cohort of response team members, and CDC would need to determine how it can ensure that public health and health care smallpox response teams maintain an adequate number of vaccinated members with the necessary expertise for each team. The commencement of phase II vaccinations may help eliminate the concern over the immediate sustainability of response teams vaccinated during phase I, but the need for new rounds of vaccinations for newly designated members of response teams should be considered once phase I and phase II vaccinations have been completed.

An important task for CDC and its medical and public health partners will be to develop an agreed-upon set of questions that must be answered satisfactorily throughout phase I and before phase II can begin. The questions, borrowed from analyses of the Swine Flu program of the late 1970s (GAO, 1977), are “What evidence on which things, when and why, would make us change the course we now propose, and to what?” (Neustadt and Fineberg, 1978).

Should a deliberation about the fundamental nature of the policy not be possible, at a minimum the committee recommends that CDC develop and communicate the criteria (e.g., types and rates of adverse reactions) that would trigger a reconsideration of the current systems in place to protect vaccinees and their contacts (e.g., the October 2002 Advisory Committee on Immunization Practices [ACIP] recommendations on contraindications, screening, care of the vaccination site, and administrative leave). For example, CDC might wish to consider from how many vaccinees it will require data and at what rate of specific serious adverse reactions (in vaccinees or their contacts) CDC would consider the program riskier than currently expected and the contraindication screening less adequate than needed (or safer than currently expected).

Hospitals and health departments will implement the first phase of the pre-event vaccination program in slightly different ways, depending upon the circumstances and needs of their communities. Much could be learned from this differential administration of the program. Since this program is a very unusual public health intervention, it will be important to gather data on which practices and techniques are most effective in different types of settings. To most effectively evaluate the progress and outcomes of the first phase, the committee recommends that CDC utilize the variation in implementation by hospitals and health departments (e.g., differences in granting administrative leave, types of bandages used, different site care instructions, degree of patient contact, adverse reaction investigation) to obtain safety data and to analyze these data before embarking on subsequent phases of the vaccination program.