This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow View responses
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Web of Science (63)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Barkley, R. A.
Right arrow Articles by Fletcher, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Barkley, R. A.
Right arrow Articles by Fletcher, K.
Related Collections
Right arrow Neurology & Psychiatry
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Facebook   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?

PEDIATRICS Vol. 111 No. 1 January 2003, pp. 97-109

Does the Treatment of Attention-Deficit/Hyperactivity Disorder With Stimulants Contribute to Drug Use/Abuse? A 13-Year Prospective Study

Russell A. Barkley, PhD*, Mariellen Fischer, PhD{ddagger}, Lori Smallish, MA{ddagger} and Kenneth Fletcher, PhD*

* Department of Psychiatry, University of Massachusetts Medical School, Worcester, Massachusetts
{ddagger} Department of Neurology, Medical College of Wisconsin, Milwaukee, Wisconsin

-->
   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 REFERENCES
 
Objective. To examine the impact of stimulant treatment during childhood and high school on risk for substance use, dependence, and abuse by young adulthood.

Methods. A total of 147 clinic-referred hyperactive children were followed approximately 13 years into adulthood (mean: 21 years old; range: 19–25). At adolescent (age 15) and adult follow-up, probands were interviewed about their use of various substances and duration of stimulant treatment.

Results. Duration of stimulant treatment was not significantly associated with frequency of any form of drug use by young adulthood. Stimulant-treated children had no greater risk of ever trying drugs by adolescence or any significantly greater frequency of drug use by young adulthood. Stimulant treatment in high school also did not influence drug use in adulthood except for greater use of cocaine. This difference was no longer significant after controlling for severity of attention-deficit/hyperactivity disorder and conduct disorder in childhood, adolescence, and adulthood. Stimulant treatment in either childhood or high school was not associated with any greater risk for any formal Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised drug dependence or abuse disorders by adulthood. Treatment with stimulants did not increase the risk of ever having tried most illegal substances by adulthood except for cocaine. Subsequent analyses showed that this elevated risk was primarily mediated by severity of conduct disorder by young adulthood and not by stimulant treatment in childhood.

Conclusion. This study concurs with 11 previous studies in finding no compelling evidence that stimulant treatment of children with attention-deficit/hyperactivity disorder leads to an increased risk for substance experimentation, use, dependence, or abuse by adulthood.

Key Words: hyperactivity • attention-deficit/hyperactivity disorder • stimulants • illicit drug use • drug abuse • conduct disorder

Abbreviations: ADHD, attention-deficit/hyperactivity disorder • CD, conduct disorder • SD, standard deviation • DSM, Diagnostic and Statistical Manual of Mental Disorders • CPRS-R, Conners Parent Rating Scale Revised • WWPARS, Werry-Weiss-Peters Activity Rating Scale • OR, odds ratio • SE, standard error


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 REFERENCES
 
Hyperactive children, or those now diagnosed as having attention-deficit/hyperactivity disorder (ADHD), manifest developmentally inappropriate levels of hyperactivity, impulsiveness, and inattention that typically have an onset in early to middle childhood, are relatively pervasive across settings, and produce impairment in major life activities.1,2 Across development, children who have hyperactivity or ADHD are at increased risk for learning disabilities and educational failure, impaired social relationships, and various psychiatric disorders, including conduct disorder (CD) and substance use disorders.3–12

One of the most effective and widespread treatments for hyperactivity/ADHD is stimulant medication.13–15 Approximately 74% to 97% of school-age children with ADHD respond positively to any given stimulant.13,16,17 These medications frequently result in marked reductions in hyperactive, impulsive, inattentive, and socially aggressive behavior while increasing vigilance, reaction time, task persistence, work productivity, working memory, handwriting, and fine motor speed and coordination more generally.13,14,18,19 The side effects are relatively benign, primarily involving appetite loss, insomnia, headaches, and stomachaches, although increased irritability, nervous mannerisms and tics, and significant weight loss may occur in a small fraction of treated children.13,16,17,20,21

The past 20 years have witnessed a marked increase in the use of stimulants for the management of ADHD.22,23 Large-scale studies suggest that 1.3% to 7.3% or more of school-age children, depending on geographic region, may be taking stimulants for behavior management24–27 with 1 locale (Norfolk, VA) purportedly having as many as 10% of their school-age population on stimulants at school.28 This rise has led to increased public concern that the United States may be overmedicating its children for behavior management.29–32 Several professional committees concluded, however, that there is little evidence to support concerns of overprescribing.33,34 Nevertheless, some social critics35 as well as the Church of Scientology’s Citizens Commission on Human Rights36 have capitalized on the public’s concern by suggesting that giving stimulants to children predisposes them to greater substance use and abuse as adolescents and adults.

The basis for this public concern may be two-fold. One is that stimulants such as methylphenidate may be chemically similar to cocaine and therefore have the potential for abuse and addiction, primarily when inhaled like cocaine or injected intravenously.37 Evidence shows, however, that the 2 drugs possess distinctly different pharmacokinetic properties. Methylphenidate enters and clears the brain more slowly than does cocaine, accounting for cocaine’s greater addictive potential.38 The second basis for concern is the potential for stimulant use to lead to increased sensitization to later stimulant exposure. Intermittent stimulant use may lead to an enduring, greater, and more rapid behavioral response on subsequent readministration.39,40 Studies in mammals suggest that repeated stimulant exposure leads to subsequently greater craving and self-administration of stimulants.41–45 Greenhill and Osmon,37 however, noted that there is little evidence of this occurring in humans. Such sensitization, they argue, seems to result from alternating between high interparenteral doses and drug-free periods between exposures. In contrast, stimulants that are given to children with ADHD are administered orally, in low doses, on a daily basis, and for long periods of time. This may explain why no cases of sensitization have been reported among stimulant-treated children with ADHD to date.

The evidence to date on the actual risks of substance use and abuse among stimulant-treated children with ADHD is relatively small. Twelve studies that addressed the issue could be located. With 1 exception, the studies have been consistent in showing that childhood stimulant treatment is not related to risk for later substance use or abuse, especially when comorbid CD is taken into account. Chilcoat and Breslau46 followed 717 children to 11 years of age and found that although ADHD was associated with an increased frequency of drug use by adolescence, particularly when associated with high levels of aggressive behavior and conduct problems, stimulant treatment of the ADHD children was unrelated to risk for drug use. The young age at follow-up may have limited the risk of exposure to abusable drugs, however. Loney et al47 followed 295 boys with ADHD from childhood to early adulthood. They found no relationship between stimulant treatment and extent of drug use for most drugs. In fact, they noted an actual protective effect such that medicated boys had less involvement with tobacco, stimulants, inhalants (glue), or opiates by follow-up. Medicated boys were also less likely to receive a diagnosis of alcoholism in adulthood. The study was more rigorous than that of Chilcoat and Breslau46 in using randomization of boys at study entry to either stimulant treatment or not, thus ensuring equivalence of the 2 groups in severity of disorder and comorbidity.

In a study of adolescents with ADHD, Biederman et al48 found that nonmedicated adolescents were at increased risk for substance use disorders at 4-year follow-up, whereas medicated adolescents were at a significantly decreased risk for such disorders. The study, nevertheless, addressed only risk of substance use disorders associated with psychopharmacological treatment generally at adolescence and not specifically stimulant treatment. However, the vast majority of medication treatment in this sample involved stimulants. Even so, the study is unable to address concerns about treatment during childhood as it affects risk of drug use later. There may be age-related differences in sensitization from stimulants such that exposure in childhood confers greater sensitization than does stimulant treatment in adolescence. Moreover, this study did not examine risk for drug use and abuse by adulthood and so cannot rule out a delayed effect of childhood treatment on adult drug use. Burke et al49 evaluated 177 clinically referred children with ADHD annually over 9 years until age 15 and assessed substance use at the teenage outcome. Although youths with ADHD were more likely to use tobacco, marijuana, alcohol, and other illicit drugs, this association became nonsignificant once childhood and adolescent CD were statistically controlled in the analyses. Important here was the finding that stimulant medication use in childhood was initially related to likelihood of tobacco use by adolescence but no longer was so once childhood and adolescent CD were taken into account in the analyses. Likewise, Milberger et al50 did not find any relationship between childhood treatments (medication and counseling) and adolescent tobacco use but did find adolescents who had ADHD and were still in treatment at adolescence to be less likely to use tobacco. Problematic here was the absence of any effort in the study to disentangle the relative effect of stimulant medication treatment in childhood on later tobacco use from the other forms of childhood treatment. In a longitudinal study of clinic-referred hyperactive children, Paternite et al51 did not find any association of stimulant duration or dosage to alcoholism or drug abuse disorder at young adulthood. However, analyses were not reported for the relationship of stimulant treatment variables to specific forms of drug abuse apart from alcohol permitting no conclusions about the relation of earlier stimulant treatment to later stimulant abuse. Earlier, smaller, and less rigorously conducted studies of the issue have likewise found no association of childhood stimulant treatment with later drug use or abuse.11,52–55

