ALCOHOL ABUSE AND DEPENDENCE CRITERIA AS PREDICTORS OF A CHRONIC COURSE OF ALCOHOL USE DISORDERS IN THE GENERAL POPULATION

Alcohol & Alcoholism Vol 40, No 5, pp 441 446, 2005 Advance Access publication 11 July 2005 doi:101093/alcalc/agh183 ALCOHOL ABUSE AND DEPENDENCE CRITERIA AS PREDICTORS OF A CHRONIC COURSE OF ALCOHOL USE DISORDERS IN THE GENERAL POPULATION CARLA DE BRUIJN*, WIM VAN DEN BRINK 1,2, RON DE GRAAF 3 and WILMA A M VOLLEBERGH 3 Department of Psychiatry, University Medical Centre, Utrecht, The Netherlands, 1 Department of Psychiatry, Academic Medical Centre, Amsterdam, The Netherlands, 2 Amsterdam Institute for Addiction Research, Amsterdam, The Netherlands and 3 Netherlands Institute of Mental Health and Addiction, Utrecht, The Netherlands (Received 12 May 2005; first review notified 1 June 2005; in revised form 20 June 2005; accepted 21 June 2005; advance access publication 11 July 2005) Abstract Aims: To investigate whether DSM-IV abuse and dependence criteria and the ICD-10 criterion for craving differentially predict a chronic course of alcohol use disorders (AUD) in the general population Methods: Data were derived from the Netherlands Mental Health Survey and Incidence Study, a large representative sample of the general Dutch population with a baseline and a 1- and 3-year follow-up assessment In the present study, a cohort of subjects with a DSM-IV AUD diagnosis at baseline was followed (n 382) Diagnostic criteria of AUD according to DSM-IV and ICD-10 were assessed using the Composite International Diagnostic Interview (CIDI) Results: In our cohort of subjects with an AUD diagnosis at baseline, the presence of all dependence criteria, except tolerance, significantly increased the risk for dependence at 1 and 3 years follow-up Abuse criteria displayed much lower and often non-significant risks for dependence at follow-up, with the exception of the criterion legal problems The ICD-10 criterion craving had the highest relative risk (RR) of all criteria for dependence at 1 year (RR 124, 95% CI 55 278) and 3 years follow-up (RR 129, 95% CI 44 377) Conclusion: With the exception of tolerance, all DSM-IV dependence criteria are useful in predicting the course of AUD in the general population INTRODUCTION The course of alcohol use disorders (AUD) in the general population differs from the course in treatment seeking populations (Cohen and Cohen, 1984; Cunningham, 1995; Caetano, 1999; Raimo et al, 1999) In a large prospective general population study, we found that 85% of subjects with DSM-IV abuse and 74% of subjects with DSM-IV alcohol dependence lost their diagnosis in the course of 3 years (De Bruijn, van den Brink, de Graaf and Vollebergh, submitted for publication) This is in contrast with the chronic relapsing course of treatment-seeking alcoholics (Finney and Moos, 1991) The challenge for primary care physicians is to identify the alcoholics who are at risk for developing chronic alcohol problems and to provide treatment or referral for these patients One of the possible strategies might be to identify AUD criteria or symptoms that are associated with an increased risk for a chronic course The presence of the DSM-IV AUD criteria and its association with severity of alcoholism have been extensively studied in both clinical and general population samples (Andreatini et al, 1994; Caroll et al, 1994; Dawson, 1998; Caetano, 1999; Chung et al, 2002; Langenbucher, 2004) In most studies the criteria using larger amounts or over a longer period (AD 3), tolerance (AD 1) and hazardous use (AA 2) have been found to occur frequently and to be associated with less severe forms of alcoholism, whereas the criteria a great deal of time is spent (AD 5), important activities are given up (AD 6) and legal problems (AA 3) are less prevalent and associated with more severe forms of alcoholism Also the order of onset of the *Author to whom correspondence should be addressed at: University Medical Centre, Utrecht, PO Box 85500, 3508 GA Utrecht, The Netherlands Tel: 31 30 2507467; Fax: 31 30 2505443; E-mail: hdebruijn-2@psychazunl symptoms has been studied quite extensively (Schuckit et al, 1993, 1995, 1998a; Langenbucher and Chung, 1995; Martin et al, 1996; Nelson et al, 1996, 1998; Chung and Martin, 2002) To sum up, hazardous use and loss of control are criteria that develop relatively