Abstract
Background and Objectives
The longitudinal risk for human immunodeficiency virus (HIV) infection following adolescent substance treatment is not known. Therefore, it is not known if adolescent substance treatment should include HIV prevention interventions. To address this important research gap, this study evaluates the longitudinal prevalence and predictors of injection drug use (IDU) and sex risk behaviors among adolescents in substance treatment.
Methods
Participants were 260 adolescents (13-18 years) in substance treatment and 201 community control adolescents (11-19 years). Participants were assessed at baseline and follow-up (mean time between assessments=6.9 years for the clinical sample and 5.6 years for the community control sample). Outcomes included self-report lifetime history of IDU, number of lifetime sex partners and frequency of unprotected sexual intercourse.
Results
At baseline, 7.5% of the clinical sample, compared to 1.0% of the community control sample had a lifetime history of IDU (χ12=10.53, p=0.001). At follow-up, 17.4% of the clinical sample compared to 0% of the community control sample had a lifetime history of IDU (χ12=26.61, p=0.0005). The number of baseline substance use disorders and onset age of marijuana use significantly predicted the presence of lifetime IDU at follow-up, after adjusting for baseline age, race, and sex. The clinical sample reported more lifetime sex partners and more frequent unprotected sex than the community control sample at baseline and follow-up.
Conclusions
Many adolescents in substance treatment develop IDU and report persistent risky sex. Effective risk reduction interventions for adolescents in substance treatment are needed that address both IDU and risky sex.
Keywords: Adolescent, Injection Drug Use, HIV, Substance treatment
Background and Objectives
Human immunodeficiency virus (HIV) infection is an important public health problem for youth because persons under 25 years of age account for 20% of all new infections.1 Two main risk factors for contracting HIV are injection drug use (IDU) and risky sex.1 IDU accounts for 8-13% of all new HIV infections while risky sex accounts for 89-95%.1
It is not clear how many adolescents in substance treatment have IDU. However, late adolescence and young adulthood is an important time for the development of IDU because: 1) among adults with current IDU, the mean age of first IDU is 20 years; 2) adolescent-onset injection drug users (<21 years) have riskier IDU behaviors (e.g. more receptive needle sharing) than adult-onset injection drug users; and 3) early IDU initiation (<21 years) compared to later initiation is associated with a 40% increase in the odds of being HIV seropositive.2-5
Risky sexual behavior is another important risk factor for HIV infection. Having sex with multiple partners and not using condoms are especially common in adolescents who use substances.6-16 These behaviors, particularly if they persist into young adulthood, represent additional risk for developing and transmitting HIV.
Despite evidence that IDU frequently develops during late adolescence and young adulthood and that adolescents in substance treatment frequently have risky sex, only half of adolescent substance treatment programs offer HIV risk assessment (56%) or HIV risk reduction interventions (57%).17 One possible explanation for the lack of HIV interventions in adolescent substance treatment is the lack of data describing the longitudinal outcome and predictors of HIV risk behaviors among these adolescents.
The existing literature reports on few longitudinal samples of HIV risk among adolescents in substance treatment. We could find no studies assessing baseline prevalence of IDU and frequency of IDU onset for adolescents in substance treatment. With respect to risky sex, behaviors such as non-condom use and having multiple sex partners persist for at least 12 months following adolescent substance treatment, and conduct disorder predicts persistence of these behaviors.18,19 However, we could not find studies that followed a sample longer than 12 months.
More is known about the longitudinal course of HIV risk in non-treatment seeking samples. Predictors of IDU from adolescent samples are not known but predictors of future IDU derived from non-treatment seeking samples of adult drug users include: early onset substance use, high school drop out, high intensity of substance use, past history of IDU, peer influence to inject and risky sex.20-24 Studies of non-treatment seeking adolescents show the following predictors of risky sex: age onset of sexual intercourse, age onset of substance use, association with deviant peers, decreased parental monitoring, depressive symptoms and baseline risky sexual behavior.25-32 However, most of these predictor variables have not been evaluated in adolescents undergoing substance treatment.
In summary, there is little information on the baseline prevalence of IDU and the long-term HIV risk and predictors of HIV risk for adolescents in substance treatment. This information is crucial to inform whether or not HIV risk should even be addressed in adolescent substance treatment. In fact, some researchers have proposed that adolescent substance treatment alone might reduce HIV risk.33 Longitudinal data would also inform the important clinical questions of which HIV risk behaviors should be addressed and which adolescents in substance treatment are most in need of HIV risk reduction interventions. To our knowledge, this study is the first long-term (greater than 12 months) evaluation of HIV risk in adolescents enrolled in substance treatment and includes a large clinical sample and a matched community control group. It also assesses HIV risk due to both IDU and risky sex and assesses not just risky sexual behaviors but attitudes and social norms. Therefore, the findings from this study represent a unique opportunity to address the important research gaps outlined above.
