Adolescent Substance Abuse


Substance abuse generally has its onset in adolescence and often occurs in the context of other psychiatric comorbidity. Dr. Riggs discussed a developmental approach to assessment and treatment of substance disorders and associated comorbidity. Dr. Lipschitz focused on Post-Traumatic Stress Disorder in adolescents and the questions of pharmacological and psychotherapeutic management to reduce self-medication of abused substances. Dr. Wilens presents data on the co-occurrence and overlap of ADHD and substance abuse.

Symposium Chairs: Robert P. Milin, M.D., University of Ottawa Faculty of Medicine, Royal Ottawa Hospital, Ottawa, Ontario; Deborah Deas, M.D., Medical University of South Carolina, Charleston, SC

Adolescent Substance Use Disorders and Comorbid Conditions: A Developmental Approach

Paula Riggs M.D., University of Colorado Health Sciences Center, Denver CO

Substance use disorders (SUDs) in adolescents are best under-stood within a multidimensional developmental framework. This is important because many risk factors associated with development of SUDs and other problems can be identified and targeted for evidence-based prevention and early intervention efforts.

The developmental risk domains begin with the individual child. There is a fairly robust genetic risk. In addition, biologically based aspects of "temperament", such as aggression, poor attention span and persistence, hyperactivity, and impulsivity, can be identified in toddlers and have been associated with development of SUDs and other behavior problems in adolescence. Children with this constellation of temperament have also been characterized as "behaviorally disinhibited". In somewhat older toddlers and early school-age children, oppositional defiant disorder and conduct disorder are also early warning signs for development of SUDs.

In addition to biological constitution, the family domain is also a critically important developmental influence imparting risk or protection against later SUDs. Family factors associated with higher risk include chronic conflict, abuse and neglect, low parental support, ineffective parental monitoring, impaired attachment, as well as parental or sibling SUD.

When "at-risk" children enter school, they can often be identified as having impaired social and coping skills, lower levels of commitment to achievement, and negative expectations about school performance. Early school failure is an easily detected "red flag", which heralds later problem behavior and substance involvement. Early identification presents an important opportunity for secondary prevention /early intervention. Such children often have treatable learning disabilities, attention deficit hyperactivity disorder (ADHD), or oppositional behavior problems.

Peers have greater influence in early to mid-adolescence. Association with deviant, substance-involved peers is an important risk factor at this age, especially when coupled with weak decision-making skills. If the larger community has a criminal or drug subculture, coupled with poor youth resources, then the risks may be amplified.

Although studies of associated risk factors do not inform us about the causes of adolescent SUDs, longitudinal studies and preliminary modeling strategies are bringing into focus developmental pathways that lead to adolescent SUDs. Children identified with such early problems should be targeted for evidence-based secondary prevention and early intervention efforts. These include Parent Management Training or other forms of empirically supported family therapy as well as school-based interventions and treatment of learning disabilities and other comorbidity, if available.1 If the developmental trajectory can be normalized prior to adolescence, the likelihood of progression of these problems to more serious conduct disorder and SUDs can be de-creased.

However, even youth with multiple risk factors, may not develop SUDs and pathological behaviors. Such youth generally have other factors that are protective and impart resilience. Such factors include effective parental monitoring, pro-social caring adult mentors, high verbal IQ, strong parent-child bonding, high commitment to prosocial values and achievement, and competence in non-drug related activities. Clinicians can be more helpful if they enhance resilience by pushing the balance toward protective factors.

Once, SUDs develop, they, adversely impact normal adolescent development by impeding identity formation, consolidation of values, and exploration of positive adult roles and vocational/career choices and positioning. Chronic substance use also impairs the development of a broader range of adroitness in the recognition and regulation of affect, which generally occurs during adolescence.

Although SUDs impede normal development, experimentation with substances of abuse is developmentally "normal" during adolescence, in the United States. Most national surveys indicate that more than 90% of adolescents experiment with nicotine, alcohol, or other substances of abuse before graduating from high school. Most "experimenters" do not progress to SUDs. Some studies indicate that "experimenters" may be psychologically healthier in some respects than either total abstainers or those with more serious involvement with substances. Therefore, an important issue for parents and clinicians is how to determine when a teenager needs more extensive evaluation and treatment.