In contrast to all other studies of this issue, the only one to have found a relationship between stimulant treatment in childhood and later drug use or abuse was by Lambert and Hartsough.39,40 They found that stimulant treatment in childhood increased the risk for cigarette smoking in adulthood as well as the risk for nicotine dependence, cocaine dependence, and possibly stimulant dependence disorders. Consequently, this study receives more careful attention here in an effort to clarify why it conflicts with 11 others on this issue. In their initial report, Lambert and Hartsough39 subdivided just their children with ADHD into groups on the basis of various durations of stimulant treatment (never, 1 year or less, and >1 year). Stimulant treatment was associated with a greater likelihood of daily smoking in adulthood and cocaine substance dependence disorder, with a marginally greater risk for tobacco dependence as well. In a later report40 of additional analyses of these same samples, the children with ADHD were collapsed in with the large group of normal children before subdividing the groups on the basis of stimulant treatment history. The same results were obtained as before except this time the risk of cocaine dependence was now significant and there was now a marginally greater likelihood of stimulant dependence disorder in adulthood. The authors concluded that taking stimulants in childhood predisposes to nicotine use in adulthood and, to a lesser extent, cocaine and stimulant dependence, possibly supporting the stimulant sensitization hypothesis.40

Several problems seem to plague the Lambert studies39,40 that raise some doubt about these conclusions. The second report40 collapsed the clinic-referred hyperactive (ADHD) children together with the large normal control sample before re-sorting them into those who had been treated with stimulants as children and those who were not. Doing so mixed up multiple confounding variables with the independent variable of interest here (stimulant treatment), resulting in the vast majority of the stimulant-treated children also having ADHD and multiple comorbid disorders associated with ADHD, such as oppositional disorder, CD, learning disabilities, and major depression. For instance, only 6% of the 172 children who served as the control group had ever been treated with a stimulant, whereas 66% to 80% of the original hyperactive groups had been. Therefore, stimulant treatment becomes a veritable proxy for having ADHD and its related disorders, several of which are known to predispose to later substance abuse,49,56,57 apart from stimulant treatment. Both of the Lambert reports39,40 failed to examine and control for the impact of severity of current ADHD and especially current or lifetime symptoms of CD on drug use and abuse at adulthood. Studies have repeatedly found that severity of ongoing ADHD and particularly CD in adolescence and adulthood (as well as the relationship of CD to affiliation with deviant, often drug-using peer groups) are among the strongest contributors to risks for adolescent and young adult substance use and abuse.49,57–59

The present longitudinal study focused on the effect of stimulant treatment in childhood on substance use, dependence, and abuse by young adulthood. This study further examined and controlled for the contribution of current severity of ADHD and lifetime severity of CD symptoms to current and lifetime substance use and risk for dependence and abuse disorders in young adulthood. If the stimulant sensitization hypothesis is correct, then duration of stimulant use in childhood should have a significant relationship with extent of cocaine and other stimulant abuse by young adulthood, even after controlling for these other mediating variables.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 CONCLUSION
 REFERENCES
 
Participants
This study originally used a group rigorously diagnosed as hyperactive in childhood (N = 158) and a matched community control group (N = 81) followed concurrently. These 2 groups were originally evaluated in 1979 to 1980 when they were ages 4 to 12 years. The majority of these participants (hyperactive, N = 123; normal, N = 66) were evaluated again in 1987 to 1988 when they were ages 12 to 20 years4 and then again in 1992 to 96, at which time all were between 19 and 25 years of age (mean: 21 years.). All of the participants in both groups were able to be located. The participation rate at this young adult follow-up was 93% (147 of 158) for the hyperactive group and 90% (73 of 81) for controls. One control subject died before the adolescent follow-up of a sudden cardiac arrest, and another died before this follow-up in a car accident. One hyperactive subject died before this follow-up by suicide. The present article uses just the clinic-referred hyperactive group because none of the control subjects had ever been treated with stimulant medication.

At childhood entry into the study, all participants were required to 1) have an IQ >80, 2) be free of gross sensory or motor abnormalities, and 3) be the biological offspring of their current mothers or have been adopted by them shortly after birth. All parents signed statements of informed consent for their own and their child’s participation in the study. The gender composition was 91% male and 9% female. The racial composition was 94% white, 5% black, and 1% Hispanic.

The hyperactive group was originally recruited from consecutive referrals to a child psychology service that specialized in the treatment of hyperactive children at Milwaukee Children’s Hospital. To be considered hyperactive and eligible for the project, these children had to 1) have scores on both the Hyperactivity Index of the Revised Conners Parent Rating Scale60 and the Werry-Weiss-Peters Activity Rating Scale61 that met or exceeded 2 standard deviations (SDs) above the mean for severity for same-age, same-gender normal children; 2) have scores on the Home Situations Questionnaire2 indicating significant behavioral problems in at least 6 of the 14 problem situations on this scale (a score exceeding +1 SD according to subsequently collected norms); 3) have parent and/or teacher complaints (as reported by parent) of poor sustained attention, poor impulse control, and excessive activity level; 4) have developed their behavior problems before 6 years of age; 5) have had their behavioral problems for at least 12 months; and 6) have no indication of autism, psychosis, thought disorder, epilepsy, gross brain damage, or mental retardation. In view of these selection criteria and the close convergence of rating scale diagnoses with the clinical diagnosis of ADHD,62 it seems highly likely that all of these subjects would have met criteria for ADHD based on Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised, or Fourth Edition (DSM-III-R or DSM-IV) guidelines (Hyperactive or Combined Types) had those been available. Indeed, >70% of them met the DSM-III-R criteria for ADHD 8 to 10 years later at the adolescent follow-up.4 Although pervasiveness of symptoms across home and school settings was not required for this study, the vast majority of children were experiencing problems in both settings. Pervasiveness in the home setting was systematically assessed using the Home Situations Questionnaire on which an explicit threshold to enter the study was specified (see above). The proportion of boys was 87% in the hyperactive group. The proportion of the hyperactive group still residing with their parents was 52% at the young adult follow-up.

Procedures
The participants were evaluated at both the adolescent (mean: 15 years of age) and the young adult follow-up (mean: 21 years of age) using measures collected in a single day for each of these evaluations. At the adolescent follow-up,4 parents were interviewed about their adolescent’s disruptive behavior disorders (ADHD, oppositional defiant disorder, and CD), history of mental health and medication treatment, history of antisocial behavior and illicit drug use, and educational history. Rating scales concerning behavioral adjustment were also collected from parents and adolescents, and psychological tests were administered to the adolescents. The adolescents were also interviewed about whether they had ever used various illicit substances.

At the young adult evaluation, the participants were evaluated using a battery of measures that assessed self-reports of psychiatric disorders, history of mental health treatments, history of adaptive functioning in major life activities (education, occupation, dating, sexual activity, driving, etc), and history of antisocial activities and drug use. Again, a battery of psychological tests and rating scales were collected. Parents were interviewed separately by telephone concerning only the participant’s current ADHD symptoms. All evaluations were conducted by a psychological assistant who held a master’s degree (L.S.) and was supervised by a board-certified clinical neuropsychologist (M.F.). This assistant was not blind to the original group membership of the participants as being in the hyperactive or control group. The present article focuses on the potential contribution of treatment with stimulant medication during childhood and adolescence in the hyperactive group to the self-reported use of illicit drugs at adolescence and adulthood and to the presence of drug dependence/abuse disorders in young adulthood.

Measures of Stimulant Medication Use
At the adolescent follow-up (mean: 15 years of age), parents were interviewed about the types of stimulant medication that their children had ever received and the duration (in months) that the children had been treated with these medications.4 Of the participants evaluated at that time, information on medication use was available for 119 of the 123 adolescents in the hyperactive group and 63 of the 66 adolescents in the control group who participated in that evaluation. None of the control children had been treated with stimulant medication. Among the hyperactive group, 96 (80%) had received methylphenidate, 4 (3%) had received d-amphetamine, and 24 (20%) had received pemoline sometime in childhood. Some children had been treated with more than 1 stimulant. Of the children who had received methylphenidate, 2 (2%) had also received d-amphetamine and 22 (22%) had received pemoline. All 4 of the children who had received pemoline had also received d-amphetamine. In total, 98 hyperactive children had received stimulants and 21 had not. The duration of treatment with each of these medications was as follows: methylphenidate (mean: 44.8 months; SD: 40.4), d-amphetamine (mean: 32.8 months; SD: 58.3), and pemoline (mean, 13.3 months; SD: 19.8). The total duration of stimulant use for the treated children was 40.2 months (SD: 43.3) with a range up to 156 months.

