Chompookham 2018.

Study characteristics
Methods	Study design: randomized controlled trial Country: Thailand Setting: inpatient alcohol treatment unit Eligibility criteria: included were inpatients with a current diagnosis of alcohol dependence. Excluded were individuals who had major psychiatric disorders, were receiving other medications not in the study protocol, had a history of using substances other than tobacco, had a medical disease (including liver disease), had a history of alcohol withdrawal seizures or delirium, had moderate to severe alcohol withdrawal symptoms, had cognitive impairment, had a history of allergy to gabapentin, or who were pregnant or breastfeeding. Duration of follow‐up: 24 weeks Informed consent: all participants provided written informed consent Ethical approvals: study procedures were approved by the PMNIDAT Research Ethics Committee
Participants	Sample size: 112 (56 gabapentin, 56 placebo) Description of the target population: Patients with alcohol dependence who are being discharged from inpatient alcohol treatment Age: 42.8 years Sex: 8.7% female Race/Ethnicity: not reported Marital status: 54.8% married Harmful alcohol use (baseline): daily consumption = 330.3 g; 77.9% reported everyday drinking Co‐occurring disorders: Not reported
Interventions	Type: pharmacologic Description: Gabapentin following herbal medication and detoxification Duration and frequency: at discharge from alcohol treatment patients received 300 mg/day for 1 week, which then increased to 600 mg and 900 mg/day if they continued drinking alcohol as reported in follow‐up sessions Delivery and provider: oral capsule provided by a pharmacist Comparison group: placebo following herbal medication and detoxification
Outcomes	Primary outcome(s): aalcohol consumption Primary outcome measurement tool(s): timeline followback, GGT Secondary outcome(s): Heavy drinking, Adverse events Secondary outcome measurement tool(s): Timeline followback Time points: 1, 2, 4, 8, 12, 20, and 24 weeks
Notes	Study funding and conflicts of interest: Princess Mother National Institute on Drug Treatment; Center for Alcohol Studies, Thailand; Fogarty International Center (National Institutes of Health) Linked study records: None
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Patients were randomly assigned to the gabapentin treatment group (n = 56) or the placebo group (n = 56) by means of block randomization with a block size of four." Pg. 36
Allocation concealment (selection bias)	Low risk	Quote: "Identical capsules of gabapentin and placebo were prepared by a pharmacist who was also in charge of the block randomization." Pg. 36
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "The patients and investigators, i.e., physicians and research nurses, were blinded to the identity of the treatment assignment." Pg. 36
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not described
Incomplete outcome data (attrition bias) All outcomes	High risk	Quote: "Twenty subjects in the gabapentin group and fourteen subjects in the placebo group completed the study regimen at the end of week 12, as shown in Fig. 1. The reasons for dropout in the gabapentin group were fever (n = 1) and loss to follow‐up (n = 35). The reasons for dropout in the placebo group were allergic reaction to another medication (trazodone, n = 1), pneumonia (n = 1), and loss to follow‐up (n = 40)." Pg. 36
Selective reporting (reporting bias)	Unclear risk	Protocol not available