SUPPLEMENT ARTICLE
doi:10.1111/add.12901
The alcohol dependence syndrome: a legacy of continuing clinical and scientific importance Tim Stockwell Department of Psychology, Centre for Addictions Research of British Columbia, University of Victoria, Victoria, Canada
ABSTRACT This paper offers some reflections on Griffith Edwards’ continuing legacy with particular reference to his and Milton Gross’s formulation of alcohol dependence as a ‘provisional’ clinical syndrome. The ideas and language from this seminal paper have heavily influenced international diagnostic classification systems. However, it is observed that there has also been significant (and increasing) divergence—in particular around the original proposal that dependence and negative alcoholrelated consequences are independent, if inevitably inter-related dimensions. This is most apparent in the conflation of alcohol-related problems and dependence phenomena implicit in DSM-V. It is also argued that the alcohol dependence syndrome (ADS) has substantial continuing influence and relevance to current clinical practice. The hypothesis that degree of alcohol dependence is a useful indicator of the possibility of a return to controlled drinking continues to receive support, and underpins the widespread implementation of brief interventions for ‘early stage’ problem drinkers. It is suggested that the kind of careful clinical observations that underpinned the original concept of alcohol dependence have continuing relevance to the formulation of improved understanding, measurement instruments, diagnostic systems and clinical responses. Keywords
Alcohol dependence syndrome, controlled drinking, craving, DSM V, Griffith Edwards, impaired control.
Correspondence to: Tim Stockwell, Director, Department of Psychology, Centre for Addictions Research of British Columbia, University of Victoria, Victoria, Canada. E-mail: [email protected]
INTRODUCTION ‘A person may, for example, develop cirrhosis, lose his job, crash his car, or break up his marriage through his drinking without suffering from the dependence syndrome’— Edwards & Gross [1] The concept of the alcohol dependence syndrome (ADS) is one of the most enduring and significant legacies from the late Griffith Edwards’ stellar career. This short retrospective is without doubt biased heavily by my luck to have worked under his and others’ mentorship at the Addiction Research Unit (ARU), Institute of Psychiatry and the Maudsley Hospital in the late 1970s and early 1980s. The year I joined the ARU in 1976 was the year of Griffith’s seminal publication with Milton Gross, laying out a ‘provisional description of a clinical syndrome’ [1]. He had also recently recruited two behavioural psychologists, Ray Hodgson and Howard Rankin, to lead research on understanding the role of learning in alcohol dependence and to develop psychological interventions. As very much the junior member of that team, I was allowed to focus my
© 2015 Society for the Study of Addiction
PhD studies on the ‘nature and measurement’ of the ADS. During those 3 years I sat in on many ward rounds at the Alcoholism Treatment Unit of the Bethlem Royal Hospital, where Griffith presided as the consultant in charge. He would hold forth on these occasions with extraordinary and often spellbinding eloquence on the plight of each patient brought up for discussion (always after the most searching interrogation of every member of the treatment team). There were many research meetings, seminars, conferences and opportunities to discuss ideas one-to-one. Early on I was asked to take detailed notes on discussions between Griffith and Ray Hodgson on how the key elements of the alcohol dependence syndrome could be operationalized into a self-completion questionnaire. Drawing up that questionnaire, the Severity of Alcohol Dependence Questionnaire or SADQ [2,3], administering it to just over 100 attendees of the Maudsley out-patient and Bethlem in-patient units and refining it became a part of my thesis. The team also undertook behavioural studies of drinking behaviour and responses to alcohol by individuals with a moderate or severe degree of alcohol dependence.