At the young adult follow-up, participants were asked whether they had ever been treated with any prescribed stimulant during high school and, if so, for how long (in months). Of the hyperactive group (N = 147), 32 (22%) reported being treated with stimulants. All of them had also been treated with stimulants in childhood. The proportion of hyperactive children who had taken each specific stimulant was not recorded, however. The duration of stimulant treatment was reported to be a mean of 26.6 months (SD: 20.0). Only 7 of the hyperactive group (5%) were currently being treated with stimulants at the young adult follow-up.

Measures of Childhood ADHD and Conduct Problems
The following parent ratings were collected at the childhood entry point into this project. They are used here as measures of childhood ADHD symptoms and conduct problems in the analyses reported below.

Conners Parent Rating Scale Revised
The Conners Parent Rating Scale Revised (CPRS-R)60 is a 48-item scale that is among the most widely used rating scales in the history of research on children with hyperactivity/ADHD.63 Each item is rated on a 4-point Likert scale (0–3 for not at all, just a little, pretty much, and very much). The scale assesses 5 behavioral factors: conduct problems, learning problems, psychosomatic, impulsive-hyperactive, and anxiety. A 10-item Hyperactivity Index is also computed and is believed to represent the most frequently occurring items in children with hyperactivity. Scores for each of the factors and the index are determined by summing the responses across all items for that factor and then dividing by the number of items to get the mean response. The Hyperactivity Index of this scale was used at the childhood study entry point to select subjects to be in the hyperactive group, as noted above. Most of these items pertain to hyperactive, impulsive, and inattentive behavior. For the present study, the Conduct Problems factor (12 items) was also used to evaluate severity of childhood conduct problems for use in these same regression analyses below. Items from this factor are a mixture of oppositional defiant disorder–and CD-related symptoms, such as sassy, carries a chip on their shoulder, destructive, lies, steals, disobedient, does not like to follow rules, bullies others, etc.60

Werry-Weiss-Peters Activity Rating Scale
The original 31-item Werry-Weiss-Peters Activity Rating Scale (WWPARS)61,63 was developed to evaluate children’s levels of hyperactive behavior in home and school situations64 but was subsequently modified to a 22-item scale for use with parents by Routh et al65 in which the school items were deleted. The modified scale was used here at study entry to select the hyperactive children on the basis of a threshold of +2 SDs above the mean for a small sample of normal children (N = 140) studied by Routh et al.65

Measures of ADHD, CD, and Drug Use at the Adolescent Follow-up
Parents were interviewed about their adolescent’s ADHD and CD using a structured interview based on DSM-III-R66 criteria created just for this adolescent follow-up project.4 For the present article, the number of symptoms of ADHD and CD reported by parents served as measures used in the various analyses described below. The coefficient {alpha} for the 14 ADHD symptoms was 0.91 and for the 13 CD items was 0.75. Both parents and adolescents were also interviewed separately concerning whether they had ever used any of the following drugs: cigarettes; alcohol; marijuana, hashish; cocaine; heroin; hallucinogens; unprescribed stimulants, sedatives, and tranquilizers. The frequency of use of these substances was not recorded, however. Adolescents’ reports are used here because adolescents reported more use of substances than did parents,4 a difference in perspective found in other studies of drug use among adolescents with ADHD.67

Measures of Young Adult ADHD, CD, Drug Use, and Drug Use Disorders
These measures were collected at the current young adult outcome evaluation.

Structured Interview of Disruptive Behavior Disorders
Participants were interviewed concerning the frequency with which they had engaged in a diversity of antisocial activities. Ten of these activities corresponded to items from the CD list of symptoms in DSM-III-R. Items not obtained were deliberately destroyed property, physical cruelty to animals, and physical cruelty to people. The DSM-III-R items pertaining to "often lies" and "often truant" were considered affirmatively endorsed when reported as occurring "frequently" or more often. The item pertaining to "often initiates physical fights" was considered to be met when it occurred 5 or more times (the mean for the control group). As in the DSM-III-R, stealing and running away from home had to occur at least twice or more to be considered as endorsed. For this article, the total number of CD items endorsed by the participant was used in the regression analyses described below. With the use of this procedure, the evaluation of CD items reflects lifetime occurrence. The internal reliability (coefficient {alpha}) for these 10 CD items was 0.64. With the use of DSM-III-R criteria, 27% of the hyperactive group met the symptom threshold for CD. The average number of CD symptoms was 1.9 for the hyperactive adolescents (SD: 1.5).

Parent Interview of ADHD Symptoms
This follow-up evaluation had been under way for several years when it became apparent that self-reports concerning ADHD symptoms might be substantially different from the reports of others, particularly their parents, about this disorder. The New York follow-up study found such disparities in their own subject interviews relative to parent interviews at their late adolescent follow-up point.68 Also, at this time, the criteria for ADHD in the DSM-IV1 were published. A structured interview was therefore created from the DSM-IV item list to be used with parents. The parents, mostly mothers, of nearly all participants in this study (134 hyperactives and 71 controls) were located and then contacted by telephone and interviewed about the extent to which the subject showed current levels of the 18 items for ADHD. Parents were asked about each item, verbatim, concerning the current functioning of the participant. The internal reliability (coefficient {alpha}) was 0.92 for the DSM-IV inattention item list and 0.91 for the hyperactive-impulsive list.

The parents endorsed significantly more current symptoms of ADHD in the hyperactive group than did the participants themselves. For the hyperactive group, parents reported an average of 9.2 symptoms (SD: 5.9), whereas for the control group, the mean was 1.7 (SD: 2.6; t = 10.23; df = 203; P < .001). As a result, 58% of the hyperactive group met criteria for any subtype of ADHD in young adulthood as compared with just 7% of the control group ({chi}2 = 50.43; df = 1; P < .001). Because parents reported more current symptoms of ADHD than did the probands and these reports were related to substantially more domains of current life impairment than were self-reports, the parental reports were believed to have greater veracity in reflecting current levels of proband ADHD.7 Therefore, parent reports are used below in the regression analyses.

Structured Clinical Interview for DSM-III-R Disorders (Nonpatient Edition)
The Structured Clinical Interview for DSM-III-R Disorders (Nonpatient Edition)69 is used to make Axis I and Axis II diagnoses from the DSM-III-R. The Structured Clinical Interview for DSM-III-R Disorders (Nonpatient Edition) provides an interview that covers the diagnostic criteria for mood disorders (major depression, mania, dysthymia), psychotic and associated symptoms, psychoactive substance use disorders, anxiety disorders, somatoform disorders, eating disorders, and adjustment disorder. Only the results for psychoactive substance use disorders are reported here. Administration procedures followed those set forth in the manual associated with these interviews. Information on interjudge agreement for the diagnoses was not collected in this project. This interview has been used extensively in research on adult mental illness.

Structured Interview on Current Illicit Drug Use at Adulthood
All participants were interviewed at the young adult outcome as to their frequency of use of alcohol and illicit drugs. Use of cigarettes or other nicotine products was not obtained, however. Specifically, participants were interviewed about 1) their typical consumption of alcohol in a typical week, 2) the number of times they had gotten drunk in the past 3 months, and 3) the number of times they had used illicit drugs during the past 3 months. They were also interviewed about the number of times they had taken each of a number of illicit drugs, including marijuana, cocaine, acid/LSD/mescaline, stimulants, narcotics, sedatives/barbiturates, and all other drugs.

A few participants (up to 7% of hyperactives) reported using some drugs so extensively that they could not estimate how often they had done so. With the exception of marijuana, participants who reported that this extreme level of use exceeded at least 100 times were recorded with a special code in the database. For the present article, these extreme scores were then recoded to the highest actual frequency endorsed by any of the other participants for that drug plus 1. This method certainly underestimated the actual frequency of drug use among these outliers yet retained them for use in the analyses while preserving their relative rank within the total pool of subjects. Excluding such outliers from the analyses would have removed important information concerning the most extreme drug users in the sample and biased the results against finding any potential link between stimulant treatment and drug use.

Marijuana was used even more extensively than the other illicit drugs. A small proportion of subjects (hyperactives = 33 [22%]) reported such frequent use of this drug that it exceeded 1000 times. As a result, these outliers were coded in the database with a special code. Again, for the analyses below, these outliers were recoded to the highest actual reported frequency in the sample plus 1, for the same reasons given above.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 CONCLUSION
 REFERENCES
 
Relationship of Childhood Stimulant Use to Adolescent Self-Reported Drug Use
The hyperactive group was subdivided into those who had (N = 98) and had not (N = 21) been treated with stimulants in childhood, as reported by parents at the adolescent follow-up. These 2 groups were compared using {chi}2 analyses on the proportion who self-reported whether they had ever used any of 10 illegal or illicit drugs. The 2 groups did not differ significantly in their frequency of having tried any of these drugs: cigarettes (52% [treated] vs 30% [untreated], alcohol (41% vs 35%), marijuana (20% vs 6%), hashish (7% vs 6%), cocaine (5% vs 0%), heroin (0% both groups), hallucinogens (3% vs 0%), unprescribed stimulants (6% vs 6%), unprescribed sedatives (3% vs 0%), and unprescribed tranquilizers (1% vs 0%). All {chi}2 tests were nonsignificant. Given that the sensitization hypothesis pertains to all forms of stimulant use or abuse, we combined cocaine use with that of amphetamines or speed and computed whether each participant had ever tried any form of these stimulants. The groups did not differ on this outcome either (6% vs 10%).