Methods
Baseline participants
Participants were recruited from 1997 to 2001 and were part of a genetics study of adolescent-onset substance use disorders in patients, families, twins and adoptees.34 The clinical sample (n = 260) consists of adolescents admitted to a Denver area substance treatment program operated by the Division of Substance Dependence at the University of Colorado. The treatment program offers both outpatient and residential treatment. Most patients in this treatment program are referred by juvenile justice or social services for serious conduct and substance use disorders. Inclusion criteria for the study were: 1) age between 13 and 19 years; 2) ability to understand and provide written, informed parental consent and minor assent, if under 18 years old, or individual consent if 18 years or older; 3) sufficient English ability to complete the assessments; and 4) enrolled in the substance treatment program. Exclusion criteria were: 1) imminent danger to self or others, 2) current psychotic symptoms and 3) no biological full sibling between the ages of 12 and 25 years.
The community control sample (n = 201) consisted of adolescents from 11 to 19 years old. Community participants were matched to 2/3 of male clinical participants and all of the female clinical participants by age (within one year), race/ethnicity, gender and ZIP code of residence. Community participants and their families were recruited through a private research firm by randomly querying public phone lists from specific ZIP codes (for a description of the sample, see Miles et al., 1998).35 Inclusion and exclusion criteria for the community participants were the same as for clinical participants, with the exception that they were not enrolled in the previously mentioned substance treatment program, and they could be under the age of 13 years.
Participants followed at time 2
The follow-up interview for the clinical sample occurred a mean of 6.9 (SD=1.5) years later (range 5-10 years). Community controls had a similar follow-up time with a mean of 5.6 (SD=0.9) years (range 4-10 years). The response rate for the clinical sample was 75.0%. Reasons for non-participation were: refused (n=18, 6.92%), dead (n=5, 1.92%) and other (e.g. missing data, unable to locate) (n=42, 16.15%). The response rate for the community controls was 68.2%. Reasons for non-participation were: refused (n=40, 19.90%), dead (n=0, 0%) and other (e.g. missing data, unable to locate) (n=24, 11.94%).
Measures
Demographic information
Demographic information such as age, ethnicity and gender was collected with a standard pen-and-paper questionnaire.
Diagnosis of conduct disorder and substance use disorder
Conduct disorder symptoms were assessed with the Diagnostic Interview Schedule for Children (DISC) based on the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) criteria.36.37 DSM-IV substance use disorders were diagnosed with the Composite International Diagnostic Interview-Substance Abuse Module (CIDI-SAM).38,39 Substance use disorders in nine categories (alcohol, amphetamines, cannabis, cocaine, hallucinogens, inhalants, opioids, PCP, and sedatives) were summed to create an aggregate measure of the number of substance abuse and dependence diagnoses. The CIDI-SAM was also used to obtain information about substance use, such as onset age of using marijuana more than five times and lifetime history of IDU. These structured instruments have been shown to be valid assessments of conduct disorder and substance use disorders in similar samples.40
Assessment of attitudes, social norms and behaviors related to HIV sex risk
Attitudes and behaviors related to risky sex were assessed at baseline and follow-up with a pen-and-paper self-report assessment. The following questions assessed attitudes and social norms: “Think of all your friends of the same sex you are. How many of them have had sex (‘gone all the way’) (almost none, some of them, most of them, all of them)?”; “When people your age have sex, do they usually use some kind of protection against disease or pregnancy (almost all do, most do, some do, almost none do)?”; “How much peer pressure is there on people your age to have sex (none, a little, a fair amount, a lot)?”; “People my age are just too young to have sex (strongly agree, agree, disagree, strongly disagree)”; and “It's better not to have sex rather than to risk getting pregnant or a disease (strongly agree, agree, disagree, strongly disagree).” At follow-up, these questions about attitudes and social norms were only administered to participants who were not currently married or living with a partner because some of the questions were less applicable to those participants (e.g. People my age are just too young to have sex).
All participants, regardless of whether or not they were married or living with a partner, completed a self-administered pen-and-paper assessment about risky sexual behaviors at baseline and follow-up. Participants answered the question, “When you had sex/if you had sex before married, did you make sure that some kind of protection was used (almost always, most of the time, about half of the time, some of the time, hardly ever, and never)?” An unprotected sex variable was created by dichotomizing responses from this item into almost always versus not almost always uses protection and never had sex was combined with the almost always category. Participants also recorded the age at which they first had sex and the number of lifetime sex partners. Due to outlying values for number of lifetime sex partners, 22 observations 2.5 standard deviations above the mean were truncated.
Three HIV risk behaviors (IDU, multiple sex partners and unprotected sex) were assessed at follow-up. The variable for multiple sex partners was dichotomized as 20 or fewer sex partners versus 21 or more sex partners. The cutoff of 20 sex partners corresponds to approximately one standard deviation above the mean number of partners in large, community samples and has been used as a cutoff for multiple lifetime partners in other studies.41,42 The variable measuring unprotected sex at follow-up was dichotomized as it was at baseline. Lifetime history of IDU at follow-up included intravenous, intramuscular and subcutaneous injection and was obtained from the computerized CIDI-SAM.
