Evaluating outcomes of pregnancy in diabetic women. Epidemiologic considerations and recommended indicators

Many programs have been applied in various settings to reduce adverse outcomes of pregnancy in women with diabetes. Efforts to standardize criteria and methods for evaluating these programs are relatively recent. Without such standardization, evaluation of the impact of many programs and comparisons among programs have not been possible. We review the suitability of available data sources for monitoring adverse outcomes of pregnancy in women with diabetes in light of epidemiological considerations relevant to selection of indicators of program impact. This article is intended to be a resource to help evaluate in a standardized fashion the impact of programs at a regional, state, or local level. We conclude that primary data (information collected by programs themselves) collected in a standardized manner are necessary for evaluation of programs for diabetes in pregnancy. Secondary data sources alone are of limited value for monitoring outcomes because of underreporting of maternal diabetes, especially in the absence of identified complications. Ultimately, the ability to rigorously assess the impact of efforts to improve outcomes of diabetes in pregnancy may depend on the creation of comprehensive statewide systems to identify women of childbearing age who have diabetes.