APA OFFICIAL ACTIONS

The American Psychiatric Association Practice

Guideline for the Pharmacological Treatment of
Patients With Alcohol Use Disorder
Victor I. Reus, M.D., Laura J. Fochtmann, M.D., M.B.I., Oscar Bukstein, M.D., M.P.H., A. Evan Eyler, M.D., M.P.H.,
Donald M. Hilty, M.D., Marcela Horvitz-Lennon, M.D., M.P.H., Jane Mahoney, Ph.D., R.N., PMHCNS-B.C.,
Jagoda Pasic, M.D., Ph.D., Michael Weaver, M.D., Cheryl D. Wills, M.D., Jack McIntyre, M.D. (Consultant),
Jeremy Kidd, M.D. (Consultant), Joel Yager, M.D. (Systematic Review), Seung-Hee Hong (Systematic Review)

At its July 2017 meeting, The APA Board of Trustees ap- Disease Control and Prevention-sponsored study (Bouchery
proved the APA Practice Guideline Writing Group’s “Practice et al., 2011), AUD and its sequelae cost the U.S. $223.5 billion
Guideline for the Pharmacological Treatment of Patients annually and account for significant excess mortality (Kendler
with Alcohol Use Disorder.” The full guideline is available at et al., 2016). Despite its high prevalence and numerous negative
APA’s Practice Guidelines website. consequences, AUD remains undertreated. Effective and
evidence-based interventions are available, and treatment is
associated with reductions in the risk of relapse (Dawson et al,
INTRODUCTION
2006) and AUD-associated mortality (Timko et al., 2006).
The goal of this guideline1 is to improve the quality of care and Nevertheless, fewer than 1 in 10 individuals in the U.S. with a
treatment outcomes for patients with alcohol use disorder 12-month diagnosis of AUD receive any treatment (Substance
(AUD), as defined by DSM-5 (American Psychiatric Association, Abuse and Mental Health Services Administration, 2014;
2013). The guideline focuses specifically on evidence-based Grant et al., 2015). Receipt of evidence-based care is even less
pharmacological treatments for AUD but also includes state- common. For example, one study found that of the 11 million
ments related to assessment and treatment planning that are people in the U.S. with AUD, only 674,000 received psycho-
an integral part of using pharmacotherapy to treat AUD. pharmacological treatment (Mark et al., 2009). Accordingly, this
AUD pharmacotherapy is a topic of increasing interest given practice guideline provides evidence-based statements aimed
the availability of several medications approved by the U.S. at increasing knowledge and the appropriate use of medica-
Food and Drug Administration (FDA) for this disorder and tions for AUD. The overall goal of this guideline is to enhance
the burden of AUD in the population. the treatment of AUD for millions of affected individuals, thereby
Worldwide, the estimated 12-month adult prevalence of reducing the significant psychosocial and public health conse-
AUD is 8.5%, with an estimated lifetime prevalence of 20% quences of this important psychiatric condition.
(Slade et al., 2016). In the United States (U.S.), AUD has es-
timated values for 12-month and lifetime prevalence of 13.9% Overview of the Development Process
and 29.1%, respectively, with approximately half of individuals Since the publication of the Institute of Medicine (IOM; now
with lifetime AUD having a severe disorder (Grant et al., 2015). known as National Academy of Medicine) report, Clinical
AUD places a significant strain on both the personal and public Practice Guidelines We Can Trust (Institute of Medicine,
health of the U.S. population. According to a 2006 Centers for 2011), there has been an increasing focus on using clearly
defined, transparent processes for rating the quality of evi-
1
Practice Guidelines are assessments of current scientific and clinical dence and the strength of the overall body of evidence in
information provided as an educational service and should not be systematic reviews of the scientific literature. This guideline
considered as a statement of the standard of care or inclusive of all was developed using a process intended to be consistent with
proper treatments or methods of care and are not continually updated the recommendations of the IOM 2011 report, the Principles
and may not reflect the most recent evidence. They are not intended
to substitute for the independent professional judgment of the treating for the Development of Specialty Society Clinical Guidelines
provider. The ultimate recommendation regarding a particular assessment, (Council of Medical Specialty Societies, 2012), and the require-
clinical procedure, or treatment plan must be made by the clinician in light ments of the Agency for Healthcare Research and Quality
of the psychiatric evaluation, other clinical data, and the diagnostic (AHRQ) for inclusion of a guideline in the National Guidelines
and treatment options available. The guidelines are available on an Clearinghouse. Parameters used for the guideline’s systematic
“as is” basis, and APA makes no warranty, expressed or implied,
regarding them. APA assumes no responsibility for any injury or review are included with the full text of the guideline. The
damage to persons or property arising out of or related to any use American Psychiatric Association (APA) website features a full
of the guidelines. description of the guideline development process.