Procedures
Approval from the Colorado Multiple Institutional Review Board was obtained before beginning the study. Written, informed consent was obtained from all participants and parents/guardians of minor participants. The baseline and follow-up interviews were conducted confidentially, and participants were told that information would not be shared with parents, guardians or treatment providers.
Statistical Analyses
Data were edited and analyzed in SPSS version 19.0.43 Pearson chi-square analyses and independent t-tests or Mann-Whitney U tests (when continuous data were not normally distributed) were used to compare the clinical and community control samples on baseline and follow-up demographic data, conduct disorder diagnosis, substance use disorder diagnoses, and HIV risk attitudes and behaviors.
To examine baseline predictors of follow-up HIV risk (IDU, multiple sex partners and unprotected sex), separate multiple logistic regressions were completed with each predictor variable (onset age of sex, onset age of marijuana use more than five times, number of conduct disorder symptoms and number of substance use disorder diagnoses, not including nicotine), while controlling for baseline age, sex and race. Candidate predictor variables were chosen based on previous literature showing them to be HIV risk factors in non-clinical adolescent samples.22,28,44 Next, significant (p < 0.05) predictor variables from the separate models were combined in multiple logistic regressions using a stepwise procedure, while adjusting for baseline age, sex and race. Each significant variable was evaluated in a model with another significant variable. When two variables remained significant in a model, a third significant predictor was added, etc. Final models evaluated whether eliminating non-significant covariates appreciably changed significance of the predictors. These multiple logistic regressions were only completed in the clinical sample because a primary goal of the study was to provide predictors that clinicians could use to identify adolescent patients most in need of HIV risk reduction interventions.
Results
Sample Description
The comparison (Table 1) of baseline demographic and clinical characteristics of the two samples revealed that, in general, the clinical sample had a higher prevalence of conduct disorder and substance use disorders. In addition, the clinical sample had a significantly younger age of marijuana use onset.
Table 1. Baseline characteristics of the clinical and community control samples.
| Variable | Clinical N = 260 % (n) or mean (SD) |
Community Controls N = 201 % (n) or mean (SD) |
Statistic | p value |
|---|---|---|---|---|
|
| ||||
| Gender | ||||
| Female | 13.5 (35) | 17.4 (35) | χ12=1.37 | 0.241 |
| Male | 86.5 (225) | 82.6 (166) | ||
|
| ||||
| Hispanic | 35.0 (91) | 34.8 (70) | χ12=0.002 | 0.969 |
|
| ||||
| Race | ||||
| White | 53.5 (139) | 54.7 (110) | χ22=0.16 | 0.923 |
| Black | 8.5 (22) | 9.0 (18) | ||
| Other (American Indian/Alaska Native, Asian and more than one race) | 38.1 (99) | 36.3 (73) | ||
|
| ||||
| Age at baseline interview (years) | 15.7 (1.2) | 15.7 (1.5) | t368.79 =0.02 | 0.988 |
|
| ||||
| Conduct disorder | 86.9 (226) | 24.9 (50) | χ12=181.65 | 0.0005 |
|
| ||||
| Number of conduct disorder symptoms | 5.8 (2.8) | 1.7 (2.2) | t458.68=-17.86 | 0.0005 |
|
| ||||
| Cannabis use disorder | 90.8 (236) | 15.4 (31) | χ12=264.05 | 0.0005 |
|
| ||||
| Alcohol use disorder | 71.9 (187) | 15.4 (31) | χ12=145.18 | 0.0005 |
|
| ||||
| Nicotine dependence | 59.2 (154) | 7.5 (15) | χ12=130.84 | 0.0005 |
|
| ||||
| Amphetamine use disorder | 18.8 (49) | 1.0 (2) | χ12=36.72 | 0.0005 |
|
| ||||
| Cocaine use disorder | 16.2 (42) | 1.0 (2) | χ12=30.17 | 0.0005 |
|
| ||||
| Opioid use disorder | 6.9 (18) | 0.5 (1) | χ12=11.85 | 0.001 |
|
| ||||
| Number of substance use disorders (not including nicotine) | 2.4 (1.5) | 0.4 (0.8) | U=4583.50 | 0.0005 |
|
| ||||
| Onset age of marijuana use (years) | 11.9 (2.0) (n=256) |
13.3 (2.3) (n=63) |
t317=4.65 | 0.0005 |
Note: Sample sizes vary because of missing data.
Baseline HIV Risk
The comparison of the baseline HIV risk factors for the clinical and community control samples revealed that, in general, the clinical sample had significantly riskier attitudes, social norms and behaviors than the community sample (Table 2). Important differences between groups were seen in the proportion with a lifetime history of IDU, the proportion ever having sex, frequency of unprotected sex, number of lifetime sex partners and age of sex initiation.