early, although several studies find that withdrawal is a symptom with a relatively late onset Few studies have been performed to investigate whether AUD criteria are able to differentially predict a chronic course of alcohol problems over time Schuckit et al studied the predictive validity of the DSM-IV AUD criteria among two samples The first study involved high functioning Caucasian men, half of whom had an alcoholic parent However, this study did not primarily investigate chronicity, since the majority of these subjects did not have an AUD diagnosis at baseline Furthermore, the specific characteristics of the sample (high functioning sons of alcoholics) limits the generalizability of these findings (Schuckit et al, 2000) The authors replicated their findings in a sample of treatment-seeking alcoholics, their relatives, and controls (Schuckit et al, 2001, 2005) In this study, subjects were divided in three groups: those with no baseline diagnosis, those with abuse at baseline, and those with dependence at baseline A comparison was made between subjects who displayed any DSM-IV AUD criterion at a 5-year follow-up and those who did not Since we are interested in the ability of the criteria to predict a chronic course, we will only discuss the data on the subjects who had an AUD diagnosis at baseline The authors found that among dependent subjects, DSM-IV criteria were not able to predict the occurrence of AUD-criteria at the 5-year follow-up (681% reported AUDcriteria at the follow-up) When abuse and dependence subjects were combined, all dependence criteria except AD 6 (important activities given up) were significantly associated with the occurrence of AUD criteria at the 5-year follow-up Of the abuse items only AA 1 (role obligations) was associated with the occurrence of AUD criteria at the follow-up 441 The Author 2005 Published by Oxford University Press on behalf of the Medical Council on Alcohol All rights reserved

442 C DE BRUIJN et al Dawson et al (2005) studied the correlates associated with recovery from alcohol dependence in the general population and found that a larger number of dependence symptoms increased the odds for abstinent recovery but decreased the odds for non-abstinent recovery This study, however, was retrospective and did not regard the individual AUD criteria The ICD-10 criteria have been studied less frequently The diagnostic criteria for alcohol dependence according to DSM-IV and ICD-10 largely overlap, but ICD-10 dependence has a criterion for craving, whereas DSM-IV does not Craving is increasingly considered to be the core symptom of alcohol dependence (Tiffany and Conklin, 2000) There is, however, no consensus on how to define this concept Research on craving as a predictor of relapse, and on the association of craving with severity of alcoholism, is inconclusive (van den Brink, 1997; Bottlender and Soyka, 2004; Mezinskis et al, 2001) In conclusion, the prevalence, order of onset, and severity of the DSM-IV AUD criteria have been studied in clinical as well as community samples However, the potential of the individual criteria to predict chronicity has only been studied among treatment-seeking alcoholics In this study, among a cohort of general population subjects with abuse or dependence, we investigate whether the presence of certain DSM-IV AUD criteria or the ICD-10 criterion for craving at baseline differentially increases the risk of still having these disorders at 1- and 3-year follow-up METHODS Subjects The data were derived from the Netherlands Mental Health Survey and Incidence Study (NEMESIS) NEMESIS is a prospective study collecting data in three waves (1996, 1997, and 1999) from a national, multistage random sample (age 18 64 years) in the Netherlands At the first wave (T 0 ) a total of 7076 people were interviewed (response rate 697%) The respondents adequately reflected the Dutch population For more detailed information, see an earlier report on NEMESIS (Bijl et al, 1998) For the present analyses we selected a cohort of the subjects who met the criteria for alcohol abuse or dependence at T 0 (n 382) Of these subjects, 298 (780% of T 0 ) were reinterviewed after 1 year (T 1 ) and 238 (623% of T 0, 799% of T 1 ) were reinterviewed after 3 years (T 2 ) Within our study cohort, the presence of the DSM-IV AUD criteria and craving predicted