Addiction, 110, 8–11
The alcohol dependence syndrome
Whether Griffith was advancing public health policy on alcohol, the idea that brief interventions for alcohol-related problems could be effective, or that controlled drinking was a realistic outcome for some problem drinkers, his ideas routinely provoked fierce debates. I will focus here briefly on two implications of the concept of alcohol dependence: (i) how is it best defined and measured and (ii) is it clinically useful; for example, does it lead to differential prediction of controlled drinking versus abstinence outcomes? These ideas were closely discussed within the small world of alcohol treatment and research in the United Kingdom in the late 1970s and 1980s. The opposing view that the ADS is but the old disease model dressed up in new clothes and is, in any case, not useful for predicting controlled drinking is well represented in a discussion series in the British Journal of Addiction (e.g. [4]) and in an excellent book by Heather & Robertson [5]. I will suggest here that while the original formulation of the ADS had profound impacts on the representation of alcohol-related problems in both the major international disease classification systems (DSM and ICD), some key insights were simply lost in translation and that present diagnostic criteria remain confused as a result. I will also suggest that degree of alcohol dependence should still be considered by all involved in the treatment process when discussing controlled drinking versus abstinence treatment goals. There is now a much more substantial literature from neuroscience and learning perspectives regarding key elements of the dependence syndrome (e.g. [6]). However, confusion about the diagnostic characteristics of ‘dependence’ is likely to hamper an integrative understanding and the development of improved interventions.
THE SEMINAL STUDY ‘Anyone concerned with treating drinking problems must find that his patients often tell him more than is in the textbooks… The varied experience that is recounted can be interpreted as the patient’s astute observation of the alcohol dependence syndrome—a condition certainly far better described by the average alcoholic than in any book’ [1] The paper by Edwards & Gross [1] on the ADS is a thoughtful and structured distillation of clinical observations and accounts from their patients. It occurs to me it that would probably be extremely difficult to publish such a paper today. It would not meet any formal guidelines for review articles or for qualitative analysis. However, the ideas and observations it contains have left a large footprint. Part of the context for the paper was the increasing input of behavioural and social scientists into the field of alcohol treatment and research and a growing dissatisfaction with the dominant concept of alcoholism as a disease. © 2015 Society for the Study of Addiction
9
‘The humane argument that alcoholism should be regarded as a disease seems to have run into difficulties recently because of semantic confusion over what is meant by the over-inclusive term ‘alcoholism’ and the social arbitrariness of the ‘disease’ label’ [1] While suggesting that biological, social and learning processes were key to the development of the ADS, the authors emphasized the provisional nature of their syndrome description and the need for future research from multiple disciplines to elucidate causal pathways. These clinical impressions were structured into seven inter-related elements, each of which is illustrated with hypothetical case examples. The list of elements will be highly familiar to most readers and, in short, comprised: 1 Narrowing of the drinking repertoire in terms of daily pattern and context of drinking. 2 Increased salience of drink-seeking, so that other interests or responsibilities may take second place. 3 Increased tolerance to alcohol. 4 Repeated withdrawal symptoms, usually observed on waking when blood alcohol levels are at their lowest. 5 Relief of withdrawal symptoms after drinking, particularly first thing upon waking. 6 Awareness of a compulsion to drink and impairment of control was emphasized as opposed to the complete ‘loss of control’ described in the disease model [7]. 7 Rapid reinstatement after abstinence, so that only a day or two after returning to drinking all the above signs might return in full force. In practice, in his clinical assessments Griffith would place great store upon whether or not an individual frequently drank to relieve withdrawal symptoms upon waking daily almost daily for a year or more. Such a drinker would be classified by him as ‘severely dependent’ and, in his clinical experience, unlikely to return to controlled drinking. That, at least, was a working hypothesis that led to many a long study and much debate (e.g. [8]).