One important indication that further evaluation is needed is when youth meet diagnostic criteria for substance abuse (SA) or sub-stance dependence (SD). The prevalence of SA/SD is in the range of 5-10% among adolescents (with the range indicating the prevalence in younger to older adolescents). Thus, SUDs are among the most prevalent psychiatric disorders in children and adolescents.

Many comorbid psychiatric disorders are also found at higher rates among youth with SUDs (e.g., conduct disorder (CD) 50-80%; ADHD 30-50%; major depression (MDD) 20-30%). Such comorbidity is often associated with more severe substance involvement. For example, MDD predicts greater substance se-verity among both adolescent males and females with CD and SUD. In addition, ADHD, in addition to CD severity, predicts more severe substance problems in adolescent males with CD and SUD.

Depression in adolescents with SUD may not remit with abstinence as has been shown in studies of adults with chronic alcohol dependence. In other words such depressions in youth may be less likely to be "secondary" to substances of abuse than in adults with SUDs.

There is also preliminary evidence that depression in adolescents with SUDs may respond to antidepressants when treatment is integrated with substance treatment, even in non-abstinent adolescents. Similarly, integrated treatment of ADHD with low abuse potential medications such as bupropion or pemoline, may also enhance substance treatment engagement. There is also evidence that treating ADHD in children before the onset of substance use may reduce risk of developing an SUD. However, once SUDs are established, substance use is not likely to "go away" by only treating the comorbid disorder without specific substance treatment.

This argues for integrated assessment and treatment of SUD and comorbidity. A thorough assessment of attitudes about substances of abuse and substance use is an essential aspect of any child/adolescent comprehensive psychiatric evaluation, which also includes assessment of family, school, peers and community across development. For each sub-stance, the history should include information regarding onset, progression to regular use, use in the last month, and last use in addition to assessing DSM-IV criteria for SA/SD. Utilizing a timeline that anchors major events as well as onset of comorbid disorders in relation to substance use is an important organizing feature of the evaluation. This can also help establish whether "comorbidity" only occurs in the context of heavy substance use or withdrawal states or whether symptoms of co-morbid disorders were present prior to onset of substance use or during periods of significant abstinence.2

Although much more research is needed, published controlled clinical trials indicate that some treatment approaches may be more efficacious in treating adolescent SUDs than others. Empirical support is found for multimodal behavioral therapy, cognitive behavioral therapy, and family-based interventions.

1) Behavioral treatments are grounded in the principles of operant conditioning. Such approaches have been shown to be more effective than "supportive therapies", especially if family members are involved in treatment. In such approaches, the environment is rear-ranged in order to positively reinforce activities incompatible with drug use (which have been identified by the patient as rewarding).

2) Cognitive behavioral therapy (CBT) has empirical support utilizing both group and individual therapy models. Generally, motivational enhancement techniques are used with adolescents, since they may not initially be at a stage of readiness to commit to behavior change and treatment. Cognitive behavioral approaches focus on correcting dysfunctional thinking related to drug use and on enhancing drug refusal skills, craving reduction, and re-lapse prevention strategies. In addition, specific CBT sessions target the most common skills' deficits in adolescents with SUDs such as communication skills, anger management, affect identification and regulation, and educational and job skills.

Virtually all empirically supported psychotherapies for adolescent SUDs emphasize the importance of family involvement in the therapy. Specifically, structural-strategic or family-systems approaches are most effective. Multisystemic family therapy (MST)—a community-based, multimodal, treatment approach—has been shown to be effective in even the most seriously conduct-disordered, substance- involved adolescents with chaotic home environments. All effective treatment approaches require regular (preferably randomized) urine toxicology assessments.

Once the adolescent and family have been effectively engaged in one of the empirically- supported psychotherapies for SUDs and either abstinence or significant reduction in substance use has been achieved, pharmacotherapy for comorbid psychiatric disorders can be considered.

