Background: Efforts to reduce preterm and low-weight births are among the leading public health objectives in the United States and the world. A necessary component of any public health endeavor is surveillance. The Department of Defense (DoD) Birth and Infant Health Registry (Registry) uses electronic healthcare utilization data to assess reproductive health outcomes among military families. Infant health outcomes are coded using the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM). The objective of this study was to determine the accuracy of using electronically derived ICD-9-CM codes for assessing gestational age and birth weight among Registry infants compared to medical records.
Methods: The authors assessed birth outcome agreement by comparing electronic Registry data for infants born at military treatment facilities (MTFs) from 1999-2002 and 1,858 randomly selected birth medical records from 17 MTFs, with descriptive statistics and measures of agreement, including the kappa statistic.
Results: Of the 1,858 reviewed infant records, 1,669 were successfully matched to the Registry analytic dataset for analyses. Despite small differences in parental demographics, this investigation established "near perfect" agreement for the primary outcomes: kappa of 0.83 for preterm and 0.87 for low birth weight. Subgroup analyses revealed no significant differences in gestational age and birth-weight agreement based on the presence of a birth defect, military parent rank, branch of military service, or specific hospital characteristics.
Conclusions: Electronically derived ICD-9-CM codes provide an accurate assessment of the gestational age and low birth weight reflected in the birth medical records of infants in a large birth and infant health registry. These findings support the integrity of Registry data for investigations assessing preterm and low-weight births among U.S. service member families.