The Validity of DSM-IV Alcohol Dependence: What Do
We Know and What Do We Need to Know?
Deborah S. Hasin, Marc A. Schuckit, Christopher S. Martin, Bridget F. Grant, Kathleen K. Bucholz, and John E. Helzer
This article presents the proceedings of a symposium at the 2002 RSA Meeting in San Francisco,
California. Deborah S. Hasin organized the symposium and co-chaired it with Marc Schuckit. The purpose
of the symposium was to provide an overview of what is known about the validity of DSM-IV and ICD-10
alcohol dependence and abuse, with a focus on work done since 1994. Presentations included: (1) Validity
of DSM-III-R alcohol dependence in adolescents, by Christopher S. Martin; (2) Reliability and validity of
DSM and ICD formulations of alcohol use disorders: findings from epidemiology, by Bridget F. Grant; (3)
Validity and reliability of the alcohol-dependence phenotype in the context of genetic studies, by Kathleen
K. Bucholz; and (4) DSM-IV and beyond: uniting the clinical utility of categories with the precision of
dimensions, by John E. Helzer. The findings supported the validity of DSM-IV alcohol dependence across
numerous study designs and samples, suggested some value in a dimensional dependence measure, and
raised questions about the validity of the diagnosis of alcohol abuse as currently defined. Marc Schuckit, as
discussant for the symposium, placed the issues in perspective for the upcoming DSM-V.
Key Words: Alcohol Dependence, Alcohol Abuse, Validity, Reliability, DSM-IV, Diagnosis.
Deborah S. Hasin
In contrast to earlier DSMs, DSM-IV decision-making
emphasized the importance of reliability and validity evi-
dence as a basis for decision-making. Major categories such
as major depression and schizophrenia did not change
markedly since DSM-III (American Psychiatric Associa-
tion, 1980), partly because reliability and validity data for
these categories had been accumulating since the 1970s. In
contrast, the criteria for alcohol abuse and dependence
changed markedly from DSM-III to DSM-III-R (American
Psychiatric Association, 1987), with the Alcohol Depen-
dence Syndrome (ADS) (Edwards, 1986; Edwards and
Gross, 1976) as the basis for the change (Rounsaville,
1987). The dependence criteria in DSM-III-R, ICD-10
(World Health Organization, 1992), and DSM-IV (Amer-
ican Psychiatric Association, 1994) were similar, reflecting
a concept of dependence combining physiologic and psy-
chological processes. Abuse, a residual category, was de-
fined in DSM-IV by social problems or hazardous use and,
in ICD-10, by hazardous use. The changes in alcohol abuse
and dependence since DSM-III stimulated many US and
international reliability and validity studies, most published
only after 1994. Comments concerning a lack of informa-
tion about alcohol-dependence validity (Meyer, 2001) indi-
cated a need to present the substantial and accumulating
reliability and validity literature on alcohol abuse and
Reliability pertains to validity because poor reliability con-
strains validity. Test–retest reliability studies of alcohol abuse
and dependence were conducted as part of the COGA study
(Bucholz et al., 1995), in the US general population (Grant et
al., 1995), in substance abuse treatment settings (Hasin et al.,
1996b,1997a), in international settings (Chatterji et al., 1997),
and in Puerto Rican medical patients (Canino et al., 1999).
These studies uniformly showed excellent test–retest reliabil-
ity for alcohol dependence but much lower reliability for
abuse. When abuse was considered independently from de-
pendence, rather than as a residual category, abuse reliability
improved, suggesting that the hierarchical relationship to de-
pendence rather than the criteria themselves contributed to
the poor reliability of DSM-IV alcohol abuse.
Several longitudinal studies addressed the stability and dis-
tinctiveness of the course of alcohol abuse and dependence
From the Mailman School of Public Health (DSH), Division of Epide-
miology, and College of Physicians and Surgeons, Department of Psychiatry
of Columbia University and the New York State Psychiatric Institute, New
York, New York; San Diego VA Healthcare System (MAS) and University of
California at San Diego Medical School, San Diego, California; Western
Psychiatric Institute and Clinic (CSM), University of Pittsburgh School of
Medicine, Pittsburgh, Pennsylvania; Division of Biometry and Epidemiology
(BFG), National Institute on Alcohol Abuse and Alcoholism, Bethesda,
Maryland; Washington University School of Medicine (KKB), St. Louis,
Missouri; and the Health Behavior Research Center (JEH), University of
Vermont, Burlington, Vermont.
Received for publication September 13, 2002; accepted October 2, 2002.
