Appendix Table 3.

Inclusion and Exclusion Criteria

Study Characteristic	Inclusion Criteria	Exclusion Criteria
Population	Adults aged ≥18 y with alcohol misuse (e.g., positive alcohol screen; KQ 2), at high risk for AUD, or with diagnosis of AUD (KQ 3).	Pregnant women
Intervention	Intervention must be a computer-based therapy adhering to evidence-based treatment principles and providing individually delivered treatment for alcohol misuse delivered by CD-ROM, Web-based, IVR, mobile phones, or in-home electronic devices (e.g., Health Buddy [Bosch Healthcare]) and may be combined with various levels of supplementary human support.	Interventions targeted at dyads (e.g., couple) or primary prevention Computerized screening only
Comparator	The comparator for KQ 2 and KQ 3 was usual care not involving psychotherapy, WL, or information or attention control. For KQ 4, the comparator was face-to-face treatment.	Any comparator where the effect of the electronic aspect of the intervention could not be isolated
Outcome	Studies must report effects on at least 1 of the following relevant outcomes: alcohol consumption, alcohol-related health problems, alcohol-related legal or social problems, HRQOL, functional status measures, medical utilization, or adverse effects from treatment.	–
Timing	Outcomes reported at ≥6 mo from randomization and initiation of intervention.	Outcomes reported at <6 mo
Setting	Outpatients in any setting (general medical, emergency department, and community) or participants not engaged in clinical care who are enrolled through self-assessments. We included studies where enrollment was inpatient but the majority of the intervention was delivered outpatient.	Inpatient settings for intervention delivery
Study design	RCTs with ≥50 participants. The sample size requirement is designed to exclude small pilot studies that typically are underpowered and have more methodological problems than larger trials. Studies with small samples and no treatment effect are also less likely to be published, increasing the risk for publication bias.	RCTs with 50 participants
Publications	English-language publication Published from 2000 to present* Peer-reviewed, full publication Study conducted in North America, the European Union, or Australia/New Zealand†	Non-English language Published before 2000 Abstract only

AUD = alcohol use disorder; HRQOL = health-related quality of life; IVR = interactive voice response; KQ = key question; RCT = randomized, controlled trial; WL = wait list.

^*Rationale is that the Internet was developed in the 1990s and not used routinely for interventions until 2000. Based on our assessment of studies included in existing systematic reviews, the earliest relevant publication was in 2004.

^†Rationale is to include economically developed countries with sufficient similarities in health care system and culture to be applicable to U.S. medical care.