This trial evaluated the clinical impact and cost-effectiveness of strategies promoting screening and brief alcohol intervention (SBI) by nurses in primary care. Randomisation was at the level of the practice and the interventions were: written guidelines (controls, n=76); outreach training (n=68); and training plus telephone-based support (n=68). After 3 months, just 39% of controls implemented the SBI programme compared to 74% of nurses in trained practices and 71% in trained and supported practices. Controls also screened fewer patients and delivered fewer brief interventions to risk drinkers than other colleagues. However, there was a trade-off between the extent and the appropriateness of brief intervention delivery with controls displaying the least errors in overall patient management. Thus cost-effectiveness ratios (cost per patient appropriately treated) were similar between the three strategies. Given the potential for anxiety due to misdirected advice about alcohol-related risk, the balance of evidence favoured the use of written guidelines to promote SBI by nurses in primary care.