loss to follow-up neither at T 1 nor at T 2 Instruments Diagnostic criteria The CIDI 11 was used to assess criteria of AUD In these analyses we assessed the presence of the criteria during the past 12 months at T 0 and T 1 and during the last 24 months at T 2 The CIDI 11 is a reliable and validated, fully structured diagnostic interview, enabling to make diagnoses according to ICD-10 and DSM-III-R criteria (Cottler et al, 1991) The interviewers had been given a 4-day training course at the WHO-CIDI training and reference centre of the Academic Medical Centre in Amsterdam Since DSM-IV uses the same AUD criteria as DSM-III-R, we could establish DSM-IV criteria based on the Composite International Diagnostic Interview (CIDI) answers Subsequently, DSM-IV abuse and dependence diagnoses were made A subject was only diagnosed as having either abuse or dependence when the subject met the full criteria Subjects who were partially remitted did not get a diagnosis Subjects with a past dependence diagnosis, who met current criteria for abuse only, were diagnosed as having abuse Data analyses At T 0, the presence of the DSM-IV AUD criteria and the ICD-10 criterion craving was assessed Among the subjects with a DSM-IV AUD at T 0, for each criterion, the proportion of subjects that still had abuse or dependence at T 1 and at T 2 was calculated The proportions were compared in crosstabs using the chi-square statistic as the indicator for statistical significance Subsequently, the relative risk (RR) for still having abuse or dependence at follow-up was calculated using the same crosstabs procedure If risk ratios were significantly 1, a forward stepwise logistic regression analysis was performed, entering both the individual AUD criterion and the total number of AUD criteria present at baseline as predictors All statistical analyses were performed with the Statistics Package for Social Sciences (SPSS for Windows, 120, 2003) RESULTS Sample characteristics Table 1 displays the characteristics of the 382 subjects with an AUD diagnosis at baseline At T 0, 299 subjects met current abuse criteria according to DSM-IV (783% of 382; 42% of 7076) and 83 subjects were diagnosed with current dependence (217% of 382; 12% of 7076) Only 8 subjects (21% of 382) had been seeking specialized treatment for their addiction in the last 12 months and 26 subjects (68%) in their lifetime AUD criteria and diagnosis at follow-up Table 2 displays the DSM-IV AUD criteria and the ICD-10 criterion craving for all 382 subjects with a DSM-IV AUD diagnosis at baseline For each criterion, the proportions and RRs for still having abuse or dependence at T 1 and T 2 are shown The risk for having dependence at follow-up was significantly increased for nearly all dependence criteria and for Table 1 Sample characteristics of the cohort of 382 subjects with a DSM-IV AUD diagnosis at baseline Characteristics x a,b (SD) % (n) Mean age a (SD) 342 (113) Sex, (male) 814 (311) Employed 647 (247) Number of drinks/week b (SD) 269 (307) Seeking specialized treatment for 21 (8) addiction in the last 12 months Seeking specialized treatment for 68 (26) addiction in lifetime Abuse 783 (299) Dependence 217 (83) Abuse of 298 subjects at T 1 178 (53) Dependence of 298 subjects at T 1 94 (28) Abuse of 238 subjects at T 2 134 (32) Dependence of 238 subjects at T 2 71 (17) a x, the age in years b x, number of drinks/week

ABUSE AND DEPENDENCE CRITERIA AS PREDICTORS OF CHRONICITY 443 Table 2 Diagnosis at 1 and 3 years follow-up for criteria of DSM-IV AUD and the ICD-10 criterion craving in the cohort of 382 subjects with a baseline DSM-IV AUD diagnosis T 1 (n 298) T 2 (n 238) Abuse Dependence Abuse Dependence T 0 (n 382) % RR % RR % RR % RR Presence of AUD criteria % (n) (n) (95% CI) (n) (95% CI) (n) (95% CI) (n) (95% CI) AD 1 Tolerance 212 (81) 224 14 179 26** 271 27** 125 22 (15) (08 23) (12) (13 52) (13) (14 51) (6) (08 55) AD 2 Withdrawal 188 (72) 182 10 236 38***,a 93 06 209 51***,a (10) (06 19) (13) (19 76) (4) (02 18) (9) (21 125) AD 3 Larger amounts or longer than 218 14 193 69***,a 140 11 161 117***,a intended 395 (151) (26) (09 24) (23) (27 177) (13) (05 21) (15) (27 500) AD 4 Persistent desire to cut down or 182 10 432 122***,a 129 10 290 75***,a control use 134 (51) (8) (05 20) (19) (59 252) (4) (04 25) (9) (31 180) AD 5 Great deal of time