Does information available at admission for delivery improve prediction of vaginal birth after cesarean?

William A. Grobman, Yinglei Lai, Mark B. Landon, Catherine Y. Spong, Kenneth J. Leveno, Dwight J. Rouse, Michael W. Varner, Atef H. Moawad, Hyagriv N. Simhan, Margaret Harper, Ronald J. Wapner, Yoram Sorokin, Menachem Miodovnik, Marshall Carpenter, Mary J. OSullivan, Baha M. Sibai, Oded Langer, John M. Thorp, Susan M. Ramin, Brian M. Mercer

Research output: Contribution to journalArticlepeer-review

98 Scopus citations

Abstract

We sought to construct a predictive model for vaginal birth after cesarean (VBAC) that combines factors that can be ascertained only as the pregnancy progresses with those known at initiation of prenatal care. Using multivariable modeling, we constructed a predictive model for VBAC that included patient factors known at the initial prenatal visit as well as those that only become evident as the pregnancy progresses to the admission for delivery. We analyzed 9616 women. The regression equation for VBAC success included multiple factors that could not be known at the first prenatal visit. The area under the curve for this model was significantly greater (p<0.001) than that of a model that included only factors available at the first prenatal visit. A prediction model for VBAC success, which incorporates factors that can be ascertained only as the pregnancy progresses, adds to the predictive accuracy of a model that uses only factors available at a first prenatal visit.

Original languageEnglish (US)
Pages (from-to)693-701
Number of pages9
JournalAmerican Journal of Perinatology
Volume26
Issue number10
DOIs
StatePublished - 2009

Keywords

  • Prediction
  • Trial of labor
  • Vaginal birth after cesarean

ASJC Scopus subject areas

  • Pediatrics, Perinatology, and Child Health
  • Obstetrics and Gynecology

Fingerprint

Dive into the research topics of 'Does information available at admission for delivery improve prediction of vaginal birth after cesarean?'. Together they form a unique fingerprint.

Cite this