Site no longer supporterd

This site is no longer supported and will not be updated with new content. You are welcome to browse and download all content already included in the site. Please note you will have to register your email address to access the site. We expect to close the site down by December 2016.

You are here

Baclofen for alcohol dependence in France: Incidence of treated patients and prescription patterns—A cohort study

Editor's comment:
This pharmacoepidemiological retrospective cohort study aims to describe the incidence of patients newly treated with baclofen for alcohol dependence in France from 2007 to 2011. Secondary aim of the study was to describe prescription patterns and prescribers. The study has found a rapid spread of baclofen use in France for alcohol dependence – an increase by a factor of 2.9 between 2007 and 2011. Median duration of baclofen treatment for alcohol dependence was 143.5 days and, in the majority of cases, with low doses (median daily dose 24.4 mg). In the majority of cases, the first prescriber was a non-hospital physician whose specialty was general medicine. In France, baclofen has received in 2014 a special authorization (Temporary Recommendation of Use) as prescription for alcohol dependence. In some other European countries, the drug is used off-label for the treatment of alcohol dependence.

European Neuropsychopharmacology 2014, Volume 24, pages 192–199

Abstract

Recently, baclofen has been widely promoted for treatment of alcohol dependence in France. Our aim was firstly to describe the incidence of patients newly treated with baclofen for alcohol dependence in France from 2007 to 2011, and secondly to describe baclofen prescription patterns and prescribers. A retrospective cohort study of patients newly treated with baclofen was conducted using the “Echantillon Généraliste des Bénéficiaires” database (EGB). Patients with a first dispensation of baclofen between 01/01/2007 and 31/12/2011, followed by a second in the next 120 days, were included. Patients were considered treated with baclofen for neurological conditions if at least one of the following conditions was found to be true: (1) presence of a neurological condition for which baclofen could be prescribed, (2) dispensation of dantrolene, another anti-spastic drug, or (3) hospitalization for a neurological condition for which baclofen could be prescribed. We assumed that all the remaining patients were treated for alcohol dependence. During the 5-year period, 676 patients were incident users. While the annual incidence rate of patients newly treated with baclofen for neurological conditions remained stable, the annual incidence rate of patients newly treated with baclofen for alcohol dependence increased by a factor of 2.9 between 2007 (0.09/1000 person-years) and 2011 (0.26/1000 person-years). In the alcohol dependence group, median duration of baclofen treatment was 143.5 [74.0; 377.0] days; median daily dose was 24.4 [14.8; 39.5] mg. This study demonstrated the rapidly increasing use of baclofen in France for treatment of alcohol dependence.

Keywords: Addictive behavior, Alcoholism, Baclofen, Cohort studies, Incidence, General practitioners.

1. Introduction

Baclofen is an anti-spastic drug used in neurological disorders, which has recently been widely promoted in alcohol dependence in France ( Rolland et al., 2012 ). It acts as a direct gamma-amino-butyric-acid-B (GABA-B) receptor agonist ( Davidoff, 1985 ). In France as in many other countries, baclofen has marketing authorization for paraplegia, multiple sclerosis and serious central or spinal neurological diseases. The usual daily dose starts at 15 mg to finally reach 40–80 mg. Adverse drug reactions (ADRs), such as sedation, drowsiness, dizziness, confusion, or hallucinations, generally occur at doses of more than 60 mg daily.

In recent years, baclofen use for alcohol dependence has been debated. Baclofen has been demonstrated to have an anti-craving action in rats (Colombo et al, 2000, Colombo et al, 2003, and Colombo et al, 2004). Clinical effects of baclofen may be improvement of withdrawal signs, abstinence achievement and decreased alcohol dose in non-abstinent patients (through its anti-craving action). Some human studies, the majority conducted by the same research team, have assessed the value of baclofen for both alcohol withdrawal and abstinence achievement. In view of methodological issues their results need to be interpreted with caution.

Regarding alcohol withdrawal, a randomized trial comparing baclofen with diazepam found that baclofen was slightly slower to act than diazepam, but found no difference in efficacy ( Addolorato et al., 2006 ). The results of this single study were included in a recent Cochrane systematic review, which concluded that there was insufficient evidence to support the use of baclofen for alcohol withdrawal ( Liu and Wang, 2013 ). Since this review was conducted, another randomized trial comparing baclofen with placebo found that baclofen led to lower doses of benzodiazepine in the management of symptomatic alcohol withdrawal syndrome ( Lyon et al., 2011 ).

Regarding abstinence achievement, three randomized clinical trials (versus placebo) conducted by the same team in the last ten years have found that baclofen use was associated with decreased craving and alcohol consumption and with increased abstinence rates (Addolorato et al, 2002, Addolorato et al, 2007, and Addolorato et al, 2011). One recent randomized, double-blind, placebo-controlled trial conducted by another team found that baclofen use did not reduce heavy drinking days and did not increase the days of abstinence, time to first drink, or time to relapse to heavy drinking ( Garbutt et al., 2010 ). A recent systematic review extracted the essential facts from these trials and concluded that no sufficient evidence supports the use of baclofen for alcohol dependence ( Muzyk et al., 2012 ). In particular, these trials were limited by their low number of subjects and selected population. A recent observational study showed positive results ( de Beaurepaire, 2012 ), with half of the patients abstinent or drinking at low-risk level after 3 months of treatment. This result was consistent with the findings of randomized trials: 70% of patients were abstinent after 4–12 weeks of treatment (Addolorato et al, 2002 and Addolorato et al, 2007). Nevertheless, this study did not include a control group and suffered from ethical issues ( Braillon, 2012 ).

