Unreliable medical information on birth certificates

OBJECTIVE: To compare selected maternal, obstetric and newborn medical information on live birth certificates to medical record abstracted data. STUDY DESIGN: The Florida Department of Health provided a computer tape containing the Hillsborough County Florida live birth certificate data for July 1, 1992, through December 31, 1996. This information was compared to selected maternal, obstetric and new-born medical data abstracted from the hospital records of women who delivered in that county. RESULTS: Three hospitals provided care for 75% (44,654) of all deliveries in the county in the study period. There were no differences in age, race, gravidity or parity of patients between the data sets. There was no concurrence between the number of patients on the state-compiled birth certificate data and the medical record data in 8 maternal medical complications, 5 complications of labor and delivery, 3 obstetric procedures, 3 abnormal neonatal conditions or 7 congenital anomalies from any hospitals. There was no difference between the three hospitals in the incidence of mismatches. CONCLUSION: The medical information on the birth certificates from this county was discrepant from that abstracted directly from the patients' medical records. Caution is required in drawing conclusions from these data sets.