Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2010 Jan 1.
Published in final edited form as: Drug Alcohol Depend. 2008 Sep 25;99(1-3):141–149. doi: 10.1016/j.drugalcdep.2008.07.016

Clinical Characteristics of Treatment Seeking Adolescents with Opioid versus Cannabis/Alcohol Use Disorders

Geetha A Subramaniam 1,, Maxine A Stitzer 2, George Woody 3, Marc J Fishman 4, Ken Kolodner 5
PMCID: PMC2758688  NIHMSID: NIHMS90559  PMID: 18818027

Abstract

Objectives

To assess the clinical characteristics of adolescents with DSM-IV opioid use disorder (OUD) and compare them to adolescents with cannabis/alcohol use disorders.

Method

94 adolescents (ages 14–18 years) with a current OUD and 74 adolescents with a current non-OUD cannabis/alcohol use disorders were recruited from admissions, predominantly residential, to a substance abuse treatment program in Baltimore, Maryland. Participants were assessed cross-sectionally using standardized interviews and self-reports. Chi-square, t-tests and ANCOVA (adjusting for age, gender and treatment setting, race and residence)were performed to determine group differences on demographic, substance use, psychiatric and HIV-risk behaviors; logistic regression analyses, both unadjusted and adjusted for the above five factors were conducted to assess the strength of associations.

Results

The OUD group was more likely to be Caucasian, to have dropped out of school and to live in the suburbs (trend). They also had greater substance use severity with higher proportion of current sedative and multiple SUDs. There were generally no differences in rates of criminal behaviors. Both groups had high rates of current psychiatric disorders (83% vs. 78%, n.s.) but the OUD adolescents reported higher depressive symptoms, mostly in the moderate range. Injection drug use (IDU) and needle sharing was almost exclusive to the OUD group, while both groups reported similar high rates of risky sexual behaviors.

Conclusions

While there were similarities between the two groups, OUD adolescents evidenced greater impairment in academic, substance use, depressive symptom and IDU-related HIV-risk areas. Findings suggest poorer long-term prognosis and highlight the need for specialized interventions for treatment seeking OUD adolescents.

Keywords: opioid dependence, heroin use, clinical characteristics, adolescent substance abuse

1. Introduction

Opioid use among adolescents has risen dramatically in the past decade. Past-year heroin use among 12th graders in the decade from1995–2005 averaged 1%, while past-year non-medical use of prescription opioids nearly doubled from 4.7% to 9%, during the same period.(Johnston et al., 2006) Similarly, the National Survey on Drug Use and Health (NSDUH) data from 2002–05 showed that 7.4% of U.S. adolescents aged 12–17years reported past year non-medical use of prescription opioids, second only to the age 18–25 year group(Substance Abuse and Mental Health Services Administration (SAMHSA) Office of Applied Studies (OAS), April 6, 2007). Correspondingly, treatment admissions for substance use disorders (SUD) among adolescents who identified opioids as the primary problem increased 196% from 1995–2005(Substance Abuse and Mental Health Services Administration (SAMHSA) Office of Applied Studies (OAS), 2006 October 3). Despite rising trends and associated health risks, including HIV and HCV infection from injection drug use(Center for Disease Control and Prevention (CDC), 2002 September), the literature is sparse on the characteristics and treatment needs of adolescents presenting with this serious form of drug use.

Four studies (Clemmey et al., 2004; Gordon et al., 2004; Hopfer et al., 2000; Marsch et al., 2005) describing adolescents with heroin use/opioid dependence entering treatment during the past decade (1996–2006) and data extracted from the Treatment Episode Data Set (Office of Applied Studies, 2007) [12–17 year olds admitted to US publicly funded treatment programs for a primary problem with heroin/opioids from 2004–2005, n=4959] analyzed by our team, forms the basis of our knowledge of the recent opioid epidemic among youth. Several findings were consistent across these studies, despite differing methods and geographic locations: the mean age of adolescents with opioid use/dependence was16–17.5 years; the proportion of females was higher than comparison populations using other substances; they were predominantly Caucasian; most were polysubstance users; and injection drug use was common, ranging from 35–64%. TEDS data suggested that very few 12–17 year-old opioid users (10%) tend to complete 12 or > years of school suggesting low school engagement (Substance Abuse and Mental Health Services Administration (SAMHSA) Office of Applied Studies , 2007). Psychiatric comorbidity was examined in only two studies: 41% of opioid-dependent adolescent patients were found to have a psychiatric disorder, the most common diagnosis being a depressive disorder (18–21%) in residential and outpatient treatment samples (Marsch et al., 2005; Gordon et al., 2004).

While these findings are informative, most studies had small sample sizes, lacked breadth in assessments and did not use standardized diagnostic interviews. Only two studies used a comparison group in an attempt to put the opioid users into the broader perspective of adolescent substance use, but both focused only on heroin use without examination of the diagnostic criteria for an opioid use disorder (Clemmey et al., 2004; Hopfer et al., 2000). To date, no single study has provided a comprehensive and systematic characterization of treatment-seeking adolescents with opioid use disorder (OUD). Understanding how adolescents with OUD differ from those having more common SUDs (e.g. cannabis use disorder) may better inform development of tailored treatments for these youth.

With this purpose in mind, we conducted a cross-sectional study to characterize treatment-seeking adolescents with OUD on a wide array of characteristics: demographic, social, substance use, psychiatric and HIV-risk behaviors; and compared them to adolescents entering treatment with cannabis and/or alcohol use disorders. We anticipated that adolescents with OUD would present with greater severity of co-occurring problems in several areas due to the health, social and financial risks associated with the rapid development of physiological dependence that often accompanies abuse of opioid drugs and injection drug use (IDU) (Center for Disease Control and Prevention (CDC), 2002 September; Inciardi, 1979; Rhodes et al., 2000 December).

2. Methods

2.1. Study Site

This study was conducted at a JCAHO-accredited adolescent SUD treatment facility in Baltimore, Maryland providing a continuum of residential and outpatient services for adolescents (13–18 years) and young adults (18–21 years), as described elsewhere in more detail.(Fishman et al., 2003) Adolescents/young adults with OUD typically enter the residential level of care and constitute approximately 10% of the 500–600 annual residential admissions. However, access to outpatients with OUD was enhanced between 2003–2005, when the NIDA Clinical Trials Network (CTN) study of buprenorphine treatment for adolescents/young adults was conducted at this site.