Finally, Tiffany and Wray (2012) recently discussed the difficulty of craving assessment in clinical trials.

Regarding safety of baclofen, a strong dose-response effect has been identified at high doses ( Addolorato et al., 2011 ). The usual daily dose ranged from 15 mg to 300 mg (de Beaurepaire, 2012 and Dore et al, 2011), with high doses of baclofen reported in a few papers (Greene, 1992 and Smith et al, 1991). Nevertheless, doses higher than 200 mg daily led to serious adverse effects ( Leung et al., 2006 ). A safety trial has found that baclofen alone had minimal abuse liability in heavy social drinkers, and that baclofen was relatively well tolerated and safe when given in combination with intoxicating doses of alcohol ( Evans and Bisaga, 2009 ). By contrast, cases of serious adverse drug reactions were reported, such as neuropsychological disorders, rashes and hepatitis (Macaigne et al, 2011, Nasti and Brakoulias, 2011, Saddichha et al, 2011, and Soufia et al, 2010).

In France, this scientific debate has rapidly escalated since Ameisen, a French physician, has widely promoted baclofen for alcohol dependence ( Ameisen, 2005 ). He published his self-case report in 2005 after he had treated his own alcohol dependence and anxiety disorder with baclofen. He reported decreasing his alcohol consumption and achieving abstinence at a dose of 270 mg/day. He then published a best-selling book widely diffused in France ( Ameisen, 2008 ). The media took up the story and presented baclofen as a miraculous drug for alcohol dependence ( Rolland et al., 2012 ). The consequence was a great increase in the demand for baclofen prescription in alcoholic patients ( Rolland et al., 2012 ).

Alcohol-dependent patients can be managed, in France, by specialist physicians (often psychiatrists, addictologists or gastroenterologists) or by general practitioners (GPs). As they are easily accessible, GPs are substantially involved in the management of alcoholic patients. In 2008–2009, 52% of a sample of French GPs reported seeing a patient for an alcohol problem in the last seven days ( Beck et al., 2011 ). Three out of 4 GPs manage their patients for alcoholic problems themselves, whether or not they are in contact with a specialized center. Alcohol is a frequent reason for consulting a GP: 16% of patients consulting their GP had an alcohol problem ( Malet et al., 2003 ). Even if GPs lack training (Charrel et al, 2010 and Malet et al, 2003), no evidence of a different prognosis has been shown for patients who underwent detoxification in a hospital or outside ( Malet et al., 2009 ). GPs are in a unique position to recognize, treat, refer and monitor patients with alcohol problems. So, physicians and GPs in particular may be in great demand for the off-label prescribing of baclofen in alcohol dependence.

To our knowledge, no study has yet explored this issue. We conducted a retrospective population-based cohort study, aiming to describe the incidence of patients newly treated with baclofen for alcohol dependence in France from 2007 to 2011. Our secondary aim was to describe prescription patterns and prescribers.

2. Experimental procedures

We performed a pharmacoepidemiological retrospective cohort study on patients newly treated with baclofen, using the “Echantillon Généraliste des Bénéficiaires” database (EGB, or permanent sample of beneficiaries) ( Tuppin et al., 2010 ). The EGB is a permanent representative sample of subjects affiliated to the French health insurance system. It is obtained by 1/97th national random sampling with control for distribution of age, gender and area of residence. The EGB database includes approximately 660,000 beneficiaries from the national health insurance scheme for salaried workers, agricultural workers and farmers, self-employed workers, retirees and patients with universal coverage (attributed to the unemployed and low income insurees) ( Tuppin et al., 2010 ). For some years, French health insurance system databases have been widely used for public health and pharmacoepidemiological purposes (Blin et al, 2012, Bongue et al, 2011, Fournier and Zureik, 2012, Frauger et al, 2011, and Pariente et al, 2010).

We used data from 01/01/2006 to 31/12/2012. As data for agricultural workers, farmers and self-employed persons were added to the EGB in 2011, we decided to include only the population of the main French health insurance system (i.e. salaried workers, retirees and patients with universal coverage, representing 86% of the French population) ( Tuppin et al., 2010 ) in order to work on the same population throughout the follow-up. We included patients who had a first dispensation of baclofen between 01/01/2007 and 31/12/2011 and who had not been exposed during the previous twelve months. Baclofen dispensation was identified with the Anatomical Therapeutical and Chemical (ATC) code M03BX01.

Patients were followed up at least one year until the end of the study period or censoring (death or dropout).

Patients were considered to be treated with baclofen for neurological conditions if at least one of the following conditions was found true:

  • 1) Presence of a neurological condition for which baclofen may be prescribed by physicians:
    • Health insurance coverage for stroke, paraplegia, multiple sclerosis and serious central or spinal neurological conditions listed as a chronic disease (affection de longue durée, ALD). In France, patients with a disease on this list receive 100% reimbursement for health care related to these conditions. The ALD codes selected were 1, 9, 20 and 25.
    • All neurological conditions identified through the International Classification of Diseases (ICD-10) for which baclofen may be prescribed. The following codes were selected with the advice of a neuropharmacologist: G04, G05, G09, G12, G13, G23, G24, G25, G26, G35, G36, G37, G46, G80, G81, G82, G83, G92, G931, G934 and G95.
  • 2) A dispensation of dantrolene, another anti-spastic drug, at least once during the study period. Dantrolene was the only anti-spastic drug available outside hospital (the EGB does not record drugs delivered in hospitals). It was identified by the ATC code M03CA01.
  • 3) Hospitalization for a neurological condition for which baclofen may be prescribed (primary or secondary diagnosis, coded in ICD 10, as above) at least once during the study period.