Alcohol dependence and problems as separate although related dimensions Importantly, and as is crystal clear in the quotation at the beginning of this paper, negative consequences of drinking behaviour were not seen as essential to the assessment of degree of dependence. Social, cultural and other factors mainly external to the individual drinker were seen as determinants of whether drinking was causing ‘problems’. This is in contrast to the descriptions of alcohol dependence contained both in DSM-IV and ICD-10, each of which in effect conflate ‘salience of drink-seeking behaviour’ with the presence of serious alcohol-related harms; for example, DSM-IV defined this as: ‘Continued drinking despite Addiction, 110, 8–11
10
Tim Stockwell
knowledge of having a persistent or recurrent physical or psychological problem that is likely to be caused or exacerbated by drinking’ [9]. DSM-V has taken this conflation to a new level by incorporating ‘alcohol abuse’ and ‘alcohol dependence’ into a single severity dimension [10], based principally on evidence of an observed correlation in surveys between alcohol-related problems and signs of dependence (e.g. [11]). While the idea that alcohol dependence exists on a continuum from mild to severe is captured in these modern diagnostic systems, the importance of understanding both pattern of consumption and problematic consequences as separate dimensions seems to have been entirely lost. Indeed, in an otherwise excellent discussion of the significance of the ADS, Li et al. [11] suggest that the apparent correlation between degree of dependence and harms is evidence that they are all part of the same fundamental continuum. For clinical and scientific purposes, however, Edwards & Gross [1] stress the importance of seeing these as inter-related but also independent dimensions. Heavy drinking can occur in the relative absence of dependence; even severe alcohol dependence can be hidden and effectively tolerated in certain social contexts without present harms being apparent; sometimes even apparently moderate drinking can be associated with serious harm. It is interesting to reflect upon how many other diagnosable mental health problems require the presence of social, work or legal problems to be worthy of a diagnosis. Does schizophrenia need to be associated with problems in the work-place or with the police; depression with relationship problems? Edwards & Gross, I think rightly, pointed out that it is not useful to equate a behavioural syndrome with the problems it may or may not cause and that to do so may be both bad clinical practice and bad science. As just one illustration of how this might make a practical difference, a person with severe alcohol dependence but no apparent problems would—according to the ADS formulation— have less chance of achieving a controlled drinking goal than someone with mild alcohol dependence and many problems. However, both might qualify for a diagnosis of a low-level alcohol use disorder according to DSM criteria. It is the degree of ‘dependence’ in the sense meant by Griffith Edwards (compulsion, impaired control, repeated drinking to relieve withdrawal) that is critical to chances of a return to controlled drinking, rather than simply volume of alcohol consumed or number of problems caused.
Is the degree of alcohol dependence relevant to the choice of treatment goal? Experimental studies on drinking behaviour as well as treatment outcome studies conducted at the Maudsley and ARU encouraged the view that increasing dependence © 2015 Society for the Study of Addiction
was associated with a reducing prospect of a successful return to controlled drinking. Patients assessed by Griffith as severely alcohol-dependent were shown to drink faster in a speed of drinking test and exhibit greater subjective and physiological signs of ‘craving’ after a priming dose of alcohol than mildly or moderately dependent drinkers [12]. This was interpreted as an indicator of ‘impaired control’. While much of the mainly US-based treatment literature on controlled drinking at the time failed to address the dimension of problem severity, the Maudsley study of 100 married men with alcohol problems indicated that among those assessed as returning successfully to ‘social drinking’ 3 or 4 years post-treatment, seven of eight scored below the cut-point for ‘severe alcohol dependence’ on the SADQ [13]. Inspection of successful controlled drinking studies of the time will often find that they focused upon groups of drinkers whose average consumption might only be in the range of 20–30 drinks per week. By contrast, for Griffith an individual would have to report daily drinking for withdrawal relief (shakes, sweats, anxiety) first thing upon waking for a period of at least 1 year to be considered severely dependent. More recently, Heather & Dawe [14] contrasted an impaired control scale and the SADQ in terms of their ability to predict success on a controlled drinking programme. The impaired drinking scale was clearly superior in this population. While this