86 ajp.psychiatryonline.org Am J Psychiatry 175:1, January 2018

 APA OFFICIAL ACTIONS

Rating the Strength of Research Evidence and harms is more difficult to judge, or either the benefits or the
Recommendations harms may be less clear. With a suggestion, patient values and
Development of guideline statements entails weighing the preferences may be more variable, and this can influence the
potential benefits and harms of the statement and then clinical decision that is ultimately made. Each guideline
identifying the level of confidence in that determination. This statement also has an associated rating for the strength of
concept of balancing benefits and harms to determine guideline supporting research evidence. Three ratings are used: high,
recommendations and strength of recommendations is a hall- moderate, or low (denoted by the letters A, B, and C, re-
mark of GRADE (Grading of Recommendations Assessment, spectively) and reflect the level of confidence that the evi-
Development and Evaluation), which is used by multiple pro- dence for a guideline statement reflects a true effect based on
fessional organizations around the world to develop prac- consistency of findings across studies, directness of the effect
tice guideline recommendations (Guyatt et al., 2013). With the on a specific health outcome, precision of the estimate of
GRADE approach, recommendations are rated by assessing effect, and risk of bias in available studies (AHRQ 2014; Guyatt
the confidence that the benefits of the statement outweigh et al., 2006; Balshem et al., 2011).
the harms and burdens of the statement, determining the
confidence in estimates of effect as reflected by the quality of
GUIDELINE STATEMENTS
evidence, estimating patient values and preferences (including
whether they are similar across the patient population), and iden- Assessment and Determination of Treatment Goals
tifying whether resource expenditures are worth the expected net
1. APA recommends (1C) that the initial psychiatric evalua-
benefit of following the recommendation (Andrews et al., 2013).
tion of a patient with suspected alcohol use disorder in-
In weighing the balance of benefits and harms for each
clude assessment of current and past use of tobacco and
statement in this guideline, our level of confidence is informed
alcohol as well as any misuse of other substances, including
by available evidence, which includes evidence from clinical
prescribed or over-the-counter medications or supplements.
trials as well as expert opinion and patient values and prefer-
2. APA recommends (1C) that the initial psychiatric evalua-
ences. Evidence for the benefit of a particular intervention within
tion of a patient with suspected alcohol use disorder in-
a specific clinical context is identified through systematic review
clude a quantitative behavioral measure to detect the
and is then balanced against the evidence for harms. In this
presence of alcohol misuse and assess its severity.
regard, harms are broadly defined and might include direct and 3. APA suggests (2C) that physiological biomarkers be used to
indirect costs of the intervention (including opportunity costs) identify persistently elevated levels of alcohol consump-
as well as potential for adverse events from the intervention. tion as part of the initial evaluation of patients with alcohol
Many topics covered in this guideline have relied on forms use disorder or in the treatment of individuals who have an
of evidence such as consensus opinions of experienced cli- indication for ongoing monitoring of their alcohol use.
nicians or indirect findings from observational studies rather 4. APA recommends (1C) that patients be assessed for co-
than research from randomized trials. It is well recognized occurring conditions (including substance use disorders,
that there are guideline topics and clinical circumstances for other psychiatric disorders, and other medical disorders)
which high-quality evidence from clinical trials is not pos- that may influence the selection of pharmacotherapy for
sible or is unethical to obtain (Council of Medical Specialty alcohol use disorder.
Societies, 2012). The GRADE working group and guidelines 5. APA suggests (2C) that the initial goals of treatment of alcohol
developed by other professional organizations have noted use disorder (e.g. abstinence from alcohol use, reduction or
that a strong recommendation or “good practice statement” moderation of alcohol use, other elements of harm reduction)
may be appropriate even in the absence of research evidence be agreed on between the patient and clinician and that this
when sensible alternatives do not exist (Andrews et al., 2013; agreement be documented in the medical record.
Brito et al, 2013; Djulbegovic et al., 2009; Hazlehurst et al., 6. APA suggests (2C) that the initial goals of treatment of
2013). For each guideline statement, we have described the alcohol use disorder include discussion of the patient’s
type and strength of the available evidence that was available legal obligations (e.g. abstinence from alcohol use, moni-
as well as the factors, including patient preferences, that were toring of abstinence) and that this discussion be docu-
used in determining the balance of benefits and harms. mented in the medical record.
The authors of the guideline determined each final rating, 7. APA suggests (2C) that the initial goals of treatment of
as described in the section “Rating the Strength of Research alcohol use disorder include discussion of risks to self (e.g.
Evidence and Recommendations,” and each statement is physical health, occupational functioning, legal involve-
endorsed by the APA Board of Trustees. A recommendation ment) and others (e.g. impaired driving) from continued
(denoted by the numeral 1 after the guideline statement) use of alcohol and that this discussion be documented in
indicates confidence that the benefits of the intervention the medical record.
clearly outweigh harms. A suggestion (denoted by the nu- 8. APA recommends (1C) that patients with alcohol use dis-
meral 2 after the guideline statement) indicates greater un- order have a documented comprehensive and person-
certainty. Although the benefits of the statement are still centered treatment plan that includes evidence-based
viewed as outweighing the harms, the balance of benefits and nonpharmacological and pharmacological treatments.