We assumed that all the remaining patients were treated for alcohol dependence. We compared these two groups of patients according to the presumed indication of baclofen: patients identified as described above composed the neurological condition group and the remaining patients composed the alcohol dependence group.

Patients were defined as incident users if they had a first dispensation followed by a second in the next 120 days. Annual incidence rate was calculated as the ratio of incident patients over person-years recorded in the EGB database for each year. Extrapolation to the whole French population was calculated for the last year of observation (2011) using a multiplier provided by the national health insurance system.

Several variables were extracted to describe baclofen users: age and gender of patient, medical specialty of first prescriber and practice settings, exposure to several drugs, hospitalization for an alcohol-related condition (principal or associated diagnosis, ICD-10 codes F10.0 to F10.9), management of the patient in an alcohol rehabilitation aftercare center, and doses of baclofen. The medical specialty of hospital physicians is not recorded in the French health insurance system databases. We assessed exposure to drugs used in alcohol dependence identified by the ATC code N07BB, exposure to anxiolytics identified by the ATC code N05B, and exposure to drugs used in opioid dependence identified by the ATC code N07BC.

Patients were considered to be treated with baclofen from the first dispensation of the drug. A period of more than 120 days between two dispensations of baclofen was considered as treatment interruption and was chosen according to the prescribing conditions of baclofen (3 months maximum, to which we added 1 month). Daily dose dispensed was calculated as follows: (total quantity of baclofen dispensed in treatment period – the last quantity of baclofen dispensed)/interval between the first and the last dispensation of baclofen. Duration of treatment was calculated as follows: duration between the first and the last dispensation of baclofen+30 days.

Continuous variables were described as means±standard deviations (SD) and medians [first quartile; third quartile]. Categorical variables were described as numbers and percentages. We compared the characteristics of patients treated for neurological conditions and for alcohol dependence using the Studentttest (or the Wilcoxon test if distribution was not normal) for continuous variables and theχ2test (or Fisher test if expected numbers were lower than 5) for categorical variables. Statistical analyses were performed using SAS Guide43®.

We performed an observational study on anonymous data. Under French law, this does not require approval by a regulatory structure or an ethics committee ( Journal Officiel de la République Française, 2012 ). The use of the EGB by research teams is therefore authorized by law and does not require the submission of a request to the national data protection commissions (CNIL and CCTIRS). The study synopsis including source of funding was submitted to INSERM, which approved the project.

3. Results

Between 01/01/2007 and 31/12/2011, 1291 patients had a first baclofen dispensation. Of these patients, 676 had a second dispensation in the next 120 days. Figure 1 presents the annual incidence rates of patients newly treated with baclofen according to the presumed indication for baclofen from 2007 to 2011. Annual incidence rates according to gender are presented in Figure S1bis in the Supplementary material. While the annual incidence rate of patients newly treated with baclofen for neurological conditions remained stable, the annual incidence rate of patients newly treated with baclofen for alcohol dependence increased by a factor of 2.9 between 2007 (0.09/1000 person-years) and 2011 (0.26/1000 person year). In 2011, 229 subjects were defined as incident (having at least one dispensation) and treated with baclofen for alcohol dependence; we extrapolated this finding to the whole French population and estimated the prevalence of patients treated with baclofen for alcohol dependence in France in 2011 to be 24,391 subjects.

gr1

Figure 1 Change in annual incidence rates (with confidence intervals) of patients newly treated with baclofen during the study period, according to the presumed indication of baclofen (alcohol dependence in red dots, neurological conditions in blue squares). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Table 1 presents the general characteristics of newly treated patients. Median age at baclofen initiation for alcohol dependence was 54 [43; 67] years, and 340 patients (50.3%) were men. No significant difference in socio-demographic characteristics was observed according to the presumed indication. During follow-up, 76 patients died (11.2%): 46 (10.8%) in the alcohol group and 30 (12.1%) in the neurological group. Mortality according to gender was 12.9% for men in the alcohol group, 8.7% for women in the alcohol group, 13.0% for men in the neurological group and 11.1% for women in the neurological group. The two groups differed by their medical characteristics. Patients treated for alcohol dependence were more frequently exposed to alcohol dependence drugs and were more frequently hospitalized for alcohol dependence or managed in alcohol rehabilitation aftercare centers than patients treated for neurological conditions. Similarly, they were more exposed to anxiolytics, hypnotics and opiate substitutes.

Table 1 Socio-demographic and medical characteristics of the patients according to the presumed indication for baclofen.

Characteristics Alcohol dependence Neurological condition p
Number of subjects, n 428 248  
Age, years
 Mean±SD 54.8±17.9 53.0±17.9 0.20
 Median [Q1;Q3] 54 [43; 67] 54 [42; 65]  
Gender (female), n (%) 219 (51.2) 117 (47.2) 0.32
Deaths during follow-up, n (%) 46 (10.8) 30 (12.1) 0.59
Drug exposure      
 Alcohol dependence drugs a , n (%) 139 (32.5) 12 (4.8) <0.001
 Anxiolytics a , n (%) 362 (84.6) 173 (69.7) <0.001
 Opiate substitutes a , n (%) 18 (4,2) 1 (0,4) 0.004
Hospitalization related to alcohol dependence, n (%) 97 (22.7) 15 (6.1) <0.001
Admission to alcohol rehab aftercare centers a , n (%) 19 (4.4) 1 (0.4) 0.003

a At least once during the study period.

