studied. For this reason and because of limited resources, this report does not include such comparisons. However, those making decisions about vaccine priorities for developing countries might wish to explore the effects of various trade-offs in the ranking process.
One example of a value judgment that could affect vaccine programs for developing countries is the belief that reducing infant and child mortality has a lower priority than alleviating adult morbidity, because the former simply perpetuates population growth and increases pressure on resources. At its extreme, this view suggests that rapid early death from infectious disease is preferable to slow starvation. Because population growth is a concern in many parts of the developing world, the committee addressed this concept directly.
Reductions in infant mortality may result in a short-term decrease in fertility because breast-feeding a surviving child (if practiced) lengthens the time of postpartum reduced fertility (Bongaarts and Menken, 1983). The death of a child may influence its parents toward further births to “replace” the lost child. Lower infant and child mortality reduces the impact of any such “make-up behavior.” Analyses of parental behavior by Cochrane and Zachariah (1983) indicate that, in the short term, preventing an infant/child death averts an average of about 0.5 births. Thus, the reduction in mortality is not fully offset by a reduction in fertility, and some population growth will occur. The study by Cochrane and Zachariah (1983) did not address longer term trends that may result from changes in general community attitudes.
Reduced child mortality is likely to reduce fertility because fewer births are required to meet family size preferences. This decrease is likely to lag behind the decrease in mortality, however, because changes in child survival may not be recognized immediately in the community, resulting in a delay that precedes widespread change in population behavior (Heer, 1983). The magnitude of population growth before the adjustment to lower fertility rates cannot be predicted with any certainty. However, it appears that mortality reduction is an essential prerequisite for fertility reduction (Gwatkin, 1984) .
That some population growth will occur as mortality declines highlights the need for integration of agricultural and health program planning to avoid food shortages.
The model proposed in this study allows the incorporation of various views on the weight that should be assigned to averting morbidity and mortality at various ages. Opinions were solicited from a range of public health professionals in developing countries; presumably, they are generally aware of the interactions between mortality reduction and fertility discussed above. None of the individual responses could be interpreted as highly favoring adult versus infant/child mortality reduction.