Indian Journal of Psychiatry, 2000, 42 (4), 378-386
ENDORSEMENT AND CONCORDANCE OF ICD-10
VERSUS DSM-IV CRITERIA FOR SUBSTANCE
DEPENDENCE : INDIAN PERSPECTIVE
DEBASISH BASU, NITIN GUPTA, NARENDRA SINGH,
SURENDRA K. MATTOO & PARMANAND KULHARA
ABSTRACT
Substance use disorders have undergone major changes in both the international (ICD-10)
and American (DSM-IV) nosological systems, thus necessitating a study of cross-system agreement
between ICD-10 and DSM-IV substance dependence, especially from a developing country setting.
Further, endorsement rates for various substance dependence criteria in the two systems need to
be studied from a similar perspective. Hence, 221 consecutive patients with 279 diagnostic categories
of substance dependence attending a de-addiction centre in Northern India were studied with regard
to endorsement of the various ICD-10 and DSM-IV criteria as well as for cross-system agreement
for severity of dependence. High endorsement rates were seen for most criteria in both the systems,
except for those related to persistence despite harm' and 'salience (neglect of various activities,
plus excessive time spent to procure the substance)'. There were some significant differences,
however, between endorsement rates across different substance categories in both the systems.
Cross-system agreement on severity of substance dependence ranged from fair to good for all
categories combined, and was good to excellent for the opioid category. The category of others'
(nicotine, cannabis and sedative-hypnotics) showed poor cross-system agreement Overall, the
results lend support to the basic theoretical construct behind both ICD-10 and DSM-IV substance
dependence syndrome from a developing country perspective.
Keywords: Substance dependence, diagnosis, endorsement, agreement, nosological systems
The diagnosis of psychiatric disorders has Substance use disorders have undergone
undergone rapid changes in the past years, major changes in various classificatory systems
These changes have been necessary as the over the years especially after the advent of DSM-
available diagnostic systems, at any given time, III. The concept of substance dependence in
did not adequately cover all of the classified DSM-III(APA,1980) centred around tolerance and
phenomena (Cottier et al ,1991) leading onto withdrawal related to any substance (Cottier et
revisions in the Diagnostic and Statistical Manual al.,1991). However, a significant change occurred
(DSM) and International Classification of from DSM-III-R (APA, 1987) onwards In this, the
diseases (ICD) systems However, another major focus shifted to dependence being a disorder
reason for development of newer version is for centred around compulsive use and loss of control
improving predictive accuracy (Cottier et al., often leading to social, physical and psychological
1991). The current classificatory systems in consequences (Cottier et al.,1991). DSM-IV
vogue are the ICD-10 (WHO, 1992) and DSM- retained the conceptual framework underlined in
IV (APA, 1994) which attempt at fulfilling the DSM-III-R, and the two editions of DSM were
above mentioned objectives. highly comparable, especially for alcohol use
378
ICD-10 VERSUS DSM-IV CRITERIA FOR SUBSTANCE DEPENDENCE
disorders (Grant et al.,1992; Rousanville et
\ al.,1993; Hasin & Grant, 1994).
DSM-lll-R's development conceptualised
and centred around the originally formulated
dependence syndrome of Edwards and Gross
(1976) and Edwards etal. (1976). Infact, the ICD-
10 also represents a version of the same criteria
by Edwards and Gross (1976). Thus, there is a
common theoretical link for the dependence
criteria in the diagnostic systems of DSM-III-R,
DSM-IV and ICD-10 (Rounsaville et al.,1986;
Hasin et al.,1996). Studies have been carried
out to validate this common theoretical construct
(Grant, 1993, Cottier, 1993; Rounsaville et al.,
1993; Langenbucher et al.,1994; Hasin et al.,
1996). Although these studies yielded results that
appeared to validate this theoretical construct,
yet apart from the studies by Langenbucher et
al. (1994) and Hasin et al. (1996), other studies
used the working draft of DSM-IV (with 11 items),
and not the final version (with 7 items). Also
many of these comparative studies on DSM-IV
and ICD-10 focussed exclusively on alcohol
(Grant,1993; Hasin et al.,1996). Further, the
samples varied from clinical to community
populations to an admixture of the two.
As has been mentioned earlier, the DSM-
III-R borrowed heavily from the concept of
dependence syndrome' given by Edwards et al.
