MYERS,

PROGRESSION
Am J Psychiatry
STEWART,FROM
155:4,
ANDCONDUCT
April
BROWN1998DISORDER

Progression From Conduct Disorder

to Antisocial Personality Disorder Following Treatment
for Adolescent Substance Abuse

Mark G. Myers, Ph.D., David G. Stewart, B.A., and Sandra A. Brown, Ph.D.

Objective: This study investigated the progression from conduct disorder to antisocial per-
sonality disorder among individuals treated for adolescent substance abuse. This issue is im-
portant because of the poor outcomes observed among individuals with antisocial pathology
after treatment for alcohol and drug problems. The utility of factors assessed at the time of
treatment in predicting progression to adult antisocial personality disorder was evaluated in
the context of developmental models of antisocial behavior. Method: This was a prospective
longitudinal study of 137 substance-abusing adolescents (53 female and 84 male), whose
average age was 15.9 years and who met the DSM-III-R criteria for conduct disorder. Con-
secutively admitted patients were recruited from two adolescent inpatient alcohol and drug
treatment facilities. Participants were interviewed again 4 years after treatment. Results: Four
years after treatment, 61% of the study group met the DSM-III-R criteria for antisocial per-
sonality disorder. Results of a logistic regression analysis indicated that onset of deviant be-
havior at or before age 10, a greater diversity of deviant behavior, and more extensive pre-
treatment drug use best predicted progression to antisocial personality disorder. At 4-year
follow-up, the subjects with an antisocial personality disorder diagnosis exhibited more in-
volvement with alcohol and drugs and poorer functioning across important life domains than
the subjects without antisocial personality disorder. Conclusions: This study found a high rate
of progression to antisocial personality disorder among substance-abusing adolescents and
identified factors predictive of this progression. Careful assessment of conduct disorder history
at the time of treatment may be valuable for treatment planning and intervention.
(Am J Psychiatry 1998; 155:479–485)

F actors associated with the development and per-

sistence of antisocial behavior have been exten-
sively researched. Of particular concern is the observed
emerge as prevalent comorbid diagnoses in clinical
samples of adolescent and adult substance abusers, re-
spectively, and have been associated with poorer treat-
concordance between antisocial behavior and sub- ment outcome (5–7). These findings suggest that adult
stance abuse. Antisocial behavior is predictive of both antisocial personality disorder and substance use disor-
adolescent and adult involvement in substance abuse ders may share common etiological pathways.
(1–3). Further, conduct disorder is a strong prognostic The consequences of comorbid antisocial behavior
indicator for both antisocial personality disorder and and substance abuse have been studied primarily in
psychoactive substance use disorders in adulthood (4). adults and serve to highlight the concern surrounding
Conduct disorder and antisocial personality disorder this issue. Findings identify high base rates of antisocial
behavior among substance abusers (1, 2), earlier and/or
more rapid onset of substance use problems among
Received March 21, 1997; revision received July 8, 1997; accepted substance abusers with antisocial personality disorder
Sept. 18, 1997. From the School of Medicine, University of California, (1, 8), and a relationship between polysubstance abuse
San Diego; the Department of Psychiatry, University of California,
San Diego; and the San Diego VA Medical Center. Address reprint and antisocial personality disorder (8, 9). Findings re-
requests to Dr. Myers, Psychology (116B), VA Medical Center, 3350 garding treatment outcomes with respect to substance
La Jolla Village Dr., San Diego, CA 92161; mgmyers@ucsd.edu (e- use and life functioning among persons with an antiso-
mail). cial personality disorder diagnosis are mixed, with
Supported by grants from the National Institute on Alcohol Abuse
and Alcoholism (AA-07033), the Research Service of the Department
some studies reporting poorer outcomes (1) and others
of Veterans Affairs, and the National Institute on Drug Abuse (DA- no differences (10) in comparison with individuals
09181). without antisocial personality disorder.