The reports by Lambert39,40 had also evaluated the contribution of stimulant treatment to drug use based on children who had received stimulants for >1 year in comparison with those who had taken them for <1 year. We therefore subdivided the hyperactive sample again into these 2 categories of childhood stimulant treatment (N = 43 treated <=1 year; N = 55 treated >1 year). Again, there were no significant differences between the groups on the proportion of adolescents who had tried each of the aforementioned medications or the composite measure of ever having tried cocaine or unprescribed amphetamines (9% vs 9%).

Contribution of Childhood Stimulant Treatment to Adult Self-Reported Substance Use
Pearson correlation coefficients were computed between the parent-reported total duration of stimulant treatment in childhood and the frequency of self-reported drug use in young adulthood on the 10 different measures of drug use. The entire hyperactive group (N = 119) was used for these analyses. None of the correlation coefficients reached or even approached the traditional level of significance (P < .05). The correlation coefficients ranged from -0.11 to 0.15. The hyperactive sample was subdivided into those who did and did not receive stimulant medication treatment anytime in childhood (N = 98 and 21, respectively). The raw frequencies of current alcohol use and lifetime use of various drugs are shown in Table 1. As can be seen, the SDs for most frequencies greatly exceeded the mean scores, particularly among the larger treated group. This would likely result in sufficiently skewed distributions that may potentially violate the assumptions of analysis of variance. To address this problem, we transformed all frequency measures using Log10 (score + 1). These transformed scores are shown in Table 1 as well. We then compared the transformed scores for the 2 groups using analysis of variance. The 2 groups did not differ significantly on any of these current measures of drug use. However, there was a marginally significant finding of increased cocaine use among the treated children (P = .059). We then reanalyzed this measure using analysis of covariance in which childhood ADHD symptoms (sum of WWPARS and Conners Index) and conduct problems (Conners factor score), adolescent ADHD and CD symptoms from DSM-III-R, and current ADHD and lifetime CD symptoms at adulthood served as covariates. Because fewer hyperactive individuals were available at the adolescent follow-up than adult follow-up, this use of variables collected at adolescence reduced the sample sizes to 16 untreated and 84 treated subjects. The resulting main effect for stimulant treatment was no longer significant (P = .165), but the analysis did reveal a significant contribution of lifetime CD symptoms by adulthood to frequency of cocaine use (F = 9.93, df = 1/92, P = .002).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Frequency of Drug Use as a Function of Stimulant Treatment in Childhood

 
These same 2 groups were then compared on the proportion of each that had ever used any of the substances listed in Table 1. The results are shown in Table 2. {chi}2 analyses were used for these comparisons. The 2 groups did not differ on any of these substances except for cocaine, for which more children who had been treated with stimulants had tried cocaine at least once. However, on the composite measure of ever having tried any stimulant (cocaine or speed/amphetamines), as Table 2 shows, the groups did not differ. Cocaine use substantially overlapped with speed/amphetamine use such that 58% of those who had ever tried cocaine had tried another stimulant, whereas just 7% of those who had not used cocaine had used another stimulant ({chi}2 = 44.87; df = 1; P < .001; E = 0.552). This comorbidity was significant in both the untreated (100% vs 0%) and treated cases (60% vs 5%).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Likelihood of Using Drugs by Adulthood as a Result of Stimulant Treatment in Childhood

 
We then used binary logistic regression analysis to determine whether stimulant treatment in childhood continued to contribute significantly to the risk of ever trying cocaine after controlling for severity of ADHD and CD symptoms. The following measures were entered into the regression equation: severity of childhood ADHD (sum of CPRS-R Index and WWPARS) and conduct problems (parent-rated CPRS-R Conduct) and severity of current ADHD (parent-reported) and CD symptoms (self-reported). Stimulant treatment was entered as a dimensional variable (duration in months). Results indicated that duration of stimulant treatment in childhood made no significant contribution to risk for cocaine use after controlling for these other potential confounding factors. This was also true when stimulant treatment was entered as a categorical variable (no/yes) rather than a dimensional one. Indeed, these analyses indicated that number of CD symptoms in young adulthood was the only significant contributor to the risk of ever trying cocaine by adulthood (ß= 0.819; standard error = 0.211; Wald = 15.07; df = 1; P < .001; odds ratio [OR] = 2.27).

The untreated and treated hyperactive groups were then compared on the proportion that received each type of DSM-III-R drug abuse or dependence diagnosis. The following drug disorders occurred in <6% of either group with their being no significant differences between the groups: sedative abuse/dependence, stimulant abuse/dependence, opiate abuse/dependence, cocaine abuse/dependence, hallucinogen abuse/dependence, and polydrug abuse/dependence. These groups also did not differ in alcohol abuse (14% [untreated] vs 9% [treated]) or dependence (19% vs 24%), or cannabis abuse (10% vs 7%) or dependence (19% vs 25%).

Once again, the hyperactive probands were subdivided on the basis of those who had (N = 66) and had not received stimulant medication for at least 1 year (N = 53). All of the above analyses involving group comparisons were repeated again using the transformed frequency scores. We found no significant group differences for any measure. There were also no significant group differences for the proportion of each group that had ever tried alcohol or any of the other drugs or for 14 of the 16 drug dependence or abuse disorders examined above. Hyperactive children who had been treated for <1 year with stimulants had a slightly but significantly greater likelihood of having a cocaine abuse disorder (6% vs 0%; {chi}2 = 3.83; df = 1; P = .05, E = 0.179) and a hallucinogen/PCP abuse disorder (9% vs 2%; {chi}2 = 3.85; df = 1; P = .05; E = 0.180) than probands who had been treated for >1 year. This could suggest a potentially protective effect of stimulant treatment as a child for risk for these 2 drug abuse disorders.

For evaluating this possibility further, logistic regression analyses were computed in which stimulant treatment was evaluated for its contribution to risk for each of these disorders while controlling for severity of child ADHD symptoms and conduct problems and adult ADHD and CD symptoms. For risk for cocaine abuse, no variables made a significant contribution after controlling for all others, suggesting no significant protective effect of using stimulants for >1 year. For hallucinogen abuse, 2 variables contributed significantly to the risk. Severity of CD symptoms had a significant association, increasing the risk for abuse of this drug (ß = 0.841; SE = 0.423; Wald = 3.95; df = 1; P = .047; OR = 2.32). In contrast, stimulant treatment for >1 year in childhood significantly decreased risk for hallucinogen abuse (ß = -2.762; SE = .1.302; Wald = 4.50; df = 1; P = .034; OR = 0.063).

Contribution of Stimulant Treatment in High School to Young Adult Drug Use
The same analyses reported above were then repeated for those hyperactive probands who had (N = 32) or had not (N = 115) been treated with stimulants in high school (as self-reported at the young adult follow-up). First, correlations were initially computed between duration of stimulant treatment in high school and the 10 measures (raw scores) reflecting frequency of alcohol and drug use (as listed in Table 1) using the entire hyperactive sample (N = 147). None of the 10 correlations was significant (range r = -.019 to .094). Then the untreated and treated groups were compared in their frequency of drug use on these same transformed measures, the results of which are shown in Table 3. One of these comparisons was significant, and that was again for frequency of cocaine use. Those who had received stimulant treatment used cocaine more frequently (P = .043). Once again, we repeated this analysis using childhood ADHD and conduct problems, adolescent ADHD and CD symptoms, and adult ADHD and lifetime CD symptoms as covariates. The main effect for stimulant treatment was no longer significant (P = .062). Once again, lifetime CD symptoms made a significant contribution to frequency of cocaine use (F = 9.29; df = 1/92; P = .003).


View this table:
[in this window]
[in a new window]
  		TABLE 3. Frequency of Drug Use Relative to Stimulant Treatment in High School

 
The two groups were then compared on the proportions that had ever taken any of these substances. Those results appear in Table 4. Again, the groups differed significantly only in the proportion that had ever tried cocaine and, as a result, those who had ever tried any stimulant. Hyperactive individuals who had been treated with stimulants in high school were more likely to have ever tried cocaine. As above, a logistic regression analysis was then used to determine whether stimulant treatment in high school contributed to risk for any cocaine use after controlling for severity of ADHD and conduct problems in childhood and ADHD and CD at adolescence and at adult outcome. When stimulant treatment was treated as a dimensional variable (duration in months), it did not make a significant contribution in this analysis. The contribution of both current severity of ADHD and CD symptoms was marginally significant (P = .07). However, when stimulant treatment in high school was treated as a categorical variable (no/yes), the contribution of both lifetime CD symptoms (ß = 0.859; SE = 0.268; Wald = 10.28; df = 1; P = .001; OR = 2.36) and high school stimulant treatment (ß = 1.41; SE = 0.637; Wald = 4.94; df = 1; P = .026; OR = 4.11) was significant.