may reflect a relatively small number of severely dependent drinkers enrolled into the study with which to thoroughly test the hypothesis across the full range of dependence severity, the finding also highlights a key weakness in the SADQ as a measure of dependence: it fails to address the key element of impaired control. An attempt to remedy this was described in an adaptation of the SADQ for community samples [15], including the description of an Impaired Control Questionnaire. In hindsight, when compared directly to Griffith’s clinical intuition, the SADQ also fails to assess duration the pattern of dependent drinking as it concentrates only on a recent period of heavy drinking lasting a ‘few months’. In any case, I suggest that the less-than-successful outcome for the SADQ in this particular study is not confused with a lack of predictive validity of the ADS—arguably, the Heather & Dawe [14] study is, in fact, evidence of continuing relevance of the ADS to the issue of controlled drinking versus abstinence given the predictive validity of their measure of impaired control. Their excellent paper also describes many other significant studies and reviews which, at the time, concluded that severity of alcohol dependence was predictive of the possibility of a return to controlled drinking. Furthermore, Heather & Dawe [14] document evidence suggesting that the idea of the ADS had, by that time, become the dominant consideration within treatment agencies for determining when controlled drinking might be the recommended goal for treatment. Addiction, 110, 8–11
The alcohol dependence syndrome
CONCLUDING THOUGHTS Reflecting back, there can be little doubt that the formulation by Edwards & Gross [1] of a clinical syndrome continues to influence thinking, research and practice, even if internationally recommended diagnostic practices diverge increasingly from the original ‘provisional description’. Some of the continuing influences may not be immediately apparent—for example, the influence of those early Maudsley studies on the now widely held view that screening and brief intervention in health-care settings constitute best practice interventions for early-stage problem drinkers (e.g. [16]). I also think Griffith would hope that studies which sensitively capture the individual’s personal experience with alcohol dependence would continue to inform the growing and increasingly sophisticated field of addictions research. As he observed, the alcohol dependence syndrome is ‘a condition certainly far better described by the average alcoholic than in any book’ [1]— perhaps a thought that could be applied usefully to underperforming measurement instruments and diagnostic systems as well as to the value of incorporating qualitative analysis of individual experience within this area of scientific inquiry. Declaration of interests None. References 1. Edwards G., Gross M. Alcohol dependence: provisional description of a clinical syndrome. BMJ 1976; 1: 1058–61. 2. Hodgson R., Stockwell T., Rankin H., Edwards G. Alcohol dependence: the concept, its utility and measurement. Br J Addict 1978; 73: 339–42.
© 2015 Society for the Study of Addiction
11
3. Stockwell T., Murphy D., Hodgson R. The Severity of Alcohol Dependence Questionnaire: its use, reliability and validity. Br J Addict 1983; 78: 145–56. 4. Shaw S. A critique of the concept of the alcohol dependence syndrome. Br J Addict 1979; 74: 339–48. 5. Heather N., Robertson I. Controlled Drinking, revised edn. London: Methuen; 1983. 6. Baker T. E., Stockwell T., Barnes G., Holroyd C. B. Individual differences in substance dependence: at the intersection of brain, behaviour, and cognition. Addict Biol 2010; 16: 458–66. 7. Jellinek E. M. The Disease Model of Alcoholism. Hillhouse Press: New Haven; 1960. 8. Booth P. G. Maintained controlled drinking following severe alcohol dependence: a case study. Br J Addict 1990; 85: 315–22. 9. American Psychiatric Association (APA). Diagnostic and Statistical Manual of Mental Disorders, 4th edn, text revision. Washington, DC: APA; 2000. 10. Babor T. F. Substance, not semantics, is the issue: comments on the proposed Addiction criteria for DSM-V. Addiction 2011; 106: 868–97. 11. Li T., Hewitt B., Grant G. The Alcohol Dependence Syndrome 30 years later: a commentary. Addiction 2007; 102: 1522–30. 12. Hodgson R., Rankin H., Stockwell T. Alcohol dependence and the priming effect. Behav Res Ther 1979; 17: 379–87. 13. Edwards G., Duckitt A., Oppenheimer E., Sheehan M., Taylor C. What happens to alcoholics? Lancet 1983; 2: 269–71. 14. Heather N., Dawe S. Level of impaired control predicts outcome of moderation-oriented treatment for alcohol problems. Addiction 2005; 100: 945–52. 15. Stockwell T., Sitharthan T., McGrath D., Lang E. The measurement of alcohol dependence and impaired control in community samples. Addiction 1994; 89: 167–74. 16. Anderson P., Chisholm D., Fuhr D. Alcohol and global health 2: effectiveness and cost-effectiveness of policies and programmes to reduce the harm caused by alcohol. Lancet 2009; 373: 2234–46.