SD, standard deviation; Q1, first quartile; Q3, third quartile.

Median duration of baclofen treatment was 161.5 [56.0; 474.0] days, and was shorter in patients treated for alcohol dependence: 143.5 [74.0; 377.0] ( Table 2 ). Median daily dose was 24.9 [15.4; 39.5] mg, with no statistical difference between the two groups. Nevertheless, the daily dose value corresponding to the 90th percentile was higher in patients treated for alcohol dependence (72.3 mg vs. 58.4 mg). In the majority of cases, the first prescriber was a non-hospital physician whose specialty was general medicine. Some medical specialties were represented only for patients treated for alcohol dependence: psychiatrists and gastroenterologists.

Table 2 Prescribing patterns according to the presumed indication of baclofen.

Characteristics Alcohol dependence Neurological condition p
Number of subjects, n 428 248  
Duration of treatment with baclofen, days
 Mean±SD 258.8±266.8 599.1±575.4 <0.001
 Median [Q1;Q3] 143.5 [74.0; 377.0] 413.0 [122.5; 887.5]  
Daily dose of baclofen, mg
 Mean±SD 35.0±39.2 30.6±20.7 0.79
 median [Q1;Q3] 24.4 [14.8; 39.5] 26.0 [15.9; 39.1]  
 90th Percentile 72.3 58.4  
 Maximum 257.1 142.9  
First prescriber of baclofen
 Hospital physicians (medical specialty not known a ), n (%) 102 (23.8) 127 (51.2) <0.001
 Non-hospital physicians, n (%) 319 (74.5) 115 (26.8) <0.001
  Medical specialty      
  General practitioners, n (%) 254 (59.3) 88 (35.5) <0.001
  Psychiatrists, n (%) 25 (5.8) 0 (0.0) <0.001
  Neurologists, n (%) 16 (3.7) 24 (9.7) <0.001
  Gastroenterologists, n (%) 4 (0.9) 0 (0.0) 0.30

a Medical specialty is not recorded in the French health insurance system databases for hospital physicians.

SD, standard deviation; Q1, first quartile; Q3, third quartile.

Regarding doses according to the medical specialty of the first prescriber of baclofen for alcohol dependence ( Table 3 ), patients received higher doses when the first prescriber was a psychiatrist (50.0 [26.1; 69.8] mg) than when the first prescriber was a GP (23.2 [13.5; 35.7] mg) or, even more notably, a neurologist (17.8 [12.9; 21.6] mg). Durations of treatment did not significantly differ according to the medical specialty of the first prescriber of baclofen for alcohol dependence.

Table 3 Prescribing patterns of baclofen for alcohol dependence according to the medical specialty of the first prescriber.

Characteristics General practitioners Psychiatrists Neurologists p
Number of subjects, n 254 25 16  
Duration of treatment, days
 Mean±SD 240.0±256.1 271.2±232.9 207.2±162.3 0.48
 Median [Q1;Q3] 132.0 [71.0; 334.0] 183.0 [111.0; 381.0] 146.5 [80.0; 332.5]  
Daily dose of baclofen, mg
 Mean±SD 33.1±42.4 54.2±34 2 19.7±11.7 <0.001
 Median [Q1;Q3] 23.2 [13.5; 35.7] 50.0 [26.1; 69.8] 17.8 [12.9; 21.6]  
 90th Percentile 62.5 102.4 46.5  
 Maximum 250.0 153.8 46.9  

SD, standard deviation; Q1, first quartile; Q3, third quartile.

4. Discussion

The annual incidence rate of patients newly treated with baclofen for alcohol dependence has increased in France since 2007. More than 22,000 patients may have been treated with baclofen for alcohol dependence in 2011.

The incidence has possibly been underestimated for two reasons. Firstly, because prescription of baclofen for alcohol dependence is off-label, physicians should write “nonrefundable” on the prescription and the French health insurance system does not reimburse off-label drugs. Some patients may therefore have purchased baclofen without reimbursement. Secondly, patients can obtain baclofen via the internet as they know that it is not an approved drug in alcohol dependence in France ( de Beaurepaire, 2012 ). Even if moderate underestimation is possible, the change in incidence was in accordance with sales data ( Rolland et al., 2012 ) and pharmacovigilance monitoring data ( Comité Technique de Pharmacovigilance de mars, 2012 ).

Figure 2 shows drug utilization data from national public or non-profit private institutions, such as national medicine agencies, health ministries and the national insurance system. The complete list is available in Table S4 in the Supplementary material ( Lapeyre-Mestre and Palmaro, 2011 ). It seems that baclofen prescription has been increasing not only in France for the last ten years, but also in many other countries such as the United Kingdom, Germany and Belgium. The phenomenon extends to at least the whole of Europe. Our data did not allow differentiation of use of baclofen for alcohol withdrawal or abstinence achievement, but it is probable that baclofen was prescribed for alcohol dependence as the patients in the alcohol group were not suffering from neurological conditions. A small proportion of patients treated for alcohol dependence were also treated with opiate substitutes. These must be poly-addicted patients. We cannot exclude the possibility that some patients might be treated with baclofen for other addictions, in particular cocaine addiction ( Shorter and Kosten, 2011 ).

gr2

Figure 2 Trends in baclofen use in Europe, 2002–2011, in defined daily doses for 1000 inhabitants per day. Source: national public or non-profit private institutions (i.e. national medicine agencies, health ministries, national insurance system). The complete list is given in Lapeyre-Mestre M., and Palmaro A., Regional Addictovigilance Center [Centre d′Evaluation et d′Information sur la Pharmacodépendence–Addictovigilance (CEIP-A)] 2011, “[Comparison of utilization data of psychoactive medications having an abuse potential in the different European countries]” available in Table S4 in the Supplementary material.

