Establishing priorities for vaccine development is complicated by large variations in the morbidity and mortality arising from diseases, the extent of knowledge about relevant pathogens and host responses, the resources and time required for vaccine development, and anticipated vaccine utilization. This report presents a comprehensive model designed at the request of the National Institute of Allergy and Infectious Diseases, National Institutes of Health, to help government decision makers set priorities for accelerated development of vaccines. It can be used to assess new vaccine candidates or to reassess current contenders as additional information becomes available.

The approach suggested uses the same (incomplete) information that could theoretically be used in other methods of decision making. Because the information is incomplete and because the method entails, in some instances, predicting the future, lacunae must be filled by estimates or judgments by experts.

The act of providing a structural framework within which information and judgments are used and combined does not of itself improve the quality of currently available information (although further research to generate new data might be guided by a framework). Nor does it reduce the range of opinion likely to be expressed in predictions, judgments, or estimates (except as issues are more precisely defined).

The committee believes that the system it proposes is the most appropriate for the desired purpose and has implemented it with the best data and estimates that it could develop. It believes that the system would improve the quality of the decision-making process by making it more accessible to evaluation/reconstruction by other decision makers and by facilitating examination of the effect of adopting different assumptions or estimates. However, some cautions and comments are needed to forestall misinterpretation of the power and precision of the method.

In setting out to identify the components on which quantitative information is desirable (but not necessarily available) the system (more than others) exposes areas of ignorance and uncertainty in which expert judgment, by necessity, must be used. Within the proposed approach the use of equations defines the way in which information or estimates are combined (something not always specified in other approaches); this does not imply that the components or the results