Addiction, 110, 8–11
doi:10.1111/add.12901
The alcohol dependence syndrome: a legacy of continuing clinical and scientific importance Tim Stockwell Department of Psychology, Centre for Addictions Research of British Columbia, University of Victoria, Victoria, Canada
ABSTRACT This paper offers some reflections on Griffith Edwards’ continuing legacy with particular reference to his and Milton Gross’s formulation of alcohol dependence as a ‘provisional’ clinical syndrome. The ideas and language from this seminal paper have heavily influenced international diagnostic classification systems. However, it is observed that there has also been significant (and increasing) divergence—in particular around the original proposal that dependence and negative alcoholrelated consequences are independent, if inevitably inter-related dimensions. This is most apparent in the conflation of alcohol-related problems and dependence phenomena implicit in DSM-V. It is also argued that the alcohol dependence syndrome (ADS) has substantial continuing influence and relevance to current clinical practice. The hypothesis that degree of alcohol dependence is a useful indicator of the possibility of a return to controlled drinking continues to receive support, and underpins the widespread implementation of brief interventions for ‘early stage’ problem drinkers. It is suggested that the kind of careful clinical observations that underpinned the original concept of alcohol dependence have continuing relevance to the formulation of improved understanding, measurement instruments, diagnostic systems and clinical responses. Keywords
Alcohol dependence syndrome, controlled drinking, craving, DSM V, Griffith Edwards, impaired control.
Correspondence to: Tim Stockwell, Director, Department of Psychology, Centre for Addictions Research of British Columbia, University of Victoria, Victoria, Canada. E-mail: [email protected]
INTRODUCTION ‘A person may, for example, develop cirrhosis, lose his job, crash his car, or break up his marriage through his drinking without suffering from the dependence syndrome’— Edwards & Gross [1] The concept of the alcohol dependence syndrome (ADS) is one of the most enduring and significant legacies from the late Griffith Edwards’ stellar career. This short retrospective is without doubt biased heavily by my luck to have worked under his and others’ mentorship at the Addiction Research Unit (ARU), Institute of Psychiatry and the Maudsley Hospital in the late 1970s and early 1980s. The year I joined the ARU in 1976 was the year of Griffith’s seminal publication with Milton Gross, laying out a ‘provisional description of a clinical syndrome’ [1]. He had also recently recruited two behavioural psychologists, Ray Hodgson and Howard Rankin, to lead research on understanding the role of learning in alcohol dependence and to develop psychological interventions. As very much the junior member of that team, I was allowed to focus my
© 2015 Society for the Study of Addiction
PhD studies on the ‘nature and measurement’ of the ADS. During those 3 years I sat in on many ward rounds at the Alcoholism Treatment Unit of the Bethlem Royal Hospital, where Griffith presided as the consultant in charge. He would hold forth on these occasions with extraordinary and often spellbinding eloquence on the plight of each patient brought up for discussion (always after the most searching interrogation of every member of the treatment team). There were many research meetings, seminars, conferences and opportunities to discuss ideas one-to-one. Early on I was asked to take detailed notes on discussions between Griffith and Ray Hodgson on how the key elements of the alcohol dependence syndrome could be operationalized into a self-completion questionnaire. Drawing up that questionnaire, the Severity of Alcohol Dependence Questionnaire or SADQ [2,3], administering it to just over 100 attendees of the Maudsley out-patient and Bethlem in-patient units and refining it became a part of my thesis. The team also undertook behavioural studies of drinking behaviour and responses to alcohol by individuals with a moderate or severe degree of alcohol dependence.