Am J Psychiatry 155:4, April 1998 479

PROGRESSION FROM CONDUCT DISORDER

The literature describing the prevalence and influ- at the time of the first conduct disorder criterion behav-
ence of antisocial behavior in clinical populations of ior and a greater number (i.e., diversity) of conduct dis-
adolescent substance abusers is currently limited. In- order behaviors independent of substance use reported
itial reports suggest a relationship between conduct at the time of treatment. We also explored the contri-
disorder and adolescent substance abuse similar to that bution of involvement in substance use prior to treat-
found for antisocial personality disorder among adult ment in predicting progression to antisocial personality
substance abusers (11). For example, the conduct dis- disorder. In addition, we expected the progression to a
order diagnosis is prevalent in clinical samples of ado- diagnosis of antisocial personality disorder to be asso-
lescent substance abusers (5, 12) and has been associ- ciated with poorer long-term substance use outcomes.
ated with poorer substance use outcomes (5, 13). Finally, we predicted that individuals diagnosed with
These studies suggest that conduct disorder behavior antisocial personality disorder as young adults would
that precedes or occurs independent of substance demonstrate poorer functioning in major life domains
abuse may indicate poorer long-term treatment out- than those without antisocial personality disorder.
come and a persistence of antisocial behavior among
substance-abusing adolescents.
Developmental models of antisocial behavior (14– METHOD
17) provide a framework for investigating the emer-
One hundred sixty-six adolescents (40% of whom were female)
gence of antisocial personality disorder in adolescent were recruited from two inpatient substance abuse treatment pro-
substance abusers. One model (14, 15) distinguishes grams in metropolitan San Diego as part of an ongoing longitudinal
two subtypes of deviant adolescents: the majority research project. Consecutively admitted patients were recruited if
whose problem behavior begins and ends in adoles- review of hospital records and structured interviews noted no evi-
dence of a DSM-III-R axis I psychiatric disorder (other than conduct
cence and the relatively few whose antisocial behavior disorder) predating the onset of substance abuse. The inclusion crite-
persists into adulthood. From this perspective, antiso- ria required 1) the participation of a resource person (usually a par-
cial behavior is likely to persist among adolescents who ent) to corroborate information regarding the adolescent and 2) resi-
demonstrate a stable history of deviant behavior since dence within 50 miles of the research facility.
The current study used data from in-treatment interviews and in-
childhood, show a wide range of antisocial behavior, terviews 4 years after treatment for those participants who met the
and fail to alter this behavior despite opportunities to criteria for conduct disorder at the time of treatment. Of the original
desist. Another etiological model (17) highlights the study group, 88% (N=146) completed 4-year interviews. Of these,
presence of deviant behavior across multiple and di- nine were excluded from the present study (five who were not 18
years old at the 4-year follow-up and four who did not meet the cri-
verse settings as predictive of a chronic and severe teria for a diagnosis of conduct disorder), resulting in a final study
course of antisocial behavior. This conceptualization group of 137 participants. The mean age of the participants at the
asserts that a diversity of antisocial behavior more 4-year follow-up time point was 20.0 years (SD=1.1). The study
likely represents enduring psychopathology. Thus, cur- group was 39% female (N=53) and predominantly Caucasian (75%),
with 6% Hispanic, 5% African American, and 14% from other eth-
rent etiological models suggest that early emergence of nic subgroups. The socioeconomic status of the participating families
antisocial behavior and display of this behavior across ranged from unskilled laborer to college-educated professional.
diverse settings are prognostic indicators of persistence. Chart reviews were used to screen consecutively admitted patients
The extensive literature on the etiology of antisocial to the inpatient treatment facilities for eligibility for the study, after
which potential participants were recruited during the second week
and delinquent behavior currently includes relatively of treatment. Written informed consent was obtained separately from
few studies that have explored the progression of anti- each adolescent and a parent or legal guardian after the procedures
social behavior in the context of substance abuse. had been fully explained. A 90-minute confidential interview was ad-
Similarly, developmental models of the persistence of ministered separately to the teenager, after which a resource person
deviant behavior have yet to be tested in clinical popu- was interviewed to provide corroborative information. Follow-up in-
terviews were conducted separately with the adolescent subjects and
lations of substance-abusing adolescents. It is therefore the resource persons.
unclear how comorbid substance use and antisocial be-
havior may alter or influence the progression from ado- Instruments
lescent conduct disorder to adult antisocial personality
disorder. Adolescents treated for substance abuse rep- The Structured Clinical Interview for Adolescents (18) was used to
resent a unique subgroup whose substance use has assess demographic and background information as well as aca-
demic, social, emotional, occupational, family, and health function-
progressed to pathological levels (18) and who exhibit ing at each interview time point.
high rates of conduct disorder behavior (5, 12). There- The Customary Drinking and Drug Use Record (19) was adminis-
fore, this population is particularly appropriate for ex- tered to the adolescent participants at each time point to assess alco-
amining the development of concomitant antisocial hol and other drug use patterns. This instrument incorporated the
personality disorder and substance abuse. The current DSM-III-R criteria for psychoactive substance abuse, dependence,
and withdrawal. This interview has demonstrated good internal con-
study extends previous investigations conducted with sistency, high test-retest reliability, excellent interrater reliability, and
the same group of adolescents to 4 years after treatment strong convergent and discriminant validity (20).
and represents an initial prospective effort to evaluate Substance use information was gathered for cigarettes, alcohol
the persistence of antisocial behavior in the context of (beer, wine, liquor), marijuana, amphetamines, hallucinogens, co-
caine, opiates, barbiturates, and inhalants. The initial interview col-
substance abuse. lected information on history (age at first use, onset of weekly regular
It was anticipated that the progression to antisocial use), lifetime use, and current (previous 3 months) quantity and fre-
personality disorder would be predicted by earlier age quency of use.