obtaining, 217 12 435 66***,a 235 19 294 54*** using or recovering from effects (5) (06 28) (10) (35 127) (4) (07 47) (5) (22 136) 79 (30) AD 6 Important activities given up or 83 05 417 52*** 200 15 300 49** reduced 42 (16) (1) (01 30) (5) (24 113) (2) (04 55) (3) (17 143) AD 7 Continued use despite physical 209 14 194 109***,a 204 25** 117 31* or psychological harm 432 (165) (27) (08 22) (25) (34 354) (21) (13 50) (12) (11 86) Craving 188 (72) 259 16 362 124***,a 125 09 271 129***,a (15) (10 28) (21) (55 278) (6) (04 21) (13) (44 377) AA 1 Failure to fulfil major role 211 12 158 18 63 04 125 19 obligations 68 (26) (4) (05 30) (3) (06 53) (1) (01 31) (2) (05 74) AA 2 Hazardous use 686 (262) 178 10 64 04* 138 11 69 09 (36) (06 17) (13) (02 08) (22) (05 22) (11) (03 23) AA 3 Legal problems 52 (20) 200 11 267 31* 182 14 364 64*** (3) (04 32) (4) (13 79) (2) (04 50) (4) (25 163) AA 4 Continued use despite social or 206 13 155 24* 203 20* 76 11 interpersonal problems 327 (125) (08 21) (15) (12 48) (16) (11 38) (6) (04 29) AD, alcohol dependence; AA, alcohol abuse; RR, relative risk for abuse and dependence Chi square *P 005; **P 001; ***P 0001 a RR for dependence RR for abuse craving The only exception was AD 1 (tolerance), which did not show a significantly increased risk for dependence at T 2 Only two of the four abuse criteria did significantly increase the risk for having dependence at follow-up: AA 3 (legal problems) and AA 4 (despite social or interpersonal problems) at T 1 and AA 3 at T 2 The risk for having abuse at T 1 was not increased for any of the AUD criteria; at T 2 it was increased only for AD 1 (tolerance), AD 7 (despite physical or psychological harm), and AA 4 (despite social or interpersonal problems) For almost all diagnostic criteria (dependence criteria, abuse criteria and craving), RRs for dependence were higher than RRs for abuse at follow-up The only exception was AA 2 (hazardous use), which displayed a significantly reduced risk of having dependence at follow-up For craving and for five of the seven DSM-IV dependence criteria, but for none of the DSM-IV abuse criteria, RRs for dependence at T 1 were significantly higher than for abuse at T 1 At T 2, the RRs for dependence were significantly higher than for abuse only for three of the seven dependence criteria (AD 2: withdrawal; AD 3: larger or longer; AD 4: quit or cut down) and for craving Craving, AD 3 (larger or longer), and AD 4 (quit or cut down) demonstrated the highest RRs for dependence at T 1 and T 2, although in most cases the confidence intervals of the RRs overlapped with the other criteria These criteria (craving and AD 4 for T 1, craving and AD 3 for T 2 ) remained significant predictors of dependence at follow-up in the logistic regression analyses, adding significantly to the effect of the total number of AUD criteria The other criteria did not significantly predict chronicity when the total number of AUD criteria was incorporated in the analyses, with the exception of AD 6 and AA 4; they significantly predicted dependence at T 1 in the logistic regression analyses The presence of AA 2 (hazardous use) significantly reduced the risk for having dependence at T 1 We, therefore, examined these subjects closer Among the 262 subjects who displayed AA 2, this was the only criterion met at baseline for 120 subjects (458% of 262) Of the 120 subjects who did not meet AA 2, only 4 (33% of 120) displayed just one AUD criterion, the other 116 subjects met several AUD criteria at baseline Withdrawal requiring tremor Earlier studies have found that withdrawal (AD 2) is associated with more severe forms and later stages of alcoholism (Schuckit et al, 1998b, 2003; Lejoyeux et al, 2001) However, hangover symptoms are sometimes misinterpreted as withdrawal symptoms Therefore, the suggestion has been made to sharpen the withdrawal definition (Langenbucher et al, 1997; Hasin et al, 2000) Hasin found that requiring tremor for withdrawal gave it a stronger prognostic meaning (Hasin et al, 2000) We recalculated the values for withdrawal when tremor was required The prevalence of withdrawal requiring tremor was reduced from 188% (n 72) to 68% (n 26) Withdrawal requiring tremor did not significantly increase the risk for abuse at either

