The WHO Initiative for Vaccine Research and Global Malaria Programme convened a joint scientific forum in June 2008 to discuss scientific, regulatory and public health perspectives on the measurement of efficacy in malaria vaccine field efficacy trials. Participants included clinical trialists, statisticians and epidemiologists from both developed and developing countries, vaccine researchers and developers from academia and industry, and representatives of regulatory agencies. The efficacy of a vaccine against a disease is a summary indication of the extent to which those vaccinated are protected. However, there are several ways of measuring this and for high incidence diseases, such as malaria, in which there is variation in exposure and susceptibility from person to person, the choice of the appropriate measure of efficacy is more complex than is the case for low incidence diseases. There was agreement amongst statisticians at the meeting that basing analyses on "time to event" is the most appropriate method to analyse both incident infection and clinical malaria data from trials. However, policymakers would need to understand that this measure is different from that based on the proportion event-free up to a defined time, which has been used in reporting clinical challenge trials of malaria vaccines. For the assessment of public health impact, data should be reported on all episodes of malaria that a trial subject experiences, not only first episodes, and on duration of efficacy. Further research is required on the analysis of such multiple episode data. It will also be important to examine end-points such as severe malaria and death, though it may be difficult for the latter to be a primary end-point in trials. In order to compare findings in trials, it was suggested that efficacy estimates are reported at different time intervals after vaccination and that data sharing should be enhanced for all malaria vaccine clinical trials. It was appreciated that the epidemiology of malaria is changing in many settings and this may affect the public health benefit of a newly available malaria vaccine, whose likely impact would have to be assessed in the context of multiple other potential control measures. Research into possible interactions between malaria control measures was highlighted as a priority.