2.2. Recruitment

Patients entering residential treatment were referred by self/parents, social agencies, or the juvenile justice system. The outpatient CTN buprenorphine study recruited from community advertising and clinic referrals and contributed 19% (n=20) of OUD participants to the study reported here. All treatment admissions to the study site were pre-screened for study eligibility and written assent (those under 18 years) and/or written consent (18 year-olds and legal guardians of minors) was obtained within 1 week of treatment entry. The protocol and all study materials were approved by Western Institutional Review Board (WIRB), a Johns Hopkins IRB designee.

2.3. Study Eligibility

Opioid users were included if they were 14–18 years of age; met DSM-IV criteria for opioid abuse or dependence; and had any recent substance use (defined as last use < 2 weeks or stay in a controlled environment < 2 weeks, prior to study entry). Patients were included in the comparison group if they met DSM-IV criteria for cannabis and/or alcohol abuse/dependence (and no OUD) and met the matching criteria described below.

2.3.1. Comparison Group Matching Procedure

For every OUD participant enrolled, subsequent admissions were pre-screened to identify a comparison participant who matched with a current OUD participant on the following criteria: age (14–16/17–18), gender (M/F), and treatment setting (residential/outpatient). Cocaine use in the past 30 days (yes/no) was initially the 4th matching criterion, which was abandoned after 100 matched participants due to low rates of cocaine use in the potential comparison participants.

2.4. Participants

183 adolescents were consented for participation from 1820 treatment admissions eligible for screening between 12/03 and 05/06. Ninety percent (n=1638) of those pre-screened were not eligible for participation for one of the following reasons: 25% did not match with a OUD group participant; 24% could not be consented within the eligibility period primarily due to staff unavailability; 23% reported no recent substance use (i.e. < 2 weeks); 12% were eligible but failed to provide consent (e.g. guardian unavailable/declined consent, participant did not keep appointment); and 5% were readmissions already enrolled in the study.

Of the 183 consented, 166 (OUD = 94, non-OUD comparison = 72) completed at least one diagnostic interview (4 missing CIDI; 16 missing DICA) and were retained for analyses; the 17 excluded participants did not differ by age, gender, race or treatment setting from those who were retained for analyses.

2.5. Study Instruments

All instruments chosen for the study have good psychometric properties and had been used in prior adolescent studies. A combination of diagnoses, total/composite scores and responses to individual items was used to provide information on patient characteristics. The assessment battery took approximately 2 ½ – 3 ½ hours and was administered over 2 or 3 sessions; participants received a $25 merchandise gift card for completing study assessments.

2.5.1. Demographic Assessment

The demographic assessment provided information on participant-matching criteria and other social characteristics: race, place of residence, guardianship status, court order for treatment, probation status. It also provided information on duration of abstinence, ages of onset of regular use of substances, and psychotropic medications at study entry. Self-reports of substance use were not validated by the use of urine toxicology results.

2.5.2. Composite International Diagnostic Interview (CIDI)

The substance abuse module (SAM) of this structured interview was used to obtain current DSM-IV diagnoses of SUD (i.e. substance dependence /abuse) for 9 substance categories: opioids, cannabis, alcohol, cocaine, sedative, other stimulants, PCP, inhalants, and any other substance.(Cottler et al., 1989) It also provided ages of onset for each of the DSM SUD diagnoses and was administered by trained research staff.

2.5.3. Diagnostic Instrument for Children and Adolescents-IV (DICA-IV)

This structured psychiatric interview for children/adolescents (Reich, 2000; Welner et al., 1987) provided information on several DSM-IV Axis-I diagnoses and their ages of onset. We selected seven psychiatric disorders commonly reported in the adolescent SUD literature (Hovens et al., 1994; Kandel et al., 1997; Stowell and Estroff, 1992) for assessment: ADHD, Major Depressive Episode (MDE), Manic Episode, Generalized Anxiety Disorder (GAD), Post Traumatic Stress Disorder (PTSD), Oppositional Defiant Disorder (ODD) and Conduct Disorder (CD). In addition, the psychosocial section provided limited information on academic functioning and family history, which were not available from other instruments. The first author (GS) administered 70% of DICA interviews; the others were administered by two psychiatrists who were trained by the first author (GS) to 100% agreement on observed interviews prior to beginning the study.

2.5.4. Beck Depression Inventory (BDI)

This 21-item self-report measure assessed depressive symptoms in the past 7 days (Beck et al., 1998). Each item is scored from 0–3 with a maximum score of 63. Total, cognitive and somatic subscale scores were examined (Steer et al., 1999).

2.5.5. Risk Behavior Survey (RBS)

This interview (Needle et al., 1995), administered by trained research staff, provided information on past 30-day HIV-risk behaviors from injection drug use (IDU) and sexual behavior. The past 30-day frequency of substance use questions was expanded to include marijuana and alcohol, in addition to the standard questions on heroin/other opioids and cocaine.

2.5.6. General Crime Scale

This component of the legal section of the Global Assessment of Individual Need (Dennis et al., 2003) was self-administered. Information on past year drug-related and other illegal activities derived both from composite scores and responses to selected individual items.

2.6. Data Analyses

To compare group differences, data were analyzed using Pearson chi-square tests for categorical characteristics and independent t-tests for characteristics with continuous data using SPSS version 13® (SPSS, 2004). In order to assess the relative strength of each correlate, we ran separate binary logistic regression analyses. Results are presented as crude odds ratios (OR) with 95% confidence intervals (CI). Each of these regressions was re-run controlling for the 3 matching criteria (age, gender and level of care), and two other demographic factors that were significantly different [race (Caucasian/other) and Baltimore City residence (yes/no)]. These results are presented as adjusted OR with 95% CI. These control factors were chosen to enhance our comparison matching procedures as we did not achieve a one-one match and to ensure that group differences were not likely due to differences on these demographic factors as they have been shown to be significantly associated with drug use patterns (Catalano et al., 1990).