Am J Psychiatry 175:1, January 2018 ajp.psychiatryonline.org 87

APA OFFICIAL ACTIONS

Selection of a Pharmacotherapy Treatment of Alcohol Use Disorder and Co-occurring

Opioid Use Disorder
9. APA recommends (1B) that naltrexone or acamprosate be
offered to patients with moderate to severe alcohol use 19. APA recommends (1C) that in patients with alcohol use
disorder who disorder and co-occurring opioid use disorder, naltrexone
• have a goal of reducing alcohol consumption or achieving be prescribed to individuals who
abstinence • wish to abstain from opioid use and either abstain from or
• prefer pharmacotherapy or have not responded to non- reduce alcohol use and
pharmacological treatments alone • are able to abstain from opioid use for a clinically appro-
• have no contraindications to the use of these medications priate time prior to naltrexone initiation.
10. APA suggests (2C) that disulfiram be offered to patients
with moderate to severe alcohol use disorder who GUIDELINE SCOPE
• have a goal of achieving abstinence
• prefer disulfiram or are intolerant to or have not responded The Agency for Healthcare Research and Quality (AHRQ)
to naltrexone and acamprosate undertook a systematic review of AUD pharmacotherapy in
• are capable of understanding the risks of alcohol con- outpatients (Jonas et al., 2014), which serves as the foun-
sumption while taking disulfiram dation of the systematic review for this practice guideline.
• have no contraindications to the use of this medication The specific medications that are discussed in the guide-
line include: acamprosate, naltrexone, disulfiram, gabapentin,
11. APA suggests (2C) that topiramate or gabapentin be of- and topiramate. The guideline does not apply to the use of
fered to patients with moderate to severe alcohol use these same medications for indications other than AUD. It
disorder who also does not address the management of individuals who are
• have a goal of reducing alcohol consumption or achieving intoxicated with alcohol, who require pharmacotherapy for
abstinence the acute treatment of alcohol withdrawal, or who are ex-
• prefer topiramate or gabapentin or are intolerant to or have periencing other acute medical problems related to alcohol
not responded to naltrexone and acamprosate use. Evidence-based psychotherapeutic treatments for AUD,
• have no contraindications to the use of these medications. including cognitive-behavioral therapy, twelve-step facili-
tation, and motivational enhancement therapy (Anton et al.,
Recommendations Against Use of Specific Medications 2006; Martin and Rehm, 2012, Project MATCH Research
12. APA recommends (1B) that antidepressant medications Group, 1998), also play a major role in the treatment of AUD,
not be used for treatment of alcohol use disorder unless but specific recommendations related to these modalities are
there is evidence of a co-occurring disorder for which an outside the scope of this guideline.
antidepressant is an indicated treatment.
13. APA recommends (1C) that in individuals with alcohol
EVIDENCE OF BENEFITS AND HARMS OF
use disorder, benzodiazepines not be used unless treating
PHARMACOTHERAPY FOR AUD
acute alcohol withdrawal or unless a co-occurring dis-
order exists for which a benzodiazepine is an indicated Naltrexone and acamprosate have the best available research
treatment. evidence as pharmacotherapy for patients with AUD. The
14. APA recommends (1C) that for pregnant or breastfeeding potential benefit of each medication was viewed as far out-
women with alcohol use disorder, pharmacological treat- weighing the harms of treatment or the harms of continued
ments not be used unless treating acute alcohol withdrawal alcohol use, particularly when nonpharmacological approaches
with benzodiazepines or unless a co-occurring disorder have not produced an effect or when patients prefer to use
exists that warrants pharmacological treatment. one of these medications as an initial treatment option. Ac-
15. APA recommends (1C) that acamprosate not be used by cordingly, APA recommends (Statement 9) that that these
patients who have severe renal impairment. medications be offered to patients with moderate to severe
16. APA recommends (1C) that for individuals with mild to alcohol use disorder in specific clinical circumstances. Both
moderate renal impairment, acamprosate not be used as a naltrexone and acamprosate have positive effects overall al-
first-line treatment and, if used, the dose of acamprosate though not all studies or outcomes show a statistically significant
be reduced compared with recommended doses in indi- benefit from these medications. Acamprosate is associated
viduals with normal renal function. with a small benefit on the outcomes of returning to any
17. APA recommends (1C) that naltrexone not be used by drinking and number of drinking days (moderate strength
patients who have acute hepatitis or hepatic failure. of research evidence). Naltrexone is associated with a small
18. APA recommends (1C) that naltrexone not be used benefit on the outcomes of returning to any drinking, returning
as a treatment for alcohol use disorder by individu- to heavy drinking, frequency of drinking days, and frequency of
als who use opioids or who have an anticipated need for heavy drinking days (moderate strength of research evidence).
opioids. In the AHRQ meta-analysis of head-to-head comparisons,

