Two examples-the "swine flu affair" in 1976 and the emergence of HIV in the blood supply in the early 1980s-illustrate the difficulties of decision-making in public health. Both cases illustrate trade-offs between product risks and public health benefits, especially with regard to uncertainty in estimates of product risks, public health risks, and the benefits of prevention. The cases also illustrate the tendency of public health policy makers to go all the way or do nothing at all, rather than consider intermediate options that can be adapted as new information emerges. This review suggests three lessons for public health policy makers: (1) be open and honest about scientific uncertainty; (2) communicate with the public, even when the facts are not clear; and (3) consider intermediate, adaptable policy options, such as obtaining more information, thus reducing uncertainty, and building in decision points to reconsider initial policies. Underlying all of these lessons is the need to commission studies to resolve important uncertainties and increase the information base for public communication, and to review regulations and other policy options in the light of the new data that emerge.