For continuous data (such as age of onset of SUD or psychiatric disorders), groups were compared using t-tests. Means and standard errors are presented for each outcome. For several of these variables, the distributions deviated from normality (e.g. a few variables displayed flat distributions and a few contained a few slight outliers). Consequently, parametric results were checked by using the Mann-Whitney U. P-values obtained using both methods were almost identical. Thus, only parametric results are presented. Using analysis of covariance, we again controlled for the three matching and two demographic factors that were significantly related to OUD status. Adjusted means and standard errors are presented for each of these outcomes.

Two of the outcomes (the number of psychiatric diagnosis and the number of substance use disorders) are “count” variables. However, neither of these variables was distributed as a typical Poisson, although the number of substance use disorders came closest. We first ran Poisson regression models (unadjusted and adjusted) and then treated these variables in the same manner as the other continuous variables and ran t-tests and ANCOVAs. The results were extremely similar using all methods. To maintain simplicity in the presentation, means and standard errors and adjusted means and standard errors are provided.

There are several variables with very small samples (less than 40 subjects). Due to the small sample and sparseness in some of the covariates, no adjusted analyses are presented for these variables.

For our sample size of 94 OUD and 72 non-OUD participants, using a range of proportions (from 10% to 50%), we determined that a minimum OR of 2.00 to 3.25 or 0.3–0.5 (when OR are reported as < 1) was needed to achieve statistical significance at p = 0.05 (95% CI) with power set at .80.

3. Results

3.1. Demographic/Social Characteristics

Table-1 shows that the OUD and non-OUD comparison groups were similar on several characteristics in addition to the matching criteria. However, the OUD group was more likely to be Caucasian, to have dropped out of school, and not to have received special education. The OUD group was also more likely to reside outside Baltimore City and have parents who were separated or divorced, although differences lost significance in the adjusted analyses. OUD group was no different from the comparison group on rates of reporting both parents as guardians (19% vs. 23%, p= 0.37)

Table 1.

Demographic and Social Characteristics in OUD and non-OUD Samples:

OUD
(n=94)
Non-OUD
(n=72)
n % n % X2 Crude OR (C.I.) Adjusted O.R. (C.I.)a
Female 42 45 25 35 1.68 1.52(0.81–2.86) 1.27(0.61–2.64)
18 year olds 31 33 24 33 0.00 0.98(0.51–1.89) 1.22(0.56–2.65)
Residential Setting 65 69 55 76 1.07 0.69(0.35–1.39) 0.45(0.19–1.07)
Caucasian Race 84 89 36 51 30.48 8.17(3.66–18.25)*** 8.68(3.50–21.53)***
Baltimore City Residence 26 30 35 50 6.88 0.42(0.22–0.81)** 0.50(0.23–1.10)
Currently in School 35 38 42 60 7.69 0.41 (0.22–0.77)** 0.36(0.16–0.81)*
Court Ordered to Treatment 17 18 19 27 1.79 0.60(0.29–1.27) 0.83(0.34–2.05)
Member at home drinks or uses drugs 38 51 30 47 0.20 1.16(0.60–2.27) 1.01(0.45–2.25)
Parents Separated/Divorced 52 69 34 53 3.85 2.00(1.00–3.99)^ 1.87(0.70–4.96)
Ever in Special Education 21 22 25 35 3.33 0.53(0.27–1.05) 0.36(0.15–0.86)*

OUD=opioid use disorder, X2 = chi-square value, O.R.= odds ratio, C.I.= 95% confiderce interval

represents matching criteria for the selection of comparison group (non-OUD)

*

p<0.05

**

p< 0.01

***

p< 0.001

^

p=0.051

a

adjusted for age, gender, race, treatment setting, place of residence

3.2. Substance Use/Diagnoses and Drug-Related Criminal Behaviors

Per group definition 100% of the OUD group met criteria for DSM-based OUD diagnoses; 88% of this group met criteria for Opioid dependence. As shown in Table-2, patterns of current DSM SUD diagnoses differed markedly between the two groups. Compared to non-OUD, the OUD group was more likely to have sedative and cocaine (unadjusted analysis only) use disorders and less likely to have a cannabis use disorder; alcohol use disorder was present at similar rates in both groups. The OUD group was more likely to have multiple SUD (3 or >) diagnoses but only when OUD was included in the total count of SUDs.

Table 2.

Substance Use/Diagnoses in OUD and non-OUD Samples:

OUD
(n=94)
Non-OUD
(n =72)
n % n % X2 Crude OR (C.I.) Adjusted O.R. (CI.)a
% Past Year DSM-IV SUD Diagnosis
Opioid 94 100 72 0 N/A - -
Cocaine 49 53 20 29 9.49 2.79(1.44–5.41)** 1.79(0.80–4.00)
Cannabis 54 57 66 92 23.83 0.12(0.05–0.31)*** 0.13(0.05–0.38)***
Alcohol 50 53 44 61 1.04 0.72(0.39–1.35) 0.68(0.32–1.48)
Sedatives 33 36 5 7 17.92 7.16(2.62–19.56)*** 7.72(2.48–24.00)***
Stimulants 11 12 3 4 2.88 2.99(0.80–11.54) 3.32 (0.76–14.42)
Multiple SUDs (3 or >) 62 67 23 33 18.35 4.13(2.13–8.03)*** 3.67(1.64–8.20)**
Multiple non-opioid SUDs (2 or >) 57 62 45 65 0.18 0.87(0.45–1.66) 0.70(0.32–1.55)
n Unadj.
mean
se n Unadj.
mean
se t value df (1) Adj. meana se Adj.
meana
se F value df(6)
Mean # Past Year DSM-IV SUD Diagnoses 92 3.5 0.2 69 2.2 0.2 5.37*** 159 2.9 0.2 1.8 0.2 16.49*** 146
n % n % X2 Crude OR (C.I.) Adjusted O.R. (C.I.)
% Anv Past 30 Dav Substance Use
Heroin 53 56 2 3 50.91 43.31(10.01–187.28)*** 39.85(8.46–187.71)***
Other Opioid 52 56 10 14 29.96 7.74(3.53–16.95)*** 7.68(3.10–19.04)***
Any Opioid 84 89 12 17 87.29 41.30(16.74–101.87)*** 42.36(14.81–121.14)***
Cocaine 43 47 16 23 10.17 3.02(1.51–6.02)*** 1.65(0.75–3.65)
Marijuana 70 76 68 96 11.96 0.14(0.04–0.49)*** 0.25(0.07–0.94)*
Alcohol 52 57 53 75 5.36 0.45(0.23–0.89)* 0.43(0.19–0.96)*
n mean se n mean se t value df(1) Adj. meana se Adj.
meana
se F value df(6)
Mean Age of Onset DSM SUDb
Opioid 92 15.3 0.1 0 N/A N/A N/A N/A N/A N/A N/A N/ A N/A N/A
Cocaine 49 15.7 0.2 20 16.3 0.2 −2.00* 67 15.8 0.2 16.3 0.3 9.67** 61
Cannabis 54 13.9 0.3 63 14.4 0.3 –1.1 115 13.8 0.4 14.2 0.3 0.67 104
Alcohol 50 14.4 0.3 43 14.4 0.3 0.12 91 14.5 0.4 14.6 0.4 0.1 81