Portions of this research were supported by NIAAA grants AA K02
AA00151 (DSH) and the New York State Psychiatric Institute (DSH).
Reprint requests: Deborah S. Hasin, PhD, Columbia University/New York
State Psychiatric Institute, 1051 Riverside Drive, Box 123, New York, NY
10032; Fax: 212-543-5386; E-mail: email@example.com.
Copyright © 2003 by the Research Society on Alcoholism.
ALCOHOLISM: CLINICAL AND EXPERIMENTAL RESEARCH
Vol. 27, No. 2
Alcohol Clin Exp Res, Vol 27, No 2, 2003: pp 244–252
based on DSM-III-R and DSM-IV criteria. These include two
national samples (Grant et al., 2001; Hasin et al., 1990), a
community sample of heavy drinkers (Hasin et al., 1997c), a
sample of male university volunteers (Schuckit et al., 2000),
and subjects in the (Schuckit et al., 2001) COGA study. These
studies consistently show that dependence is likely to remain
chronic, while abuse is likely to remit and unlikely to progress
to dependence. These findings support the validity of depen-
dence and its distinction from abuse.
When different assessment methods agree well in iden-
tifying cases of a given condition, this supports the validity
of the condition (Hasin, 2002). Methods can include clas-
sification systems whose criteria vary or instruments whose
probes and coding systems vary. Studies comparing DSM-
III-R, DSM-IV, and ICD-10 assessments included samples
of substance abuse patients and nonpatients (Rounsaville
et al., 1993), alcoholism patients and their relatives (Schuc-
kit et al., 1994), community heavy drinkers (Hasin et al.,
1996a), a US national survey (Grant, 1996), and a 9-country
World Health Organization/NIH study (Hasin et al.,
1997b). Cross-instrument comparisons (AUDADIS, CIDI,
and SCAN) were conducted in the context of the WHO
study, one using ICD-10 (Pull et al., 1997) and the other
using DSM-IV (Cottler et al., 1997). All studies showed
excellent cross-method agreement for alcohol dependence,
much lower agreement for abuse, and better agreement for
abuse when considered nonhierarchically.
whether abuse and dependence diagnoses offer clinically
useful information in general population samples of
heavier than average drinkers. One study addressed this
question in a community sample (Hasin et al., 1997c) and
another in a national sample (Hasin and Paykin, 1999).
Both showed that dependence was strongly associated with
clinically relevant characteristics such as drinking level,
family history of alcoholism, treatment, and suicidality,
supporting its validity. Abuse was weakly and inconsistently
associated with these variables.
Finally, data from the National Longitudinal Alcohol Epi-
demiologic Survey (NLAES) (Grant, 1997) were used to ex-
amine whether alcohol dependence can more validly be seen
as a categorical or continuous measure (Hasin et al., 2002).
Based on early drinking onset, family history of alcoholism,
comorbid drug use, and treatment for alcohol problems, evi-
dence failed to support any specific symptom count (including
three or more) as demarcating a categorical threshold. This
suggests validity for alcohol dependence as a dimensional
measure, consistent with its original conceptualization (Ed-
wards, 1986; Edwards and Gross, 1976).
VALIDITY OF DSM-IV ALCOHOL DEPENDENCE
Christopher S. Martin
Patterns of alcohol use and related problems tend to
differ in adults and adolescents. Compared with adults,
adolescent drinking tends to involve relatively infrequent
but high quantity binge drinking (Deas et al., 2000), and
most adolescents with DSM-IV alcohol dependence drink,
on average, only every other day (Martin et al., 1995). In
addition, almost all adolescent problem drinkers are poly-
drug users (Martin et al., 1996a), which complicates the
assessment of DSM-IV alcohol abuse and dependence
The meaning of the DSM-IV dependence criteria may
differ in adolescents and adults. For example, during ado-
lescence, onset of tolerance is probably a normal develop-
mental phenomenon (Martin and Winters, 1998). We
found that change-based definitions of tolerance (such as
the DSM-IV definition) are very poor at distinguishing
adolescents with and without alcohol dependence (Chung
et al., 2001). Another example is the criterion: “Using more
or longer than intended.” Adolescents often do not have
fixed intentions regarding drinking limits, so drinking more
than intended in this group often seems to be due to social
reasons rather than a compulsion to drink (Chung and
Martin, 2002; Cooper, 1994).
