A cascade of interventions: A classification tree analysis of the determinants of primary cesareans in Australian public hospitals

Abstract

Background Both globally and in Australia, there has been a sharp rise in cesarean births (CB). Commonly, this rise has been attributed to the changing epidemiology of women giving birth. A significant body of knowledge exists on the risk factors associated with a greater need for cesarean. Yet, we have little information on the reasons recorded by clinicians as to why cesareans are provided. This study aimed to explore the drivers of primary cesareans in Australian public hospitals. Methods Using a linked administrative data set, the frequency and percent of mothers’ characteristics were compared between those who had a cesarean birth and those who had a vaginal birth (n = 98 967) with no history of previous cesareans in Queensland public hospitals between July 1, 2012, and June 30, 2015. The top 10 reasons recorded by clinicians for a primary cesarean were reported. Using a machine‐learning algorithm, two decision trees were built to determine factors driving primary cesarean birth. Results “Labour and delivery complicated by fetal heart rate anomaly” (23%) and “primary inadequate contractions” (22.8%) were the top two reasons for a primary cesarean birth. The most common characteristics among mothers who had fetal heart rate anomalies were as follows: artificial rupture of membranes (39%), oxytocin (32%), no obstruction of labor (42%), and epidural (52%). For women who had primary inadequate contractions, the most common characteristics were as follows: epidural (33%), oxytocin (49%), artificial rupture of membranes (45%), and fetal stress (56%). Conclusions Efforts should be made by health practitioners during the antenatal period to maximize the use of preventative measures that minimize the need for medical interventions.

Sociodemographic risk factors of under-five stunting in Bangladesh: Assessing the role of interactions using a machine learning method

“…While classification trees are being increasingly used in many areas of public health research recently [ 32 , 33 ], its applications in the field of childhood nutrition are still few [ 34 – 36 ]. To our knowledge, only one paper [ 37 ] used classfication trees to discover risk factors of childhood stunting in the context of Bangladesh.…”

Response to: The “Cascade of interventions”: Does it really exist?

“…We thank the author for their reply to our previously published manuscript "A cascade of interventions: A classification tree analysis of the determinants of primary cesareans in Australian public hospitals". 1 The findings of our study show that the top two reasons for primary cesareans in Queensland public hospitalsabnormal fetal heart rate and inadequate contractionswere strongly associated with artificial rupture of membranes, induction of labor, and epidural analgesia.…”

“…Australia, cesarean birth, labor and birth Haylee Fox BN, MPH 1 Stephanie M. Topp PhD 1 Daniel Lindsay PhD…”

Use of pudendal nerve block among midwives in Norway: A national cross-sectional study

“…However, women in such units could benefit from being offered a PNB for suturing 27 , 28 . Interventions are reported to lead to more interventions 29 . In specialized obstetric units, midwives are likely to be more familiar with performing other interventions and may thus use PNB more easily 30 .…”