480 Am J Psychiatry 155:4, April 1998

 MYERS, STEWART, AND BROWN

The Conduct Disorder/Antisocial Personality TABLE 1. Demographic, Substance Use, and Conduct Disorder Variables at Study Intake
Disorder Questionnaire (5) is a structured interview for Subjects With and Without a Diagnosis of Antisocial Personality Disorder 4 Years
designed to comprehensively assess the DSM-III-R After Treatment
criteria for conduct disorder and antisocial person-
ality disorder. The questionnaire was independently Subjects With Subjects Without
administered to each adolescent participant and a Antisocial Antisocial
resource person (usually a parent). The diagnoses of Personality Personality
conduct disorder and antisocial personality disor- Variable Disorder (N=84) Disorder (N=53)
der were determined according to standard DSM-
N % N %
III-R criteria. The questionnaire also evaluated each
conduct disorder criterion behavior in relation to
Male gendera 60 71 24 45
involvement with alcohol or drug use. Behavior that
Caucasian race 60 71 42 79
occurred exclusively during periods of substance in-
Exhibited conduct disorder criterion be-
toxication was scored as being directly related to
haviors at or before age 10a 63 75 23 43
substance use; behavior that occurred while the
subject was trying to obtain substances (e.g., steal-
ing money to buy drugs) was scored as being indi- Mean SD Mean SD
rectly related to substance use, and behavior that
occurred in the absence of any substance involve- Age at intake (years) 15.93 1.15 16.22 1.07
ment was scored as independent of substance use. Hollingshead socioeconomic status index 31.10 12.54 28.38 12.63
The reliability and validity of the Conduct Disor- Lifetime days of alcohol use 600.90 579.74 380.11 530.00
der/Antisocial Personality Disorder Questionnaire Lifetime days of drug use 1269.82 666.07 960.58 606.35
have been demonstrated through associations with Lifetime drug types useda 4.64 1.14 3.89 1.24
other assessment procedures (5, 18). The use of a Recent alcohol use (quantity/frequency in-
corroborative interview with a resource person to dex for prior 3 months) 34.16 53.76 17.20 41.25
provide objective information on conduct disorder Recent drug use (average days per month
and antisocial personality disorder behaviors adds in prior 3 months)a 41.19 16.33 32.11 18.38
to confidence in the validity of the Conduct Disor- Age at first use of hard drugs (years) 13.44 1.55 13.91 2.02
der/Antisocial Personality Disorder Questionnaire Number of conduct disorder criteria met
for diagnosing antisocial personality disorder. In independent of substance usea 5.26 2.16 3.36 2.04
support of the present approach, Hare (21) found
aSignificant difference between groups reflecting a 5% probability of type I error.
that clinical-behavioral measurement of antisocial
personality disorder with the use of DSM criteria
was reliable when compared to the use of other
checklists and ratings. daily functioning and by whether individuals received counseling for
emotional problems. The data on anxiety and depression were based on
participants’ and collateral reports and do not reflect formal diagnoses.
Measurement of 4-Year Treatment Outcome Finally, legal issues were represented by participants’ reports of arrests
or jail time during the prior 2 years.
Substance use. Alcohol use at intake and at the 4-year time point
was represented by a quantity/frequency index of recent alcohol use