444 C DE BRUIJN et al T 1 or T 2, but the RRs for dependence at T 1 (RR 53, 95% CI 26 105) and T 2 (RR 76, 95% CI 32 178) were higher than the RRs when tremor was not required (AD 2) (T 1 : RR 38, 95% CI 19 76; T 2 : RR 51, 95% CI 21 125) In the logistic regression analyses, including the total number of AUD criteria, withdrawal requiring tremor did significantly add to the prediction of dependence at T 2 but not at T 1 DISCUSSION In this prospective general population study on the predictive validity of the DSM-IV AUD criteria and craving, we found that nearly all dependence criteria and craving significantly increased the risk for still having alcohol dependence at 1 and 3 years follow-up The abuse criteria showed much lower and often non-significant risks for having dependence at follow-up The risk for having abuse at T 1 was not increased for any of the AUD criteria and was increased for only a few criteria at T 2 Craving, AD 3 (larger or longer), and AD 4 (quit or cut down) at T 0 had relatively high RRs for predicting dependence at both 1 and 3 years follow-up These criteria remained significant predictors of chronicity even after correcting for overall severity (total number of AUD criteria met at baseline) In contrast, the physiological dependence criteria, tolerance (AD 1) and withdrawal (AD 2), had relatively low RRs for having dependence at follow-up in comparison with the other DSM-IV criteria However, requiring tremor for the presence of withdrawal considerably increased the risk for dependence at follow-up Methodological issues The present study included all subjects with a baseline DSM- IV AUD diagnosis from a large representative general population sample and used a well-validated instrument (CIDI) to assess DSM-III-R and DSM-IV AUD criteria (Compton et al, 1996; Cottler et al, 1997; Pull et al, 1997; Üstun et al, 1997) Despite the large sample size, some criteria (AD 5, AD 6, AA 1, and AA 3) resulted in small numbers (30, 16, 26, and 20, respectively) because they occur infrequently in the general population Even with these small numbers we found significant RRs for predicting dependence at follow-up, but the confidence intervals for the RRs were rather broad Earlier findings In general, our results converge with the findings by Schuckit et al among treatment-seeking subjects with either abuse or dependence at baseline, demonstrating that DSM-IV dependence criteria are useful in predicting the course of AUD, whereas most abuse criteria are not (Schuckit et al, 2001) There are, however, several differences in the designs of the studies First, the outcome in the Schuckit et al study was the occurrence of any AUD criterion at follow-up, whereas we studied the presence of a full diagnosis of either abuse or dependence at follow-up Furthermore, Schuckit et al divided their subjects according to their baseline diagnosis (dependence, abuse, or no AUD) and studied the associations between the baseline AUD criteria and the outcome for each of these three groups separately They did the same analyses for the combined groups of subjects with either abuse or dependence at baseline The authors found less significant associations when dependence subjects and abuse subjects were considered separately, than when abuse and dependence subjects were combined Since we were interested to see whether the criteria could predict a chronic course of AUD in the general population, we confined the present analyses to subjects with an AUD diagnosis at baseline In an earlier report on NEMESIS, predictors of incidence of substance use disorders were assessed In these analyses, a much broader range of possible predictors was investigated The only AUD criterion that was assessed was AD 4 (quit or cut down), which proved to be a strong predictor of the first incidence of substance use disorders (De Graaf et al, 2002) In order to avoid very small numbers and thus lose power, we did not consider subjects with abuse and subjects with dependence at baseline separately Furthermore, the clinical relevance of this distinction is of limited value in our general population sample Our goal was to provide primary care physicians with indicators of a chronic course of AUD Most of the time, the DSM-IV diagnostic status is not extensively assessed in primary care settings Screening instruments for primary care settings Several screening instruments have been recommended to screen for alcohol