Concurrent Validity. Among teens, those with DSM-IV
alcohol dependence, alcohol abuse, and no alcohol use
disorder (AUD) differ in alcohol use and alcohol problem
severity, as well as comorbid psychiatric disorders and psy-
chosocial functioning, indicating concurrent validity of the
diagnoses. Data from a clinical sample of adolescents stud-
ied in the Pittsburgh Adolescent Alcohol Research Center
(PAARC) serve as an example. Teens in this sample with
current dependence (n ? 204), abuse (n ? 169), and no
AUD (n ? 148) differed significantly from each other on
average drinks per occasion in the past year, the past-year
average frequency of drinking, the number of illicit drugs
ever used, and the number of other substance use disorders
(SUD). This pattern of results has been observed repeat-
edly with adolescent clinical and community samples.
Teens with alcohol dependence also have significantly
greater substance use severity, more comorbid psychopa-
thology, and poorer psychosocial functioning than nonde-
pendent teens (Lewinsohn et al., 1996; Martin et al., 1995;
Winters et al., 1999).
Despite this evidence for concurrent validity of DSM-IV
alcohol dependence among adolescents, alternative algo-
rithms can also produce robust group differences in alcohol
problem severity and related functioning. For example, in
the PAARC sample, grouping participants by the number
of current dependence criteria (0–7) produced a clear grad-
uated increase across all seven groups for variables indicat-
ing current drinking, without evidence for a clear cutpoint
indicating the most valid threshold for dependence. A
threshold of 1, 2, or 4 dependence criteria produced group
differences as large as the DSM-IV threshold of 3 out of 7.
This apparent arbitrariness for the diagnostic threshold
suggests limits in the validity of this diagnostic category.
Predictive Validity. Predictive validity is a critical compo-
nent of a useful diagnostic system, as diagnoses should
VALIDITY OF DSM-IV ALCOHOL DEPENDENCE
Grant BF (1996) DSM-IV, DSM-III-R, and ICD-10 alcohol and drug
abuse/harmful use and dependence, United States, 1992: a nosological
comparison. Alcohol Clin Exp Res 20:1481–1488.
Grant BF, Harford TC, Dawson DA, Chou PS, Pickering R (1995) The
Alcohol Use Disorder and Associated Disabilities Schedule (AUDA-
DIS): reliability of alcohol and drug modules in a general population
sample. Drug Alcohol Depend 39:37–44.
Grant BF, Stinson FS, Harford TC (2001) Age at onset of alcohol use and
DSM-IV alcohol abuse and dependence: a 12-year follow-up. J Subst
Harford TC, Grant BF (1994) Prevalence and population validity of
DSM-III-R alcohol abuse and dependence: the 1989 National Longitu-
dinal Survey on Youth. J Subst Abuse 6:37–44.
Hasin D (2002) Classification of alcohol use disorders. Alcohol Res
Health, in press.
Hasin D, Carpenter KM, McCloud S, Smith M, Grant BF (1997a) The
Alcohol Use Disorder and Associated Disabilities Interview Schedule
(AUDADIS): reliability of alcohol and drug modules in a clinical
sample. Drug Alcohol Depend 44:133–141.
Hasin D, Grant BF, Cottler L, Blaine J, Towle L, Üstün B, Sartorius N
(1997b) Nosological comparisons of alcohol and drug diagnoses: a
multisite, multi-instrument international study. Drug Alcohol Depend
Hasin D, Grant BF, Endicott J (1990) The natural history of alcohol
abuse; implications for definitions of alcohol use disorders. Am J Psy-
Hasin D, Li Q, McCloud S, Endicott J (1996a) Agreement between
DSM-III, DSM-III-R, DSM-IV and ICD-10 alcohol diagnoses in US
community-sample heavy drinkers. Addiction 91:1517–1527.
Hasin D, Paykin A (1999) Alcohol dependence and abuse diagnoses:
concurrent validity in a nationally representative sample. Alcohol Clin
Exp Res 23:144–150.
Hasin D, Trautman K, Miele G, Samet S, Smith M, Endicott J (1996b)
Psychiatric Research Interview for Substance and Mental Disorders
(PRISM): reliability for substance abusers. Am J Psychiatry 153:1195–
Hasin D, Van Rossem R, McCloud S, Endicott J (1997c) Alcohol depen-
dence and abuse diagnoses: validity in community sample heavy drink-
ers. Alcohol Clin Exp Res 21:213–219.
Hasin DS, Muthen B, Grant BF (1993) The dimensionality of DSM-IV
alcohol abuse and dependence: factor analysis in a clinical sample. Drug
Alcohol Depend 88:1079–1090.