OUD=opioid use disorder, X2 = chi-square value, O.R.= odds ratio, C.l.= 95% confidence interval, se= standard error, df=degrees of freedom, N/A = not applicable; adj=adjusted

*

p<0.05

**

p< 0.01

***

p< 0.001

a

adjusted for age, gender, race, treatment setting, place of residence

b

mean age of onset of each SUD for those with a past year DSM-IV SUD Diagnosis

c

mean age of onset of regular use of each substance for those with a current SUD diagnosis for that substance

3.2.1. Recency of Use

Consistent with their diagnosis, the OUD group was more likely to have used opioids and cocaine (unadjusted analysis only), while the comparison group was more likely to have used marijuana and alcohol in the 30 days prior to interview.

3.2.2. Onset of Use

Examining the data for those meeting criteria for each SUD, the mean age at which participants reported first regular use of each of the substances and age of onset of each SUD diagnosis (except for cocaine use disorder) did not differ between the two groups. For OUD and non-OUD groups combined, the age of onset of regular marijuana use was approximately 13 years, alcohol 13.5 years, any opioid 15.1 years and cocaine 16 years; the two groups did not differ on age of onset of regular use (results not shown in table). Both groups combined, the onset of SUD diagnosis was approximately 14 years for cannabis and alcohol (about 1 year after onset of regular use), 15–16 years of age for opioids, cocaine and sedatives (within months after onset of regular use.

3.2.3. Criminal Behaviors (results not shown in table)

The groups did not differ on composite scores for property, interpersonal or drug-related crimes committed in the past year. Nevertheless, the OUD group was more likely to have forged checks/documents (20% vs. 7%, p = 0.018) and reported any kind of stealing (87% vs. 71%, p=0.013). They were not different on other commonly reported criminal behaviors such as driving under the influence of any substance (DUI, 63% vs. 51%), selling drugs (60% vs. 57%), and trading sex for drugs (8% vs. 6%). Probation status rates were no different for the two groups (OUD 35% vs. Comparison 44%, p= 0.17)

3.3. Psychiatric History/Diagnoses

Both the OUD and the comparison groups had high rates of any DSM-IV psychiatric disorders (83% vs. 78%, respectively, p=0.313) and virtually identical profiles of psychiatric diagnoses as seen in Table-3. Over 50% had CD, 40% MDE or GAD and approximately a third had ADHD while PTSD, Mania and ODD had lower prevalence. Mean ages of onset of these disorders (ADHD not included as the diagnoses requires onset before age 7) were also identical for both groups and ranged from 10 to14 years.

Table 3.

Psychiatric History/Diagnoses in OUD and non-OUD Samples

OUD
(n=94)
Non-OUD
(n=72)
n % n % X2 Crude OR (C.I.) Adjusted O.R.
(CI.)a
% Admitted on any psychotropic Meds 33 35 27 38 0.1 0.90(0.48–1.71) 0.68(0.31–1.47)
% suicide attempt in the past 18 22 11 17 0.57 1.38(0.60–3.17) 0.78(0.28–2.15)
% Current DSM-IV Axis-I Psychiatric Diagnosis
Conduct Disorder 44 53 33 52 0.03 1.06(0.55–2.04) 1.14(0.53–2.48)
Major Depressive Episode 33 40 20 31 1.14 1.45(0.73–2.89) 1.39(0.62–3.14)
Generalized Anxiety Disorder 33 40 19 30 1.6 1.56(0.78–3.13) 1.3(0.54–3.15)
ADHD 27 33 25 39 0.68 0.75(0.38–1.49) 0.80(0.36–1.81)
PTSD 21 26 23 36 1.82 0.61(0.30–1.25) 0.44(0.18–1.07)
Manic/Hypomanic Episode 15 18 14 22 0.33 0.79(0.35–1.78) 0.61(0.23–1.62)
ODD 12 15 14 22 1.37 0.60(0.26–1.42) 0.57(0.21–1.52)
n Unadj.
mean
se n Unadj.
mean
se t
value
df
(1)
Adj.
mean*
se Adj. mean* se F
value
df
(6)
Mean # Axis-I Psychiatric Diagnoses 83 2.2 0.2 64 2.3 0.2 −0.25 145 2.5 0.2 2.7 0.2 0.41 133
Aqe of Onset Of DSM-IV Psychiatric Diagnosis Symptoms b
Conduct Disorder 40 10.8 0.5 32 9.8 0.5 1.46 70 11.1 0.6 9.7 0.5 3.06 62
Major Depressive Episode 27 13.1 0.6 23 11.6 0.5 1.81 48 12.9 0.7 12 0.7 1.13 59
Generalized Anxiety Disorder 28 11.8 0.6 33 12.2 0.7 −0.4 45 11.1 0.9 11.6 0.8 0.25 38
Age at Significant Trauma 29 13.9 0.7 18 12.8 0.6 1.12 43 13 0.8 12.5 0.6 0.29 37
Manic/Hypomanic episodec 12 11.1 0.9 14 12.2 0.9 −0.93 24
ODDc 12 10.3 0.5 14 10.3 0.5 0.04 24

OUD=opioid use disorder, X2 = chi-square value, O.R.= odds ratio, C.l.= 95% confidence interval, se= standard error, df =degrees of freedom; adj = adjusted

*

p<0.05

a

adjusted for age, gender, race, treatment setting, place of residence

b

mean age of onset for those with a current DSM-IV Psychiatric Diagnosis

c

no adjusted means were computed due to the small sample size

3.3.1. BDI Results

While the rates of MDE were not different, the OUD groups reported higher mean total scores on the BDI (17.3 vs. 14.2), with differences approaching significance (p=0.06). Correspondingly, a greater proportion of OUD adolescents reported depressive symptoms in the moderately severe range (i.e. >16, 48% vs. 33%, p=0.04 but the differences lost significance in the adjusted analyses.