problems in primary care settings The two most commonly studied screening instruments are the CAGE [acronym based on its four questions: Cut down drinking, Annoyed by criticism about drinking, Guilty feelings about drinking, and Eye opener (drinking in the morning), Mayfield et al, 1974] and the alcohol use disorder identification test (AUDIT, Babor et al, 1989) The AUDIT is more effective in identifying subjects with at-risk, hazardous drinking, whereas the CAGE is superior for detecting alcohol abuse and dependence (Aertgeerts et al, 2004; Fiellin et al, 2000) The criteria that we found to have the highest ability to predict chronicity ( craving, larger or longer and quit or cut down ) match the CAGE questions rather well These findings might implicate that the CAGE is not only useful for the detection of AUD in the general population and in primary care, but especially for the identification of subjects who are at risk for a chronic subtype of alcohol dependence This hypothesis calls for further examination, taking into account all items of the CAGE and the AUDIT Tolerance and withdrawal According to DSM-IV, a diagnosis of alcohol dependence should be accompanied by the specification with or without physiological dependence Physiological dependence is specified when either withdrawal or tolerance is present The severity and predictive validity of these criteria have been studied quite extensively Most studies find that tolerance is not strongly associated with severity or worse prognosis, whereas the findings for withdrawal depend on the way in which this symptom is defined (Caroll et al, 1994; Schuckit et al, 1998, 2003; Hasin et al, 2000; Langebucher et al, 1997; Lejoyeux et al, 2001) Our results converge with these findings; tolerance demonstrated relatively low risk for dependence at T 1 and was the only dependence criterion that

ABUSE AND DEPENDENCE CRITERIA AS PREDICTORS OF CHRONICITY 445 did not significantly increase the risk for dependence at T 2 Withdrawal as defined in DSM-IV performed relatively weak among the dependence criteria However, requiring the presence of tremor or withdrawal substantially increased the RR for having dependence at follow-up Abuse Alcohol abuse according to DSM-IV requires the presence of just one criterion This results in a heterogeneous group of subjects The validity of alcohol abuse, especially of the hazardous use criterion, is often criticized (Hasin et al, 1999, 2003; Rounsaville, 2002; De Bruijn et al, 2004; De Bruijn et al, 2005) Our results show that the abuse criteria, with the exception of legal problems (AA 3) demonstrated no or just slightly increased risks for having either abuse or dependence at 1 and 3 years follow-up The finding that AA 3 had a stronger prognostic meaning links up with earlier findings, demonstrating that this criterion is relatively rare and associated with more severe forms of alcoholism (Muthen et al, 1993; Muthen, 1995; Chung and Martin, 1996) A remarkable finding is the fact that the criterion hazardous use (AA 2) significantly lowered the risk of having dependence at T 1 AA 2 was the most prevalent of all AUD criteria For nearly half of the subjects who displayed AA2, this was the only criterion they met It might well be that hazardous use represents some kind of time-limited risky behaviour, but that it does not belong to the prototype characteristics of an AUD per se Craving Craving is not considered among the dependence criteria in DSM-IV As stated in the introduction, we decided to include this symptom in the study, since it is increasingly considered to be the core symptom of alcoholism We found that craving was associated with the highest risk for dependence at followup among the dependence criteria Our findings, therefore, support its usefulness as a dependence criterion CONCLUSION In the general population, DSM-IV dependence criteria and craving are able to predict the 3 year course of AUD The abuse criteria are less useful in predicting the AUD course The criteria craving, larger or longer (AD 3), and persistent desire to quit or cut down alcohol use (AD 4) are relatively strongly associated with a chronic course of dependence Primary care physicians should be alert on the existence or development of a chronic alcohol problem among subjects who endorse these criteria REFERENCES Aertgeerts, B, Buntinx, F