OBJECTIVE: To predict a woman's risk of postpartum hemorrhage at labor admission using machine learning and statistical models. METHODS: Predictive models were constructed and compared using data from 10 of 12 sites in the U.S. Consortium for Safe Labor Study (2002-2008) that consistently reported estimated blood loss at delivery. The outcome was postpartum hemorrhage, defined as an estimated blood loss at least 1,000 mL. Fifty-five candidate risk factors routinely available on labor admission were considered. We used logistic regression with and without lasso regularization (lasso regression) as the two statistical models, and random forest and extreme gradient boosting as the two machine learning models to predict postpartum hemorrhage. Model performance was measured by C statistics (ie, concordance index), calibration, and decision curves. Models were constructed from the first phase (2002-2006) and externally validated (ie, temporally) in the second phase (2007-2008). Further validation was performed combining both temporal and site-specific validation. RESULTS: Of the 152,279 assessed births, 7,279 (4.8%, 95% CI 4.7-4.9) had postpartum hemorrhage. All models had good-to-excellent discrimination. The extreme gradient boosting model had the best discriminative ability to predict postpartum hemorrhage (C statistic: 0.93; 95% CI 0.92-0.93), followed by random forest (C statistic: 0.92; 95% CI 0.91-0.92). The lasso regression model (C statistic: 0.87; 95% CI 0.86-0.88) and logistic regression (C statistic: 0.87; 95% CI 0.86-0.87) had lower-but-good discriminative ability. The above results held with validation across both time and sites. Decision curve analysis demonstrated that, although all models provided superior net benefit when clinical decision thresholds were between 0% and 80% predicted risk, the extreme gradient boosting model provided the greatest net benefit. CONCLUSION: Postpartum hemorrhage on labor admission can be predicted with excellent discriminative ability using machine learning and statistical models. Further clinical application is needed, which may assist health care providers to be prepared and triage at-risk women.
OBJECTIVE: To predict a woman's risk of postpartum hemorrhage at labor admission using machine learning and statistical models. METHODS: Predictive models were constructed and compared using data from 10 of 12 sites in the U.S. Consortium for Safe Labor Study (2002-2008) that consistently reported estimated blood loss at delivery. The outcome was postpartum hemorrhage, defined as an estimated blood loss at least 1,000 mL. Fifty-five candidate risk factors routinely available on labor admission were considered. We used logistic regression with and without lasso regularization (lasso regression) as the two statistical models, and random forest and extreme gradient boosting as the two machine learning models to predict postpartum hemorrhage. Model performance was measured by C statistics (ie, concordance index), calibration, and decision curves. Models were constructed from the first phase (2002-2006) and externally validated (ie, temporally) in the second phase (2007-2008). Further validation was performed combining both temporal and site-specific validation. RESULTS: Of the 152,279 assessed births, 7,279 (4.8%, 95% CI 4.7-4.9) had postpartum hemorrhage. All models had good-to-excellent discrimination. The extreme gradient boosting model had the best discriminative ability to predict postpartum hemorrhage (C statistic: 0.93; 95% CI 0.92-0.93), followed by random forest (C statistic: 0.92; 95% CI 0.91-0.92). The lasso regression model (C statistic: 0.87; 95% CI 0.86-0.88) and logistic regression (C statistic: 0.87; 95% CI 0.86-0.87) had lower-but-good discriminative ability. The above results held with validation across both time and sites. Decision curve analysis demonstrated that, although all models provided superior net benefit when clinical decision thresholds were between 0% and 80% predicted risk, the extreme gradient boosting model provided the greatest net benefit. CONCLUSION: Postpartum hemorrhage on labor admission can be predicted with excellent discriminative ability using machine learning and statistical models. Further clinical application is needed, which may assist health care providers to be prepared and triage at-risk women.
Authors: Michael S Kramer; Cynthia Berg; Haim Abenhaim; Mourad Dahhou; Jocelyn Rouleau; Azar Mehrabadi; K S Joseph Journal: Am J Obstet Gynecol Date: 2013-07-16 Impact factor: 8.661
Authors: Sarah J Kilpatrick; M Kathryn Menard; Christopher M Zahn; William M Callaghan Journal: Am J Obstet Gynecol Date: 2019-07-25 Impact factor: 8.661
Authors: Robert M Rossi; Erin W Requarth; Carri R Warshak; Kevin Dufendach; Eric S Hall; Emily A DeFranco Journal: Obstet Gynecol Date: 2019-09 Impact factor: 7.661
Authors: Alison E Fohner; John D Greene; Brian L Lawson; Jonathan H Chen; Patricia Kipnis; Gabriel J Escobar; Vincent X Liu Journal: J Am Med Inform Assoc Date: 2019-12-01 Impact factor: 4.497
Authors: J Eric Jelovsek; Audra Jolyn Hill; Kevin M Chagin; Michael W Kattan; Matthew D Barber Journal: Obstet Gynecol Date: 2016-02 Impact factor: 7.661
Authors: Holly B Ende; M James Lozada; David H Chestnut; Sarah S Osmundson; Rachel L Walden; Matthew S Shotwell; Jeanette R Bauchat Journal: Obstet Gynecol Date: 2021-02-01 Impact factor: 7.661
Authors: Amanda B Zheutlin; Luciana Vieira; Ryan A Shewcraft; Shilong Li; Zichen Wang; Emilio Schadt; Susan Gross; Siobhan M Dolan; Joanne Stone; Eric Schadt; Li Li Journal: J Am Med Inform Assoc Date: 2022-01-12 Impact factor: 7.942
Authors: Amanda B Zheutlin; Luciana Vieira; Ryan A Shewcraft; Shilong Li; Zichen Wang; Emilio Schadt; Yu-Han Kao; Susan Gross; Siobhan M Dolan; Joanne Stone; Eric Schadt; Li Li Journal: J Am Med Inform Assoc Date: 2022-01-12 Impact factor: 4.497
Authors: Joshua Cohen; Jennifer Wright-Berryman; Lesley Rohlfs; Douglas Trocinski; LaMonica Daniel; Thomas W Klatt Journal: Front Digit Health Date: 2022-02-02