Table 2. Baseline risk factors for HIV infection comparing clinical and community control samples.
| Item | Clinical N = 260 % (n) or mean (SD) |
Community Controls N = 201 % (n) or mean (SD) |
Statistic | p value |
|---|---|---|---|---|
| Lifetime history of IDU | 7.5 (19) (n=255) |
1.0 (2) (n=199) |
χ12=10.53 | 0.001 |
| Proportion of friends having sex (all and most) | 88.7 (227) (n=256) |
28.0 (54) (n=193) |
χ12=173.10 | 0.0005 |
| People my age are just too young to have sex (strongly disagree and disagree) | 82.0 (210) (n=256) |
44.9 (89) (n=198) |
χ12=68.28 | 0.0005 |
| It's better not to have sex rather than to risk getting pregnant or a disease (strongly disagree and disagree) | 47.1 (120) (n=255) |
22.3 (44) (n=197) |
χ12=29.39 | 0.0005 |
| When people your age have sex, do they usually use some kind of protection (almost none and some) | 33.5 (85) (n=254) |
14.5 (28) (n=193) |
χ12=20.86 | 0.0005 |
| How much peer pressure is there on people your age to have sex (a lot and a fair amount) | 50.4 (129) (n=256) |
50.8 (100) (n=197) |
χ12=0.01 | 0.938 |
| Unprotected sex (any response other than almost always uses protection or no sex) | 49.6 (128) (n=258) |
10.6 (21) (n=198) |
χ12=77.48 | 0.0005 |
| Ever had sex | 92.7 (240) (n=259) |
33.8 (67) (n=198) |
χ12=176.11 | 0.0005 |
| Onset age of sex (years) | 13.1 (1.6) (n=239) |
14.6 (1.6) (n=67) |
U=4103.50 | 0.0005 |
| Number of lifetime sex partners | 6.2 (5.5) (n=256) |
1.3 (3.0) (n=196) |
U=6798.00 | 0.0005 |
Note: Some participants failed to answer all questions; n = the number of participants who responded to the question.
Follow-up HIV Risk
Of the 260 clinical participants, 195 (75%) were re-interviewed. Of 201 community control participants, 137 (68.2%) were re-interviewed. Neither the clinical participants nore the community control participants who completed the follow-up interview differed from those who did not complete based on baseline age, gender, ethnicity, number of lifetime sex partners, lifetime IDU, race or unprotected sex. However, the percentage of white participants in the follow-up clinical group was greater than the percentage of white participants in clinical group that did not complete the follow-up interview (56.4% vs. 44.6%; χ22 = 8.61, p = 0.014).
Table 3 compares the clinical and community samples with respect to HIV risk attitudes, social norms and behaviors at follow-up. Overall, the clinical sample continued to have riskier attitudes, social norms and behaviors than the community control sample.
Table 3. HIV risk at follow-up in clinical and community control samples.
| Item | Clinical N = 195 % (n) or mean (SD) |
Comparison Sample N = 137 % (n) or Mean (SD) |
Statistic | p value |
|---|---|---|---|---|
| Lifetime history of IDU | 17.4 (34) | 0 (0) | χ12=26.61 | 0.0005 |
| Amount of friends having sex† (all and most) | 94.6 (139) (n=147) |
82.7 (86) (n=104) |
χ12=9.24 | 0.002 |
| People my age are just too young to have sex† (strongly disagree and disagree) | 91.8 (135) (n=147) |
88.5 (92) (n=104) |
χ12=0.80 | 0.370 |
| It's better not to have sex rather than to risk getting pregnant or a disease† (strongly disagree and disagree) | 48.3 (71) (n=147) |
41.7 (43) (n=103) |
χ12=1.05 | 0.306 |
| When people your age have sex, do they usually use some kind of protection† (almost none and some) | 46.9 (69) (n=147) |
19.2 (20) (n=104) |
χ12=20.43 | 0.0005 |
| How much peer pressure is there on people your age to have sex† (a lot and a fair amount) | 40.1 (59) (n=147) |
55.8 (58) (n=104) |
χ12=5.98 | 0.014 |
| Unprotected sex (any response other than almost always uses protection or no sex) | 73.7 (140) (n=190) |
41.5 (51) (n=123) |
χ12=32.59 | 0.0005 |
| Number of lifetime sex partners | 16.3 (13.7) (n=186) |
7.4 (10.9) (n=120) |
U=4910.00 | 0.0005 |
Notes: sample sizes vary because some participants failed to answer all questions and n = the number of participants who responded to the question;
These items were only completed by participants who were not currently married or living with a partner at follow-up.
Baseline Predictors of Follow-up HIV Risk
Only the significant results of the multiple logistic regressions to determine baseline predictors of follow-up HIV risk are discussed. All analyses adjust for baseline age, sex and race. With respect to predicting IDU, the separate multiple logistic regressions revealed the following. There is a 21% decrease in the odds of ever injecting drugs at follow-up for every year sexual intercourse is delayed (AOR = 0.79, p = 0.044). The odds of ever injecting drugs are decreased by 27% for every delayed year of marijuana initiation (AOR = 0.73, p = 0.001). For one additional conduct disorder symptom at baseline, the odds of ever injecting drugs at follow-up are increased by about 20% (AOR = 1.20, p = 0.020). One additional substance use disorder diagnosis (not including nicotine) at baseline increases the odds of ever injecting drugs at follow-up by approximately 63% (AOR = 1.63, p = 0.0005). The number of clinical participants with a lifetime history of IDU at follow-up was 34 (17.4%).
