Advanced analytics could help reduce rate of C-section births
Raleigh, NC. At North Carolina State University, research engineers have taken a major step toward developing a computer system that would provide doctors in maternity wards with a bedside tool to evaluate risk in real time and determine the best possible option for delivery.
The precipitous rise in the rate of cesarean section (C-section) births in the U.S. in recent years has health care professionals concerned. In 2014, the American College of Obstetricians and Gynecologists (ACOG) along with the Society for Maternal-Fetal Medicine issued new guidelines to prevent unnecessary C-sections due to the risks to both mother and newborn.
Dr. Julie Ivy, associate professor at NC State’s Edward P. Fitts Department of Industrial and Systems Engineering and a Fitts Faculty Fellow in Health Systems Engineering, and Ph.D. student Karen Hicklin are using advanced analytics to develop delivery room protocols so pregnant women and their doctors can make the best possible choices and safely lower the rate of C-sections.
“Sometimes the data is very clear that a C-section is the best option, but there’s a large grey area and during the progression of labor, risk is always evolving,” said Ivy. “Doctors need tools to translate those risks in real time so we can know with much greater certainty when intervention is necessary.”
For their research, Ivy and Hicklin consider first-time mothers who did not have preexisting indications for a C-section and who are prime candidates for a vaginal delivery. The study examines data that includes tipatient preferences, patient safety, the chance of vaginal delivery, dilation states, labor progression, and the risk thresholds that influence a doctor’s decision to end a trial of labor and perform a C-section. Stochastic modeling, factoring in a wide range of variables from 8,500 patient simulations, helped to identify opportunities to improve the real-time decision-making process.
“The modeling can be used for training purposes to evaluate the trade-off between the risk of prolonging labor and the short- and long-term effects of a C-section,” said Hicklin. “Ultimately, this information can be used to create a computer-aided guide for monitoring labor progression and the associated risks of complications.”
Ivy has a personal motivation to help improve delivery room decision making. Her first pregnancy was complicated by the effects preeclampsia, a condition marked by high blood pressure and fluid retention that can ultimately lead to seizures in the mother and a high risk of death for the baby. Her son, now a healthy, thriving 14 year-old, was delivered at 33 weeks via emergency C-section under general anesthesia, but only after an emotionally exhausting series of conflicting opinions from physicians that changed with every shift, compounded by terse communications from the doctors on duty.
The experience has led Ivy on a 10-year academic quest. Healthcare providers, she asserts, need tools to quickly and thoroughly assess the risks of any given situation in labor, and to accurately convey the options available, including the risks of those options and any drug treatments that may be administered.
“Our primary objective with this project was to give structure to a very complex process where information is critical and ill-informed decisions can be life-changing, and even life-threatening,” said Ivy.
Currently, almost 33% of all births in the U.S. are C-sections, up from 4.5% in 1970. The World Health Organization suggests a rate of 10 to 15% is associated with favorable health outcomes. In addition to the increased risk of respiratory problems for the newborn and the risks of major surgery for the mother, C-sections can cause complications in future pregnancies and births. They also increase the hospital stay, recovery time, and associated costs, and take away from early mother-child bonding.
As the ACOG states, clinical practice patterns and the rates of C-sections vary greatly from state to state and hospital to hospital. Ivy and Hicklin have moved health technology closer to a uniform data analysis tool that doctors can use in collaboration with their patients to recommend the best mode of delivery, thereby improving outcomes for both mother and baby.
Their research, one example of how systems engineering is being used to improve healthcare, will be presented at the 2015 INFORMS Healthcare conference in Nashville, July 29-31.