View this table:
[in this window]
[in a new window]
  		TABLE 4. Likelihood of Using Any Drug by Adulthood as a Result of Stimulant Treatment in High School

 
Finally, the 2 groups were compared in the proportion who met criteria for any of the drug dependence or abuse disorders (DSM-III-R). As before, no significant group differences emerged in any of these comparisons. The frequency of all disorders except alcohol and cannabis dependence placed below 10% in both groups. For alcohol dependence, the results were 20% (untreated) versus 34% (treated), whereas for cannabis dependence, they were 21% and 34%, respectively.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
CONCLUSION
 REFERENCES
 
The results of the present >13-year longitudinal study of hyperactive children followed into young adulthood (age 20–21 years) finds very little support for the sensitization hypothesis or the assertion that treatment with stimulant medication, either in childhood or in adolescence, contributed to a significant risk for lifetime substance use, dependence, or abuse. These results are consistent with 11 previous studies (see above) that likewise found no association of stimulant treatment to an increased risk of later drug use of any sort among children with ADHD but particularly for stimulants or cocaine. We found that the duration of stimulant treatment in childhood and in high school was not significantly related to any measures of current alcohol use or drug use or lifetime frequency of drug use. It specifically was not related to stimulant use (amphetamines or cocaine). We also found that whether a hyperactive child had been treated with stimulants was not associated with the likelihood of ever having used tobacco, alcohol, marijuana, cocaine, stimulants, or other illegal or illicit substances by midadolescence. Whether children had been treated with stimulants, either in childhood, or in high school, was not associated with risk for their ever having used these substances by young adulthood, with 1 exception. That exception was the risk of using cocaine. There, we found that stimulant treatment in childhood or in high school increased the risk of ever using cocaine. High school stimulant treatment was also associated with a greater frequency of cocaine use.

At first blush, these results might support the stimulant sensitization hypothesis. Additional analyses, however, showed that this relationship between childhood or high school stimulant treatment and risk of ever using cocaine was mediated by severity of lifetime CD symptoms. Controlling for severity of childhood, adolescent, and adult ADHD and CD resulted in childhood stimulant treatment’s no longer contributing to the risk of ever using cocaine and high school stimulant treatment’s no longer contributing to the frequency of using cocaine. Only when considering stimulant treatment in high school did it make a significant contribution to this risk even after controlling for childhood, adolescent, and current ADHD and CD. Finally, this study found no relationship between stimulant treatment, either in childhood or high school, and risk for any DSM-III-R substance dependence or abuse disorders in adulthood.

So as to better compare our results to those of Lambert and Hartsough, 39,40 who found such an association, we re-sorted the hyperactive sample into those who had and had not been treated with stimulants for 1 year or more as they had done. No significant differences were found between these groups for risk of trying any substances as an adolescent, including cigarettes; for frequency of use of any substances by young adulthood, including cocaine; for risk of ever having tried these substances by young adulthood; or for risk of having 14 of the 16 drug dependence and abuse disorders evaluated in young adulthood. Only 2 differences were evident among all of these analyses. Children who had been treated with stimulants for >1 year were significantly less likely to have a cocaine abuse or a hallucinogen/PCP abuse disorder in young adulthood. This finding seems to contradict the Lambert study.40 It also suggests the possibility of a potential protective effect against these 2 disorders from stimulant treatment sustained for >1 year as a child. Additional analyses were done to control for severity of childhood ADHD and conduct problems and adult ADHD and CD symptoms while examining the contribution of stimulant treatment to each risk. For cocaine abuse disorder, stimulant treatment was no longer protective of that risk. For hallucinogen abuse disorder, however, we found that whereas severity of CD symptoms significantly increased the odds of having such a disorder, stimulant treatment for 1 year or more significantly decreased the odds. Thus, there may be some protective effect of stimulant treatment against developing this hallucinogen abuse. However, the risk for either disorder was small, and these 2 group differences were the only ones that were significant out of all of the multiple comparisons of these treatment groups. It is therefore possible that these may be chance findings (type I errors), hence the need for future research on this issue of a protective effect against abuse of hallucinogens.

These findings seem to conflict with the conclusions drawn in the Lambert and Hartshough reports39,40 that stimulant treatment for 1 year or longer in childhood leads to sensitization such that there is a greater risk of cocaine and stimulant dependence disorders. Indeed, the most direct evidence against this hypothesis came from the simplest of our analyses. There was no significant correlation between duration of childhood or high school stimulant treatment and frequency of any form of drug use. One would expect that the longer children had stayed on stimulants, the greater their risk for such sensitization. This did not occur in our study. Stimulant treatment for 1 year or longer also was not associated with risk for substance use or abuse in any form. The Lambert study did find a greater likelihood of adult smoking and nicotine dependence disorder among stimulant-treated children. The present study did not document smoking activities to the same extent as the Lambert study, so the relationship of stimulant treatment to this type of drug use could not be fully evaluated here. Nevertheless, at least by adolescence (mean: 15 years of age), stimulant-treated children were not more likely to have ever smoked than untreated children, those who had been treated for 1 year or more were not significantly more likely to have ever smoked than those hyperactive children who had been treated with stimulants for <1 year.

This study, however, did document a significant relationship between being treated with stimulants in high school and risk of ever trying cocaine. That risk remained significant even after controlling for severity of childhood ADHD and conduct problems, adolescent ADHD and CD, and adult ADHD and CD symptoms. It is certainly possible that this reflects a sensitizing effect of stimulants resulting in increased risk of ever trying cocaine. We believe that there are good reasons from other results in this study to pose strong reservations about that conclusion. First, there was no significant relationship between duration of either childhood or high school stimulant treatment and frequency of cocaine use, directly contradicting the sensitization hypothesis. Second, the duration of childhood or high school stimulant treatment did not make a significant contribution to the risk of ever trying cocaine. Third, neither childhood or high school stimulant treatment status was associated with any greater risk for cocaine dependence or abuse disorders—only with ever having used cocaine at least once. Consequently, only when high school stimulant treatment was treated as a categorical variable (yes/no) in the regression analysis and only when cocaine use was studied categorically (yes/no) was a significant contribution evident for treatment status to the risk of ever trying cocaine. This singular finding that was not evident in any other analyses of cocaine use and abuse suggests that some other variable associated with stimulant treatment status in high school may mediate this risk of ever using cocaine. One that has been suggested in recent research is affiliation with deviant, substance-abusing peers.70 That this was 1 of only 2 significant group differences to emerge out of all of the statistical analyses computed for treated versus untreated children also creates the possibility that those may be simply chance findings. In fact, had we applied a more conservative level of significance (say, P < .01) or a family-wise Bonferroni correction to (reducing P to <.005) to control for type I errors, none of these findings would have reached these more conservative levels of significance. Clearly, more research is needed on this issue of high school stimulant treatment and risk of ever trying cocaine.

The present study also provides a possible explanation for the marked disparity between the Lambert study and the present study apart from other differences in methods, such as different sources of subjects (school/community versus clinic referral, respectively). As noted earlier, 1 of the Lambert reports40 collapsed its normal control group in with its ADHD group before re-sorting the entire sample on the basis of stimulant treatment. However, only 6% of the control children had ever been treated with stimulants, whereas up to 80% of the children with ADHD had been so treated in childhood. Such a procedure automatically confounds stimulant treatment status with the diagnosis of ADHD and its severity and associated comorbid disorders. As a result, when stimulant-treated and -untreated children are compared with each other, the stimulant-treated group has a vastly disproportionate share of the ADHD participants and consequently a greater share of the CD individuals. Both of these diagnoses, not to mention other variables associated with them (deviant peer groups, substance-using and -abusing parents, etc.), pose confounding variables that could contribute to the higher risk found by Lambert for adult smoking and for cocaine, nicotine, and stimulant dependence disorders apart from treatment in childhood with stimulants. The present study addressed this issue by examining only hyperactive children separately from our control group and then comparing subsets of stimulant-treated and -untreated probands in lifetime drug use. We then went further and statistically partialled out the influence of childhood, adolescent, and adult ADHD and CD/conduct problems on any differences that were evident. Doing so resulted in our finding no association between stimulant treatment and lifetime use of any substance or risk for having used most substances except for ever trying cocaine in relation to high school stimulant treatment, as discussed above. Although the Lambert40 study attempted to statistically control for some of these confounding factors (severity of childhood ADHD and childhood conduct problems), it failed to examine others that were more important.