(average days per month on which drinking occurred multiplied by
the average number of drinks per occasion for the past 3 months RESULTS
summed across alcohol types). The drug use scale reflected the aver-
age number of days per month of substance use summed across drug
types for the 3 months preceding the interview. Substance dependence Eighty-four (61%) of the participants met the DSM-
at the 4-year interview was assessed for alcohol and for the other drug III-R criteria for the antisocial personality disorder di-
most frequently used by each participant on the basis of DSM-III-R agnosis. The significance level for comparisons of base-
criteria. In addition, at the 4-year time point a time-line follow-back line characteristics was set at 0.006 so as to provide a
procedure (22) was used to provide an estimate of the number of days
on which alcohol and/or other drug use occurred in the 2-year inter- type I error rate of ≤5% for multiple analyses. Table 1
val before the interview (23, 24). Although this time-line follow-back displays demographic data, baseline alcohol and drug
interval is longer than the one that is typically used, these data were use characteristics, and variables included in the logistic
included to provide an estimate of aggregate substance use over the
entire time period.
regression for the two diagnostic outcome groups. Male
Data on drug and alcohol use were based on the adolescent’s self- subjects were disproportionately represented in the an-
report and independent corroborative interviews with a resource per- tisocial personality disorder group (71%, N=60) com-
son. Previous studies have established that alcohol and drug abusers pared to female subjects (29%, N=24) (χ2=9.37, df=1,
can provide reliable drinking and drug use data with use of similar p<0.005); however, the two groups were comparable in
procedures (assurance of confidentiality, multiple sources of data,
corroborative interviews). race/ethnicity, age, and socioeconomic status at the
Major domains of functioning. We compared the functioning of the time of admission to treatment. The subjects meeting
subjects with and without antisocial personality disorder in the follow- the criteria for antisocial personality disorder had sig-
ing major life domains: school/work, interpersonal functioning, emo- nificantly greater drug use but not alcohol use during
tional well-being, and illegal behavior. These domains reflect function-
ing over the 2 years before the 4-year interview time point. The influence
the 3 months before treatment; they did not differ on
of involvement with drugs and alcohol on functioning was assessed for lifetime use of alcohol or drugs.
each domain. School/work functioning was represented by high school
graduation status, current employment, and the presence of alcohol Prediction of Antisocial Personality Disorder
and/or drug related problems at school or at work. Interpersonal func-
tioning was measured by evaluating marital status, interpersonal prob-
lems, and problems with partner/spouse that were related to alcohol The antisocial personality disorder diagnosis was
and/or other drug use. Emotional health was examined by using vari- used as the criterion measure in a hierarchical logistic
ables assessing substantial depression and anxiety that interfered with regression analysis to evaluate the hypothesized predic-

