Liu 2011.

Methods	RCT
Participants	Country of origin:Tawain N=616; Age 18‐65 years (mean age 41.4 years); Sex: male Clinical setting: medical and surgical wards Inclusion criteria: Hospital inpatients consuming more than 14 drinks per week and 168g alcohol in the previous 30 days Exclusion criteria: psychotic disorders or symptoms, bipolar disorder, major suicide risk, serious medical illness or current psychiatric treatment or speciality treatment for an alcohol use disorder
Interventions	Intervention delivered by: Social workers trained in brief interventions (5 days skilled based training) Brief Intervention: two 30 minute sessions (one week apart) of brief intervention based on the principles of motivational interviewing together with a brochure for use as a reference for cutting back or stopping alcohol use. An optional third booster session was provided for drinkers with alcohol use disorders referred to specialist care for alcohol assessment or treatment (n=308). Control group: Usual treatment though the physician in charge may have advised modifying alcohol if that was his/her normal practice (n=308)
Outcomes	Follow up by telephone at 4 months, 9 months and 1 year a) Change in alcohol consumption using time line follow up, i.e. self reported weekly alcohol consumption b) Number of drinking days per week c) Number of heavy drinking episodes (5 drinks or more on one occasion) d) Self reported alcohol problems e) Number of hospital days d) Number of emergency department visits in previous 3 months e) Self reported receipt of specialist alcohol treatment in 12 month study period
Notes	paper reports 14 drinks = 168g alcohol (thus each drink = 12 g and was used to convert data presented from drinks to g of alcohol)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer generated allocation blocked procedure (block of 4) to ensure balanced group assignments
Allocation concealment (selection bias)	Low risk	Off site data management group
Incomplete outcome data (attrition bias) All outcomes	Low risk	Intention to treat analysis undertaken
Other bias	Low risk	No additional sources of bias identified
Blinding of assessors?	Low risk	Assessors were blinded to group allocation