The Lambert40 study noted that severity of childhood conduct problems was not related to risk of smoking or of drug use disorders in adulthood. This was true in our sample in which most correlations between parent-rated childhood conduct problems and lifetime drug use were nonsignificant except for marijuana (r = 0.14; N = 219; P = .043) and amphetamine use (r = 0.16; N = 219; P = .017). Not addressed in either of the Lambert reports39,40 were variables known to have a high association with drug dependence and abuse by adulthood, particularly severity of CD symptoms through adolescence and young adulthood. For example, self-reports of lifetime CD symptoms in this study were significantly related to all 10 forms of substance use assessed here (r range 0.16–0.53; N = 220; all P <= .001 except narcotics, P = .019). As the present study showed when the treatment group differences in risk for cocaine use were evaluated further, after controlling for severity of lifetime CD symptoms, childhood stimulant treatment status no longer makes a significant contribution. Previous studies49,71 also found this to be the case in that severity of CD symptoms mediated all significant relations of ADHD to various types of substance use, except tobacco use, for which the inattentive dimension of ADHD made a significant contribution to risk. In our entire sample, severity of adolescent ADHD, oppositional disorder, and CD all are significantly correlated with duration of stimulant treatment (r range = 0.36–0.43; all P < .001; N = 169) as was severity of current ADHD (r = 0.39; P < .001; N = 175) and CD in adulthood (r = 0.17; P = .02; N = 182). Also, twice as many adolescents in the stimulant-treated ADHD group qualified for a diagnosis of CD at adolescence than did the untreated ADHD group (49% vs 25%; {chi}2 = 3.74; df = 1; P = .05). Therefore, the failure to control for adolescent or adult ADHD and CD symptoms left the Lambert reports with potentially confounding variables in their comparisons of stimulant-treated and -untreated children.

The present study, as noted above, could not test for the association of stimulant treatment with adolescent or adult smoking frequency or adult nicotine dependence disorder as did Lambert. Only the risk of adolescents’ trying smoking could be studied here, and it was unrelated to childhood stimulant treatment. There remains the possibility that childhood stimulant treatment may predispose to nicotine use or dependence in adulthood, which was not examined here. However, this can be determined only after severity of adolescent and adult CD and ADHD symptoms have been taken into account–something not done in the Lambert reports.39,40 Burke et al,49 among others,71 found that severity of adolescent inattention was related to risk for tobacco use in addition to adolescent CD symptoms, so both disorders at adolescence must be controlled in any attempt to study stimulant treatment and risk for tobacco use. The Lambert findings on nicotine use therefore seem to remain in some doubt until the issue can be investigated further.

The limitations of the present study surely deserve consideration in interpreting its results. We did not attempt to verify the reports of the parents or participants through medical records as to the type, dosage, and duration of medication use. These results are based solely on the subjective reports of these individuals, which could be inaccurate in some respects. Nevertheless, this same procedure was used in all previous studies, so it is not likely to affect the comparison of these results to those studies. Another limitation was the relatively small sample of hyperactive children who had not been treated with stimulants in childhood (N = 21) as well as the relatively small sample of this same group that had received stimulant treatment in high school (N = 32). Such sample sizes may have limited the statistical power of the present study to detect small to moderate effect sizes in the various group comparisons. Nevertheless, when the relationship of the duration of stimulant treatment in childhood or high school using all hyperactive subjects (N = 119 and 147, respectively) was examined, no significant associations were found. This suggests that any relationship of drug use to stimulant treatment is of an exceptionally low magnitude. Also, when children who had been treated with stimulants for 1 year or more were compared with those who had received less treatment, sample sizes in each group were larger and more evenly balanced, yet no significant contribution of stimulant treatment to substance use, dependence, or abuse could be documented.

Other qualifications pertain to how CD was assessed and then used in the analyses. Lifetime CD by adulthood was determined by converting self-reported frequencies of 10 of the 13 CD symptoms and not by structured DSM-III-R interview for all 13 CD items and thus serves only as a proxy for the diagnosis of CD. However, all DSM-III-R symptoms were assessed at adolescence and were used as a control in the analyses as well. Also, we treated CD in our analyses as though it were a potential confound that is likely to precede and contribute to later substance use and abuse. It is possible that early substance use and abuse precedes and contributes to CD. Our approach was predicated on research that shows that CD in fact precedes the onset of and contributes to the risk for substance use and abuse72,73 while recognizing that once developed, adolescent substance use and abuse can interact with and further exacerbate later antisocial conduct.57,59,74 Because we did not collect data on the onset of substance use and abuse, we cannot directly examine this issue. However, the mean age of onset of CD symptoms (DSM-III-R) in our hyperactive group at the adolescent follow-up by parent report was 8.2 years (SD: 2.3),4 in keeping with other studies of hyperactive children,72,73 which is considerably earlier than the average age at which substance use and dependence is likely to develop (13–16 years.).72

This study also used a clinic-referred sample rather than a community/school-based sample like that of Lambert and Hartsough, which could have led to some differences in our results. Nonetheless, clinic-referred samples of ADHD children are often more severe cases with higher rates of comorbid disorders, which may have a higher proportion receiving stimulant treatment and for longer periods. All of this suggests that this study may have been a stronger test of the hypothetical association of stimulant treatment to later drug use/abuse.


   CONCLUSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
REFERENCES
 
The present study found no consistent or convincing evidence that stimulant treatment in childhood or during high school was associated with risk for adolescent or adult substance use, the frequency of such use in adulthood, or the likelihood of having a substance dependence or abuse disorder. It also did not find that duration of such treatment was associated with any of these risks. The greater risk for using cocaine in adulthood among children or adolescents who had been treated with stimulants found here was largely explained by severity of current CD symptoms, a previously well-established risk factor for substance use and abuse among hyperactive/ADHD probands. Indeed, stimulant treatment for 1 year or more in childhood may have contributed a protective effect to risk of a hallucinogen abuse disorder in adulthood. Coupled with the similar results of 11 other studies, the present results indicate that clinicians probably need not fear that the stimulant treatment of children with ADHD is predisposing those children to later drug use, dependence, or abuse.


   ACKNOWLEDGMENTS
 
This project was supported by a grant (MH42181) from the National Institute of Mental Health. The contents of this article, however, are solely the responsibility of the authors and do not necessarily represent the official views of this institute. We appreciate the comments of Rachel Klein and Nadine Lambert on an earlier draft of this manuscript.


   FOOTNOTES
 
Received for publication Jan 2, 2002; Accepted Jul 31, 2002.

Reprint requests to (R.A.B.) College of Health Professions, 19 Hagood Ave, Charleston, SC 29425. E-mail: barkleyr{at}musc.edu