Am J Psychiatry 155:4, April 1998 481

PROGRESSION FROM CONDUCT DISORDER

TABLE 2. Substance Involvement of Groups With and Without Antisocial Personality Disorder 4 Years After Treatment
Subjects Subjects
With Without
Antisocial Antisocial
Personality Personality
Disorder (N=84) Disorder (N=53) Analysis
Variable Mean SD Mean SD F df p
Recent alcohol use (quantity/frequency index for prior 3 months) 40.40 54.93 9.92 17.78 11.04 1, 109 <0.01a
Recent drug use (days per month in prior 3 months) 10.75 15.30 4.60 12.07 4.82 1, 111 <0.03
Alcohol dependence (number of symptoms) 1.56 2.14 0.56 1.21 7.21 1, 110 <0.01a
Drug dependence (number of symptoms) 1.93 2.63 0.60 1.60 8.36 1, 111 <0.01a
Total number of days on which alcohol and/or drugs were used in
past 2 years 381.68 248.95 158.83 214.63 28.52 1, 134 <0.001a
aSignificant difference reflecting a 5% probability of type I error corrected for multiple analyses (p<0.01).

tors of progression to antisocial personality disorder. der classification. As predicted, the subjects in the anti-
Gender and age were entered as covariates in the first social personality disorder group had significantly
step of the logistic regression. The second step, reflect- more alcohol involvement at the 4-year time point, re-
ing early onset and diversity of antisocial behavior, in- ported more alcohol and drug dependence symptoms,
cluded the lifetime number of conduct disorder crite- and tended to report greater recent drug use than those
rion behaviors reported to have occurred independent not meeting the criteria for antisocial personality disor-
of substance use at the initial interview and a dichoto- der. In addition, the participants with antisocial person-
mous variable reflecting whether the earliest conduct ality disorder reported using alcohol and/or drugs on
disorder criterion behavior was reported as having oc- more days during the preceding 2 years than did those
curred at or before age 10 or after age 10. The latter without antisocial personality disorder.
variable was dichotomized to reflect childhood onset of
antisocial behavior and is similar to the distinction used Major Domains of Functioning
in DSM-IV. The third step of the regression included
frequency of drug use and a quantity/frequency index To compare the participants with and without anti-
of alcohol use in the 3 months preceding admission to social personality disorder on important domains of
treatment. The distributional properties of all variables young adult functioning, chi-square analyses were con-
were assessed and found acceptable. ducted with the alpha level set so as to provide a 0.95
Each step in the hierarchical regression added signifi- probability of avoiding type I error.
cantly (p≤0.05) to the prediction of membership in the In examining work and school status, we found no
groups with and without antisocial personality disor- significant differences between groups in current em-
der. The model that best fitted the data (–2 log likeli- ployment or completion of high school education.
hood χ2=134.26; model χ2=12.50, p<0.01; goodness of However, the group with antisocial personality disor-
fit index=123.58, df=130) correctly classified 77% of der more frequently reported school and/or work prob-
the participants as having or not having antisocial per- lems related to their alcohol and drug involvement
sonality disorder (87% of the subjects with antisocial (54% versus 20%; χ2=15.64, df=1, p<0.001).
personality disorder and 62% of those without antiso- In the realm of interpersonal functioning, a signifi-
cial personality disorder). Of the variables in the final cantly larger proportion of the group with antisocial
model, onset of conduct disorder behavior at age 10 or personality disorder reported interpersonal problems at
earlier (beta=–1.37, p<0.01), greater diversity of con- home, school, or work (76% versus 50%; χ2=9.84, df=
duct disorder behavior (beta=0.30, p<0.05), and heav- 1, p<0.001). Although no differences between groups
ier drug use prior to admission (beta=0.04, p≤0.01) were evident for marital status (single versus mar-
emerged as the best predictors of the antisocial person- ried/living with partner), the group with antisocial per-
ality disorder diagnosis. sonality disorder had a five times greater incidence of
problems with a spouse/partner that were related to al-
Substance Use cohol and/or drug use (61% versus 13%; χ2=29.39, df=
1, p≤0.001).
The hypothesis that the antisocial personality disor- Significant differences between groups were also
der diagnosis would be associated with poorer substance found across diverse measures of emotional function-
use outcome was tested by examining the diagnostic ing. Individuals in the antisocial personality disorder
groups’ recent substance involvement and dependence group were four times more likely to report requiring
symptoms with the use of analysis of variance, sepa- professional help for emotional problems overall (51%
rately for alcohol and drugs. Alpha was set at 0.01 to versus 12%; χ2=20.60, df=1, p<0.001), but not for
account for multiple analyses. Table 2 displays 4-year problems unrelated to alcohol or drug use (26% versus
substance use outcome by antisocial personality disor- 11%; χ2=4.64, df=1, p=0.03). The participants who