(1976). DSM-III-R listed nine criteria which were
analyzed in a study by Cottier et al. (1991) so as
to assess the impact of the individual criteria in
arriving at a diagnosis of substance dependence.
It was seen that endorsement rates for the criteria
related to 'withdrawal' and tolerance' were low
for the substances - alcohol and cannabis; and
high for opiates and nicotine. However, no other
study has commented upon the endorsement
rates for various criteria needed for substance
dependence even among different substances.
This is important to assess so as to evaluate the
relative importance of each criteria in arriving at
a diagnosis of substance dependence.
There is a need to study the above
mentioned issues especially in the context of
substance dependence in developing countries.
Substance use patterns and concepts of social
use vis-a-vis pathological use vary substantially
between different cultures (Babor & Mendelson,
1986; Miller,1986). Further, as noted by the
originator of the concept of alcohol dependence
syndrome himself, "its presentation will be
shaped by the pathoplastic influence of
personality and culture "(Edwards, 1986, p. 172).
Hence, criteria developed for diagnosis of
substance dependence in developed western
culture may or may not necessarily be highly
endorsed in people from the developing
countries constituting different cultural
backgrounds. For a similar reason, cross-system
agreement between DSM-IV and ICD-10 needs
to be studied from a developing country's
perspective. No such effort has been made so
far, to the best of our knowledge.
Objectives :
1. To study the rates of endorsement of various
diagnostic criteria for substance dependence
using ICD-10 and DSM-IV in a consecutive
sample of substance dependence patients
attending a de-addiction centre in Northern India.
2. To find out the extent of endorsement of
various diagnostic criteria, using ICD-10 and
DSM-IV, across different categories of
substances in the sample studies.
3. To examine the degree of agreement between
ICD-10 and DSM-IV diagnostic systems with
regard to the severity of substance dependence.
MATERIAL AND METHOD
Sample : The sample consisted of patients
attending the outpatient clinic or admitted to the
inpatient unit of the Drug De-addiction and
Treatment Centre (DDTC), Department of
Psychiatry, PGIMER, Chandigarh over the period
1996-97. The sample recruited was a
consecutive type meeting the inclusion and
exclusion criteria.
Inclusion criteria :
1. Attending the DDTC for treatment of
substance dependence, i.e., meeting atieast 3
criteria of either ICD-10 or DSM-IV.
379
DEBASISH BASU ef a/.
2. Cooperative patient
3. Duration of substance use for at least twelve
months
4. Accompanied by a reliable and key informant
e.g. family member staying with the patient.
Exclusion criteria :
1. Psychotic illness
2. Marked cognitive impairment
3. Acute intoxication at time of assessment
Design: Assessment was done at a single time,
i.e., it was a cross-sectional (survey) design.
Instruments :
1. ICD-10(WHO, 1992)
2. DSM-IV (APA.1994)
3. Sociodemographic profile sheet developed by
Department of Psychiatry, PGIMER, Chandigarh.
Assessment Procedure : A semi-structured
proforma incorporating all the basic and relevant
clinical details was used for eliciting details of
the substance(s) use. A diagnosis of substance
dependence (and any other comorbid illness)
was thereafter established by one of the two
investigators (SKM, DB). Following this, another
investigator (NS) assessed the patient as regards
the fulfilment and endorsement of diagnostic
criteria for ICD-10 and DSM-IV. Only one
investigator (NS) was used for assessment of
the diagnostic criteria so as to remove the
confounding variability induced by presence of
two separate raters.
Operational criteria : To determine the degree of
agreement between ICD-10 and DSM-IV
diagnostic criteria as regards the severity of
substance dependence, criteria were
operationalised for varying grades of severity viz.
1. Mild degree of severity : fulfilment of three
diagnostic criteria in either ICD-10 or DSM-IV
2. Moderate : fulfilment of either 4-5 diagnostic
criteria in ICD-10 or fulfilment of 4-6 diagnostic
criteria in DSM-IV.
3. Severe : fulfilment of all six diagnostic criteria
in ICD-10 or all seven criteria in DSM-IV.
Categorization of severity of substance
dependence is not incorporated in the diagnostic
systems and lacks adequate validation.
However, this exercise of categorization was
alluded to being possibly helpful by Edwards and
Gross (1976). They stated " the need to see
dependence in terms of degrees rather than
absolutes may have some message for clinical
practice". Hence, the authors felt the need to
attempt the same.