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 REFERENCES
 

  1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 1987
  2. Barkley RA. Attention Deficit Hyperactivity Disorder: A Handbook for Diagnosis and Treatment. 2nd ed. New York, NY: Guilford; 1998
  3. August GJ, Stewart MA, Holmes CS. A four-year follow-up of hyperactive boys with and without conduct disorder. Br J Psychiatry.1983; 143 :192 –198[Abstract/Free Full Text]
  4. Barkley RA, Fischer M, Edelbrock CS, Smallish L. The adolescent outcome of hyperactive children diagnosed by research criteria, I: an 8-year prospective follow-up study. J Am Acad Child Adolesc Psychiatry.1990; 29 :546 –557[Web of Science][Medline]
  5. Claude D, Firestone P. The development of ADHD boys: a 12-year follow-up. Can J Behav Sci.1995; 27 :226 –249[Web of Science]
  6. Klein R, Mannuzza S. Long-term outcome of hyperactive children: a review. J Am Acad Child Adolesc Psychiatry.1991; 30 :383 –387[Web of Science][Medline]
  7. Barkley RA, Fischer M, Smallish L, Fletcher K. The persistence of attention-deficit/hyperactivity disorder into young adulthood as a function of reporting source and definition of disorder. J Abnorm Psychol.2002; 111 :279 –289[CrossRef][Web of Science][Medline]
  8. Mannuzza S, Klein R, Bessler A, Malloy P, LaPadula M. Adult outcome of hyperactive boys: educational achievement, occupational rank, and psychiatric status. Arch Gen Psychiatry.1993; 50 :565 –576[Abstract/Free Full Text]
  9. Mannuzza S, Klein R, Bessler A, Malloy P, LaPadula M. Adult psychiatric status of hyperactive boys grown up. Am J Psychiatry.1998; 155 :493 –498[Abstract/Free Full Text]
  10. Mannuzza S, Klein RG, Bonagura N, Malloy P, Giampino H, Addalli KA. Hyperactive boys almost grown up: replication of psychiatric status. Arch Gen Psychiatry.1991; 48 :77 –83[Abstract/Free Full Text]
  11. Weiss G, Hechtman LT. Hyperactive Children Grown Up. 2nd ed. New York, NY: Guilford Press; 1993
  12. Fischer M, Barkley RA, Smallish L, Fletcher K. Young adult follow-up of hyperactive children: self-reported psychiatric disorders, comorbidity, and the role of childhood conduct problems. J Abnorm Child Psychol.2002; 30 :463 –475[CrossRef][Web of Science][Medline]
  13. DuPaul GJ, Barkley RA, Connor, DF. Stimulants. In: Barkley RA, ed. Attention Deficit Hyperactivity Disorder: A Handbook for Diagnosis and Treatment. New York, NY: Guilford; 1998:510–551
  14. Solanto MV, Arnsten AFT, Castellanos FX. Stimulant Drugs and ADHD: Basic and Clinical Neuroscience. New York, NY: Oxford University Press; 2001
  15. The MTA Cooperative Group. A 14-month randomized clinical trial of treatment strategies for attention-deficit/hyperactivity disorder. Arch Gen Psychiatry.1999; 56 :1073 –1086[Abstract/Free Full Text]
  16. Findling RL, Short EJ, Manos MJ. Developmental aspects of psychostimulant treatment in children and adolescencets with attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc Psychiatry.2001; 40 :1441 –1447[CrossRef][Web of Science][Medline]
  17. James RS, Sharp WS, Bastain TM, et al. Double-blind, placebo-controlled study of single-dose amphetamine formulations in ADHD. J Am Acad Child Adolesc Psychiatry.2001; 40 :1268 –1276[CrossRef][Web of Science][Medline]
  18. Greenhill LL, Halperin JM, Abikoff A. Stimulant medications. J Am Acad Child Adolesc Psychiatry.1999; 38 :503 –512[CrossRef][Web of Science][Medline]
  19. Rapport MD, Kelly KL. Psychostimulant effects on learning and cognitive functioning. In: Matson JL, ed. Handbook of Hyperactivity in Children. Boston, MA: Allyn & Bacon; 1993:97–136
  20. Barkley RA, McMurray MB, Edelbrock CS, et al. The side effects of Ritalin: a systematic placebo controlled evaluation of two doses. Pediatrics.1990; 86 :184 –192[Abstract/Free Full Text]
  21. Gillberg C, Melander H, von Knorring AL, et al. Long-term stimulant treatment of children with attention-deficit hyperactivity disorder symptoms. Arch Gen Psychiatry.1997; 54 :857 –864[Abstract/Free Full Text]
  22. Safer DJ, Krager JM. A survey of medication treatment for hyperactive/inattentive students. JAMA.1988; 260 :2256 –2258[Abstract/Free Full Text]
  23. Safer DJ, Zito JM, Fine EM. Increased methylphenidate usage for attention deficit disorder in the 1990s. Pediatrics.1996; 98 :1084 –1088[Abstract/Free Full Text]
  24. Angold A, Erkanli A, Egger HL, Costello EJ. Stimulant treatment for children: a community perspective. J Am Acad Child Adolesc Psychiatry.2000; 39 :975 –984[CrossRef][Web of Science][Medline]
  25. Safer DJ, Krager M. Stimulant treatment in Maryland public schools. Pediatrics.2000; 106 :533 –539[Abstract/Free Full Text]
  26. Jensen PS, Kettle L, Roper MT, et al. Are stimulants overprescribed? Treatment of ADHD in four U.S. communities. J Am Acad Child Adolesc Psychiatry.1999; 38 :797 –804[CrossRef][Web of Science][Medline]
  27. Zito JM, Safer DJ, dosReis S, Gardner JF, Boles M, Lynch F. Trends in the prescribing of psychotropic medications to preschoolers. JAMA.2000; 283 :1025 –1030[Abstract/Free Full Text]
  28. Lefever GB, Dawson KV, Morrow AL. The extent of drug therapy for attention deficit-hyperactivity disorder among children in public schools. Am J Public Health.1999; 89 :1359 –1364[Abstract/Free Full Text]
  29. Pear R. White house seeks to curb pills used to calm the young. New York Times.2000; March 20:A1 , A16
  30. Hancock L. Mother’s little helper. Newsweek.1996; March 18:51 –56
  31. Pekkanen J. Making sense of Ritalin. Reader’s Digest.2000; June:153 –158
  32. Rogers P. Drawing the line. People.2001; July 23:50 –55
  33. Committee on Children with Disabilities and Committee on Drugs. Medication for children with attentional disorders. Pediatrics.1996; 98 :301 –304[Abstract/Free Full Text]
  34. Goldman LS, Genel M, Bezman RJ, Slanetz PJ. Diagnosis and treatment of attention-deficit/hyperactivity disorder in children and adolescents. JAMA.1998; 279 :1100 –1107[Abstract/Free Full Text]
  35. Breggin P. Talking Back to Ritalin. Monroe, ME: Common Courage Press; 1998
  36. Citizens Commission on Human Rights (CCHR). Ritalin: A Warning to Parents. Los Angeles, CA: Church of Scientology; 1987
  37. Greenhill L, Osmon B. Ritalin: Theory and Patient Management. New York, NY: Mary Ann Liebert; 1999
  38. Volkow ND, Ding Y, Fowler JS, et al. Is methylphenidate like cocaine? Studies on their pharmacokinetics and distribution in the human brain. Arch Gen Psychiatry.1995; 52 :456 –463[Abstract/Free Full Text]
  39. Lambert NM, Hartsough CS. Prospective study of tobacco smoking and substance dependencies among samples of ADHD and non-ADHD participants. J Learn Disabil.1998; 31 :533 –544
  40. Lambert NM. Stimulant treatment as a risk factor for nicotine use and substance abuse. In Jensen PS, Cooper JR, eds. Diagnosis and Treatment of Attention Deficit Hyperactivity Disorder: An Evidence-Based Approach. New York, NY: American Medical Association Press; in press
  41. Strakowski SM, Szx KW. Progressive behavioral response to repeated d-amphetamine challenge: further evidence for sensitization in humans. Biol Psychiatry.1998; 44 :1171 –1177[CrossRef][Web of Science][Medline]
  42. Robinson TE, Berridge KC. The neural basis of drug craving: an incentive-sensitization theory of addiction. Brain Res Rev.1993; 18 :247 –291[CrossRef][Medline]
  43. Horger BA, Giles MK, Schenk S. Pre-exposure to amphetamine and nicotine predisposes rats to self-administer a low dose of cocaine. Psychopharmacology (Berl).1992; 107 :271 –276[CrossRef][Medline]
  44. Lett BW. Repeated exposures intensify rather than diminish the rewarding effects of emphetamine, morphine, and cocaine. Psychopharmacology (Berl).1989; 98 :357 –362[CrossRef][Medline]
  45. Schenk S, partridge B. Sensitization and tolerance in psychostimulant self-administration. Pharmacol Biochem Behav.1997; 57 :543 –550[CrossRef][Web of Science][Medline]
  46. Chilcoat HD, Breslau N. Pathways from ADHD to early drug use. J Am Acad Child Adolesc Psychiatry.1999; 38 :1347 –1354[CrossRef][Web of Science][Medline]
  47. Loney J, Kramer JR, Salisbury H. Medicated versus unmedicated ADHD children: adult involvement with legal and illegal drugs. In: Jensen PS, Cooper JR, eds. Diagnosis and Treatment of Attention Deficit Hyperactivity Disorder: An Evidence-Based Approach. New York, NY: American Medical Association Press; in press
  48. Biederman J, Wilens T, Mick E, Spencer T, Faraone SV. Pharmacotherapy of attention-deficit/hyperactivity disorder reduces risk for substance use disorder. Pediatrics.1999; 104(2) . Available at: www.pediatrics.org/cgi/content/full/104/2/e20
  49. Burke JD, Loeber R, Lahey BB. Which aspects of ADHD are associated with tobacco use in early adolescence? J Child Psychol Psychiatry.2001; 42 :493 –502[CrossRef][Web of Science][Medline]
  50. Milberger S, Biederman J, Faraone SV, Chen L, Jones J. ADHD is associated with early initiation of cigarette smoking in children and adolescents. J Am Acad Child Adolesc Psychiatry.1997; 36 :37 –44[CrossRef][Web of Science][Medline]
  51. Paternite CE, Loney J, Salisbury H, Whaley MA. Childhood inattention-overactivity, aggression, and stimulant medication history as predictors of young adult outcomes. J Child Adolesc Psychopharmacol.1999; 9 :169 –184[Web of Science][Medline]
  52. Blouin AGA, Bornstein R, Trite R. Teenage alcohol use among hyperactive children: a 5-year follow-up study. J Pediatr Psychol.1978; 3 :188 –194[Abstract/Free Full Text]
  53. Beck L, Langford WS, Mackay M, Sum G. Childhood chemotherapy and later drug abuse and growth curve: a follow-up study of 30 adolescents. Am J Psychiatry.1975; 132 :436 –438[Web of Science]
  54. Henker B, Whalen C, Bugental DB, Barker C. Licit and illicit substance use patterns in stimulant-treated children and their peers. In: Gadow K, Loney J, eds. Psychosocial Aspects of Drug Treatment for Hyperactivity. Boulder, CO: Westview Press; 1981
  55. Feldman S, Denhoff, E, Denhoff J. The attention disorders and related syndromes: outcome in adolescence and young adult life. In: Stern L, Denhoff E, eds. Minimal Brain Dysfunction: A Developmental Approach. New York, NY: Masson Publishing; 1979
  56. Mick E, Biederman J, Faraone SV. Comment on Lambert and Hartsough (1998). J Learn Dis.2000; 33 :314
  57. Brook JS, Whiteman M, Finch SJ, Cohen P. Young adult drug use and delinquency: childhood antecedents and adolescent mediators. J Am Acad Child Adolesc Psychiatry.1996; 35 :1584 –1592[CrossRef][Web of Science][Medline]
  58. Molina BSG, Smith BH, Pelham WE. Interactive effects of attention deficit hyperactivity disorder and conduct disorder on early adolescent substance use. Psychol Addict Behav.1999; 13 :348 –358[CrossRef][Web of Science]
  59. White HR, Xie M, Thompson W, Loeber R, Southamer-Loeber M. Psychopathology as a predictor of adolescent drug use trajectories. Psychol Addict Behav.2001; 15 :210 –218[CrossRef][Web of Science][Medline]
  60. Goyette CH, Conners CK, Ulrich RF. Normative data for Revised Conners Parent and Teacher Rating Scales. J Abnorm Child Psychol.1978; 6 :221 –236[CrossRef][Web of Science][Medline]
  61. Barkley RA. Hyperactive Children: A Handbook for Diagnosis and Treatment. New York, NY: Guilford; 1981
  62. Edelbrock CS, Costello A. Convergence between statistically derived behavior problem syndromes and child psychiatric diagnoses. J Abnorm Child Psychol.1988; 16 :219 –231[CrossRef][Web of Science][Medline]
  63. Barkley RA. Attention Deficit Hyperactivity Disorder: A Handbook for Diagnosis and Treatment. New York, NY: Guilford; 1990
  64. Werry J, Sprague R. Hyperactivity. In: Costello CG, ed. Symptoms of Psychopathology. New York, NY: Wiley; 1970:397–417
  65. Routh DK, Schroeder CS, O’Tuama L. The development of activity level in children. Dev Psychol.1974; 10 :163 –168[CrossRef][Web of Science]
  66. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed, Revised. Washington, DC: American Psychiatric Association; 1987
  67. O’Donnell D, Biederman J, Jones J, et al. Informativeness of child and parent reports on substance use disorders in a sample of ADHD probands, control probands, and their siblings. J Am Acad Child Adolesc Psychiatry.1998; 37 :752 –758[CrossRef][Web of Science][Medline]
  68. Mannuzza S, Gittelman R. Informant variance in the diagnostic assessment of hyperactive children as young adults. In: Barrett JE, Rose RM, eds. Mental Disorders in the Community. New York, NY: Guilford; 1986:243–254
  69. Spitzer RL, Williams J, Gibbon M, First MB. The Structured Clinical Interview for DSM-III-R (SCID). New York, NY: Biometric Research Department, New York State Psychiatric Institute; 1989
  70. Fergusson DM, Swain-Campbell NR, Horwood LJ. Deviant peer affiliations, crime and substance use: a fixed effects regression analysis. J Abnorm Child Psychol.2002; 30 :419 –430[CrossRef][Web of Science][Medline]
  71. Whalen CK, Jamer LD, Henker B, Delfino RJ, Lozano JM. The ADHD spectrum and everyday life: experience sampling of adolescent moods, activities, smoking, and drinking. Child Dev.2002; 73 :209 –227[CrossRef][Web of Science][Medline]
  72. Kuperman S, Schlosser SS, Kramer JR, et al. Developmental sequence from disruptive behavior diagnosis to adolescent alcohol dependence. Am J Psychiatry.2001; 158 :2022 –2026[Abstract/Free Full Text]
  73. Loeber R, Burke JD, Lahey BB, Winters A, Zera M. Oppositional defiant and conduct disorder: a review of the past 10 years, part I. J Am Acad Child Adolesc Psychiatry.2000; 39 :1 –17[CrossRef][Web of Science][Medline]
  74. Barkley RA, Fischer M, Smallish L, Fletcher KR. Young adult follow-up of hyperactive children: antisocial activity and drug use. J Child Psychol Psychiatry. In press
PEDIATRICS (ISSN 1098-4275). ©2003 by the American Academy of Pediatrics