482 Am J Psychiatry 155:4, April 1998

 MYERS, STEWART, AND BROWN

met the criteria for antisocial personality disorder were the group with an antisocial personality disorder diag-
also significantly more likely to report experiencing se- nosis. A reliance on alcohol and drug use for coping
rious depression (38% versus 14%; χ2=8.93, df=1, p< with life stress may be particularly likely given the his-
0.01) and anxiety (68% versus 37%; χ2=12.46, df=1, tory of polysubstance abuse in this study group (18,
p<0.001). However, these differences were not signifi- 27). This finding also appears consistent with the no-
cant when we considered only depression and anxiety tion of “snares” as discussed by Moffitt (15) and her
reported as occurring independent of alcohol and/or colleagues, in which certain consequences or features of
drug use (χ2=4.05, df=1, p=0.06, and χ2=3.57, df=1, antisocial behavior serve to limit options for escaping
p=0.45, respectively). an antisocial lifestyle. In the present study, heavier in-
Finally, antisocial personality disorder was signifi- volvement with alcohol and drugs following treatment
cantly associated with legal problems. Specifically, the may have served to limit options for engaging in more
subjects with antisocial personality disorder had a sig- prosocial activities or behavior, restricted environ-
nificantly greater likelihood of being arrested (48% ver- mental contexts to more “risky” settings, and func-
sus 8%; χ2=23.58, df=1, p<0.001) and serving time in tioned to perpetuate a deviant lifestyle.
jail (36% versus 6%; χ2=17.56, df=1, p<0.001) than Examination of antisocial personality disorder in re-
those without the disorder. However, diagnostic status lation to functioning in major life domains yielded re-
was not significantly associated with arrests or incar- sults largely consistent with our predictions. In the do-
cerations that were unrelated to alcohol or drug in- mains of school/work, interpersonal relations, and legal
volvement (χ2=2.68, df=1, p=0.10, and χ2=2.46, df=1, p= difficulties, the greater extent of problems reported by
0.12, respectively). the subjects with antisocial personality disorder ap-
pears largely related to involvement with alcohol
and/or drugs. While completion of schooling and cur-
DISCUSSION rent employment status did not differ between the
groups, problems attributed to alcohol and/or drug use
This study examined prospectively the development were significantly more common among the partici-
of antisocial personality disorder during the 4 years fol- pants with antisocial personality disorder in each of the
lowing treatment for adolescent substance abuse. The aforementioned domains. Thus, the greater substance
results largely supported hypotheses derived from de- involvement by those progressing to antisocial person-
velopmental models for the persistence of antisocial be- ality disorder appears to exacerbate difficulties in func-
havior and previous empirical evidence of the relation- tioning across these important domains of young adult
ship between substance abuse and progression of functioning. When we examined emotional function-
antisocial behavior. In addition, these findings are con- ing, subjects in the group with antisocial personality
sistent with retrospective studies of adult substance disorder reported more overall help seeking for psycho-
abusers that have demonstrated an association between logical problems and greater overall anxiety and de-
early onset of antisocial behavior and a greater extent pression. Therefore, it may be that the persistence of
and severity of antisocial behavior (25, 26). antisocial behavior in this study group was influenced
Of the variables included in the logistic regression, by additional psychopathology. The substance abuse
early onset of conduct disorder behavior (at age 10 or and emotional difficulties apparent in the group with
earlier), greater diversity of conduct disorder behavior antisocial personality disorder may well be reciprocally
occurring independent of substance involvement, and related in such a way that each serves to exacerbate the
greater recent use of drugs were significant predictors other. The current results indicate the prominence of
of the antisocial personality disorder diagnosis in late substance use in relation to poorer overall functioning,
adolescence/early adulthood. Our findings are largely suggesting that alcohol and drug abuse may be inextri-
consistent with theoretical models of the persistence of cably involved in the progression of antisocial pathol-
antisocial behavior that emphasize the importance of ogy among young people with a history of substance
early, severe, and diverse deviant behavior in predicting abuse.
lifelong antisocial behavior (14, 16). In addition, the The present data on the progression of antisocial be-
fact that involvement with drugs prior to treatment was havior suggest continued difficulties for the participants
found to add to the prediction of antisocial personality diagnosed with antisocial personality disorder. Moffitt
disorder highlights the role of substance use in the pro- (15) has proposed that the persistence of antisocial be-
gression of antisocial pathology among youths with a havior reflects in part the limited options for change
history of substance abuse. available to antisocial individuals. The results of the
As anticipated, study participants who developed an- present study can be interpreted to support this concep-
tisocial personality disorder by 4 years after treatment tualization. For example, the poorer relationship func-
for adolescent substance abuse had poorer alcohol and tioning that was apparent among the subjects with an-
drug use outcomes than did those not diagnosed with tisocial personality disorder may reflect various deficits
antisocial personality disorder. Our results indicated in interpersonal and cognitive functioning as well as the
significantly higher overall levels of involvement with negative consequences of continued substance involve-
alcohol, more symptoms of alcohol and drug depen- ment. The likelihood of remediation or improvement of
dence, and more days on which substances were used in skills critical to successful functioning appears rather

