Given the increased use of economic evaluation to inform decision making in the health service, it is particularly important that the research methods used are critically assessed and, where possible, improved. The systematic handling of uncertainty in economic evaluation is an important area that remains methodologically underdeveloped. With the increased use of the clinical trial as a vehicle for economic evaluation, there has been recent interest in how the statistical methods routinely employed to handle uncertainty in clinical research might be applied to economic evaluation. This paper reviews the types of uncertainty that exist in economic evaluation and argues that some forms of uncertainty are not amenable to statistical methods. Sensitivity analysis is not a single approach but can take a number of different forms. The different types of sensitivity analysis are reviewed, with an indication of their strengths and weaknesses in relation to the different types of uncertainty in economic evaluation.