Statistical analysis : For determining the rates
of endorsement, frequency values were calculated
Extent of endorsement was determined by
applying chi-square test. Agreement between
ICD-10 and DSM-IV classificatory systems was
assessed with the Kappa (K) coefficient. This
coefficient indicates the level of agreement
beyond the level expected due to chance. K'
values can range from +1.00 (perfect agreement)
to -1.00 (total disagreement). K' of 0.75 and
above indicates excellent agreement, from 0.40
to 0.74 indicates fair to good agreement, and
below 0.39 indicates poor agreement (Fleiss,
1981). A' K' of zero indicates agreement no better
than chance. Standard errors and 95% confidence
intervals were computed for all K' values.
RESULTS
A total of 221 patients were taken up for
the study as per the inclusion and exclusion
criteria. All the patients were male. The
sociodemographic profile is presented in table 1.
It was seen that there was a
preponderance of married patients in the age
groups of 25-44 years belonging to Hindu
religion, of nuclear families and of urban
background.
The diagnostic break-up of the whole
sample is shown in table 2 It should be noted
here that the overall diagnostic number (n=279)
exceeded the sample size (n=221) as there were
58 comorbid diagnosis for the sample of patients.
Henceforth, the analysis further presented was
done for the overall diagnostic number (n=279)
rather than for the total sample of patients
(n=221).
Regarding the duration of substance
dependence, more than half the diagnosis
categories (n=279) had substance use of 10
380
ICD-10 VERSUS DSM-IV CRITERIA FOR SUBSTANCE DEPENDENCE
TABLE 1
SOCIODEMOGRAPHIC PROFILE OF SAMPLE (N=221)
Variable
Age (in years)
15-24
25-34
35-44
>45
Marital status
Married
Not married
Education
Under matric
Matnc/Diploma
Graduate/Postgraduate
Occupation
Professional/Semi-professiona
Clerk/Fqrmer/Shopowner
Semi-skilled/Unskilled/
Skilled worker
Unemployed/Others
Religion
Hinduism
Sikhism
Others
Family type
Nuclear
Non-nuclear
Locality
Urban
Rural
Number of
patients
33
72
76
40
165
56
70
86
55
36
84
53
48
119
95
7
131
80
167
54
Percentage
14.93
32.58
34.39
18.10
74.66
2534
31.75
38.84
29.41
16.29
38.01
23.98
21.71
53.85
42.99
3.16
59.28
40.72
75.57
24!43
TABLE 3
PERCENTAGE ENDORSEMENT FOR ICD-10
CRITERIA FOR ALL DIAGNOSIS (N=279)
TABLE 2
DIAGNOSTIC BREAKDOWN OF SAMPLE (N=221)
Diagnosis Number
Alcohol dependence syndrome 124
Opioid dependence syndrome 109
Nicotine dependence syndrome 28
Cannabis dependence syndrome 6
Sedative/Hypnotic dependence
syndrome 12
Total number of diagnostic categories 279
Percentage
44.44
39.06
10:05
2.15
4.30
100.00
* Total number of diagnostic categories exceeds the
sample (N=221) because there were 58 concurrent
diagnosis of dependence on an other substance.
years or more (n=145).
Thereafter, the percentage endorsement
(i.e. number of diagnostic categories fulfilling a
criteria) for the diagnostic criteria of ICD-10
(Table 3) and DSM-IV (Table 4) were calculated.
It was seen that endorsement rates in ICD-
10 were low for criterion number 5 (neglect of
alternative pleasures and excessive time spent
Criterion
Desire/compulsion for use
Impaired control
Withdrawal
Tolerance
Neglect of alternative
Endorsement rate (%)
99 64
92.11
81.00
93.19
pleasureS+time spent excessively 73.84
Persistence despite harm 65 95
TABLE 4
PERCENTAGE ENDORSEMENT FOR DSM-IV
CRITERIA FOR ALL DIAGNOSIS (N=279)
Criterion
Use over longer periods/in
larger amount
Impaired control
Withdrawal
Tolerance
Social/OccupationaL/Recreational
activities given up
Use despite physical/psychological
Excessive t;
me spent to obtain
Endorsement rate (%)
harm
99 28
81 72
80 65
94 62
73.84
7491
56 63
in procuring the substance; 73.84%) and criterion
number 6 (persistence despite harm; 65.95%).