Am J Psychiatry 155:4, April 1998 483

PROGRESSION FROM CONDUCT DISORDER

limited for individuals with antisocial personality disor- stance use or appears following the onset of substance
der who have passed the critical period during which abuse, and who demonstrate such behavior across few
many cognitive and interpersonal skills are typically ac- settings, are likely to desist from antisocial behavior
quired and refined. In addition, continued substance in- and thus may not require as much intervention beyond
volvement exhibited by youths with antisocial person- treatment for substance abuse. In contrast, those who
ality disorder may serve to limit future opportunities appear to be on a lifetime course of antisocial behavior
for acquisition of skills and is likely to create new prob- may benefit from more intensive targeted interventions.
lems, the consequences of which may further promote For example, it has been shown that multisystemic ther-
substance involvement and restrict opportunities for apy (29) is effective in producing long-term reductions
emerging from an antisocial lifestyle. in violent and criminal offenses among serious juvenile
The findings of this study must be considered in light offenders, and as such it may represent a useful adjunct
of several limitations. First, the current method of de- intervention for substance-abusing youths at risk for
termining an antisocial personality disorder diagnosis persistent antisocial behavior.
from the Conduct Disorder/Antisocial Personality Dis- In summary, this investigation is one of the first to
order Questionnaire has not been validated in compari- examine prospectively the development of antisocial
son with standardized diagnostic measures. However, personality disorder in the context of concomitant ado-
this instrument has demonstrated validity and utility lescent substance abuse and conduct disorder. These
for obtaining a diagnosis of conduct disorder (5, 18), data illustrate some of the developmental factors that
thereby increasing our confidence in the antisocial per- underlie the poor prognosis that accompanies comor-
sonality disorder diagnoses in this study. Another issue bid antisocial personality disorder and substance abuse;
concerns the accuracy of retrospective information used they also demonstrate how the matrix of problems en-
to assess the onset of early conduct disorder behavior. countered by antisocial youths may function to sustain
The utilization of multiple reporters has been found problem behavior and limit options for change. How-
useful in retrospective assessment of childhood problem ever, our findings should not be interpreted to suggest
behavior and may mitigate some of the concerns re- that antisocial youths do not benefit from treatment.
garding potential recall bias (28). The hypothesized Those with persistent antisocial psychopathology may
variables successfully predicted progression to antiso- enter treatment with poorer overall functioning, and
cial personality disorder on the basis of information thus between-group differences may reflect more lim-
collected during adolescence, suggesting that careful as- ited improvement rather than treatment failure. Future
sessment of retrospective information can be used to research on the factors that interact to maintain and
predict continuity of antisocial behavior. Further, the reinforce the concordance between antisocial behavior
results of this study are supported by the convergence and substance abuse is needed to further our under-
of findings across our three investigations that have standing of mechanisms underlying this process.
demonstrated a relationship between early conduct dis-
order behavior and treatment outcome. The reliance on REFERENCES
self-report data for assessment of anxiety, depression,
and illegal behavior suggests caution in interpreting our 1. Liskow B, Powell BJ, Nickel E, Penick E: Antisocial alcoholics:
findings. Future studies that use more formal diagnostic are there clinically significant diagnostic subtypes? J Stud Alco-
procedures for assessing these variables are needed to hol 1991; 52:62–69
2. Schuckit MA: The clinical implications of primary diagnostic
replicate the present findings. Another limitation of this groups among alcoholics. Arch Gen Psychiatry 1985; 42:1043–
study is that not all factors believed to be important to 1049
the progression of antisocial behavior are considered in 3. Windle M: A longitudinal study of antisocial behaviors in early
the present model. Finally, the inclusion of male and adolescence as predictors of late adolescent substance use: gen-
der and ethnic group differences. J Abnorm Psychol 1990; 99:
female subjects in the present analyses represents a dis- 86–91
tinction from the predominantly male-based body of 4. Robins LN, Price RK: Adult disorders predicted by childhood
literature on the development of antisocial behavior. conduct problems: results from the NIMH Epidemiologic Catch-
The ability of variables assessed during treatment to ment Area project. Psychiatry 1991; 54:116–132
significantly predict antisocial personality disorder 5. Brown SA, Gleghorn A, Schuckit M, Myers MG, Mott MA: Con-
duct disorder among adolescent substance abusers. J Stud Alco-
status 4 years later has important clinical implications. hol 1996; 57:314–324
Adolescents who present for treatment of substance 6. Carroll KM, Power MD, Bryant K, Rounsaville BJ: One-year fol-
abuse may appear to be uniformly conduct disordered, low-up status of treatment-seeking cocaine abusers—psychopa-
and thus attention to conduct disorder behavior may thology and dependence severity as predictors of outcome. J
Nerv Ment Dis 1993; 181:71–79
seem to add little information for prognosis or treat- 7. Rounsaville BJ, Dolinsky ZS, Babor TF, Meyer RE: Psychopa-
ment planning. However, the findings of this study sug- thology as a predictor of treatment outcome in alcoholics. Arch
gest that careful assessment of the extent of conduct Gen Psychiatry 1987; 44:505–513
disorder behavior and identification of behavior that 8. Carroll KM, Ball SA, Rounsaville BJ: A comparison of alternate
occurs prior to and independent of substance abuse systems for diagnosing antisocial personality disorder in cocaine
abusers. J Nerv Ment Dis 1993; 181:436–444
may serve to identify individuals at risk of persisting in 9. O’Boyle M: Personality disorder and multiple substance depen-
antisocial behavior. Adolescents whose conduct disor- dence. J Personality Disorders 1993; 7:342–347
der behavior occurs primarily in the context of sub- 10. Cacciola JS, Alterman AI, Rutherford MJ, Snider EC: Treatment

