Appendix A. Clinical Questions and Search Strategies

Articles were screened by one reviewer (L. J. F.) for relevance based on whether the patient population was primarily individuals with AUD and whether specific preferences for AUD treatments were discussed. Three articles were identified but were of limited relevance because one addressed only patients who were undomiciled, one was in a primary care setting, and one was based on a survey of the Swedish general population. None of the articles commented on preferences for specific pharmacotherapies.

Articles were screened by one reviewer (L. J. F.) for relevance based on whether the quantitative measure was used to support a diagnosis of AUD and establish its severity. Articles were excluded if they focused on the use of quantitative measures for screening purposes in community samples or primary care settings. Three articles were identified, one of which was a systematic review of properties of the AUDIT.

Articles were screened by one reviewer (L. J. F.) for relevance based on whether the laboratory biomarker was used as part of an initial evaluation of AUD or for ongoing monitoring of alcohol consumption patterns during treatment. Articles were included if they focused on the impact of quantitative measures on patient outcomes and used a randomized controlled design or a controlled or prospective cohort design with at least 50 individuals. Articles that were aimed primarily at establishing threshold values to optimize sensitivity and specificity or optimizing laboratory assay methodologies were excluded. Three articles were identified, of which one was a systematic review that included articles on use of phosphatidylethanol as a possible marker for chronic alcohol consumption or binge drinking. Two articles addressed the utility of biomarkers in identifying relapse of AUD in individuals who had received a liver transplant.

Articles were screened by one reviewer (L. J. F.) for relevance based on whether treatment using the medications listed above was at least 3 weeks in duration and not just at delivery or on an as-needed basis (e.g., for intermittent nausea). Included articles were randomized controlled trials, clinical trials of at least 50 women, or data from registries (e.g., MotherRisk). On the basis of these criteria, 11 articles were identified for full text review for possible citation in the discussion of evidence for guideline Statement 14.