Important clinical principles in utilizing pharmacotherapy for comorbidity such as ADHD or depression in adolescents with SUDs include: 1) Utilize medications with low abuse-potential when possible; 2) Implement a plan for monitoring medication compliance and safety as well as ongoing monitoring of substance use by history and urine toxicology assessment; 3) Educate the patient and family about how little is known of possible inter-actions of drugs of abuse and illicit drugs; 4) Emphasize with both patient and family members that, for reasons of safety, behavioral contracting is required for continuing Pharmacotherapy; 5) Evaluate the ability of the family to provide a safe, substance-free environment with appropriate monitoring. Inability to comply with these principles coupled with on-going or escalation in substance use indicates that a more restrictive intervention may be needed.

In summary, substance use disorders in adolescents have developmental antecedents that, if identified early, may be targets of effective prevention and early intervention efforts. However, even if treatment is not begun until an SUD is firmly entrenched, advancements in adolescent substance treatment demonstrate that multimodal family-based treatment coupled with integrated treatment of comorbidity may improve outcomes and help restore a more normalized developmental trajectory.


  1. See National Institute on Drug Abuse Web site: www.nida.nih.gov
  2. Riggs PD, Whitmore EA. Substance use disorders and disruptive behavior disorders. Chapter 4. In : Disruptive Behavior Disorders in Children and Adolescents . RL Hendren (Ed) pp. 133 -174; JM Oldhan, MB Riba (Series Eds) In Review of Psychiatry Vol l8 American Psychiatric Press, 1999.


Trauma, Post-Traumatic Stress Symptoms and Substance Use in Adolescents

Deborah S. Lipschitz, M.D., VA Connecticut Health Care System, West Haven, CT

There is little empirical literature on the relationship between trauma exposure, post-traumatic stress disorder (PTSD) and substance use disorder (SUD) in adolescents. Not much PTSD is diagnosed in adolescents with SUD because the DSM-IV criteria do not apply well so most work has been done with adults.

Examining several case studies will demonstrate the complexities of treating this population of adolescents. 1) Stacy is a 16 year-old white female who lives with her parents. Her substance history included starting social drinking and trying marijuana at 14, weekend binge drinking at 15, blackout and suicide attempt at 16. Her trauma was a date rape at 15, leading to criteria for PTSD. Alcohol use was linked to the PTSD. Her trauma was identified early, treatment duration was about a year, her family supported treatment, and she has a good prognosis.

2) Annie is a 15 year-old white female who is in a correctional facility. Her substance use history included cigarette smoking at 11; drinking and marijuana (MJ) use at 12; joined gang, sold drugs, ran away at 13; daily MJ use at 13. Her trauma history included physical abuse by father at 4/5 years; witnessed domes-tic violence by a gang of girls at age 5-10; attacked at 14 (a retraumatization leading to PTSD); attacked another girl at 15 and arrested for assault. In the correctional facility her delinquency and MJ were addressed, but not her trauma or PTSD. That was identified when she volunteered for a study, although she refused an offer of treatment.

3) Rita is a 15 year-old Hispanic female who is homeless. Her substance use history includes: alcohol at age 7; MJ at age 9; daily alcohol use with blackouts over the past year; MJ, speed, diet pills, Tylenol with codeine, cocaine, and ecstasy over past year. Her trauma history includes physical abuse by both parents at ages 4-6; raped by an uncle at 9; foster care at 10-13, where she was beaten; robbed at gun point at 15. SUD and PTSD were interrelated and it is unclear what comes first. Both conditions must be addressed. She is not ready for treatment.

An overview of PTSD begins with examining the DSM-IV criteria. DSM requires that there be a characteristic response to a Criterion A stressor which means the event must be life-threatening and the emotional response one of extreme horror and a feeling of helplessness. Types of these traumas for girls are childhood sexual abuse, adolescent sexual assault, witnessed domestic violence, and community violence. Types of traumas for boys are community violence, childhood abuse, and witnessed domestic violence. Girls are more likely to develop PTSD than boys. It may be partly the nature of the trauma that increases girl’s vulnerability, but hormonal factors may play a part as well.

