 |
The 130th Annual Meeting of APHA |
|
The 130th Annual Meeting of APHA |
Charles E. Denk, Ph D, Maternal and Child Health Epidemiology, New Jersey Department of Health and Senior Services, P.O. Box 364, Trenton, NJ 08625-0364, 609-292-5656, Charles.Denk@doh.state.nj.us and Lakota K. Kruse, MD, MPH, Division of Family Health Services, New Jersey Department of Health and Senior Services, P O Box 364, Trenton, NJ 08625.
Objective: Examine the accuracy of fields on the NJ electronic birth certificate (EBC) that represent persistent medical risk factors, such as prior caesarian section, prior low or high birth weight, and hepatitis-B status. Methodology: Probabilistic matching (via AutoMatch) of EBCs 1996-2001 yielded a cohort of 75,516 distinct women with multiple pregnancies resulting in birth. The longitudinal sequence of EBCs is used to calculate the accuracy of “historical” variables and recurrence rates for chronic risk factors and complications. Results: The EBC flag for prior caesarian section was highly reliable: 94% of deliveries with prior caesarian section were correctly flagged, and few false positives occurred. Most other flags performed poorly in longitudinal validation. The flag for prior birth weight >4000 grams only captured 12% of actual prior events. Flags for hemoglobinopathy (a genetic trait of the mother) and positive hepatitis B status were consistent across pregnancies less than half the time (16% and 37%, respectively). Flags for incompetent cervix and insulin-dependent diabetes demonstrated relatively low recurrence across pregnancies (33% and 58%, respectively). Implications: The EBC is compiled at the hospital at time of delivery, capturing information from prenatal care records, hospital charts and interviews after delivery. Quality improvement activities should attempt to identify all points where data capture is vulnerable to break down: continuity of providers and patient records, availability of necessary records to EBC clerks, usability of forms and software. Researchers using birth certificate risk factor fields should be concerned about underreporting that may attenuate predictive power for adverse birth outcomes.
Learning Objectives:
Keywords: Birth Outcomes, Data/Surveillance
Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.