Brief intervention (BI) is intended as an early intervention for non-treatment-seeking, non-alcohol-dependent, hazardous and harmful drinkers. This text provides a brief summary of key BI research findings from the last three decades and discusses a number of knowledge gaps that need to be addressed. Five areas are described: patient intervention efficacy and effectiveness; barriers to BI implementation by health professionals; individual-level factors that impact on BI implementation; organization-level factors that impact on BI implementation; and society-level factors that impact on BI implementation. BI research has focused largely upon the individual patient and health professional levels, with the main focus upon primary health care research, and studies are lacking in other settings. However, research must, to a larger degree, take into account the organizational and wider context in which BI occurs, as well as interaction between factors at different levels, in order to advance the understanding of how wider implementation of BI can be achieved in various settings and how different population groups can be reached. It is also important to expand BI research beyond its current parameters to investigate more ambitious long-term educational programmes and new organizational models. More widespread implementation of BI will require many different interventions (efforts, actions, initiatives, etc.) at different interlinked levels, from implementation interventions targeting individual health professionals' knowledge, skills, attitudes and behaviours concerning alcohol issues, BI and behaviour change counselling to efforts at the organizational and societal levels that influence the conditions for delivering BI as part of routine health care.