3.4. HIV/Hepatitis C-Risk Characteristics

As shown in Table - 4, past 30-day IDU (41%) was almost exclusively present among OUD adolescents. Among OUD needle users, 49% reported sharing needles or “works” in the past 30 days (not shown in table). Over 3/4th of both groups were sexually active in the past month. There were no group differences on rates of multiple sexual partners or reports of always engaging in unprotected sexual activity. The OUD group was less likely to use protection at all times, but only in the unadjusted analyses.

Table 4.

Sexual and Injection Drug Use (IDU) Related HIV-Risk Behaviors in OUD and non-OUD Samples

OUD (n=94)
Non-OUD (n=72)
n % n % X2 Crude OR (C.I.) Adjusted O.R. (C.I.)a
        Past 30 Day Sexual 1 HIV-Risk Behaviors
Sexually Active 71 76 58 81 0.42 0.78(0.37–1.66) 0.99(0.40–2.43)
    Always unprotected sex 29 41 17 29 1.85 1.67(0.80–3.48) 0.99(0.41–2.36)
    Never unprotected sex 6 9 15 26 7.1 0.27(0.1–0.74)** 0.43(0.12–1.52)
    2 or > sexual partners 27 38 20 35 0.17 1.17(0.57–2.40) 2.22(0.85–5.81)
Any Lifetime IDU 45 49 2 3 39.27 31.12(7.19–134.69)*** 32.98(6.74–161.33)***
% Past 30 day IDU
Any IDU 38 41 1 1 35.03 49.06(6.53–368.52)*** 63.37(7.27–552.66)***
Heroin 36 39 1 1 31.21 42.95(5.71–323.08)*** 59.96(6.76–531.57)***
Cocaine 15 17 0 0 N/A N/A
Other Opioid 4 4 0 0 N/A N/A

OUD=opioid use disorder, X2 = chi-square value, O.R.= odds ratio, C.I.=confidence interval

*

p<0.05

**

p< 0.01

***

p< 0.001

a

adjusted for age, gender, race, treatment setting, place of residence

4. Discussion/Conclusions

This study is unique in its in-depth examination of adolescents with opioid dependence/abuse on a wide range of characteristics: demographic, social, substance use, psychiatric and HIV/HCV-risk using standardized assessments; and in its use of a marijuana/alcohol use disorder comparison group matched for age, gender and treatment setting. Overall, the two groups had similar and multiple comorbidities; but the OUD youth were distinguished by being mostly Caucasian, having dropped out of school, living outside Baltimore city (approaching significance), having greater substance use severity and injection-related HIV/HCV-risk; the latter two suggest a more serious manifestation of a SUD and signal poor long-term prognosis for these OUD youth.

Some of these findings are consistent in the adolescent treatment literature from the past decade (Clemmey et al., 2004; Gordon et al., 2004; Marsch et al., 2005; Substance Abuse and Mental Health Services Administration (SAMHSA) Office of Applied Studies (OAS), 2007) including predominance of Caucasian race, and suburban residence (Gordon et al., 2004). Greater opioid use among Caucasian teenagers has also been reported in an analysis of the community-based Monitoring the Future Study (McCabe et al., 2005) supporting a unique vulnerability for this racial group. These characteristics raise issues of drug access, economic means, social networks and perhaps lower opioid-risk perception among suburban Caucasian youth that are worthy of exploration in future studies. An alarming finding was the lower rate of current school enrollment among OUD youth, a finding consistent with lower educational achievement among OUD young adults (Gandhi et al., 2006), and our analyses of TEDS adolescent data (Office of Applied Studies, 2007). This is a serious circumstance, independently associated with substantial lifelong impediments including poverty (Iceland, 2003) and increased criminal justice involvement (Harlow, 2003 January). While the reasons for lower academic engagement remain unclear, we speculate that not receiving special educational services addressing their early onset behavioral problems may have influenced this outcome. Further research may help clarify whether OUD youth are more likely to drop out of school either due to the effect of opioids or the need to support their expensive opioid habit by working or engaging in certain criminal behaviors.

Profiles of substance use were markedly different across the groups, with OUD adolescents showing greater substance severity, even after controlling for residential level of care, which is associated with higher substance severity (Fishman et al., 2003). A greater proportion of the treatment-seeking OUD youth had sedative, cocaine (approaching significance) and multiple SUDs, while more of the comparison group, as expected, had cannabis use disorder. The impact of multiple SUDs is likely to be detrimental; cocaine use signals a path of adverse health, criminal and social outcomes (Leri et al., 2003); concurrent sedative use (mainly benzodiazepines) has been linked to reports of death from overdose (Darke et al., 2002; Zanis and Woody, 1998) and multiple-substanceabusing youth generally have poor treatment outcomes (Ciraulo et al., 2003). These findings also reinforce the common clinical observation that setting goals for total abstinence can be difficult when multiple substances are being abused.

Our data did not support our hypothesis that OUD youth would demonstrate higher rates of criminal activity. The drug-related criminal behaviors were equally severe for both the OUD and non-OUD samples: over 1/3rd reported being on probation and virtually all engaged in criminal activities. These similarities are potentially explained by the predominantly residential sample with juvenile justice involvement. However, the possible higher socioeconomic status of suburban OUD youth and better access to prescription opioids from medicine cabinets of families and neighbors (data not collected in this study); and higher rates of stealing and forging among them, may offer some explanation to their means of support their opioid habit. Also, almost half of both groups reported past-year drug dealing or driving under the influence of any substance, behaviors that increase their risk exposure and pose a risk to their communities.

