Cesarean delivery among women with low-risk pregnancies: a comparison of birth certificates and hospital discharge data.

OBJECTIVE: To examine the effect of data source (birth certificate compared with hospital discharge records) and the definition of risk on the prevalence of cesarean deliveries thought to have "no indicated risk"; eg, the fetus is full-term, singleton, and in the vertex position, and the mother has no reported medical risk factors or complications of labor and/or delivery identified on the birth certificate.
METHODS: The study is based on data from 565,767 women who delivered singleton, vertex neonates with gestational ages of 37-41 weeks in Georgia hospitals between 1999 and 2004 and for whom data from birth certificates and hospital discharge records could be linked. The percentages of women with primary cesarean deliveries who did not have risk indicated on the birth certificate and on the hospital discharge record were compared. We also calculated the agreement between data sources overall and for each risk indicator.
RESULTS: Among 40,932 women with primary cesarean deliveries and no risk indicated on the birth certificate, 35,761 (87.4%) had a risk identified in the hospital discharge data. The overall agreement between data sources on the presence of any risk indicator was low (kappa=0.18). Among primary cesarean deliveries, the percentage without indicated risk was 58.3% when using birth certificate data alone and 3.9% when using hospital discharge data in combination with the birth certificate.
CONCLUSION: Using birth certificate information alone overestimated the proportion of women who had no-indicated-risk cesarean deliveries in Georgia. Evidence of many indications for cesarean delivery can be found only in the hospital discharge data. The construct of no indicated risk as determined from birth certificates should be interpreted with caution, and the use of linked data should be considered whenever possible. LEVEL OF EVIDENCE: III.