Addiction, 110, 8–11
The alcohol dependence syndrome
Whether Griffith was advancing public health policy on alcohol, the idea that brief interventions for alcohol-related problems could be effective, or that controlled drinking was a realistic outcome for some problem drinkers, his ideas routinely provoked fierce debates. I will focus here briefly on two implications of the concept of alcohol dependence: (i) how is it best defined and measured and (ii) is it clinically useful; for example, does it lead to differential prediction of controlled drinking versus abstinence outcomes? These ideas were closely discussed within the small world of alcohol treatment and research in the United Kingdom in the late 1970s and 1980s. The opposing view that the ADS is but the old disease model dressed up in new clothes and is, in any case, not useful for predicting controlled drinking is well represented in a discussion series in the British Journal of Addiction (e.g. [4]) and in an excellent book by Heather & Robertson [5]. I will suggest here that while the original formulation of the ADS had profound impacts on the representation of alcohol-related problems in both the major international disease classification systems (DSM and ICD), some key insights were simply lost in translation and that present diagnostic criteria remain confused as a result. I will also suggest that degree of alcohol dependence should still be considered by all involved in the treatment process when discussing controlled drinking versus abstinence treatment goals. There is now a much more substantial literature from neuroscience and learning perspectives regarding key elements of the dependence syndrome (e.g. [6]). However, confusion about the diagnostic characteristics of ‘dependence’ is likely to hamper an integrative understanding and the development of improved interventions.
THE SEMINAL STUDY ‘Anyone concerned with treating drinking problems must find that his patients often tell him more than is in the textbooks… The varied experience that is recounted can be interpreted as the patient’s astute observation of the alcohol dependence syndrome—a condition certainly far better described by the average alcoholic than in any book’ [1] The paper by Edwards & Gross [1] on the ADS is a thoughtful and structured distillation of clinical observations and accounts from their patients. It occurs to me it that would probably be extremely difficult to publish such a paper today. It would not meet any formal guidelines for review articles or for qualitative analysis. However, the ideas and observations it contains have left a large footprint. Part of the context for the paper was the increasing input of behavioural and social scientists into the field of alcohol treatment and research and a growing dissatisfaction with the dominant concept of alcoholism as a disease. © 2015 Society for the Study of Addiction
9
‘The humane argument that alcoholism should be regarded as a disease seems to have run into difficulties recently because of semantic confusion over what is meant by the over-inclusive term ‘alcoholism’ and the social arbitrariness of the ‘disease’ label’ [1] While suggesting that biological, social and learning processes were key to the development of the ADS, the authors emphasized the provisional nature of their syndrome description and the need for future research from multiple disciplines to elucidate causal pathways. These clinical impressions were structured into seven inter-related elements, each of which is illustrated with hypothetical case examples. The list of elements will be highly familiar to most readers and, in short, comprised: 1 Narrowing of the drinking repertoire in terms of daily pattern and context of drinking. 2 Increased salience of drink-seeking, so that other interests or responsibilities may take second place. 3 Increased tolerance to alcohol. 4 Repeated withdrawal symptoms, usually observed on waking when blood alcohol levels are at their lowest. 5 Relief of withdrawal symptoms after drinking, particularly first thing upon waking. 6 Awareness of a compulsion to drink and impairment of control was emphasized as opposed to the complete ‘loss of control’ described in the disease model [7]. 7 Rapid reinstatement after abstinence, so that only a day or two after returning to drinking all the above signs might return in full force. In practice, in his clinical assessments Griffith would place great store upon whether or not an individual frequently drank to relieve withdrawal symptoms upon waking daily almost daily for a year or more. Such a drinker would be classified by him as ‘severely dependent’ and, in his clinical experience, unlikely to return to controlled drinking. That, at least, was a working hypothesis that led to many a long study and much debate (e.g. [8]).