Table 4 reports the combined multiple logistic regression model predicting IDU at follow-up. After adjusting for all the variables in the model, the odds of ever injecting drugs at follow-up are decreased by about 20% for every year of delayed marijuana initiation. One additional substance use disorder diagnosis at baseline increases the odds of ever injecting drugs at follow-up by approximately 48%, after adjusting for all the variables in the model.
Table 4. Multiple logistic regression in the clinical sample predicting lifetime IDU at follow-up, after adjusting for baseline age, race and sex.
| Predictor variable | B (SE) | AOR | 95% C.I. for AOR | p value |
|---|---|---|---|---|
| Onset age of marijuana use | -0.22 (0.10) | 0.80 | 0.65, 0.98 | 0.028 |
| Number of substance use diagnoses (not including nicotine) | 0.39 (0.14) | 1.48 | 1.13, 1.94 | 0.004 |
Note: AOR=Adjusted Odds Ratio
With respect to predicting multiple lifetime sex partners, one additional conduct disorder symptom at baseline increases the odds of having 21 or more lifetime sex partners by approximately 20% (AOR = 1.20, p = 0.007).
Conclusions and Scientific Significance
Summary of findings
These results show that, compared to a community control sample, adolescents in substance treatment more frequently engage in IDU both as adolescents and young adults. Most alarming, at follow-up, 17% of the clinical sample had a lifetime history of IDU. The baseline number of substance use disorder diagnoses and the onset age of marijuana use significantly predicted follow-up IDU. The clinical sample also had worse baseline and follow-up attitudes, beliefs and behaviors related to risky sex than the community control sample. The baseline predictor of multiple sex partners was the number of conduct disorder symptoms.
Relationship to previous findings
Concerning onset of IDU, previous research shows that 24-30% of non-treatment seeking adult drug users initiate IDU during a 2-year follow-up.5,24 Our current findings contribute to the understanding of IDU onset by showing that a substantial proportion of adolescents with substance use disorders develop IDU even after undergoing substance treatment.
With respect to risky sexual behaviors, prior research shows that following residential substance treatment, adolescents frequently continue having sex with multiple partners and not using condoms at one year follow-up.18,19,45 Our current findings extend the understanding of risky sex among youth in substance treatment by showing that these behaviors persist for up to 8 years and by evaluating various predictors of persistent sex risk such as onset age of sex, onset age of marijuana use, number of conduct disorder symptoms and number of substance use disorder diagnoses.
Clinical implications
Our findings have several important clinical implications. First, adolescents in substance treatment need IDU prevention. Currently, IDU prevention is not a routine part of adolescent substance treatment likely because, as our data show, few adolescents in substance treatment have IDU. However, our data show that almost one in five adolescents in substance treatment will develop IDU within approximately 8 years. Our data also show that youth with multiple substance use disorders and early marijuana use onset are especially in need of effective IDU prevention interventions.
Second, our data underscore the need for effective interventions to reduce risky sexual behaviors for adolescents in substance treatment. Prior research shows that substance treatment alone may reduce sex risk.33 However, our findings show that substance treatment alone certainly does not eliminate risky sex. In fact, at follow-up 74% of our sample report not using protection almost always, and the mean number of lifetime sex partners was 16. Therefore, even if adolescent substance treatment alone reduces sex risk, adolescents in substance treatment need additional interventions to further reduce their risky sexual behaviors. Unfortunately, there is no proven sex risk behavior intervention for teens in outpatient substance treatment, where 86% of adolescent substance treatment takes place.48
Study limitations
There are several limitations to the current study. First, this study did not use a comprehensive, validated HIV risk assessment that obtained details about frequency of IDU, needle sharing or type of sexual intercourse. Second, there was not a variable specifically measuring condom use. Therefore, caution should be used when drawing conclusions about HIV risk from the unprotected sex variable. Finally, while those lost to follow-up appear similar to those not lost to follow-up at baseline, it is possible that those lost to follow-up developed more severe substance use disorders.
There are several factors which could influence the interpretation of the study findings. For example, while the study accounted for participants who were married or living with a partner, it did not ask participants if they were trying to get pregnant. Furthermore, sexual norms change over time. For example, from 1991 to 2009, the proportion of adolescents reporting ever having sexual intercourse (54 to 46%) and the proportion reporting intercourse with four or more persons (18.7 to 13.8%) decreased while the proportion reporting condom use during last intercourse increased (46.2 to 61.1%).49
Significance and Future Directions
There are several important and new clinical implications from this study. First, many adolescents in substance treatment develop IDU by young adulthood. Second, many adolescents in substance treatment have persistent, risky sexual behaviors. Finally, substance treatment alone does not eliminate these risky behaviors. Future research is needed to develop effective HIV risk reduction interventions that target both IDU and risky sex among youth in substance treatment, especially youth with multiple substance use disorders, early marijuana initiation, and many conduct disorder symptoms.