For all other criteria, endorsement rates ranged
from 81% to 99.64%.
In DSM-IV, the endorsement rates were
low for criterion number 6 (use despite physical
or psychological harm; 74.91%), criterion
number 5 (neglect of social, occupational and
recreational activities; 73.84%) and criterion
number 7 (time spent to procure the substance;
56.63%). For all other criteria, endorsement rates
ranged from 80.65% to 99.28%
However, to determine the extent of
endorsement across different categories of
substance dependence in the ICD-10 /Table 5)
and DSM-IV (Table 6). the endorsement values
per criterion for each substance were calculated
and subjected to chi-square analysis.
In the ICD-10 system, there was no
significant difference across the substances -
alcohol, opioids and others' for criterion number
1 (desire/compulsion for use) and criterion
number 4 (tolerance)
In the DSM-IV system, there was no
significant difference across the substances -
381
DEBASISH BASU ef a/.
TABLE 5
ENDORSEMENT RATES FOR ICD-10 CRITERIA
ACROSS DIAGNOSIS OF DEPENDENCE ON ALCOHOL
(N=124), OPIOIDS (N=109) AND OTHERS' (N=46)a
Criterion
Desire/ compulsion
for use
Impaired control
Withdrawal
Tolerance
Neglect of
Endorsement rate (%)
Alcohol
100.00
96.77
86.29
95.16
alternative pleasures
+Time spent
excessively
Persistence
despite harm
80.65
72.58
Opioids
99.08
87.16
97.25
91.74
80.73
66.97
Others
100.00
91.30
28.26
91.30
39.13
45 65
Chi-square
value
1.75NS
7.72*
104.06"*
1.39NS
34.32***
11.73"
a = Cannabis, Nicotine, Sedative & Hypnotics
NS = Not significant; *p < 0 05; "*p p< 0.01; ***p<0.001
TABLE 6
ENDORSEMENT VALUES FOR DSM-IV CRITERIA
ACROSS DIAGNOSIS OF DEPENDENCE ON ALCOHOL
(N=124), OPIOIDS (N=109) AND OTHERS' (N=46)a
Criterion
Use over longer
periods/in larger
amounts
Impaired control
Withdrawal
Tolerance
Endorsement rate (%)
Alcohol
100.00
80.65
85.48
95.97
Social/Occupational/
Recreational
activities given up 83.87
Use despite physical/
psychological harm
Excessive time
spent to obtain
78.23
5081
Opioids
99.08
87.16
96.33
92.66
82.87
76.15
71.56
Others
97.83
71.74
30.43
95.65
26.09
63.04
36.96
Chi-square
value(df=2)
2.47NS
5.31NS
93.36"*
1.36NS
65.10*"
4.26NS
18 8 6 " *
a = Cannabis, Nicotine, Sedative & Hypnotics
NS = Not significant; * " p < 0 001
alcohol, opioid and 'others' for criterion number 1
(use over long periods OR in excessive amount),
criterion number 2 (impaired control), criterion
number 4 (tolerance) and criterion number 6 (use
despite physical/psychological harm).
Finally, degree of agreement between
ICD-10 and DSM-IV was determined for all the
diagnostic categories as well for the individual
ones (Table 7). Agreement was calculated
related to two subsets of dependence i.e.
Moderate (Mild + Moderate of operational
criteria) and severe - as per operational definition
of 'severe'.
The cross-system agreement was fair to
TABLE 7
CONCORDANCE BETWEEN ICD-10 AND DSM-IV
SYSTEMS REGARDING SEVERITY OF DEPENDENCE
Diagnostic
group
All categories
combined
Alcohol
dependence
syndrome
Opioid
dependence
syndrome
Others'
dependence
syndrome
Kappa(K)
value
0.55
0.39
0.68
-0.10
Standard
error of K'
0.05
0.08
0.07
0.46.
95%
confidence
Interval of'K'
0.45 to 0.65
0.23 to 0.55
0.54 to 0.82
-1.00 to 0.80
good for all diagnostic categories combined
together; even with the 95% confidence interval.
As regards individual substances, the results were
varied. For alcohol, level of agreement was
bordering on fair but the confidence interval
showed a wide range (from poor to fair agreement).