and Kester, A (2004) The value of the CAGE in screening for alcohol abuse and alcohol dependence in general clinical populations: a diagnostic meta-analysis Journal of Clinical Epidemiology 57, 30 39 Andreatini, R, Galduroz, J C, Ferri, C P et al (1994) Alcohol dependence criteria in DSM-III-R: presence of symptoms according to degree of severity Addiction 89, 1129 1134 Babor, T F, Krantzler, H R and Lauerman, R J (1989) Early detection of harmful alcohol consumption: comparison of clinical, laboratory, and self-report screening procedures Addictive Behaviours 14, 139 157 Bijl, R V, van Zessen, G, Ravelli, A et al (1998) The Netherlands Mental Health Survey and Incidence Study (NEMESIS): objectives and design Social Psychiatry and Psychiatric Epidemiology 33, 581 586 Bottlender, M and Soyka, M (2004) Impact of craving on alcohol relapse during, and 12 months following, outpatient treatment Alcohol and Alcoholism 39, 357 361 Caetano, R (1999) The identification of alcohol dependence criteria in the general population Addiction 94, 255 267 Carroll, K M, Rounsaville, B J and Bryant, K J (1994) Should tolerance and withdrawal be required for substance dependence disorders? Drug and Alcohol Dependence 36, 15 22 Chung, T and Martin, C S (2002) Concurrent and discriminant validity of DSM-IV symptoms of impaired control over alcohol consumption in adolescents Alcoholism: Clinical and Experimental Research 26, 485 492 Chung, T, Martin, C S, Armstrong, T D et al (2002) Prevalence of DSM-IV alcohol diagnoses and symptoms in adolescent community and clinical samples Journal of the American Academy of Child and Adolescent Psychiatry 41, 546 554 Cohen, P and Cohen, J (1984) The clinician s illusion Archives of General Psychiatry 41, 1178 1182 Compton, W M, Cottler, L B, Dorsey, K B et al (1996) Comparing assessments of DSM-IV substance dependence disorders using CIDI-SAM and SCAN Drug and Alcohol Dependence 41, 179 187 Cottler, L B, Robins, L N, Grant, B F et al (1991) The CIDI-core substance abuse and dependence questions: cross-cultural and nosological issues The WHO/ADAMHA Field Trial British Journal of Psychiatry 159, 653 658 Cottler, L B, Grant, B F, Blaine, J et al (1997) Concordance of DSM-IV alcohol and drug use disorder criteria and diagnoses as measured by AUDADIS-ADR, CIDI and SCAN Drug and Alcohol Dependence 47, 195 205 Cunningham, J A, Sobell, L C, Sobell, M B et al (1995) Resolution from alcohol problems with and without treatment: reasons for change Journal of Substance Abuse 7, 365 372 Dawson, D A (1998) Symptoms and characteristics of individuals with different types of recovery from DSM-IV alcohol dependence Journal of Substance Abuse 10, 127 142 Dawson, D A, Grabnt, B F, Stinson, F S et al (2005) Recovery from DSM-IV alcohol dependence: United States, 2001 2002 Addiction 100, 281 292 De Bruijn, C, Korzec, A, Koerselman, F et al (2004) Craving and withdrawal as core symptoms of alcohol dependence Journal of Nervous and Mental Disease 192, 494 502 De Bruijn, C, van den Brink, W, de Graaf, R et al (2005) The craving withdrawal model for alcoholism: towards the DSM-V Improving the discriminant validity of alcohol use disorder diagnosis Alcohol and Alcoholism 40, 314 322 De Graaf, R, Bijl, R V, Ravelli, A et al (2002) Predictors of first incidence of DSM-III-R psychiatric disorders in the general population: findings from the Netherlands Mental Health Survey and Incidence Study Acta Psychiatrica Scandinavia 106, 303 313 Fiellin, D A, Reid, M C and O Connor, P G (2000) Screening for alcohol problems in primary care: a systematic review Archives of Internal Medicine 160, 1977 1989 Finney, J W and Moos, R H (1991) The long-term course of treated alcoholism: I Mortality, relapse and remission rates and comparisons with community controls Journal of Studies on Alcohol 52, 44 54 Hasin, D, Paykin, A, Endicott, J et al (1999) The validity of DSM- IV alcohol abuse: drunk drivers versus all others Journal of Studies on Alcohol 60, 746 755 Hasin, D, Paykin, A, Meydan, J et al (2000) Withdrawal and tolerance: prognostic significance in DSM-IV alcohol dependence Journal of Studies on Alcohol 61, 431 438 Hasin, D S, Schuckit, M A, Martin, C S et al (2003) The validity of DSM-IV alcohol dependence: what do we know and what do we need to know? Alcoholism: Clinical and Experimental Research 27, 244 252