In addition, other DSM criteria for PTSD include: 1) Re-experiencing symptoms. These include recurrent, intrusive and distressing thoughts or dreams. Classic flashbacks (i.e., those of veterans) or nightmares don’t seem to happen to adolescents. However, younger children might have trauma reenactments by developing regressive behaviors, e.g., cowering under a desk when someone yells at them. Adolescents might get very upset by exposure to reminders, and will try to avoid talking about the event. 2) Hyperarousal symptoms. These tend to be the hallmark of PTSD for combat veterans. Adolescents can complain of insomnia, irritability, and poor concentration. These symptoms overlap with major depression and with substance use. Studies with inner city adolescents show that they reported hypervigilance, but not exaggerated startle response, which is probably an adaptive behavior (e.g., if frequent gun shots are heard it wouldn’t be functional to be startled every time). 3) Common avoidance symptoms for adolescents include avoidance of talking or thinking about the trauma, and avoiding people and places that are reminders. More severe trauma patients report feeling isolated and detached with no idea about what the future holds for them.

Community and inner city studies show that PTSD is quite common among adolescents: study findings ranged from 9% - 27%; 14% (full), 12% (partial), and 22% in psychiatric settings. It is more common in girls. Some months full DSM criteria will be met and some months only partial criteria will be met.

Most prevalence studies of comorbidity of PTSD and SUDs have been done with adults. 73% of PTSD vets were found to meet criteria for lifetime SUD; the number is higher for treatment seeking veterans. 52% of men and 34% of women with PTSD were found to meet criteria for lifetime alcohol abuse/dependence; 35% of men and 27% of women met criteria for drug abuse/dependence. When one looks at the data for adult substance abusers, some-where between 30-60% of women in SUD treatment have lifetime PTSD, compared to 11% for adult women in general. Patients with both have more pathology and medical problems, drop out of treatment, and have more re-lapse so they have more severe SUDs. They are at risk for further traumatization, homelessness and having their children taken away. Compared to controls, one study found a group of adolescents with SUD to be 6-12 times more likely to have histories of physical abuse and 20 times more likely to have histories of sexual abuse. Also compared to controls, a second study of adolescents with SUDs found gender differences in comorbidity of various disorders. It found that 25% of girls with SUDs had concurrent PTSD; and about 12% of the boys had PTSD. Another study showed that girls with SUDs are two to three times as likely to have PTSD. It is not known which of the disorders comes first.

There is a neurobiological aspect that might make people more vulnerable to developing PTSD and SUD. Neuropeptide Y (NPY) is a 36 amino acid peptide, the most abundant neurotransmitter in the central nervous system. It co-localizes with norepinephrine in the sympathetic nervous system fibers. It is released when norepinephrine is released. Very early thinking is that with PSTD the sympathetic nervous system is hyperactive with higher blood pressure and higher pulse rates in response to trauma stimuli. The NPY levels in combat veterans were examined and found to be lower at base-line compared to healthy controls. NPY puts the brakes on the norepenephrine system. PTSD individuals can’t slow down their sym-pathetic nervous systems when they were traumatized.

In addition there are other substances that affect NPY release such as nicotine. For example, there is a higher rate of smoking in patients with PTSD. Also, mice with NPY deficits consume more alcohol than those with normal levels. High NPY level mice consume less alcohol and are also more susceptible to the effects of alcohol. This is just one neurobiological factor out of many that might ac-count for susceptibility to PTSD and SUD.

In conclusion, in trauma exposed populations, the comorbidity of PTSD and SUD is extremely common, and girls are more vulnerable than boys. There are no prospective studies to document the onset of each disorder, and just a few retrospective studies—only one with adolescents. There are no empirical studies of effective treatment for the comorbidity of these two conditions. Although there is a need to treat both simultaneously. When working with an adolescent with SUD, one must take a comprehensive trauma history and inquire specifically about PTSD symptoms.

ADHD and The Substance Use Disorders

Timothy E. Wilens, M.D., Massachusetts General Hospital, Harvard Medical School, Boston, MA

There are over 25 studies worldwide, demonstrating that the lifetime rate of ADHD in children is 4-5%. Three quarters of ADHD in children persists into adolescence and approximately half persists into adulthood. The estimate of ADHD in adults is about 2%.

