhD, Kimberly Hoagwood, PhD, Heather Ringeisen, PhD, and Eugene Griffin, PhD. Jacques Normand, PhD, Helen Cesari, MS, Richard Needle, PhD, Robert Booth, PhD, David Huizinga, PhD, and David Ostrow, MD, PhD, generously offered their expertise in developing our instruments. David Stoff, PhD, Grayson Norquist, MD, and Delores Parron, PhD, provided support and encouragement. Celia Fisher, PhD, guided our human subject procedures. We thank Frank Palella, Jr, MD, and the anonymous reviewers for their many helpful comments. We also thank all of the project staff, especially Amy Mericle, PhD, for instrumentation development and project leadership and Lynda Carey, MA, for her superb data management. We thank Kate Elkington, PhD, for literature review. We also greatly appreciate the cooperation of everyone working in the Cook County and State of Illinois systems. Without their cooperation, this study would not have been possible. Finally, we thank the participants for their time and willingness to participate and our field interviewers for their commitment to excellence.
|
FOOTNOTES |
|---|
Address correspondence to Linda A. Teplin, PhD, Department of Psychiatry and Behavioral Sciences, Feinberg School of Medicine, Northwestern University, 710 N Lakeshore Dr, Suite 900, Chicago, IL 60611. E-mail: psycho-legal{at}northwestern.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||