For opioids, agreement was good with variability
range extending from fair to excellent agreement.
For other substances, agreement levels appeared
to be due to chance only.
DISCUSSION
Endorsement rates across the two systems
Endorsement rates in the ICD-10 were
highest for criteria of compulsion for use',
'tolerance' and impaired control' and lowest for
'neglect of alternative pleasures and excessive
time spent' and persistence despite harm'.
These values show that the underlying
theoretical construct in ICD-10, based on the
concept by Edwards et al. (1976), appears to be
the central component of substance
dependence. However, the consequences
arising out of substance dependence were also
382
ICD-10 VERSUS DSM-IV CRITERIA FOR SUBSTANCE DEPENDENCE
L part of the theoretical construct, which as
r observed, showed a relatively low endorsement
rate. Interestingly, the central concept of DSM-
III physiological dependence manifested by
presence of tolerance' and withdrawal', which
was modified and not felt necessary in the
revised DSM versions and ICD-10, appeared to
be a common feature amongst the endorsed
criteria for dependence; especially tolerance'.
Therefore these findings show that ICD-10,
though based heavily on conqept of Edwards et
al. (1976), appears to give credence to the DSM-
III concept too.
On the other hand, endorsement rates in
DSM-IV were highest for criteria of 'use over
longer periods/in longer amounts' and tolerance'
and lowest for use despite physical/
psychological harm', "giving up of various
activities' and 'excessive time spent', with
intermediate values for the other criteria. Thus,
it appears that both ICD-10 as well as DSM-IV
have gainfully used the theoretical construct of
dependence syndrome as envisaged by Edwards
et al. (1976). While this remains overall true, it
must be stressed here that two of the elements
of dependence syndrome described by Edwards
and Gross (1976) were not endorsed highly in
either system : persistence despite harm' and
salience' (neglect of various activities, plus
excessive time spent to procure the substance)'.
The term salience' had been originally proposed
by Edwards and Gross (1976) in which the
individual gives priority to maintaining his alcohol
intake and the same is reflected in criterion 5
(neglect of alternative pleasures and excessive
time spent in procuring the substance) of ICD-
10 and criteria 5 (neglect of social, occupational
and recreational activities) and 7 (time spent to
procure the substance) of DSM-IV. The reasons
for this need further investigation in India.
The relatively higher endorsement rates
for the construct of physiological dependence'
(see above) appear to lend more credence to
the DSM-III based criteria for substance
dependence. This appears to support the
specifier added for sub-types of substance
dependence in DSM-IV i.e. with physiological
dependence' and "without physiological
dependence'. Thus, there may be the need for
developing specifiers of with/without
physiological dependence' in ICD-10 on the
same lines as DSM-IV.
Endorsement rates across different
substance categories
Although there appeared to be some
similar trends, in both ICD-10 and DSM-IV,
regarding endorsement of various diagnostic
criteria for substance dependence in general but
it was necessary to evaluate the presence of
such trends across the major groups of
substances in our sample i.e alcohol, opioids
and others'. Using ICDr10, comparable
endorsement levels were reported across the
three groups for criteria of desire/compulsion
for use', and 'tolerance'. There was a relatively
lower level of endorsement for' impaired control'
by patients using opioids' compared to other
substances. Altnough this appears unusual but
it could be due to the fact that opioids' are a
heterogeneous group (including heroin, opium,
codeine, pentazocine, dextropropoxyphene and
buprenorphine) of substances including some
that are used for therapeutic purposes and more
quantifiable. This may have led to the cognitive
set in patients that "we can reduce the amount
by ourselves and/or substance used is not that
harmful". However, this is a mere speculation
and couid not be confirmed by the authors.
As compared to alcohol' and opioid'
groups, significantly low levels of endorsement
were reported for others' group as regards the
other three criteria (Table 5) The criteria of
withdrawal' showed low endorsement as it could
be due to lack of characteristic withdrawal
syndrome with cannabis (Kaplan & Sadock, 1995)
or because no/little attempt was ever made by
this group to abstain/drastically and abruptly
reduce their intake of the substances. The latter
hypothesis appears more tenable if the low
endorsement rates for neglect of alternative
pleasures' and persistence despite harm' are
383
DEBASISH BASU ef a/.
considered. This is so because the drugs
comprising 'others' group (cannabis, nicotine,
sedatives and hypnotics) have a social sanction
in this part of the world and are generally not
considered to be very harmful by the general
population. Also, these are freely available in
India leading to the regular usage of these drugs
without actually having to put in extra efforts in
procuring them Infact, these findings reinforce
the serious need in future to provide stringent
rules and regulations as regards procurement
and usage of these group of substances. On the
other hand, it does not imply that the criteria
showing low endorsement should be deleted for
diagnosing these disorders of substance
dependence.