ADHD is a chronic disorder characterized by developmentally inappropriate degrees of inattention, hyperactivity, and impulsivity. There is very robust evidence of the overlap of ADHD and substance use disorders (SUDs) throughout the lifespan. Roughly 20% of adults with SUDs have ADHD. There are fewer studies in adolescents, but the rate is probably closer to 40-45%. Family studies show excess rates of ADHD in relatives of those with SUDs and conversely there are excess rates of SUDs in relatives of those with ADHD. This is good evidence that the disorders are tracking together and one would suspect they are familial and may be genetic. There is also a third variable to consider which includes the other environmental effects of the community and the inutero-exposure, including the teratogenic effects of substances on ADHD. The two most problematic sub-stances are cigarettes and alcohol. If the parent has ADHD there is a 25% risk that the child will have it. A study we have compiled shows the risk for ADHD if the parent has SUD is 25%; and if the parent has both ADHD and SUD, 50% of the children will have ADHD.

In evaluating the developmental effect of ADHD on SUD, retrospective data from adults indicate that adults with ADHD have twice the risk of developing SUDs, compared to the con-trols. While noncomorbid ADHD is an inter-mediate risk factor for SUD, conduct or and bipolar disorders are the two most malevolent disorders in producing SUD in individuals with ADHD. Comorbidity is driving much of the SUDs in ADHD individuals. Among adults with ADHD marijuana was the most abused drug. Age of onset of drug abuse from our adult data is approximately 19 in these ADHD individuals, compared to age 22 in the control group.

In prospective studies in youth with ADHD, the data for smoking shows twice the risk across the lifespan . By the mean age of 15, almost 10% of non- ADHD controls were addicted to smoking; and 20% in the ADHD youth . When we look at children growing up, by age 14-15, ADHD itself wasn’t a huge risk factor yet for SUDs; however, CD in ADHD is really what is driving early onset of SUD. Moreover, girls with ADHD were developing SUDs at an early age compared to their male counterparts—a finding replicated in other of our datasets and epidemiologically.

Contemporary treatment issues include the important question of whether ADHD pharmacotherapy predisposes youth to SA. There is some pre-clinical animal data that would indicate that exposure of rats at developmental times may cause their brains to kindle for speiific dopaminergic substances, e.g., methylphenidate. But these studies employ some-times 20-100 times the dose we use in children because of the route of administration.

The aggregate human data suggests that medication treatment of ADHD reduces SUD. Data in our study were examined using a family genetic study of ADHD. There were 120 ADHD families and 109 non-ADHD families. Medication exposure at age 10-11 was examined and follow over five years. Structured interviews without objective measures of SA were used in this phase. The unmedicated group had the most SA; the medicated group showed lower abuse with alcohol, marijuana, cocaine, and hallucinogens and was similar to or a little higher than the controls. There was almost a five times greater chance of having SUDs in the unmedicated group vs. the controls.

In reviewing other similar studies, one needs to be concerned with the baseline se-verity of illness and comorbidity. The more severely ill one is, the more aggressive the treatment will be. Medication can erroneously be said to correlate with poorer results as it is well known throughout medicine that more severe illness begets more perturbed out-comes. Therefore, it is necessary to evaluate the baseline severity of the illness.

There were six studies evaluating risk of SUDs in ADHD individuals in treated vs. un-treated samples. Three studies followed youth into adolescence and three into adulthood. Our study was the only one to have a control group. In five of the studies (all with matched baseline groups) reduced risk for SA was ob-served in treated vs. untreated children with ADHD. Only one study of the six showed an increased risk of SUD with nicotine and cocaine, but did not control for baseline severity. All three studies with adolescents indicated a similar magnitude of risk reduction in treated vs. untreated subjects. 12% of 16 year-olds with treated ADHD had reduced SA compared to 30% in the untreated ADHD group.

The conclusions in the majority of studies indicate that ADHD youth treated with medications have reduced risk of SUDs. These data are particularly robust when treated and untreated groups are matched at baseline for severity of illness. These data are among the strongest in child psychiatry indicating the preventive influence of pharmacotherapy on the development of SUDs.

2000 Proceedings

2000 AAAP Annual Meeting Proceedings Copyright 2001 AAAP