In our population of patients treated with baclofen for alcohol dependence, the gender ratio was 1 whereas in the general European population, the gender ratio for alcohol dependence has been estimated at 0.3 in an exhaustive review ( Wittchen et al., 2011 ). However, alcohol dependence may not have been considered appropriately ( Effertz and Mann, 2013 ). Rumpf and colleagues, in addition, emphasized the lower sensitivity of screening questionnaires for alcohol use disorders in women than in men ( Rumpf et al., 2002 ), which may mean that women are less screened in a general population. To explore this issue, we examined the demographic characteristics of patients treated with baclofen for alcohol dependence, restricting this population to patients with a strong presumption of alcohol dependence (patients exposed to alcohol dependence drugs or hospitalized in relation with an alcohol dependence condition or in alcohol rehabilitation aftercare centers,n=162). Sixty-three patients were women (38.9%) and the gender ratio was 0.6, which is more consistent with the data shown above.

Our study yielded interesting results regarding doses and duration of treatment of baclofen in a large population. Patients treated for alcohol dependence were treated for a median of 4 months. This duration is in accordance with data of clinical trials (patients were treated for 3 months) and lower than in the study of de Beaurepaire (2012) . In this study, 43% of the 132 patients included at baseline had stopped baclofen at 6 months, and 32% had been treated at least 2 years. This may be explained either by the heterogeneity of the treated population, the safety of baclofen (occurrence of ADRs could lead to premature discontinuation), inefficacy (and dropouts), or efficacy at 3 months (and discontinuation of the treatment). Whereas high-dose treatment has been promoted by some authors (Ameisen, 2005 and Ameisen, 2008), we identified only 10% of patients receiving baclofen at doses superior to 100 mg a day. High-dose baclofen may be prescribed by few physicians, as the majority may be afraid to increase doses, possibly for fear of ADRs. Another explanation could be that baclofen may be effective at low doses (Addolorato et al, 2002 and Colombo et al, 2000). Finally, we observed a crude mortality rate of 10% in our cohort, which was higher than that reported in another French observational study of patients treated with baclofen (4%; de Beaurepaire, 2012 ) and in alcoholic patients managed in general practice in France (6.6%; Malet et al., 2009 ). This higher mortality rate must be interpreted with caution since the other studies were performed on very small samples of patients. The causes of death could not be determined as they are not recorded in the EGB. We lack information to discuss this issue further.

Three out of four prescribers were non-hospital physicians, and the majority were general practitioners, followed by psychiatrists and neurologists. As expected, GPs were the main first prescribers of baclofen. This could be explained by their central position in the French health care system. Psychiatrists, although less numerous than GPs, nevertheless represented more than 5% of the non-hospital first prescribers. They also may prescribe higher doses, as being probably more experienced in the care of addicted patients.

The increased use of baclofen for alcohol dependence highlights the strong influence of the media in this unusual topic. This has led to an untypical turn in the process of drug evaluation: for the safety of patients, drugs are normally fully evaluated before being used in the general population. Moreover, other drugs have also been investigated for alcohol dependence, but they have never received so much attention. Regarding the legal aspects of baclofen prescription for alcohol dependence, prescribers should bear in mind that this remains an off-label prescription. It thus carries risks for patients as well as a medico-legal risk for the prescriber. To limit these risks for both patients and prescribers, specific care networks have been created to assess the benefit/risk ratio of baclofen for alcohol dependence using a multidisciplinary approach ( Rolland et al., 2010 ).

This observational study is based on the data of the French health insurance system. Use of such databases has been widespread in France for more than 10 years, and use of the EGB for 5 years (Blin et al, 2012, Bongue et al, 2011, Fournier and Zureik, 2012, Frauger et al, 2011, and Pariente et al, 2010). Some studies have shown a satisfactory level of agreement between drug consumption estimated through health insurance system databases and self-reported drug consumption ( Noize et al., 2009 ). However, we cannot exclude the possibility that even if the drug is prescribed and reimbursed, it is not consumed by the patient.

The main strength of our study is the strict definition of patients treated for neurological conditions. This definition was needed to avoid patient misclassification. Precise identification of patients treated for alcohol dependence was not possible, as a specific drug is lacking (e.g. benzodiazepines). The higher proportion of hospital physicians as first prescribers in the neurological group is a strong argument for the accurate classification in the two groups: the neurological conditions concerned are burdensome and patients suffering from them are more often in need of inhospital care than alcoholic patients. Nevertheless, some patients may present both alcohol dependence and neurological conditions, as shown by the rate of alcohol-related hospitalizations and treatments in the patients of the neurological group. We did not take this non-differential information bias into account.