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Facebook Facebook   Add to Reddit Reddit   Add to Technorati Technorati   Add to Twitter Twitter    What's this?


This article has been cited by other articles:


Home page
 J Atten DisordHome page
A. Halmoy, O. B. Fasmer, C. Gillberg, and J. Haavik
Occupational Outcome in Adult ADHD: Impact of Symptom Profile, Comorbid Psychiatric Problems, and Treatment: A Cross-Sectional Study of 414 Clinically Diagnosed Adult ADHD Patients
J Atten Disord, September 1, 2009; 13(2): 175 - 187.
[Abstract] [PDF]

Home page
 J Atten DisordHome page
C. Advokat
What Exactly Are the Benefits of Stimulants for ADHD?
J Atten Disord, May 1, 2009; 12(6): 495 - 498.
[PDF]

Home page
 J Atten DisordHome page
B. S. G. Molina, K. Flory, O. G. Bukstein, A. R. Greiner, J. L. Baker, V. Krug, and S. W. Evans
Feasibility and Preliminary Efficacy of an After-School Program for Middle Schoolers With ADHD: A Randomized Trial in a Large Public Middle School
J Atten Disord, November 1, 2008; 12(3): 207 - 217.
[Abstract] [PDF]

Home page
 Arch Pediatr Adolesc MedHome page
T. E. Wilens, J. Adamson, M. C. Monuteaux, S. V. Faraone, M. Schillinger, D. Westerberg, and J. Biederman
Effect of Prior Stimulant Treatment for Attention-Deficit/Hyperactivity Disorder on Subsequent Risk for Cigarette Smoking and Alcohol and Drug Use Disorders in Adolescents
Arch Pediatr Adolesc Med, October 1, 2008; 162(10): 916 - 921.
[Abstract] [Full Text] [PDF]

Home page
 FocusHome page
J. Biederman, M. C. Monuteaux, T. Spencer, T. E. Wilens, H. A. MacPherson, and S. V. Faraone
Stimulant Therapy and Risk for Subsequent Substance Use Disorders in Male Adults With ADHD: A Naturalistic Controlled 10-Year Follow-Up Study
Focus, July 1, 2008; 6(3): 358 - 365.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. PsychiatryHome page
J. Biederman, M. C. Monuteaux, T. Spencer, T. E. Wilens, H. A. MacPherson, and S. V. Faraone
Stimulant Therapy and Risk for Subsequent Substance Use Disorders in Male Adults With ADHD: A Naturalistic Controlled 10-Year Follow-Up Study
Am J Psychiatry, May 1, 2008; 165(5): 597 - 603.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. PsychiatryHome page
S. Mannuzza, R. G. Klein, N. L. Truong, J. L. Moulton III, E. R. Roizen, K. H. Howell, and F. X. Castellanos
Age of Methylphenidate Treatment Initiation in Children With ADHD and Later Substance Abuse: Prospective Follow-Up Into Adulthood
Am J Psychiatry, May 1, 2008; 165(5): 604 - 609.
[Abstract] [Full Text] [PDF]

Home page
 Arch Pediatr Adolesc MedHome page
S. E. McCabe
Screening for Drug Abuse Among Medical and Nonmedical Users of Prescription Drugs in a Probability Sample of College Students
Arch Pediatr Adolesc Med, March 1, 2008; 162(3): 225 - 231.
[Abstract] [Full Text] [PDF]

Home page
 J Atten DisordHome page
C. Advokat
Literature Review: Update on Amphetamine Neurotoxicity and Its Relevance to the Treatment of ADHD
J Atten Disord, July 1, 2007; 11(1): 8 - 16.
[Abstract] [PDF]

Home page
 PediatricsHome page
M. Ernst, D. A. Luckenbaugh, E. T. Moolchan, M. K. Leff, R. Allen, N. Eshel, E. D. London, and A. Kimes
Behavioral predictors of substance-use initiation in adolescents with and without attention-deficit/hyperactivity disorder.
Pediatrics, June 1, 2006; 117(6): 2030 - 2039.
[Abstract] [Full Text] [PDF]

Home page
 PediatricsHome page
L. Furman and R. A. Barkley
Attention-Deficit/Hyperactivity Disorder Treatment and Later Drug Use
Pediatrics, December 1, 2003; 112(6): 1459 - 1460.
[Full Text] [PDF]

eLetters:

Read all eLetters

relationship between untreated adhd and substance abuse
joanne shorser-gentile
Pediatrics Online, 14 Jun 2009 [Full text]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow View responses
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Web of Science (63)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Barkley, R. A.
Right arrow Articles by Fletcher, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Barkley, R. A.
Right arrow Articles by Fletcher, K.
Related Collections
Right arrow Neurology & Psychiatry
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Facebook   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?