484 Am J Psychiatry 155:4, April 1998

 MYERS, STEWART, AND BROWN

response of antisocial substance abusers. J Nerv Ment Dis 1995; PW: Psychometric evaluation of the Customary Drinking and
183:166–171 Drug Use Record (CDDR): a measure of adolescent alcohol and
11. McKay JR, Buka SL: Issues in the treatment of antisocial adoles- drug involvement. J Stud Alcohol (in press)
cent substance abusers. J Child and Adolescent Substance Abuse 21. Hare RD: Comparison of procedures for the assessment of psy-
1994; 3:59–81 chopathy. J Consult Clin Psychol 1985; 53:7–16
12. Stowell RJA, Estroff TW: Psychiatric disorders in substance- 22. Sobell LC, Sobell MB: Time-line follow-back: a technique for
abusing adolescent inpatients: a pilot study. J Am Acad Child assessing self-reported alcohol consumption, in Measuring Alco-
Adolesc Psychiatry 1992; 31:1036–1040 hol Consumption: Psychosocial and Biochemical Methods. Ed-
13. Myers MG, Brown SA, Mott MA: Preadolescent conduct disor- ited by Litten RZ, Allen JP. Totowa, NJ, Humana Press, 1992,
der behaviors predict relapse and progression of addiction for pp 73–98
alcohol and drug abusers. Alcohol Clin Exp Res 1995; 19:1528– 23. Babor TF, Del Boca FK: Just the facts: enhancing measurement
1536 of alcohol consumption using self-report methods. Ibid, pp 3–19
14. Moffitt TE, Caspi A, Dickson N, Silva P, Stanton W: Childhood- 24. Sobell LC, Maisto SA, Sobell MB, Cooper AM: Reliability of
onset versus adolescent-onset antisocial conduct problems in alcohol abusers’ self-reports of drinking behaviour. Behav Res
males: natural history from ages 3 to 18 years. Developmental Ther 1979; 17:157–160
Psychopathology 1996; 8:399–424 25. Brooner RK, Schmidt CW, Felch LJ, Bigelow GE: Antisocial be-
15. Moffitt TE: Adolescence-limited and life-course persistent anti- havior of intravenous drug abusers: implications for diagnosis of
social behavior: a developmental taxonomy. Psychol Rev 1993; antisocial personality disorder. Am J Psychiatry 1992; 149:482–
100:674–701 487
16. Loeber R: Antisocial behavior: more enduring than changeable? 26. Cacciola JS, Rutherford MJ, Alterman AI, Snider EC: An exami-
J Am Acad Child Adolesc Psychiatry 1990; 30:393–397 nation of the diagnostic criteria for antisocial personality disor-
17. Loeber R, Wung P, Keenan K, Giroux B, Stouthamer-Loeber M, der in substance abusers. J Nerv Ment Dis 1994; 182:517–523
Van Kammen W, Maughan B: Developmental pathways in dis- 27. Mott MA, Brown SA, Myers MG: Alcohol and polydrug relapse:
ruptive child behavior. Developmental Psychopathology 1993; evaluation of relapse models. Alcohol Clin Exp Res 1992; 16:
5:103–133 403
18. Stewart DG, Brown SA: Withdrawal and dependency symptoms 28. Windle M: A retrospective measure of childhood behavior prob-
among adolescent alcohol and drug abusers. Addiction 1995; lems and its use in predicting adolescent problem behaviors. J
90:627–635 Stud Alcohol 1992; 54:422–431
19. Brown SA, Vik PW, Creamer VA: Characteristics of relapse fol- 29. Borduin CM, Mann BJ, Cone LT, Henggeler SW, Fucci BR,
lowing adolescent substance abuse treatment. Addict Behav Blaske DM, Williams RA: Multisystemic treatment of serious ju-
1989; 14:291–300 venile offenders: long-term prevention of criminality and vio-
20. Brown SA, Myers MG, Lippke L, Tapert SF, Stewart DG, Vik lence. J Consult Clin Psychol 1995; 63:569–578

Am J Psychiatry 155:4, April 1998 485