To conclude, this study has demonstrated the rapid spread of baclofen use in France for alcohol dependence. Patients are treated for a brief period and, in the majority of cases, with low doses. Two randomized clinical trials are ongoing in France (ClinicalTrialsgov-a, and ClinicalTrialsgov-b,). One aims to assess the effectiveness of a year of baclofen versus placebo on the proportion of outpatients who achieve low-risk alcohol consumption (according to WHO criteria) or abstention, and the other aims to assess the efficacy of baclofen (180 mg/ day, 20 weeks of treatment) compared with placebo on continuous abstinence rate, after withdrawal, in patients receiving BRENDA therapy sessions. These studies should add new perspectives to our results.

Role of the funding source

No funding (direct or indirect) was received.

Contributors

Author JD co-designed the study, co-wrote the protocol, co-undertook the statistical analysis, co-managed the literature searches and analyses, and wrote the first draft of the manuscript and the final version of the article. Author JPF co-designed the study, co-undertook the statistical analysis, co-wrote the draft, and contributed to writing the final version of the article. Author EJ co-managed the literature searches and analyses and contributed to writing the final version of the article. Author AP provided European data, analyzed them and contributed to writing the final version of the article. Authors JCP and SO contributed to writing the final version of the article. MLM co-designed the study and co-wrote the protocol and contributed to writing the final version of the article. All authors contributed to and have approved the final manuscript.

Conflicts of interest

None declared.

Acknowledgments

The authors would like to thank Dr Christine Brefel for assistance with selection of diagnostic codes for neurological conditions, the health insurance system (Caisse Nationale d′Assurance Maladie des Travailleurs Salariés Cnam-TS) for training two authors (J.D. and J.P.F.) and for enabling researchers to access their database, with particular thanks to Dr Philippe Ricordeau and his team for their advice. Finally, the authors would like to thank Dr Cédric Collin for valuable advice on use of the database.

Appendix A. Supporting materials

 

Download file

Supplementary Material. Figureure S1bis: Change in annual incidence rates (with confidence intervals) of patients newly treated with baclofen during the study period, according to the presumed indication of baclofen and to gender (alcohol dependence in red, neurological conditions in blue, men in squares, women in dots).

