Preventability of infant mortality in a rural southern county was examined with a Delphi technique using case summaries of infant deaths during a selected four-year period. The first two rounds were aimed at developing a consensus of panelists' opinions about problems leading to the high infant mortality rate in the study area. From these opinions, an Infant Mortality Preventability Decision Tree and a Problem List was developed. Panelists used these in Rounds III and IV to evaluate the case summaries. There were significant differences in the preventability ratings between physicians and nurses, indicating the importance of assessing individuals' philosophies of preventability when working with an interdisciplinary team of health care providers.