Alcohol dependence and problems as separate although related dimensions Importantly, and as is crystal clear in the quotation at the beginning of this paper, negative consequences of drinking behaviour were not seen as essential to the assessment of degree of dependence. Social, cultural and other factors mainly external to the individual drinker were seen as determinants of whether drinking was causing ‘problems’. This is in contrast to the descriptions of alcohol dependence contained both in DSM-IV and ICD-10, each of which in effect conflate ‘salience of drink-seeking behaviour’ with the presence of serious alcohol-related harms; for example, DSM-IV defined this as: ‘Continued drinking despite Addiction, 110, 8–11
10
Tim Stockwell
knowledge of having a persistent or recurrent physical or psychological problem that is likely to be caused or exacerbated by drinking’ [9]. DSM-V has taken this conflation to a new level by incorporating ‘alcohol abuse’ and ‘alcohol dependence’ into a single severity dimension [10], based principally on evidence of an observed correlation in surveys between alcohol-related problems and signs of dependence (e.g. [11]). While the idea that alcohol dependence exists on a continuum from mild to severe is captured in these modern diagnostic systems, the importance of understanding both pattern of consumption and problematic consequences as separate dimensions seems to have been entirely lost. Indeed, in an otherwise excellent discussion of the significance of the ADS, Li et al. [11] suggest that the apparent correlation between degree of dependence and harms is evidence that they are all part of the same fundamental continuum. For clinical and scientific purposes, however, Edwards & Gross [1] stress the importance of seeing these as inter-related but also independent dimensions. Heavy drinking can occur in the relative absence of dependence; even severe alcohol dependence can be hidden and effectively tolerated in certain social contexts without present harms being apparent; sometimes even apparently moderate drinking can be associated with serious harm. It is interesting to reflect upon how many other diagnosable mental health problems require the presence of social, work or legal problems to be worthy of a diagnosis. Does schizophrenia need to be associated with problems in the work-place or with the police; depression with relationship problems? Edwards & Gross, I think rightly, pointed out that it is not useful to equate a behavioural syndrome with the problems it may or may not cause and that to do so may be both bad clinical practice and bad science. As just one illustration of how this might make a practical difference, a person with severe alcohol dependence but no apparent problems would—according to the ADS formulation— have less chance of achieving a controlled drinking goal than someone with mild alcohol dependence and many problems. However, both might qualify for a diagnosis of a low-level alcohol use disorder according to DSM criteria. It is the degree of ‘dependence’ in the sense meant by Griffith Edwards (compulsion, impaired control, repeated drinking to relieve withdrawal) that is critical to chances of a return to controlled drinking, rather than simply volume of alcohol consumed or number of problems caused.
Is the degree of alcohol dependence relevant to the choice of treatment goal? Experimental studies on drinking behaviour as well as treatment outcome studies conducted at the Maudsley and ARU encouraged the view that increasing dependence © 2015 Society for the Study of Addiction
was associated with a reducing prospect of a successful return to controlled drinking. Patients assessed by Griffith as severely alcohol-dependent were shown to drink faster in a speed of drinking test and exhibit greater subjective and physiological signs of ‘craving’ after a priming dose of alcohol than mildly or moderately dependent drinkers [12]. This was interpreted as an indicator of ‘impaired control’. While much of the mainly US-based treatment literature on controlled drinking at the time failed to address the dimension of problem severity, the Maudsley study of 100 married men with alcohol problems indicated that among those assessed as returning successfully to ‘social drinking’ 3 or 4 years post-treatment, seven of eight scored below the cut-point for ‘severe alcohol dependence’ on the SADQ [13]. Inspection of successful controlled drinking studies of the time will often find that they focused upon groups of drinkers whose average consumption might only be in the range of 20–30 drinks per week. By contrast, for Griffith an individual would have to report daily drinking for withdrawal relief (shakes, sweats, anxiety) first thing upon waking for a period of at least 1 year to be considered severely dependent. More recently, Heather & Dawe [14] contrasted an impaired control scale and the SADQ in terms of their ability to predict success on a controlled drinking programme. The impaired drinking scale was clearly superior in this population. While this may reflect