446 C DE BRUIJN et al Langenbucher, J W and Chung, T (1995) Onset and staging of DSM- IV alcohol dependence using mean age and survival-hazard methods Journal of Abnormal Psychology 104, 346 354 Langenbucher, J, Chung, T, Morgenstern, J et al (1997) Physiological alcohol dependence as a specifier of risk for medical problems and relapse liability in DSM-IV Journal of Studies on Alcohol 58, 341 350 Langenbucher, J W, Labouvie, E, Martin, C S et al (2004) An application of item response theory analysis to alcohol, cannabis, and cocaine criteria in DSM-IV Journal of Abnormal Psychology 113, 72 80 Lejoyeux, M, Claudon, M, McLoughlin, M et al (2001) Comparison of alcohol-dependent patients with and without physiological dependence European Addiction Research 7, 198 201 Martin, C S, Langenbucher, J W, Kaczynski, N A et al (1996) Staging in the onset of DSM-IV alcohol symptoms in adolescents: survival/ hazard analyses Journal of Studies on Alcohol 57, 549 558 Mayfield, D, Mcleod, G and Hall, P (1974) The CAGE questionnaire: validation of a new alcoholism screening instrument American Journal of Psychiatry 131, 1121 1123 Mezinskis, J P, Honos-Webb, L, Kropp, F et al (2001) The measurement of craving Journal of Addictive Disorders 20, 67 85 Muthen, B O (1995) Factor analysis of alcohol abuse and dependence symptom items in the 1988 National Health Interview Survey Addiction 90, 637 645 Muthen, B O, Grant, B and 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, 1079 1090 Nelson, C B, Little, R J, Heath, A C et al (1996) Patterns of DSM-III-R alcohol dependence symptom progression in a general population survey Psychological Medicine 26, 449 460 Nelson, C B, Heath, A C and Kessler, R C (1998) Temporal progression of alcohol dependence symptoms in the US household population: results from the National Comorbidity Survey Journal of Consulting and Clinical Psychology 66, 474 483 Pull, C B, Saunders, J B, Mavreas, V et al (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 and Alcohol Dependence 47, 207 216 Raimo, E B, Daeppen, J B, Smith, T L et al (1999) Clinical characteristics of alcoholism in alcohol-dependent subjects with and without a history of alcohol treatment Alcoholism: Clinical and Experimental Research 23, 1605 1613 Rounsaville, B J (2002) Experience with ICD-10/DSM-IV substance use disorders Psychopathology 35, 82 88 Schuckit, M A, Smith, T L, Anthenelli, R et al (1993) Clinical course of alcoholism in 636 male inpatients American Journal of Psychiatry 150, 786 792 Schuckit, M A, Anthenelli, R M, Bucholz, K K et al (1995) The time course of development of alcohol-related problems in men and women Journal of Studies on Alcohol 56, 218 225 Schuckit, M A, Daeppen, J B, Tipp, J E et al (1998a) The clinical course of alcohol-related problems in alcohol dependent and nonalcohol dependent drinking women and men Journal of Studies on Alcohol 59, 581 590 Schuckit, M A, Smith, T L, Daeppen, J B et al (1998b) Clinical relevance of the distinction between alcohol dependence with and without a physiological component American Journal of Psychiatry 155, 733 740 Schuckit, M A, Smith, T L and Landi, N A (2000) The 5-year clinical course of high-functioning men with DSM-IV alcohol abuse or dependence American Journal of Psychiatry 157, 2028 2035 Schuckit, M A, Smith, T L, Danko, G P et al (2001) Five-year clinical course associated with DSM-IV alcohol abuse or dependence in a large group of men and women American Journal of Psychiatry 158, 1084 1090 Schuckit, M A, Danko, G P, Smith, T L et al (2003) A 5-year prospective evaluation of DSM-IV alcohol dependence with and without a physiological component Alcoholism: Clinical and Experimental Research 27, 818 825 Schuckit, MA, Smith, TL, Danko, GP et al (2005) Prospective evaluation of the four DSM-IV criteria for alcohol abuse in a large population American Journal of Psychiatry 162, 350 360 Tiffany, S T and Conklin, C A (2000) A cognitive processing model of alcohol craving and compulsive alcohol use Addiction 95, S145 S153 Üstün, B, Compton, W, Mager, D, et al (1997) WHO study on the reliability and validity of the alcohol and drug use disorder instruments: overview of methods and results Drug and Alcohol Dependence 47, 161 169 van den Brink W (1997) Editorial: craving and relapse prevention European Addiction Research 3, 197 199