Should we be prescribing medication to help people with alcohol use disorders?


Alcohol Use Disorders (AUDs) are common, chronic conditions which cause substantial harm to the individual and society. They are associated with substantial morbidity (Schuckit, 2009) and responsible for approximately 4% of all deaths annually. The approximate cost of AUDs to the National Health Service is upwards of £3 billion per year.

Less than one-third of patients with AUDs will receive any treatment and an even smaller number (<10%) will be prescribed medications to assist in reducing their alcohol use. There are a number of FDA approved medications for AUDs such as acamprosate, disulfiram and naltrexone. However, various barriers exist for health-care professionals prescribing these medications including a lack of familiarity and a lack of confidence in their effectiveness.

A recent systematic review and meta-analysis published in the Journal of the American Medical Association, set out to examine the benefits and harms of pharmacotherapy for adults with AUDs (Jones et al, 2014).


The authors searched databases (including the Cochrane Library, Pubmed etc) for relevant studies from January 1st 1970 to October 11, 2013. They also searched for unpublished studies using various trial registration platforms and requested data from medication manufacturers.

In order to be included in the analyses studies had to include adults with AUDs who were treated with an FDA approved or off-label medication for at least 12 weeks in an outpatient setting. The studies were required to assess either:

  1. Alcohol consumption,
  2. Health outcomes (i.e. accidents, mortality etc), or
  3. Adverse effects of the medication.


One hundred and twenty three studies were included in the meta-analysis, with a total of 22,803 participants. All the studies except one were randomised controlled trials. The majority of the studies assessed acamprosate or naltrexone alone or in combination with repeated behavioural interventions.

Both acamprosate and naltrexone were associated with complete abstinence from alcohol

Both acamprosate and naltrexone were associated with complete abstinence from alcohol

Consumption measures

  • Acamprosate was associated with:
    • An increase in total abstinence from drinking (Number needed to treat (NNT) = 12, 95% Confidence Interval (CI) 8 to 26; Risk difference (RD) -0.09; 95%CI, -0.14 to -0.04)
    • But not in abstinence from heavy drinking
  • Oral naltrexone was also associated with:
    • An improvement in total abstinence from drinking (NNT = 20, 95% CI, 11 to 500; RD, -0.05; 95%CI -0.10 to -0.002)
    • Abstinence from heavy drinking (NNT = 12, 95% CI, 8 to 26; RD -0.09; 95%CI, -0.13 to -0.04)
  • Injectable naltrexone was only associated with a reduction in heavy drinking days (Weighted Mean Difference = -4.6%, 95% CI, -8.5% to -0.56%)
  • A comparison between acamprosate and naltrexone found no statistically significant differences between the two medications on alcohol consumption measures
  • Disulfiram demonstrated no significant improvements in alcohol consumption measures
  • There was some evidence that off-label drugs such as nalmefene and topiramate were associated with reductions in alcohol consumption measures (fewer heavy drinking days)

Health outcomes

There was insufficient evidence to examine whether treatments lead to an improvement in health outcomes.

Adverse effects

Again, there was insufficient evidence to examine potential adverse effects. However, the authors noted that for most of the specific adverse effects in head-to-head studies, estimates favoured placebo.

Do these results give clinicians and patients the evidence they need to

Do these results give clinicians and patients the evidence they need to make prescribing decisions?


The use of pharmacotherapy as a treatment for AUDs is underutilized (Harris et al, 2010) but may lead to clinically beneficial results. Both acamprosate and naltrexone were associated with complete abstinence from alcohol, and naltrexone was also associated with a reduction in heavy drinking days. There was also some limited evidence for off-label prescriptions of topiramate and nalmefene.

A recent commentary (Bradley et al, 2014) suggests that ‘no single treatment is superior to all others’ and many health-care professionals do not fully explore the treatment options for AUDs. These findings demonstrate that patients should be offered individual treatment options, including medications shown to be effective through evidence.

There are some limitations to the current research. Only trials with at least 12 weeks of treatment were eligible. However, such short treatment periods may yield misleading conclusions about benefits in the long term. Furthermore, the authors suggest that most studies included were at a moderate risk of bias, i.e. selective or incomplete reporting of important variables within studies.

In conclusion, the authors suggest their findings have the potential to inform clinicians and health-care providers who are reluctant to prescribe medication for AUDs. Future research should focus on establishing the efficacy of pharmacotherapy for AUDs in patients for which controlled-drinking, rather than complete abstinence, is a realistic goal.


Jones, D.E., Amick, H.R., Feltner, C., et al (2014). Pharmacotherapy for adults with alcohol use disorders in outpatient settings: a systematic review and meta-analysis. The Journal of the American Medical Association, 311(18), 1889-900.

Bradley, K.A., Kivlahan, D.R. (2014). Bringing patient-centered care to patients with alcohol use disorders. The Journal of the American Medical Association, 311(18), 1861-2

Harris, A.H., Kivlahan, D.R., Bowe, T., Humphreys, K.N. (2010). Pharmacotherapy of alcohol use disorders in the Veterans Health Administration. Psychiatric Services, 61(4), 392-8.

Schuckit, M. Alcohol-use disorders. Lancet, 2009 Feb 7, 373 (9662), 492-501.

Share on Facebook Tweet this on Twitter Share on LinkedIn Share on Google+