88 ajp.psychiatryonline.org Am J Psychiatry 175:1, January 2018

 APA OFFICIAL ACTIONS

neither acamprosate nor naltrexone showed superiority to the appetite or weight loss were also reported more often with
other medication in terms of return to heavy drinking (mod- topiramate in placebo-controlled trials in AUD.
erate strength of research evidence), return to any drinking Gabapentin was associated with moderate benefit on
(moderate strength of research evidence), or percentage of rates of abstinence from drinking and abstinence from heavy
drinking days (low strength of research evidence). However, drinking (low strength of research evidence). Gabapentin was
in the U.S. COMBINE study (but not the German PREDICT not associated with an increased likelihood of adverse events
study), naltrexone was associated with better outcomes than relative to placebo (low strength of research evidence); however,
acamprosate. in studies that examined side effects of the medication in other
For both acamprosate and naltrexone, the harms of treat- conditions, side effects are typically mild and have included
ment are considered minimal, particularly compared with the dizziness and somnolence. Although gabapentin had a small
harms of continued alcohol use, as long as there is no con- positive effect, the harm of treatment was seen as being minimal,
traindication to the use of the medication (e.g. pregnancy, renal particularly compared with the harms of continued alcohol
impairment for acamprosate, acute hepatitis/hepatic failure use, as long as there was no contraindication to the use of the
for naltrexone). Harms of acamprosate are small in magnitude, medication (e.g. pregnancy).
with slight overall increases in diarrhea and vomiting as The full text of the practice guideline includes a detailed
compared with placebo (moderate strength of research description of research evidence related to effects of medi-
evidence). Harms of naltrexone are also small in magni- cation in individuals with AUD. It also describes aspects of
tude, with slight overall increases in dizziness, nausea, and guideline implementation that are relevant to individual pa-
vomiting relative to placebo (moderate strength of research tients’ circumstances and preferences.
evidence). Alterations in hepatic function are also possible
with naltrexone. For many other potential harms, including
mortality, evidence was not available or was rated by the AUTHOR AND ARTICLE INFORMATION
AHRQ review as insufficient. However, withdrawals from From the APA Practice Guideline Writing Group (Victor I. Reus, M.D., Chair)
the studies due to adverse events did not differ from placebo Address correspondence to Jennifer Medicus (jmedicus@psych.org).
for acamprosate (low strength of research evidence) and APA wishes to acknowledge the contributions of APA staff (Jennifer
were only slightly greater than placebo for naltrexone al- Medicus, Seung-Hee Hong, Samantha Shugarman, Michelle Dirst, Kristin
Kroeger Ptakowski). APA and the Guideline Writing Group especially thank
though statistically significant (moderate strength of re-
Laura J. Fochtmann, M.D., M.B.I., Jeremy Kidd, M.D., Seung-Hee Hong,
search evidence). and Jennifer Medicus for their outstanding work and effort on developing
