Scott 1990.

Methods	Parallel RCT: brief intervention versus control. ITT: yes.
Participants	Setting: England, general practice. Participants: 17 to 69 years, consuming > 350 g alcohol for men and > 168 g for women in previous week; screened by Health Survey Questionnaire including quantity frequency measure of alcohol; excluded if consumption > 1050 g/week for men or 560 g/week for women, or previous advice to cut down during the previous year. Number randomised = 226; 68% male; mean age = 44.7 years. At baseline: mean alcohol consumption for previous week (from interview) = 526 g for men, 293 g for women; mean quantity frequency drinking for previous week (from HSQ) = 439 g for men, 247 g for women; binge drinkers = 43% for men (defined as consumption of 140 g on at least two occasions in previous three months), 18% for women (defined as consumption of > 14 units on at least two occasions in previous three months).
Interventions	Intervention group (N = 113) asked at the end of the assessment interview to make an appointment with their own general practitioner. Advice lasted ten minutes and consisted of feedback of the assessment interview and results of blood tests, information on the risks of excessive drinking, information on the benefits of drinking less, information on how the patient's weekly alcohol consumption compared with that of the general population using a histogram, and advice to reduce alcohol consumption to below four units per week. Advice was supplemented with a self‐help booklet (the 'Cut Down on Drinking' booklet) designed for the study. Control group (N = 113) received no advice from their general practitioner, except at their own request.
Outcomes	Change in weekly alcohol consumption. Assessed at one year.
Funding source	The pilot study was supported by the Alcohol and Education Research Council and the main study by the Mental Health Foundation.
Declaration of interests	Not reported.
Notes	'Cut Down on Drinking' trial. Loss to follow‐up: Intervention group = 33/113 (29%). Control group = 43/113 (38%).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random numbers table used to randomise patients (p. 891).
Allocation concealment (selection bias)	Low risk	Allocated at assessment interview before making an appointment with doctor (p. 891).
Blinding of treatment providers	Low risk	Researcher carried out baseline assessment and only intervention patients were asked to make an appointment with GP so low risk of contamination between groups.
Blinding of participants	Low risk	No mention was made that the assessment interview was about alcohol consumption, and alcohol questions were embedded amongst other health‐related questions.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Follow‐up interview done by a researcher blinded to patient allocation.
Incomplete outcome data (attrition bias) All outcomes	High risk	Loss to follow‐up > 30%.
Selective reporting (reporting bias)	Low risk	Outcomes specified in methods are reported.