Acknowledgments
Funding source: NIDA grants 5P60DA011015, 5R01DA012845, 5R01DA21913, T32AA007464
References
- 1.Centers for Disease Control and Prevention. HIV Surveillance Report 2009. Atlanta, GA: Centers for Disease Control and Prevention; 2011. p. 21. [Google Scholar]
- 2.Battjes RJ, Leukfeld CG, Pickens RW. Age at first injection and HIV risk among injection drug users. Am J Drug Alcohol Abuse. 1992;18:263–274. doi: 10.3109/00952999209026066. [DOI] [PubMed] [Google Scholar]
- 3.Booth R, Koester S, Brewster JT, Weibel WW, Fritz RB. Intravenous drug users and AIDS: risk and behaviors. Am J Drug Alcohol Abuse. 1991;17:337–354. doi: 10.3109/00952999109027557. [DOI] [PubMed] [Google Scholar]
- 4.Des Jarlais DC, Friedman SR, Perlis T, Chapman TF, Sotheran JL, Paone D, Monterroso E, Neaigus A. HIV risk behavior and HIV infection among new drug injectors in the era of AIDS in New York City. J Acquir Immune Defic Syndr. 1999;20:67–72. doi: 10.1097/00042560-199901010-00010. [DOI] [PubMed] [Google Scholar]
- 5.Garfein RS, Vlahov D, Galai N, Doherty MC, Nelson KE. Viral infections in short-term injection drug users: the prevalence of the hepatitis C, hepatitis B, human immunodeficiency, and human T-lymphotrophic viruses. Am J Public Health. 1996;86:655–661. doi: 10.2105/ajph.86.5.655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ammon L, Sterling S, Mertens J, Weisner C. Adolescents in private chemical dependency programs: who are most at risk for HIV? J Subst Abuse Treat. 2005;29:39–45. doi: 10.1016/j.jsat.2005.03.003. [DOI] [PubMed] [Google Scholar]
- 7.Deas-Nesmith D, Brayd KT, White R, Campbell S. HIV-risk behaviors in adolescent substance abusers. J Subst Abuse Treat. 1999;16:169–172. doi: 10.1016/s0740-5472(98)00040-3. [DOI] [PubMed] [Google Scholar]
- 8.Elkington KS, Teplin LA, Mericle AA, Welty LJ, Romeron EG, Abram KM. HIV/sexually transmitted infection risk behaviors in delinquent youth with psychiatric disorders: a longitudinal study. J Am Acad Child Adolesc Psychiatry. 2008;47:901–911. doi: 10.1097/CHI.0b013e318179962b. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Elkington KS, Bauermeister JA, Zimmerman MA. Psychological distress, substance use, and HIV/STI risk behaviors among youth. J Youth Adolesc. 2010;39:514–527. doi: 10.1007/s10964-010-9524-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hou SI, Basen-Engquist K. Human immunodeficiency virus risk behavior among white and Asian/Pacific Islander high school students in the United States; does culture make a difference? J Adol Health. 1997;20:68–74. doi: 10.1016/S1054-139X(96)00323-0. [DOI] [PubMed] [Google Scholar]
- 11.Malow RM, Dévieux JG, Jennings T, Lucenko BA, Kalichman SC. Substance-abusing adolescents at varying levels of HIV risk: Psychosocial characteristics, drug use, and sexual behavior. J Subst Abuse Treat. 2001;13:103–117. doi: 10.1016/s0899-3289(01)00069-4. [DOI] [PubMed] [Google Scholar]
- 12.Murphy DA, Brecht ML, Herbeck DM, Huang D. Trajectories of HIV risk behavior from age 15 to 25 in the national longitudinal survey of youth sample. J Youth Adolesc. 2009;38:1226–1239. doi: 10.1007/s10964-008-9323-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Rounds-Bryant JL, Kristiansen PL, Hubbard RL. Drug abuse treatment outcome study of adolescents: a comparison of client characteristics and treatment modalities. Am J Drug Alcohol Abuse. 1999;25:573–591. doi: 10.1081/ada-100101880. [DOI] [PubMed] [Google Scholar]
- 14.Teplin LA, Elkington KS, McClelland GM, Abram KM, Mericle AA, Washburn JJ. Major mental disorder, substance use disorders, comorbidity, and HIV-AIDS risk behaviors in juvenile detainees. Psychiatr Serv. 2005;56:823–828. doi: 10.1176/appi.ps.56.7.823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Thurstone C, Riggs PD, Klein C, Mikulich-Gilbertson SK. A one-session human immunodeficiency virus risk-reduction intervention in adolescents with psychiatric and substance use disorders. J Am Acad Child Adolesc Psychiatry. 2007;46:1179–1186. doi: 10.1097/chi.0b013e31809fe774. [DOI] [PubMed] [Google Scholar]
- 16.Tubman JG, Gil AG, Wagner EF, Artigues H. Patterns of sexual risk behaviors and psychiatric disorders in a community sample of young adults. J Behav Med. 2003;26:473–500. doi: 10.1023/a:1025776102574. [DOI] [PubMed] [Google Scholar]
