BACKGROUND Although the efficacy of alcohol pharmacotherapy has been widely investigated, little is known about real-world prescription patterns. Population-based dispensing data can provide an understanding of prescription patterns and characteristics of treatment in nonexperimental settings. METHODS A retrospective cohort study of patients (aged 15-84) treated with acamprosate or naltrexone between July 2009 and June 2013 was conducted using dispensing claims from the Australian Pharmaceutical benefits Scheme Database. Only individuals with prescriptions from September 2009 onwards were included. RESULTS We identified 61,904 individuals (40% female, 32% in 35-44 age bracket,) with a total number of 198,247 dispensings. There were 23,452 naltrexone-treated and 38,452 acamprosate-treated patients. For naltrexone, 42% of initial dispenses were followed by a second dispense with only 25% receiving at least 3 months of treatment. For acamprosate, 28% of dispenses were followed by a third dispense with only 15% receiving at least 3 months of treatment. Patients in older age groups were more likely to be dispensed a repeat script than those in younger age groups (e.g., for the 75-84 vs 15-24 age bracket OR's=2.27 and 2.98 for naltrexone and acamprosate respectively). CONCLUSION Current national guidelines in Australia recommend alcohol pharmacotherapy for a minimum period of 3 months yet only 15-25% are receive this duration of treatment. Naltrexone-treated patients were more likely to return for a second and third dispense than acamprosate-treated patients. Prevalence and prescribing patterns change with age.