APA suggests (Statement 10) that disulfiram be offered this guideline. APA also thanks the APA Steering Committee on Practice
to patients with moderate to severe alcohol use disorder Guidelines (Michael Vergare, M.D., Chair), liaisons from the APA Assembly
in specific clinical circumstances. Although the bulk of the for their input and assistance, and APA Councils and others for providing
feedback during the comment period.
research evidence for benefits and harms of disulfiram was
from randomized open-label studies, the potential benefits of Am J Psychiatry 2018; 175:86–90; doi: 10.1176/appi.ajp.2017.1750101
disulfiram were viewed as likely to outweigh the harms for
most patients given the medium to large effect size for the REFERENCES
benefit of disulfiram and particularly compared with the Agency for Healthcare Research and Quality: Methods Guide for
harms of continued alcohol use. With carefully selected Effectiveness and Comparative Effectiveness Reviews. AHRQ
patients in clinical trials, adverse events (e.g. drowsiness, Publication No. 10(14)-EHC063-EF. Rockville, MD, Agency for
Healthcare Research and Quality. Jan 2014. Available at: http://
increased levels of hepatic enzymes, drug-drug reactions) www.effectivehealthcare.ahrq.gov/search-for-guides-reviews-and-
were somewhat greater with disulfiram. However, serious reports/?pageaction5displayproduct&productid5318. Accessed on Feb
adverse events were few and comparable in numbers to 15, 2017
serious adverse events in comparison groups consistent American Psychiatric Association: Diagnostic and Statistical Manual of
with the long history of safe use of disulfiram in clinical Mental Disorders, 5th ed. Arlington, VA, American Psychiatric Pub-
lishing, 2013
practice. Andrews JC, Schünemann HJ, Oxman AD, et al: GRADE guidelines: 15.
Topiramate and gabapentin are also suggested as med- Going from evidence to recommendation-determinants of a recom-
ications to be offered to patients with moderate to severe mendation’s direction and strength. J Clin Epidemiol 66(7):726–735,
alcohol use disorder in specific clinical circumstances (State- 2013
Anton RF, O’Malley SS, Ciraulo DA, et al: Combined pharmacotherapies
ment 11). It was noted that even small effect sizes for these
and behavioral interventions for alcohol dependence: the COMBINE
medications may be clinically meaningful because of the study: a randomized controlled trial. JAMA 2006; 295:2003–2017.
significant morbidity associated with AUD. A moderate Available at doi: https://doi.org/10.1001/jama.295.17.2003
strength of research evidence from multiple randomized Balshem H, Helfand M, Schünemann HJ, et al: GRADE guidelines:
controlled trials showed moderate benefit of topiramate 3. Rating the quality of evidence. J Clin Epidemiol 64(4):401–406,
on drinks per drinking day, percentage of heavy drinking 2011
Bouchery EE, Harwood HJ, Sacks JJ, et al: Economic costs of excessive
days, and percentage of drinking days. Despite the bene- alcohol consumption in the U.S., 2006. Am J Prev Med 2011; 41:
fits, adverse events such as an increased likelihood of cognitive 516–524. Available at doi: https://doi.org/10.1016/j.amepre.2011.06.
dysfunction, dizziness, taste abnormalities, and decreased 045