a relatively small number of severely dependent drinkers enrolled into the study with which to thoroughly test the hypothesis across the full range of dependence severity, the finding also highlights a key weakness in the SADQ as a measure of dependence: it fails to address the key element of impaired control. An attempt to remedy this was described in an adaptation of the SADQ for community samples [15], including the description of an Impaired Control Questionnaire. In hindsight, when compared directly to Griffith’s clinical intuition, the SADQ also fails to assess duration the pattern of dependent drinking as it concentrates only on a recent period of heavy drinking lasting a ‘few months’. In any case, I suggest that the less-than-successful outcome for the SADQ in this particular study is not confused with a lack of predictive validity of the ADS—arguably, the Heather & Dawe [14] study is, in fact, evidence of continuing relevance of the ADS to the issue of controlled drinking versus abstinence given the predictive validity of their measure of impaired control. Their excellent paper also describes many other significant studies and reviews which, at the time, concluded that severity of alcohol dependence was predictive of the possibility of a return to controlled drinking. Furthermore, Heather & Dawe [14] document evidence suggesting that the idea of the ADS had, by that time, become the dominant consideration within treatment agencies for determining when controlled drinking might be the recommended goal for treatment. Addiction, 110, 8–11
The alcohol dependence syndrome
CONCLUDING THOUGHTS Reflecting back, there can be little doubt that the formulation by Edwards & Gross [1] of a clinical syndrome continues to influence thinking, research and practice, even if internationally recommended diagnostic practices diverge increasingly from the original ‘provisional description’. Some of the continuing influences may not be immediately apparent—for example, the influence of those early Maudsley studies on the now widely held view that screening and brief intervention in health-care settings constitute best practice interventions for early-stage problem drinkers (e.g. [16]). I also think Griffith would hope that studies which sensitively capture the individual’s personal experience with alcohol dependence would continue to inform the growing and increasingly sophisticated field of addictions research. As he observed, the alcohol dependence syndrome is ‘a condition certainly far better described by the average alcoholic than in any book’ [1]— perhaps a thought that could be applied usefully to underperforming measurement instruments and diagnostic systems as well as to the value of incorporating qualitative analysis of individual experience within this area of scientific inquiry. Declaration of interests None. References 1. Edwards G., Gross M. Alcohol dependence: provisional description of a clinical syndrome. BMJ 1976; 1: 1058–61. 2. Hodgson R., Stockwell T., Rankin H., Edwards G. Alcohol dependence: the concept, its utility and measurement. Br J Addict 1978; 73: 339–42.
© 2015 Society for the Study of Addiction
11
3. Stockwell T., Murphy D., Hodgson R. The Severity of Alcohol Dependence Questionnaire: its use, reliability and validity. Br J Addict 1983; 78: 145–56. 4. Shaw S. A critique of the concept of the alcohol dependence syndrome. Br J Addict 1979; 74: 339–48. 5. Heather N., Robertson I. Controlled Drinking, revised edn. London: Methuen; 1983. 6. Baker T. E., Stockwell T., Barnes G., Holroyd C. B. Individual differences in substance dependence: at the intersection of brain, behaviour, and cognition. Addict Biol 2010; 16: 458–66. 7. Jellinek E. M. The Disease Model of Alcoholism. Hillhouse Press: New Haven; 1960. 8. Booth P. G. Maintained controlled drinking following severe alcohol dependence: a case study. Br J Addict 1990; 85: 315–22. 9. American Psychiatric Association (APA). Diagnostic and Statistical Manual of Mental Disorders, 4th edn, text revision. Washington, DC: APA; 2000. 10. Babor T. F. Substance, not semantics, is the issue: comments on the proposed Addiction criteria for DSM-V. Addiction 2011; 106: 868–97. 11. Li T., Hewitt B., Grant G. The Alcohol Dependence Syndrome 30 years later: a commentary. Addiction 2007; 102: 1522–30. 12. Hodgson R., Rankin H., Stockwell T. Alcohol dependence and the priming effect. Behav Res Ther 1979; 17: 379–87. 13. Edwards G., Duckitt A., Oppenheimer E., Sheehan M., Taylor C. What happens to alcoholics? Lancet 1983; 2: 269–71. 14. Heather N., Dawe S. Level of impaired control predicts outcome of moderation-oriented treatment for alcohol problems. Addiction 2005; 100: 945–52. 15. Stockwell T., Sitharthan T., McGrath D., Lang E. The measurement of alcohol dependence and impaired control in community samples. Addiction 1994; 89: 167–74. 16. Anderson P., Chisholm D., Fuhr D. Alcohol and global health 2: effectiveness and cost-effectiveness of policies and programmes to reduce the harm caused by alcohol. Lancet 2009; 373: 2234–46.
Addiction, 110, 8–11