- 17.Knudsen HK, Oser CB. Availability of HIV-related health services in adolescent substance treatment programs. AIDS Care. 2009;21:1238–1246. doi: 10.1080/09540120902803182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Jainchill N, Yagelka J, Hawke J, De Leon G. Adolescent admissions to residential drug treatment: HIV risk behaviors pre-and post-treatment. Psychol Addict Behav. 1999;13:163–173. [Google Scholar]
- 19.Joshi V, Hser Y, Grella CE, Houlton R. Sex-related HIV risk reduction behavior among adolescents in DATOS-A. J Adolesc Res. 2001;16:642–660. [Google Scholar]
- 20.Des Jarlais DC, Casriel C, Friedman SR, Rosenblum A. AIDS and the transition to illicit drug injection – results of a randomized trial prevention program. Br J Addict. 1992;87:493–498. doi: 10.1111/j.1360-0443.1992.tb01950.x. [DOI] [PubMed] [Google Scholar]
- 21.Fuller CM, Vlahov D, Ompad DC, Shah N, Arria A, Strathdee SA. High-risk behaviors associated with transition from illicit non-injection to injection drug use among adolescent and young adult drug users: a case-control study. Drug Alcohol Depend. 2002;66:189–198. doi: 10.1016/s0376-8716(01)00200-9. [DOI] [PubMed] [Google Scholar]
- 22.Neaigus A, Miller M, Friedman SR, Hagen DL, Sifaneck SJ, Ildefonson G, Des Jarlais DC. Potential risk factors for the transition to injecting among non-injecting heroin users: a comparison of former injectors and never injectors. Addiction. 2001;96:847–860. doi: 10.1046/j.1360-0443.2001.9668476.x. [DOI] [PubMed] [Google Scholar]
- 23.Neaigus A, Gyarmathy VA, Miller M, Frajzyngier V, Friedman S, Des Jarlais DC. Transitions to injecting drug use among noninjecting heroin users: social network influence and individual susceptibility. J Acquir Immune Defic Syndr. 2006;41:493–503. doi: 10.1097/01.qai.0000186391.49205.3b. [DOI] [PubMed] [Google Scholar]
- 24.Van Ameijden EJ, van den Hoek JA, Hartgers C, Coutinho RA. Risk factors for the transition from noninjection to injection drug use and accompanying AIDS risk behavior in a cohort of drug users. Am J Epidemiol. 1994;139:1153–1163. doi: 10.1093/oxfordjournals.aje.a116962. [DOI] [PubMed] [Google Scholar]
- 25.Sherman SG, Fuller CM, Shah N, Ompad DV, Vlahov D, Strathdee SA. Correlates of initiation of injection drug use among young drug users in Baltimore, Maryland: the need for early intervention. J Psychoactive Drugs. 2005;37:437–443. doi: 10.1080/02791072.2005.10399817. [DOI] [PubMed] [Google Scholar]
- 26.Tubman JG, Windle M, Windle RC. The onset and cross-temporal patterning of sexual interourse in middle adolescence: prospective relations with behavioral and emotional problems. Child Dev. 1996;67:327–343. [PubMed] [Google Scholar]
- 27.Tolou-Shams M, Brown LK, Houck C, Lescano CM, Project SHIELD Study Group The association between depressive symptoms, substance use, and HIV risk among youth with an arrest history. J Stud Alcohol Drugs. 2008;69:58–64. doi: 10.15288/jsad.2008.69.58. [DOI] [PubMed] [Google Scholar]
- 28.Capaldi DM, Stoolmiller M, Clark S, Owen LD. Heterosexual risk behaviors in at-risk young men from early adolescence to young adulthood: prevalence, prediction, and association with STD contraction. Dev Psychol. 2002;38:394–406. doi: 10.1037//0012-1649.38.3.394. [DOI] [PubMed] [Google Scholar]
- 29.Guo J, Chung IJ, Hill KG, Hawkins JD, Catalano RF, Abbott RD. Developmental relationships between adolescent substance use and risky behavior in young adulthood. J Adolesc Health. 2002;31:354–362. doi: 10.1016/s1054-139x(02)00402-0. [DOI] [PubMed] [Google Scholar]
- 30.Lehrer JA, Shrier LA, Gortmaker S, Buka S. Depressive Symptoms as a Longitudinal Predictor of Sexual Risk Behaviors Among US Middle and High School Students. Pediatrics. 2006;118:189–200. doi: 10.1542/peds.2005-1320. [DOI] [PubMed] [Google Scholar]
- 31.Shrier LA, Harris SK, Beardslee WR. Temporal associations between depressive symptoms and self-reported sexually transmitted disease among adolescents. Arch Pediatr Adolesc Med. 2002;156:599–606. doi: 10.1001/archpedi.156.6.599. [DOI] [PubMed] [Google Scholar]
