Phone-based Safety Monitoring of Baclofen Prescriptions for Alcohol Use Disorder (BACLOPHONE)

Phone-based Safety Monitoring of the First Year of Baclofen Treatment for Alcohol Use Disorder: the BACLOPHONE Cohort Study

BACLOPHONE is a prospective multicenter cohort study, conducted in two nearby French regions (Hauts-de-France and Normandie). BACLOPHONE consists of the monthly phone-based monitoring of 792 patients during their first year of baclofen prescription for alcohol use disorder. The main objective of the study is to determine the rate of patients who stop baclofen due to an adverse event (AE) in the first year of treatment.The BACLOPHONE study also aims to determine which types of AEs and serious AEs are actually liable to baclofen, and which other types are more likely the consequence of confounding factors, e.g., concomitant alcohol, psychotropic medications or substance uses, and comorbidities.

Inclusions are performed by any previously-labeled baclofen-prescribing physician in the Hauts-de-France or Normandie French regions. INITIAL ASSESSMENT PHONE CALLS In the 15 days following the inclusion, participants will receive two successive phone calls. The first call will be performed by a physician investigator of the study who will: re-explain the principles of the study to patients, check the absence of non-inclusion criteria, and note the medical history, date of baclofen initiation and current dose, undergoing medications and current doses, occurrence of AEs since the initiation of baclofen, birthdate, weight, and height. The physician will also check whether specific safety elements important with regards to the baclofen treatment are noticeable in the patient: history of seizures, suicide, and other concurrent substance use. The second phone call is performed by a clinical research associate (CRA) who will check the absence of non-inclusion criteria, and collects the 30 last-days reported alcohol use according to alcohol timeline follow-back (A-TLFB) method, date of initiation and current dose of baclofen, severity alcohol dependence questionnaire (SADQ), alcohol use disorder identification test (AUDIT), and A-B neuropsychological assessment schedule (ABNAS) which scores the current level of sedation in patients with psychotropic drugs. FOLLOW-UP PHONE CALLS During the subsequent one-year follow-up, two types of phone interviews are performed: 1) Monthly Standardized Interviews (MSI). MSIs consist in collecting the previous-month daily use of alcohol using the A-TLFB, the daily dose of baclofen across the previous month, the current ABNAS sedation score, any change in any associated medication, any change in any substance use, and any AE reported by the patient. No predefined list of AEs is used, as the investigator only asks the patient if he/she has noticed or experienced any unexpected symptoms of health issues since the last call. The daily dose distribution of baclofen, i.e., hours and doses of baclofen intakes, are also systematically investigated in the patient. The average subjective level of craving for alcohol is self-scored by the patient using a 0-10 verbal rating scale (i.e., 0 for no craving at all, to 10 for the most severe possible craving). The daily hours of maximum craving is also noted. 2) Semi-standardized Pharmacovigilance Interviews (SPI) which are unscheduled, and are performed in the case of an AE reported by the patient to the CRA, or in the case of a direct phone call to the pharmacovigilance center by the patient or their baclofen prescriber, with the aim to report an AE. Data collected are: current baclofen dose and recent changes in baclofen dosing, recent and current drinking patterns, associated medications, type of AE, 'serious' feature of the AE according to the definition of the Food and Drugs Administration, onset date and occurrence conditions. Patients can be secondarily re-contacted to assess the final outcome of the AE, and to determine the causality score of baclofen regarding the AE, using both the French causality method and the Naranjo's algorithm if applicable. DELAYED PHONE CALL IN THE CASE OF BACLOFEN CESSATION If baclofen is stopped for any reason during the year following the initiation, a last phone interview will be conducted by the CRA 3 months after baclofen cessation.

Patient with current alcohol use disorder included by any baclofen prescriber located in the French region of Nord - Pas-de-Calais - Picardie.

The link between AE and the discontinuation of baclofen will be defined by the results of the analysis of the pharmacovigilance team. the one-year probability (with it 95% confidence interval) of baclofen discontinuation due to the occurrence of an AE will be estimated using the Kaplan-Meier method. Drop-outs, death, or baclofen discontinuation for other reasons than an AE will be treated as censoring events. A sensitivity analysis treating death or baclofen discontinuation for other reasons than an AE as competing risk will be performed using the approach of Kalbfleisch and Prentice.

The AEs will be identified and reported by the "Monthly Standardized CRA Phone Interviews" and by the "Unscheduled pharmacovigilance Phone Interviews".

To determine whether there is an association between the occurrence of baclofen-related sedation and the concurrent doses of baclofen, alcohol and psychoactive drugs. , bivariate and multivariate time-varying Cox's regression models will be used to assess the impact of doses of baclofen and alcohol on occurrence of first sedative AE by using available exposition measures.

To determine the causality of baclofen in the occurrence of each AE, using the French method for assessing causality, and the Naranjo's algorithm (if applicable). For each AE, the determined causality of baclofen will thus be "doubtful", "possible", "probable", and "definite"

Proportion of patients whose prescription meets the official prescription requirements of the Temporary Recommendation for Use (TRU)

This proportion of patients will be calculated from data collected during "Monthly Standardized CRA Phone Interviews", based on whether the patient has received the official TRU document which should systematically be given together with the prescription

Inclusion Criteria: Aged 18 or more and under 80 years Receiving an off-label baclofen treatment for alcohol use disorder for less than 15 days Mentally and physically able to participate in telephone interviews Reachable by phone Living in the Hauts-de-France or Normandie French regions Beneficiary of a health insurance plan Informed and signed consent before the beginning of the study. Exclusion Criteria: Baclofen prescriber not registered as investigator or not previously labeled for receiving consent, or not located in the regions of Hauts-de-France or Normandie Patient with ICD-10 criteria for other substance dependence (except tobacco) in the past three months. Patient not reachable by phone or unable to correctly understand the French language Pregnancy or breastfeeding.