Am J Psychiatry 175:1, January 2018 ajp.psychiatryonline.org 89

APA OFFICIAL ACTIONS

Brito JP, Domecq JP, Murad MH, et al: The Endocrine Society guide- MD, Agency for Healthcare Research and Quality, 2014. Available from
lines: when the confidence cart goes before the evidence horse. J Clin http://www.ncbi.nlm.nih.gov/books/NBK208590/
Endocrinol Metab 98(8):3246–3252, 2013 Kendler KS, Ohlsson H, Sundquist J, et al: Alcohol use disorder and
Council of Medical Specialty Societies (CMSS): Principles for the De- mortality across the life-span: a longitudinal cohort and co-relative
velopment of Specialty Society Clinical Guidelines. Chicago, IL, analysis. JAMA Psychiatry 2016; 73:575–581. Available at doi: https://
Council of Medical Specialty Societies, 2012 doi.org/10.1001/jamapsychiatry.2016.0360
Dawson DA, Grant BF, Stinson FS, et al: Estimating the effect of help- Mark TL, Kassed CA, Vandivort-Warren R, et al: Alcohol and opioid
seeking on achieving recovery from alcohol dependence. Addiction 2006; dependence medications: prescription trends, overall and by phy-
101:824–834. Available at doi: https://doi.org/10.1111/j.1360-0443. sician specialty. Drug Alcohol Depend 2009; 99:345–349. Available at
2006.01433.x doi: https://doi.org/10.1016/j.drugalcdep.2008.07.018
Djulbegovic B, Trikalinos TA, Roback J, et al: Impact of quality of evidence on Martin GW, Rehm J: The effectiveness of psychosocial modalities in the
the strength of recommendations: an empirical study. BMC Health Serv treatment of alcohol problems in adults: a review of the evidence. Can J
Res 9:120, 2009 https://doi.org/10.1186/1472-6963-9-120 Psychiatry 2012; 57:350–358. Available at doi: https://doi.org/10.1177/
Grant BF, Goldstein RB, Saha TD, et al: Epidemiology of DSM-5 alcohol 070674371205700604
use disorder: results from the national epidemiologic survey on alcohol Project MATCH Research Group: Matching alcoholism treatments to
and related conditions III. JAMA Psychiatry 2015; 72:757–766. client heterogeneity: treatment main effects and matching effects on
Available at doi: https://doi.org/10.1001/jamapsychiatry.2015.0584 drinking during treatment. J Stud Alcohol 1998; 59:631–639. Available at
Guyatt G, Gutterman D, Baumann MH, et al: Grading strength of rec- doi: https://doi.org/10.15288/jsa.1998.59.631
ommendations and quality of evidence in clinical guidelines: report Slade T, Chiu WT, Glantz M, et al: A cross-national examination of
from an American College of Chest Physicians Task Force. Chest 129 differences in classification of lifetime alcohol use disorder between
(1):174–181, 2006 https://doi.org/10.1378/chest.129.1.174 DSM-IV and DSM-5: findings from the world mental health survey.
Guyatt G, Eikelboom JW, Akl EA, et al: A guide to GRADE guidelines for Alcohol Clin Exp Res 2016; 40:1728–1736. Available at doi: https://doi.
the readers of JTH. J Thromb Haemost 11(8):1603–1608, 2013 org/10.1111/acer.13134
Hazlehurst JM, Armstrong MJ, Sherlock M, et al: A comparative quality Substance Abuse and Mental Health Services Administration: Results
assessment of evidence-based clinical guidelines in endocrinology. from the 2013 National Survey on Drug Use and Health: Summary of
Clin Endocrinol (Oxf ) 78(2):183–190, 2013 https://doi.org/10.1111/ National Findings. Rockville, MD, NSDUH Series H-48, HHS Publi-
j.1365-2265.2012.04441.x cation No (SMA) 14-4863, 2014
Institute of Medicine: Clinical Practice Guidelines We Can Trust. Timko C, Debenedetti A, Moos BS, et al: Predictors of 16-year mortality
Washington, DC, National Academies Press, 2011 among individuals initiating help-seeking for an alcoholic use disorder.
Jonas DE, Amick HR, Feltner C, et al: Pharmacotherapy for Adults With Alcohol Clin Exp Res 2006; 30:1711–1720. Available at doi: https://doi.
Alcohol-Use Disorders in Outpatient Settings [Internet]. Rockville, org/10.1111/j.1530-0277.2006.00206.x

90 ajp.psychiatryonline.org Am J Psychiatry 175:1, January 2018