Analysis of DSM-IV for the different
substances showed similar trends for' prolonged
use' and 'tolerance' with comparable
endorsements for impaired control' and
persistence despite harm'. Withdrawal',
'excessive time spent' and activities given up'
were poorly endorsed by the 'others' group.
Overall, the findings appear similar to those
derived from ICD-10 but more comparability
across the different groups in DSM-IV could be
due to its criteria being less broadly defined with
respect to ICD-10 (Rounsaville et al.,1993).
Therefore, it appeared that pattern for
endorsement for each criterion (which is
comparable in the ICD-10 and DSM-IV) was on
similar lines across the various groups of
substance dependence This showed that ICD-
10 and DSM-IV criteria are comparable
However, the concordance for severity of
dependence in the two systems was required to
establish the actual comparability of ICD-10 and
DSM-IV (see below).
Cross-system agreement for grades of
severity of dependence
High level of cross-system concordance
was evident for all diagnostic categories i.e.
whole sample and the opioid group; though with
a wide range. The wide range in opioids could
be due to the use of various types of opioids by
patients ranging from natural to synthetic opioids
(mentioned earlier in discussion) which coula
have influenced the total sample agreement too
A barely moderate level of concordance for
alcohol sub-group could be due to the presence
of greater number of cases with "less severity of
dependence" as compared to those for opioids
However, due to design of the current study, it
was not possible to confirm this observation. The
presence of poor concordance for the "others'
group provides support to the varying rates of
endorsement for both ICD-10 and DSM-IV
criteria. This supports the observations of
researchers (Rounsaville et al.,1993; Hasin et
al.,1996) that a polythetic approach to diagnosis
of substance dependence may not be actually
helpful in evaluating the severity of dependence
of a particular substance Therefore, there may
be merit in trying to sub-classify severity of
dependence not on the basis of mere number of
criteria fulfilled but probably on the basis of
presence of certain specific criteria or factors.
Although previous studies have
commented upon the good to excellent
concordance values between DSM-IV and ICD-
10 (Rounsaville et al.,1993; Grant,1993; Hasin
et al.,1996), but these have generally been on
community samples and have evaluated for both
'dependence' and harmful use/abuse'. In
keeping with these two broad categories, this
can be extrapolated to the two degrees of
severity of substance dependence defined in this
study. In the two previous studies (Grant, 1993:
Hasin et al.,1996) purely alcohol was studied,
on different samples, with different instruments
and different criteria were used These could be
some of the factors leading on to different results
in this current study
Conclusions and implicaticns
Overall, it can be concluded that ICD-10
and DSM-IV share a similar underlying basic
theoretical construct and inasmuch, a specifier
for physiological dependence' (as in DSM-IV)
may need to be added to ICD-10 to bring these
two systems closer. Also, there is a need to
384
ICD-10 VERSUS DSNUV CRITERIA FOR SUBSTANCE DEPENDENCE
further evaluate the degree of concordance for
the two systems. But, the need of the hour
appears to be to develop measures so as to
classify the severity of substance dependence.
This would help in a more rational approach to
the evaluation and treatment of patients with
possible integration to the psychosocial
implications (insurance claims, financial aspects
etc.) of substance dependence.
However, this study had certain
limitations, viz. the sample being clinic-based,
studying only 'dependence' and not abuse/
harmful use', small sample size for substances
other than alcohol and opioids, and lack of
structured instruments for arriving at initial
clinical diagnosis.
Nevertheless, this study assumes
importance as being the first of its kind in
assessing endorsement of criteria and
concordance of severity of dependence;
especially from a developing country. This could
help in further refining the concept of substance
dependence.
REFERENCES
APA (1980) Diagnostic and Statistical
Manual of Mental Disorders, Edn.,3rd
(DSM-III).
Washington DC : American Psychiatric
Association.
