Is Controlled Drinking an Acceptable Goal in the Treatment of Alcohol Dependence?

A Survey of French Alcohol Specialists

A. Luquiens; M. Reynaud; H.J. Aubin


Alcohol Alcohol. 2011;46(5):586-591. 

In This Article

Abstract and Introduction


Aim: Assessing acceptance of controlled drinking (CD) among French alcohol specialists.
Methods: On-line survey of 547 French alcohol specialists. We searched factors associated with acceptance of CD, and factors that affected the specialists' selection of treatment goal. Criteria for success used by specialists in clinical practice were compared with criteria expected to be used in clinical trials.
Results: CD was accepted as a treatment goal by 48.6% of alcohol specialists (n = 105, n = 216), and 61.9% practiced CD for their own patients (n = 130, n = 210). Factors in selecting outcome goals were: patient's choice, perceived self-efficacy, relapse history and severity of dependence. Age, profession and basis of specialists' opinion on CD were associated with acceptance of CD.
Conclusion: Half of French alcohol specialists accept CD as a goal. Acceptance was associated with specialists' personal and professional characteristics. The criteria for success specialists use in their clinical practice differ from those they expect to be used in clinical trials.


Controlled drinking (CD) has been a controversial subject since the early 1960s in the USA (Ambrogne, 2002), and subsequently in the rest of the world, and at that time divided the field of alcohol treatment into two entities with different therapeutic goals, abstinence or non-abstinence.

There is still no agreed definition of CD as an outcome. Control can be understood as either a quantitative or qualitative meaning. One could control how much one drinks, or alternatively control drinking consequences, and some authors have proposed integrating these two dimensions in the definition of CD as an outcome (Rosenberg, 2002). Quality of life is increasingly being assessed in studies of alcohol dependence (Foster et al., 2000) in line with the current World Health Organization definition of health in a way that is closer to well-being, rather than a simple absence of disease.

In addition, there is currently no consensus on outcomes that should be used in the treatment of alcohol dependence, which further complicates comparison of abstinent versus non-abstinent goals (Heather and Tebbutt, 1992). In 2001, the National Institute on Alcohol Abuse and Alcoholism determined a sentinel criterion to be the percentage of heavy drinking days (Allen, 2003), whereas the European Medicines Agency recently defined two possible primary endpoints in the assessment of the efficacy of drugs for alcohol dependence: 'a full abstinence goal', considered to be 'relapse prevention after detoxification', and 'an intermediate harm reduction goal', considered to be a moderation of the quantity of alcohol consumed. CD should not be a final goal (EMEA, 2010).

Several early studies that reported resumed normal drinking outcomes from abstinence-oriented treatment led to controversy. The first study, by Davies, reported on seven English alcohol-dependent patients who returned to CD, out of a group of 93 subjects (Davies, 2011). A second study (the Rand report) presented longitudinal 18 months data from a cohort of 600 men treated for alcohol dependence: at 18 months, 22% were drinking moderately, whereas 24% were abstinent (38% were lost to follow up) (Polich et al., 1980). In this study, the criteria for remission were either abstinence or 'normal drinking', defined as'(a) daily consumption less than three oz of ethanol, (b) typical quantities on drinking days less than five oz, (c) no tremor reported and (d) no evidence of frequent episodes of ≥3 of the following: (i) blackout, (ii) missing words, (iii) morning drinking, (iv) missing meals and (v) being drunk'. A study by Sobell and Sobell (1973) probably provoked most debate. It was a randomized clinical trial of 70 subjects, who followed either an abstinence-oriented or a CD-oriented therapeutic programme; the latter focused on loss of control. Opponents of a CD goal objected to various methodological problems in these studies, such as small sample sizes, short follow-up periods or biased samples (e.g. male-only cohorts). At the time, this controversy had little impact on clinical practice, at least in the USA (Peele, 1987). This impact has nevertheless been more important in other countries (Robertson and Heather, 1982).

More recent comparisons of abstinence and CD goals have given conflicting results. The main argument for the goal of CD is that imposition of abstinence as the sole treatment aim could pose an obstacle to treatment access (Sobell and Sobell, 1995). Furthermore, some people spontaneously resolve their alcohol dependence with CD outcomes (Dawson et al., 2005). Some authors have attempted to identify criteria that may select people who could benefit from a mode of treatment aimed at CD: young age, marital status, dependence severity and confidence in the therapeutic goal may be associated with a positive outcome (Heather and Robertson, 1983; Hodgins, 2005). Self-selection of the therapeutic goal by the patient himself has been described as having a positive impact on the outcome. However, other authors take the view that abstinence should be the only goal, highlighting neurobiological arguments such as neural sensitization (Owen and Marlatt, 2001). Cultural and societal factors that influence assessment and perception of therapeutic goals in the treatment of alcohol dependence have also been described (Peele, 1987).

Acceptance of CD as a goal, and its use in clinical practice, varies internationally. The acceptability of CD as a goal has been investigated in several countries since the early 1990s. In Australia, Britain, Norway and—more recently—Switzerland, CD as a goal is widely accepted by those treating alcohol dependence (Rosenberg et al., 1992; Donovan and Heather, 1997; Klingemann and Rosenberg, 2009), whereas most treatment services in the USA do not accept CD as a goal (Rosenberg and Davis, 1994). Acceptability of CD as a goal appears to be mixed in Canada; one survey reported that 40% of treatment services questioned accepted CD as a goal (Rosenberg et al., 1996). As far as we are aware, no data on the acceptability of CD as a goal in France have been published. France has a special relationship with alcohol through its history and societal habits; alcohol production and marketing is one of the most emblematic and traditional mainstays of the French economy (INSERM, 2003). It is therefore of particular interest to assess acceptance of CD as a goal, and determine its use in clinical practice, in France.

Our study had four main objectives: to assess the acceptance of CD as a goal among French specialists in the treatment of alcohol dependence; to investigate patient characteristics that influence the specialists' selection of non-abstinence as a therapeutic goal; to identify independent factors associated with specialists' acceptance of CD as a goal; to assess outcome criteria used by specialists in clinical practice, compared with the criteria they expect to be used in clinical trials.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: