The Alabama Perinatal Outcome Project: some methodological issues

As practice networks are built to facilitate the study of unselected patient populations, special advantages and problems of these systems are emerging. The Alabama Perinatal Outcome Project is an example of such a network which has been functioning for over three years. Key methodological questions of sample selection, completeness and reliability of data, and research definitions have been addressed in the initial phase of the project. The evolution of an approach to these methodological issues and some preliminary data are presented for consideration by those designing practice networks.