- 32.Stiffman AR, Earls F, Doré P, Cunningham R. Changes in acquired immunodeficiency syndrome-related risk behavior after adolescence: relationships to knowledge and experience concerning human immunodeficiency virus infection. Pediatrics. 1992;89:950–956. [PubMed] [Google Scholar]
- 33.Hops H, Ozechowski TJ, Waldron HB, et al. Adolescent health-risk sexual behaviors: effects of a drug abuse intervention. AIDS Behav. 2011;15:1664–1676. doi: 10.1007/s10461-011-0019-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Stallings MC, Corley RP, Hewitt JK, et al. A genome-wide search for quantitative trait loci influencing substance dependence vulnerability in adolescence. Drug Alcohol Depend. 2003;70:295–307. doi: 10.1016/s0376-8716(03)00031-0. [DOI] [PubMed] [Google Scholar]
- 35.Miles DR, Stallings MC, Young SE, Hewitt JK, Crowley TJ, Fulker DW. A family history and direct interview study of the familial aggregation of substance abuse: the adolescent substance abuse study. Drug Alcohol Depend. 1998;49:105–114. doi: 10.1016/s0376-8716(97)00156-7. [DOI] [PubMed] [Google Scholar]
- 36.Costello EJ, Edelbrock CS, Costello AJ. Validity of the NIMH Diagnostic Interview Schedule for Children: a comparison between psychiatric and pediatric referrals. J Abnorm Child Psychol. 1985;13:579–595. doi: 10.1007/BF00923143. [DOI] [PubMed] [Google Scholar]
- 37.American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4thedition: DSM-IV. Washington, DC: American Psychiatric Association; 1994. [Google Scholar]
- 38.Cottler LB, Robins LN, Helzer JE. The reliability of the CIDI-SAM: a comprehensive substance abuse interview. Br J Addict. 1989;84:801–814. doi: 10.1111/j.1360-0443.1989.tb03060.x. [DOI] [PubMed] [Google Scholar]
- 39.American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th. Washington, DC: American Psychiatric Association; 1994. [Google Scholar]
- 40.Crowley TJ, Mikulich SK, Ehlers KM, Whitmore EA, MacDonald MJ. Validity of structured clinical evaluations in adolescents with conduct and substance problems. J Amer Acad Child Adolesc Psychiatry. 2001;40:265–273. doi: 10.1097/00004583-200103000-00005. [DOI] [PubMed] [Google Scholar]
- 41.Billy JO, Tanter K, Grady WR, Klepinger DH. The sexual behavior of men in the United States. Family Planning Perspectives. 1993;25:52–60. [PubMed] [Google Scholar]
- 42.Michael RT, Gagnon JH, Laumann EO, Kolata G. Sex in America: A definitive survey. Boston, MA: Little, Brown and Company; 1994. [Google Scholar]
- 43.IBM SPSS Statistics. Statistical Package for the Social Sciences (IBM SPSS Statistics) for Windows, release 19.0.0.1. Chicago, Illinois: IBM SPSS Statistics; 2010. [Google Scholar]
- 44.Brown LK, Hadley W, Stewart A, Lescano C, Whiteley L, Donenberg G, DiClemente R, Project STYLE Study Group Psychiatric disorders and sexual risk among adolescents in mental health treatment. J Consult Clin Psychol. 2010;78:590–597. doi: 10.1037/a0019632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Brodbeck J, Vilén UL, Bachmann M, Znoj H, Alsaker FD. Sexual risk behavior in emerging adults: gender-specific effects of hedonism, psychosocial distress, and sociocognitive variables in a 5-year longitudinal study. AIDS Educ Prev. 2010;22:148–159. doi: 10.1521/aeap.2010.22.2.148. [DOI] [PubMed] [Google Scholar]
- 46.StLawrence JS, Jefferson KW, Alleyne E, Brasfield TL. Comparison of education versus behavioral skills training interventions in lowering sexual HIV-risk behavior of substance-dependent adolescents. J Consult Clin Psychol. 1995;63:154–157. doi: 10.1037//0022-006x.63.1.154. [DOI] [PubMed] [Google Scholar]
- 47.St Lawrence JS, Crosby RA, Brasfield TL, O'Bannon RE., 3rd Reducing STD and HIV risk behavior of substance-dependent adolescents: a randomized controlled trial. J Consult Clin Psychol. 2002;70:1010–1021. [PubMed] [Google Scholar]
- 48.Drug and Alcohol Services Information System. Substance Abuse and Mental Health Services Administration; 2006. [last accessed 18.7.11]. Characteristics of young adult (aged 18-25) and youth (Aged 12-17) admissions: 2004. http://oas.samhsa.gov/2k6/youngtx/youngtx.htm. [Google Scholar]
- 49.Centers for Disease Control and Prevention. Trends in the Prevalence of Sexual Behaviors, National YRBS 1991-2009. Atlanta, GA: Centers for Disease Control and Prevention; 2011. [Google Scholar]