High rates of Axis-I psychiatric disorders found in the OUD and non-OUD samples, (83% vs. 78%) were similar to other reports about treatment samples of adolescents with non-opioid SUDs (Hovens et al., 1994; Kandel et al., 1997; Stowell and Estroff, 1992). The exception is the low (41%) rate reported by Gordon et al. (Gordon et al., 2004), a finding most likely influenced by the fact that not all patients received a psychiatric evaluation. However, our findings are derived from a predominantly residential treatment sample for whom admission criteria may include problematic coexisting emotional and behavioral conditions (Mee-Lee et al., 2001) and thus do not necessarily reflect prevalence of psychiatric disorders in the outpatient treatment populations of adolescents with SUDs. Relative prevalence among specific psychiatric disorders was similar to previous reports with highest rates for conduct disorder (50%), followed by MDE, GAD, ADHD and PTSD (Bukstein et al., 1992; Riggs et al., 1995). Since having co-occurring psychiatric disorders has been shown to negatively impact outcomes among adolescents with SUD (Grella et al., 2001), these data further highlight the need for routine mental health assessments and integrated psychiatric care with SUD treatment. The high rates of depressive and anxiety disorders in the OUD youth, also seen in opioid dependent adults (Brooner et al., 1997; Rounsaville et al., 1982; Teesson et al., 2005). While the rates of MDE diagnoses were not different, the OUD youth had significantly higher depressive symptoms consistent with previous studies (Clemmey et al., 2004; Subramaniam et al., 2007), with half scoring in the moderately depressed range. This may serve as a marker for poorer prognosis as shown in the authors’ previous work.(Subrmaniam et al., 2007)

The risk for HIV/HCV infection (Center for Disease Control and Prevention (CDC), 2007 May 14) was almost exclusively present among OUD adolescents; 40% reported recent IDU and half of them reported sharing needles. While risky sexual behaviors were high in both groups, they take on greater significance among OUD youth who may contact HIV/HCV through IDU and subsequently serve as vectors for potential sexual transmission to their partners.

Our data on onset of SUD diagnoses versus psychiatric disorders provides information on the developmental time course of these disorders, for both groups. In general, onset of psychiatric disorders at ages 10–14 years preceded the onset of OUD and other SUD diagnoses (range: age 14–16 years). This lends support to the concept that psychiatric disorders are risk factors for the development of SUD among adolescents (Fergusson et al., 2007; Kessler et al., 2005) Our data confirmed the typical pattern of progression of substance initiation, from marijuana or alcohol to cocaine and/or opioids (Kandel, 1975). The progression to having a SUD diagnoses from time of regular use was about a year for alcohol and marijuana. However, in the case of cocaine and opioids, this progression was within a year. These findings suggest prevention efforts may best target those experimenting with cocaine or opioid cautioning against their regular use which leads rapidly to abuse/dependence.

4.1. Limitations

This study recruited from a single treatment site and may not generalize to other adolescent populations. However, as previously mentioned, there were notable similarities between this study sample and the larger national TEDS adolescent sample; this increases confidence that our sample is representative rather than aberrant. The use of a predominantly residential sample may also skew results toward adolescents with higher severity characteristics. However, matching the comparison sample on this variable means that differences observed should be valid and robust. It is also conceivable that group differences were missed due to sample size. The validity of information could have been enhanced if we had interviewed legal guardians/parents, particularly in the determination of psychiatric disorders, even though well-trained psychiatrists made diagnoses using a standardized diagnostic interview. Similarly, the absence of confirmatory drug screens may have limited the validity of self-reports on substance use, even though adolescent self-reports of substance use have been shown to be valid and reliable (Lennox et al., 2006). Finally, the cross-sectional design did not permit inferences on causality.

4.2. Implications and Future Directions

This study further characterizes the complex and high severity comorbidities of OUD youth. The OUD sample was fairly heterogeneous with several subtypes that emerged (e.g. male vs. female; heroin vs. prescription opioid users; community treatment vs. study referred; court-ordered vs. self-referred; those with opioid dependence vs. abuse). This warrants further research with larger samples to determine how these characteristics are inter-related and how demographic and psychosocial profiles compare across youth with drug use patterns that do not involve opioids. Another important step is to determine the influence of these characteristics on treatment retention and outcomes. While evidence is emerging for the efficacy for agonist treatments for opioid dependent adolescents (Marsch et al., 2005), future trials are needed to target comorbidities such as academic disengagement, multiple substance use and risks from overdose, HIV/HCV infections and psychiatric disorders.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Contributor Information

Geetha A. Subramaniam, Department of Psychiatry, Johns Hopkins University, C/O Mountain Manor Treatment Center, 3800 Frederick Ave, Baltimore, MD 21229, Phone: 410-233-1400; Fax: 410-233-1666; E-mail: gsubram@jhmi.edu

Maxine A. Stitzer, Johns Hopkins University, Clinical Trials Network, 5510 Nathan Shock Dr., Ste 3040, Baltimore, MD 21224

George Woody, Treatment Research Institute, 600 Public Ledger Building, 150 South Independence Mall, Philadelphia, PA 19106.

Marc J. Fishman, Department of Psychiatry, Johns Hopkins University; Mountain Manor Treatment Center, 3800 Frederick Ave, Baltimore, MD 21229

Ken Kolodner, 3806 Fenchurch Rd, Baltimore, MD 21218.