References

  • Addolorato et al., 2002 G. Addolorato, F. Caputo, E. Capristo, M. Domenicali, M. Bernardi, L. Janiri, R. Agabio, G. Colombo, G.L. Gessa, G. Gasbarrini. Baclofen efficacy in reducing alcohol craving and intake: a preliminary double-blind randomized controlled study. Alcohol and Alcoholism. 2002;37:504-508 Crossref
  • Addolorato et al., 2006 G. Addolorato, L. Leggio, L. Abenavoli, R. Agabio, F. Caputo, E. Capristo, G. Colombo, G.L. Gessa, G. Gasbarrini. Baclofen in the treatment of alcohol withdrawal syndrome: a comparative study vs diazepam. American Journal of Medicine. 2006;119:276.e13-276.e18 Crossref
  • Addolorato et al., 2011 G. Addolorato, L. Leggio, A. Ferrulli, S. Cardone, G. Bedogni, F. Caputo, G. Gasbarrini, R. Landolfi. Dose-response effect of baclofen in reducing daily alcohol intake in alcohol dependence: secondary analysis of a randomized, double-blind, placebo-controlled trial. Alcohol and Alcoholism. 2011;46:312-317 Crossref
  • Addolorato et al., 2007 G. Addolorato, L. Leggio, A. Ferrulli, S. Cardone, L. Vonghia, A. Mirijello, L. Abenavoli, C. D′Angelo, F. Caputo, A. Zambon, P.S. Haber, G. Gasbarrini. Effectiveness and safety of baclofen for maintenance of alcohol abstinence in alcohol-dependent patients with liver cirrhosis: randomised, double-blind controlled study. Lancet. 2007;370:1915-1922 Crossref
  • Ameisen, 2005 O. Ameisen. Complete and prolonged suppression of symptoms and consequences of alcohol-dependence using high-dose baclofen: a self-case report of a physician. Alcohol and Alcoholism. 2005;40:147-150
  • Ameisen, 2008 O. Ameisen. Le Dernier Verre. (Denoël, Paris, 2008)
  • Beck et al., 2011 F. Beck, R. Guignard, I. Obradovic, A. Gautier, L. Karila. Increasing trends in screening for addictives behaviors among general practitioners in France. Revue d'Epidémiologie et de Santé Publique. 2011;59:285-294 Crossref
  • Blin et al., 2012 P. Blin, R. Lassalle, C. Dureau-Pournin, B. Ambrosino, M.A. Bernard, A. Abouelfath, H. Gin, C. Le Jeunne, A. Pariente, C. Droz, N. Moore. Insulin glargine and risk of cancer: a cohort study in the French National Healthcare Insurance Database. Diabetologia. 2012;55:644-653 Crossref
  • Bongue et al., 2011 B. Bongue, M.L. Laroche, S. Gutton, A. Colvez, R. Guéguen, J.J. Moulin, L. Merle. Potentially inappropriate drug prescription in the elderly in France: a population-based study from the French National Insurance Healthcare system. European Journal of Clinical Pharmacology. 2011;67:1291-1299 Crossref
  • Braillon, 2012 A. Braillon. Baclofen and protecting the patients during research. Alcohol and Alcoholism. 2012;47:631-632
  • Charrel et al., 2010 C.L. Charrel, C.V. Cuervo-Lombard, M. Miron, V. Fruntes, C. Béra-Potelle, F. Limosin. Alcohol dependence in women: difficulty of its assessment in general practice. Journal of Women's Health (Larchmt). 2010;19:343-348 Crossref
  • ClinicalTrials.gov-a, ClinicalTrials.gov-a. Baclofen for the Treatment of Alcohol Drinkers. 〈http://www.clinicaltrials.gov/ct2/show/NCT01604330?term=bacloville&rank=1〉 .
  • ClinicalTrials.gov-b, ClinicalTrials.gov-b. Efficacy and Safety of Baclofen for Maintenance of Abstinence in Alcohol Dependent Patients. 〈http://www.clinicaltrials.gov/ct2/show/NCT01738282?term=alpadir&rank=1〉 .
  • Colombo et al., 2004 G. Colombo, G. Addolorato, R. Agabio, M.A.M. Carai, F. Pibiri, S. Serra, G. Vacca, G.L. Gessa. Role of GABA(B) receptor in alcohol dependence: reducing effect of baclofen on alcohol intake and alcohol motivational properties in rats and amelioration of alcohol withdrawal syndrome and alcohol craving in human alcoholics. Neurotoxicity Research. 2004;6:403-414 Crossref
  • Colombo et al., 2000 G. Colombo, R. Agabio, M.A. Carai, C. Lobina, M. Pani, R. Reali, G. Addolorato, G.L. Gessa. Ability of baclofen in reducing alcohol intake and withdrawal severity: I—Preclinical evidence. Alcoholism: Clinical and Experimental Research. 2000;24:58-66
  • Colombo et al., 2003 G. Colombo, S. Serra, G. Brunetti, G. Vacca, M.A.M. Carai, G.L. Gessa. Suppression by baclofen of alcohol deprivation effect in Sardinian alcohol-preferring (sP) rats. Drug Alcohol Depend. 2003;70:105-108 Crossref
  • Comité Technique de Pharmacovigilance de mars, 2012 Comité Technique de Pharmacovigilance de mars 2012. Effets indésirables du baclofène dans le traitement des addictions. Suivi national de pharmacovigilance: année 2011. 〈http://ansm.sante.fr/var/ansm_site/storage/original/application/264282cafe82f9e1f08323df4722fda1.pdf〉 .
  • Davidoff, 1985 R.A. Davidoff. Antispasticity drugs: mechanisms of action. Annals of Neurology. 1985;17:107-116 Crossref
  • de Beaurepaire, 2012 R. de Beaurepaire. Suppression of alcohol dependence using baclofen: a 2-year observational study of 100 patients. Frontiers in Psychiatry. 2012;3:103
  • Dore et al., 2011 G.M. Dore, K. Lo, L. Juckes, S. Bezyan, N. Latt. Clinical experience with baclofen in the management of alcohol-dependent patients with psychiatric comorbidity: a selected case series. Alcohol and Alcoholism. 2011;46:714-720 Crossref
  • Effertz and Mann, 2013 T. Effertz, K. Mann. The burden and cost of disorders of the brain in Europe with the inclusion of harmful alcohol use and nicotine addiction. European Neuropsychopharmacology. 2013;23:742-748 Crossref
  • Evans and Bisaga, 2009 S.M. Evans, A. Bisaga. Acute interaction of baclofen in combination with alcohol in heavy social drinkers. Alcoholism: Clinical and Experimental Research. 2009;33:19-30 Crossref
  • Fournier and Zureik, 2012 A. Fournier, M. Zureik. Estimate of deaths due to valvular insufficiency attributable to the use of benfluorex in France. Pharmacoepidemiology and Drug Safety. 2012;21:343-351 Crossref
  • Frauger et al., 2011 E. Frauger, V. Pauly, V. Pradel, F. Rouby, J. Arditti, X. Thirion, M. Lapeyre-Mestre, J. Micallef. Evidence of clonazepam abuse liability: results of the tools developed by the French Centers for Evaluation and Information on Pharmacodependence (CEIP) network. Fundamental and Clinical Pharmacology. 2011;25:633-641 Crossref