Kranzler HR, Tennen H, Babor TF, Kadden RM, Rounsaville BJ (1997)
Validity of the longitudinal, expert, all data procedure for psychiatric
diagnosis in patients with psychoactive substance use disorders. Drug
Alcohol Depend 45:93–104.
Lewinsohn P, Rodhe P, Seeley J (1996) Alcohol consumption in high
school students: frequency of use and dimensional structure of related
problems. Addiction 91:375–390.
Lynch K, Martin C, Chung T (2000) Factor structure of DSM-IV alcohol
use disorder symptoms in a clinical sample of adolescents. Alcohol Clin
Exp Res 24S:S75A.
Martin C, Kaczynski N, Maisto S, Bukstein O, Moss H (1995) Patterns of
DSM-IV alcohol abuse and dependence symptoms in adolescent drink-
ers. J Stud Alcohol 56:672–680.
Martin C, Kaczynski N, Maisto S, Tarter R (1996a) Polydrug use in
adolescent drinkers with and without DSM-IV alcohol abuse and de-
pendence. Alcohol Clin Exp Res 20:1099–1108.
Martin C, Langenbucher J, Kaczynski N, Chung T (1996b) Staging in the
onset of DSM-IV alcohol abuse and dependence symptoms in adoles-
cents: survival/hazard analyses. J Stud Alcohol 57:549–558.
Martin C, Winters K (1998) Diagnosis and assessment of alcohol use
disorders among adolescents. Alcohol Health Res World 22:95–105.
Meyer RE (2001) Finding paradigms for the future of alcoholism re-
search: an interdisciplinary perspective. Alcohol Clin Exp Res 25:1393–
Muthen B, Grant B, Hasin D (1993) The dimensionality of alcohol abuse
and dependence: factor analysis of DSM-III-R and proposed DSM-IV
criteria in the 1988 National Health Interview Survey. Addiction 88:
Nelson CB, Rehm J, Üstün B, Grant BF, Chatterji S (1999) Factor
structure for DSM-IV substance disorder criteria endorsed by alcohol,
cannabis, cocaine and opiate users: results from the World Health
Organization Reliability and Validity Study. Addiction 94:843–855.
Pull CB, Saunders JB, Mavreas V, Cottler LB, Grant BF, Hasin DS,
Blaine J, Mager D, Üstün B (1997) Concordance between ICD-10
alcohol and drug use disorder criteria and diagnoses as measured by the
AUDADIS-ADR, CIDI and SCAN: results of a cross-national study.
Drug Alcohol Depend 47:207–216.
Robins LN, Helzer JE, Croughan J, Ratcliff K (1981) National Institute of
Mental Health Diagnostic Interview Schedule: its history, characteris-
tics, and validity. Arch Gen Psychiatry 38:381–389.
Robins LN, Helzer JE, Ratcliff KS, Seyfried W (1982) Validity of the
Diagnostic Interview Schedule, version II: DSM-III diagnoses. Psychol
Rounsaville BJ (1987) An evaluation of the DSM-III substance-use dis-
orders, in Diagnosis and Classification in Psychiatry (Tischler G ed), pp
175–194, Cambridge University Press, New York.
Rounsaville BJ, Bryant K, Babor TF, Kranzler H, Kadden R (1993) Cross
system agreement for substance use disorders: DSM-III-R, DSM-IV
and ICD-10. Addiction 88:337–348.
Schuckit MA, Hesselbrock V, Tipp JE, Anthenelli R, Bucholz KK, Rad-
ziminski S (1994) A comparison of DSM-III-R, DSM-IV and ICD-10
substance use disorders diagnoses in 1922 men and women subjects in
the COGA study. Addiction 89:1629–1638.
Schuckit MA, Smith TL, Danko GP, Bucholz KK, Reich T, Bierut L
(2001) Five-year clinical course associated with DSM-IV alcohol abuse
or dependence in a large group of men and women. Am J Psychiatry
Schuckit MA, Smith TL, Landi NA (2000) The 5-year clinical course of
high-functioning men with DSM-IV alcohol abuse or dependence. Am J
Spitzer RL (1983) Psychiatric diagnosis: are clinicians still necessary?
Compr Psychiatry 24:399–411.
Winters K, Latimer W, Stinchfield R (1999) The DSM-IV criteria for
adolescent alcohol and cannabis use disorders. J Stud Alcohol 60:337–
World Health Organization (1992) The ICD-10 Classification of Mental
and Behavioral Disorders. World Health Organization, Geneva.
HASIN ET AL.