REFERENCES

  1. Beck A, Steer R, Gabin M. Psychometric properties of the BDI: Twenty-five years of evaluation. Clinical Psychological Review. 1998;8:77–100. [Google Scholar]
  2. Brooner RK, King VL, Kidorf M, Schmidt CW, Jr, Bigelow GE. Psychiatric and substance use comorbidity among treatment-seeking opioid abusers. Arch Gen Psychiatry. 1997;54:71–80. doi: 10.1001/archpsyc.1997.01830130077015. [DOI] [PubMed] [Google Scholar]
  3. Bukstein OG, Glancy LJ, Kaminer Y. Patterns of affective comorbidity in a clinical population of dually diagnosed adolescent substance abusers. J Am Acad Child Adolesc Psychiatry. 1992;31:1041–1045. doi: 10.1097/00004583-199211000-00007. [DOI] [PubMed] [Google Scholar]
  4. Catalano RF, Hawkins JD, Wells EA, Miller J, Brewer D. Evaluation of the effectiveness of adolescent drug abuse treatment, assessment of risks for relapse, and promising approaches for relapse prevention. Int J Addict. 1990;25:1085–1140. doi: 10.3109/10826089109081039. [DOI] [PubMed] [Google Scholar]
  5. Center for Disease Control and Prevention (CDC) Viral Hepatitis and injection drug users. [Accessed on September 19, 2007];2002 September; http://www.cdc.gov/IDU/hepatitis/viral_hep_drug_use.htm.
  6. Center for Disease Control and Prevention (CDC) HIV prevention among injection drug users. [Accessed on September 19, 2007];2007 May 14; http://www.cdc.gov/IDU/default.htm.
  7. Ciraulo DA, Piechniczek-Buczek J, Iscan EN. Outcome predictors in substance use disorders. Psychiatr Clin North Am. 2003;26:381–409. doi: 10.1016/s0193-953x(02)00106-5. [DOI] [PubMed] [Google Scholar]
  8. Clemmey P, Payne L, Fishman M. Clinical characteristics and treatment outcomes of adolescent heroin users. J Psychoactive Drugs. 2004;36:85–94. doi: 10.1080/02791072.2004.10399726. [DOI] [PubMed] [Google Scholar]
  9. 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]
  10. Darke S, Topp L, Ross J. The injection of methadone and benzodiazepines among Sydney injecting drug users 1996–2000: 5-year monitoring of trends from the Illicit Drug Reporting System. Drug Alcohol Rev. 2002;21:27–32. doi: 10.1080/09595230220119318. [DOI] [PubMed] [Google Scholar]
  11. Dennis ML, Titus JC, White M, Unsicker J, Hodgkins D. Global Appraisal of Individual Needs (GAIN): Administration guide for the GAIN and related measures. 2006 October;24 http://www.chestnut.org/li/gain. Accessed on. [Google Scholar]
  12. Fergusson DM, Horwood LJ, Ridder EM. Conduct and attentional problems in childhood and adolescence and later substance use, abuse and dependence: results of a 25-year longitudinal study. Drug Alcohol Depend. 2007;88(Suppl 1):S14–S26. doi: 10.1016/j.drugalcdep.2006.12.011. [DOI] [PubMed] [Google Scholar]
  13. Fishman M, Clemmey P, Adger H. The adolescent residential substance abuse treatment program at Mountain Manor Treatment Center, Baltimore, Maryland. In: Stevens SJ, Morral AR, editors. Adolescent Drug Treatment: Theory Implementationin Ten National Projects. New York: Haworth Press; 2003. [Google Scholar]
  14. Gandhi DH, Kavanagh GJ, Jaffe JH. Young heroin users in Baltimore: a qualitative study. Am J Drug Alcohol Abuse. 2006;32:177–188. doi: 10.1080/00952990500479290. [DOI] [PubMed] [Google Scholar]
  15. Gordon SM, Mulvaney F, Rowan A. Characteristics of adolescents in residential treatment for heroin dependence. Am J Drug Alcohol Abuse. 2004;30:593–603. doi: 10.1081/ada-200032300. [DOI] [PubMed] [Google Scholar]
  16. Grella CE, Hser YI, Joshi V, Rounds-Bryant J. Drug treatment outcomes for adolescents with comorbid mental and substance use disorders. J Nerv Ment Dis. 2001;189:384–392. doi: 10.1097/00005053-200106000-00006. [DOI] [PubMed] [Google Scholar]
  17. Harlow CW. Bureau of Justice Statistics Special Report. Washington, DC: Department of Justice; 2003. Jan, Education and correctional populations. [Google Scholar]
  18. Hopfer CJ, Mikulich SK, Crowley TJ. Heroin use among adolescents in treatment for substance use disorders. J Am Acad Child Adolesc Psychiatry. 2000;39:1316–1323. doi: 10.1097/00004583-200010000-00021. [DOI] [PubMed] [Google Scholar]
  19. Hovens JG, Cantwell DP, Kiriakos R. Psychiatric comorbidity in hospitalized adolescent substance abusers. J Am Acad Child Adolesc Psychiatry. 1994;33:476–483. doi: 10.1097/00004583-199405000-00005. [DOI] [PubMed] [Google Scholar]
  20. Iceland J. Dynamics of economic well-being: Poverty 1996–1999. [Accessed on September 19, 2007];2003 www.census.gov/hhes/www/poverty/sipp96/table059697.html.
  21. Inciardi JA. Heroin and street crime. Crime and Delinquency. 1979;25:333–346. [Google Scholar]
  22. Johnston LD, O'Malley PM, Bachman JG, Schulenberg JE. Bethesda, MD: National institute on Drug Abuse; 2006. Monitoring the Future national survey results on drug use, 1975–2005 Volume I Secondary school students. (NIH Publication No. 06-5883) [Google Scholar]
  23. Kandel DB. Stages in adolescent involvement in drug use. Science. 1975;190:912–914. doi: 10.1126/science.1188374. [DOI] [PubMed] [Google Scholar]
  24. Kandel DB, Johnson JG, Bird HR, Canino G, Goodman SH, Lahey BB, Regier DA, Schwab-Stone M. Psychiatric disorders associated with substance use among children and adolescents: findings from the Methods for the Epidemiology of Child and Adolescent Mental Disorders (MECA) Study. J Abnorm Child Psychol. 1997;25:121–132. doi: 10.1023/a:1025779412167. [DOI] [PubMed] [Google Scholar]