APA (1987) Diagnostic and Statistical
Manual of Mental Disorders, Edn.3rd
, revised
(DSM-III-R). Washington DC : American
Psychiatric Association.
APA (1994) Diagnostic and Statistical
Manual of Mental Disorders, Edn.,^ (DSM-IV).
Washington DC : American Psychiatric
Association.
Babor.T.F. & Mendelson.J.H. (1986)
Ethnic/religious differences in the manifestation
and treatment of alcoholism. Annals of New York
Academy of Sciences, All, 46-59.
Cottier,!..B., Helzer.J.E., Mager.D.,
Spitznagel.E.L. & Compton.W.M. (1991)
Agreement between DSM-III and lll-R substance
use disorders. Drug and Alcohol Dependence,
29, 17-25.
Cottler.L.B. (1993) Comparing DSM-III-
R and ICD-10 substance use disorders.
Addiction, 88, 689-696.
Edwards, G. (1986) The alcohol
dependence syndrome : a concept as stimulus
to enquiry. British Journal of Addiction, 81,
171-183.
Edwards,G. & Gross,M. (1976) Alcohol
dependence: provisional description of a clinical
syndrome. British Medical Journal, 1,1058-1061.
Edwards,G., Arif.A. & Hodgson,R.
(1976) Nomenclature and classification of drug
and alcohol-related problems : a WHO
memorandum. Bulletin of World Health
Organization, 59, 225-242.
Fleiss.J. (1981) Statistical Methods for
Rates and Proportions, Edn.2. New York: Wiley.
Grant,B., HarfordJ., Hasin,D., Chou.P.
& Pickering.R. (1992) DSM-III-R and proposed
DSM-IV alcohol use disorders, U.S. 1988 : a
nosological comparison. Alcohol and Clinical
Experimental Research, 16, 215-221.
Grant.B.F. (1993) ICD-10 and proposed
DSM-IV harmful use of alcohol/alcohol abuse
and dependence, United States 1988 : a
nosological comparison. Alcohol and Clinical
Experimental Research, 17, 1093-1101
Hasin.D. & Grant.B.F. (1994) Nosological
comparisons of DSM-II-R and DSM-IV alcohol
abuse and dependence in a clinical facility :
comparison with the 1988 National Health
Interview Survey results Alcohol and Clinical
Experimental Research, 18, 272-279
Hasin.D., Li,Q., Mccloud.S. & Endicott,
385
DEBASISH BASU et al.
J. (1996) Agreement between DSM-III, DSM-III-
R, DSM-IV and ICD-10 alcohol diagnoses in US
community-sample heavy drinkers. Addiction,
91, 1517-1527.
Kaplan,H.I. & Sadock.B.J (1995)
Comprehensive Textbook of Psychiatry, Edn.6*\
Baltimore : Williams & Wilkins.
the United States. Annals of the New York
Academy of Sciences, 472, 110-129.
Rounsaville,B.J., SpitzenR. & Williams,
J. (1986) Proposed changes in DSM-III
substance use disorders : description and
rationale. American Journal of Psychiatry, 143,
463-468.
Langenbucher.J., Morgenstern.J.,
Labouvei,E. & Nathan.P.E. (1994) Diagnostic
concordance of substance use disorders in DSM-
III, DSM-IV and ICD-10. Drug and Alcohol
Dependence, 36, 193-203.
Miller.W.R. (1986) Haunted by the
Zeitgeist : reflections on contrasting treatment
goals and concepts of alcoholism in Europe and
Rounsaville.B.J., Bryant,K., Babor.T.,
Kranzler.H. & Kadden.R. (1993) Cross system
agreement for substance use disorders :
DSM-III-R, DSM-IV & ICD-10. Addiction, 88,
337-348.
WHO (1992) The ICD-10 classification of
mental and behavioural disorders. Geneva :
World Health Organization.
DEBASISH BASU'. MD. DNB. Associate Professor. NITIN GUPTA, MD, Assistant Professor, NARENDRA SINGH, MD, Ex-
Senior Resident, S.KMATTOO, MD, Additional Professor & PARMANAND KULHARA, MD, FAMS. FRCPsych. Professor &
Head. Department of Psychiatry, Drug De-addiction and Treatment Centre, Postgraduate Institute of Medical Education and
Research. Chandigarh-160012 (E-mail: ab_sm@satyam.net.in).
* Correspondence
386