  • Garbutt et al., 2010 J.C. Garbutt, A.B. Kampov-Polevoy, R. Gallop, L. Kalka-Juhl, B.A. Flannery. Efficacy and safety of baclofen for alcohol dependence: a randomized, double-blind, placebo-controlled trial. Alcoholism: Clinical and Experimental Research. 2010;34:1849-1857 Crossref
  • Greene, 1992 P. Greene. Baclofen in the treatment of dystonia. Clinical Neuropharmacology. 1992;15:276-288 Crossref
  • Journal Officiel de la République Française, 2012 Journal Officiel de la République Française. Arrêté du 11 juillet 2012 relatif à la mise en oeuvre du système national d'information interrégimes de l'assurance maladie. < http://www.legifrance.gouv.fr/affichTexte.do?cidTexte=JORFTEXT000026221180&dateTexte=&categorieLien=id>
  • Lapeyre-Mestre and Palmaro, 2011 Lapeyre-Mestre, M., Palmaro, A, 2011. Comparison of Utilization Data of Psychoactive Medications Having an Abuse Potential in the Different European Countries. Centre d′Evaluation et d′Information sur la Pharmacodépendence–Addictovigilance (CEIP-A). Available on specific request to the authors.
  • Leung et al., 2006 N.Y. Leung, I.M. Whyte, G.K. Isbister. Baclofen overdose: defining the spectrum of toxicity. Emergency Medicine Australasia. 2006;18:77-82 Crossref
  • Liu and Wang, 2013 J. Liu, L. Wang. Baclofen for alcohol withdrawal. Cochrane Database of Systematic Reviews. 2013; (CD008502)
  • Lyon et al., 2011 J.E. Lyon, R.A. Khan, C.E. Gessert, P.M. Larson, C.M. Renier. Treating alcohol withdrawal with oral baclofen: a randomized, double-blind, placebo-controlled trial. Journal of Hospital Medicine. 2011;6:469-474 Crossref
  • Macaigne et al., 2011 G. Macaigne, N. Champagnon, F. Harnois, S. Cheiab, C. Chayette. Baclofen-induced acute hepatitis in alcohol-dependent patient. Clinics and Research in Hepatology and Gastroenterology. 2011;35:420-421 Crossref
  • Malet et al., 2003 L. Malet, P.M. Llorca, D. Boussiron, R. Schwan, F. Facy, M. Reynaud. General practitioners and alcohol use disorders: quantity without quality. Alcoholism: Clinical and Experimental Research. 2003;27:61-66
  • Malet et al., 2009 L. Malet, M. Reynaud, P.M. Llorca, N. Chakroun, O. Blanc, B. Falissard. Outcomes from primary care management of alcohol dependence in France. Journal of Substance Abuse Treatment. 2009;36:457-462 Crossref
  • Muzyk et al., 2012 A.J. Muzyk, S.K. Rivelli, J.P. Gagliardi. Defining the role of baclofen for the treatment of alcohol dependence: a systematic review of the evidence. CNS Drugs. 2012;26:69-78 Crossref
  • Nasti and Brakoulias, 2011 J.J. Nasti, V. Brakoulias. Chronic baclofen abuse and withdrawal delirium. Australian and New Zealand Journal of Psychiatry. 2011;45:86-87 Crossref
  • Noize et al., 2009 P. Noize, F. Bazin, C. Dufouil, N. Lechevallier-Michel, M.L. Ancelin, J.F. Dartigues, C. Tzourio, N. Moore, A. Fourrier-Réglat. Comparison of health insurance claims and patient interviews in assessing drug use: data from the Three-City (3C) Study. Pharmacoepidemiology and Drug Safety. 2009;18:310-319 Crossref
  • Pariente et al., 2010 A. Pariente, M. Pinet, Y. Moride, Y. Merlière, N. Moore, A. Fourrier-Réglat. Factors associated with persistence of cholinesterase inhibitor treatments in the elderly. Pharmacoepidemiology and Drug Safety. 2010;19:680-686 Crossref
  • Rolland et al., 2012 B. Rolland, R. Bordet, O. Cottencin. Alcohol-dependence: the current French craze for baclofen. Addiction. 2012;107:848-849 Crossref
  • Rolland et al., 2010 B. Rolland, S. Deheul, T. Danel, R. Bordet, O. Cottencin. [A system of prescriptions without drug approval: example of baclofen.]. Therapie. 2010;65:511-518 Crossref
  • Rumpf et al., 2002 H.J. Rumpf, U. Hapke, C. Meyer, U. John. Screening for alcohol use disorders and at-risk drinking in the general population: psychometric performance of three questionnaires. Alcohol and Alcoholism. 2002;37:261-268 Crossref
  • Saddichha et al., 2011 S. Saddichha, N. Jayaram, N. Manjunatha, V. Benegal. Baclofen-induced morbiliform rashes: a case series. Journal of Clinical Pharmacology. 2011;51:1733-1734 Crossref
  • Shorter and Kosten, 2011 D. Shorter, T.R. Kosten. Novel pharmacotherapeutic treatments for cocaine addiction. BMC Medicine. 2011;9:119 Crossref
  • Smith et al., 1991 C.R. Smith, N.G. LaRocca, B.S. Giesser, L.C. Scheinberg. High-dose oral baclofen: experience in patients with multiple sclerosis. Neurology. 1991;41:1829-1831
  • Soufia et al., 2010 M. Soufia, M. Plaze, B. Gueguen, G. Demigneux, J.P. Olié, R. Gaillard. Behavioral disinhibition with baclofen. Journal of Clinical Psychopharmacology. 2010;30:759-760 Crossref
  • Tiffany and Wray, 2012 S.T. Tiffany, J.M. Wray. The clinical significance of drug craving. Annals of the New York Academy of Sciences. 2012;1248:1-17 Crossref
  • Tuppin et al., 2010 P. Tuppin, L. de Roquefeuil, A. Weill, P. Ricordeau, Y. Merlière. French national health insurance information system and the permanent beneficiaries sample. Revue d'Epidémiologie et de Santé Publique. 2010;58:286-290 Crossref
  • Wittchen et al., 2011 H.U. Wittchen, F. Jacobi, J. Rehm, A. Gustavsson, M. Svensson, B. Jönsson, J. Olesen, C. Allgulander, J. Alonso, C. Faravelli, L. Fratiglioni, P. Jennum, R. Lieb, A. Maercker, J. van Os, M. Preisig, L. Salvador-Carulla, R. Simon, H.C. Steinhausen. The size and burden of mental disorders and other disorders of the brain in Europe 2010. European Neuropsychopharmacology. 2011;21:655-679 Crossref

Footnotes

a UMR1027 Inserm—Université Paul Sabatier Toulouse III, Toulouse, France

b Département Universitaire de Médecine Générale, Faculté de Médecine, Université Paul Sabatier Toulouse III, Toulouse, France

c CEIP-Addictovigilance, CHU Toulouse, Toulouse, France

lowast Corresponding author at: Inserm UMR1027, Université de Toulouse III, Faculté de Médecine, 37 allées Jules Guesde, 31000 Toulouse, France. Tel.: +33 561145606; fax: +33 561145928.