  25. Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62:593–602. doi: 10.1001/archpsyc.62.6.593. [DOI] [PubMed] [Google Scholar]
  26. Lennox R, Dennis ML, Ives M, White MK. The construct and predictive validity of different approaches to combining urine and self-reported drug use measures among older adolescents after substance abuse treatment. Am J Addict. 2006;15(Suppl 1):92–101. doi: 10.1080/10550490601006089. [DOI] [PubMed] [Google Scholar]
  27. Leri F, Bruneau J, Stewart J. Understanding polydrug use: review of heroin and cocaine co-use. Addiction. 2003;98:7–22. doi: 10.1046/j.1360-0443.2003.00236.x. [DOI] [PubMed] [Google Scholar]
  28. Marsch LA, Bickel WK, Badger GJ, Stothart ME, Quesnel KJ, Stanger C, Brooklyn J. Comparison of pharmacological treatments for opioid-dependent adolescents: a randomized controlled trial. Arch Gen Psychiatry. 2005;62:1157–1164. doi: 10.1001/archpsyc.62.10.1157. [DOI] [PubMed] [Google Scholar]
  29. McCabe SE, Boyd CJ, Teter CJ. Illicit use of opioid analgesics by high school seniors. J Subst Abuse Treat. 2005;28:225–230. doi: 10.1016/j.jsat.2004.12.009. [DOI] [PubMed] [Google Scholar]
  30. Mee-Lee D, Shulman GD, Fishman M, Gastfriend D, editors. Chevy Chase MD: American Society of Addiction Medicine, Inc; 2001. ASAM Patient placement criteria for the treatment of substance-related disorders, Second edition-revised (ASAM PPC-2R) [Google Scholar]
  31. Needle R, Fisher DG, Weatherby N, Chitwood D, Brown B, Cesari H, Booth R, Williams ML, Watters J, Anderson M, Braunstein M. Reliability of self-reported HIV risk behaviors of drug users. Psychology of Addictive Behaviors. 1995;9:242–250. [Google Scholar]
  32. Office of Applied Studies. Treatment Episode Dataset (TEDS), 1992–2005 [Computer file] [Accessed on September 19, 2007];2007 http://www.icpsr.umich.edu/cocoon/SAMHDA/SERIES/00056.xml?token=1.
  33. Reich W. Diagnostic interview for children and adolescents (DICA) J Am Acad Child Adolesc Psychiatry. 2000;39:59–66. doi: 10.1097/00004583-200001000-00017. [DOI] [PubMed] [Google Scholar]
  34. Rhodes W, Layne M, Johnston P, Hozik L. What America's users spend on illegal drugs 1988–1998. [Accessed on September 19, 2007];2000 December; http://www.whitehousedrugpolicy.gov/publications/drugfact/american%5Fusers%5Fspend/credit.html. [Google Scholar]
  35. Riggs PD, Baker S, Mikulich SK, Young SE, Crowley TJ. Depression in substance-dependent delinquents. J Am Acad Child Adolesc Psychiatry. 1995;34:764–771. doi: 10.1097/00004583-199506000-00017. [DOI] [PubMed] [Google Scholar]
  36. Rounsaville BJ, Weissman MM, Crits-Christoph K, Wilber C, Kleber H. Diagnosis and symptoms of depression in opiate addicts. Course and relationship to treatment outcome. Arch Gen Psychiatry. 1982;39:151–156. doi: 10.1001/archpsyc.1982.04290020021004. [DOI] [PubMed] [Google Scholar]
  37. SPSS. Chicago: SPSS, Inc; 2004. SPSS Base 13.0 User's Guide. [Google Scholar]
  38. Steer RA, Ball R, Ranieri WF, Beck AT. Dimensions of the Beck Depression Inventory-II in clinically depressed outpatients. J Clin Psychol. 1999;55:117–128. doi: 10.1002/(sici)1097-4679(199901)55:1<117::aid-jclp12>3.0.co;2-a. [DOI] [PubMed] [Google Scholar]
  39. Stowell RJ, Estroff TW. Psychiatric disorders in substance-abusing adolescent inpatients: a pilot study. J Am Acad Child Adolesc Psychiatry. 1992;31:1036–1040. doi: 10.1097/00004583-199211000-00006. [DOI] [PubMed] [Google Scholar]
  40. Subramaniam GA, Stitzer MA, Clemmey P, Kolodner K, Fishman MJ. Baseline depressive symptoms predict poor substance use outcome following adolescent residential treatment. J Am Acad Child Adolesc Psychiatry. 2007;46:1062–1069. doi: 10.1097/chi.0b013e31806c7ad0. [DOI] [PubMed] [Google Scholar]
  41. Substance Abuse and Mental Health Services Administration (SAMHSA) Office of Applied Studies (OAS) [Accessed on September 19, 2007];October 3. Admissions aged 12–17 by primary substance of abuse: TEDS 1995–2005. 2006 Table 5.1a. http://wwwdasis.samhsa.gov/teds05/TEDSAd2k5Tbl5.1a.htm.
  42. Substance Abuse and Mental Health Services Administration (SAMHSA) Office of Applied Studies (OAS) [Accessed on September 19, 2007];Treatment Episode Dataset (TEDS), 1992–2005 [Computer file] 2007 http://www.icpsr.umich.edu/cocoon/SAMHDA/SERIES/00056.xml?token=1.
  43. Substance Abuse and Mental Health Services Administration (SAMHSA) Office of Applied Studies(OAS) Rockville, MD: 2007. Apr 6, The NSDUH Report: Patterns and Trends in Nonmedical Prescription Pain Reliever Use: 2002 to 2005. [Google Scholar]
  44. Teesson M, Havard A, Fairbairn S, Ross J, Lynskey M, Darke S. Depression among entrants to treatment for heroin dependence in the Australian Treatment Outcome Study (ATOS): prevalence, correlates treatment seeking. Drug Alcohol Depend. 2005;78:309–315. doi: 10.1016/j.drugalcdep.2004.12.001. [DOI] [PubMed] [Google Scholar]
  45. Welner Z, Reich W, Herjanic B, Jung KG, Amado H. Reliability, validity, and parent-child agreement studies of the Diagnostic Interview for Children and Adolescents (DICA) J Am Acad Child Adolesc Psychiatry. 1987;26:649–653. doi: 10.1097/00004583-198709000-00007. [DOI] [PubMed] [Google Scholar]
  46. Zanis DA, Woody GE. One-year mortality rates following methadone treatment discharge. Drug Alcohol Depend. 1998;52:257–260. doi: 10.1016/s0376-8716(98)00097-0. [DOI] [PubMed] [Google Scholar]

RESOURCES