The ARRIVE study, funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network and published in the American Journal of Obstetrics & Gynecology, January 2018 Supplement, looked at composite outcomes comparing induction of labor at 39 weeks gestation, to expectant management, allowing women to spontaneously go into labor through 40 weeks 5 days. A paper with the findings was read at a conference of the Society for Maternal-Fetal-Medicine prepublication.
The content of this article is derived from a commentary on the ARRIVE Study by Dr. R. Dekker of Evidence-Based Birth.com.
The ARRIVE study, a randomized controlled trial, looked at various outcomes on neonatal and maternal health including c-sections. Criteria for inclusion were: pregnant women having their first baby at 38 weeks, singleton, vertex, with no plans for induction or cesarean, and no medical, neonatal, or maternal complications. 6016 women were included in the study. The early induction group went through cervical ripening, then induction, and were given an extraordinarily long time for the induction to work.
The outcomes were: cesarean section rate 18.6% for the induction group compared to 22.22% in the expectant management group. This was a 3.6% drop but this is not statistically significant. That means there is no real difference in outcomes between the groups and it could have happened randomly by chance.
There were no differences in neonatal deaths or illness, very low Apgar scores, brain damage or seizures. There was a 1.2% higher rate of needing resuscitation in the expectant management group but no other differences. The only maternal outcome was 5% less incidence of Pregnancy Induced Hypertension (PIH) in the early induction group, probably due to the fact they were induced before they had a change to manifest PIH. There were no other maternal health differences. None of the researchers looked at differences in breastfeeding rates or epidural rates.
The effect on the cesarean section rate could have been due to the observer effect: people who know they are being watched behave differently. The doctors in this study knew they were being studied and waited a long time for the women to deliver, much longer than they ordinarily would, doing everything they could to have a successful induction and vaginal delivery rather than call for a C/S. With this in mind, if we could enroll everybody in a study, we would probably lower the C/S rate just from being observed!
Other ways to explain the outcomes:
This study took place in a heavily medicalized large hospital system with MD management, continuous EFM, and epidurals, the medical model. Women in this study stayed in bed on continuous fetal monitoring and had epidurals. There were no natural births and no midwives. In this medicalized setting, it reduced the cesarean section rate 3-4%. The study did not look at cesarean section rates with natural labor or with the midwifery model of care (for more on the midwifery model of care, see: http://cfmidwifery.org/pdf/mmoc_brochure.pdf, and https://www.ncbi.nlm.nih.gov/pubmed/10466283).
This study is not GENERALIZABLE: that is, we cannot apply the results of this study to other types of maternity patients or maternity care models, such as those having a natural birth or women not having their first baby or institutions that do not allow a long period of time for inductions to be successful. It is not generalizable to women who do not want a medicalized birth or who have the midwifery model of care (natural labor, labor support, and few interventions).
The number of subjects needing to undergo the intervention in order to save one cesarean section, in this case, is called the number to treat. You get the number to treat by dividing 1 by the difference in cesarean section rates between the two groups, 3.6, and you get the number to treat, which is 28. That means you will save 1 cesarean section for every 28 people that are induced at 39 weeks. That’s a lot of intervention to a lot of people to save 1 cesarean section. What do we know we can do that is much less interventive yet is known to reduce cesarean section?
Regaia’s study: Walking in the first stage of labor reduced the cesarean section rate from 16% to 5%. The number to treat is only 9. That means that for every 9 women who walk in labor, you prevent 1 cesarean section. Walking is easy and cheap, and is healthy for mothers and babies.
McGrath & Kennel’s study: having a doula reduced the cesarean section rate from 25% to 13.4%. The number to treat is only 9. So for every 9 women who have a doula help them in labor saves one cesarean section. And there are MANY more advantages to having a doula (less use of narcotics, higher breastfeeding rates, faster labors, etc.).
Allowing longer pushing time for women with an epidural: Allowing 4 hours pushing time for women with an epidural instead of limiting it to 3 hours decreased the cesarean section rate from 43% to 20% in one study. The number needed to treat was 4, so allowing 4 women to push longer will save 1 cesarean section.
Waterbirth: A 2013 study showed a cesarean section rate of 36% for land births but a cesarean section rate of 5% for waterbirths. The number needed to treat was 9, so for every 9 planned waterbirths, there is one less cesarean section.
Free standing birth centers: We know the cesarean section rate for a free standing birth center is 6%. Care in free standing birth centers involves midwifery care, doula services, water labor/birth, and intermittent monitoring (rather than continuous monitoring). All these things lower the cesarean section rate and are low intervention, low cost, and healthy for mothers and babies.
There are so many ways to lower the cesarean section rate without having to induce everybody.
Bell 2017: This study showed that implementing the American College of Obstetricians and Gynecologists’s new guidelines for diagnosing arrested labor, training nurses to provide labor support and use birthing balls, dropped cesarean section rates from 28% to 20%.
Chavenick’s 2017 study showed a drop in cesarean section rates from 29% to 12% in 12 months when they stopped elective inductions below 41 weeks, waited to admit until laboring women were 4 cm, used intermittent auscultation (rather than continuous electronic fetal monitoring), and watched each provider’s cesarean section rate AND THEN PUBLISHED THEM!
CMQCC has a whole toolkit with ideas for care to lower cesarean section rates without inducing women (https://www.cmqcc.org/qi-initiatives/promoting-vaginal-birth).
None of these studies or interventions used early inductions and got much better results in lowering cesarean section rates.
As I have outlined in this article, there are a lot of things we can do with a much lower number needed to treat to successfully reduce cesarean section rates.
Induction of labor without medical indication is a big intervention with a host of implications for the health of mothers and babies in addition to birth outcomes and effects on the societal and personal view of the competence of the birth process. The ARRIVE study is very short-sighted at best.
Other links:
The AJOG paper: http://www.ajog.org/article/S0002-9378(17)32491-2/pdf
The ACNM response: http://www.midwife.org/induced-labor-study-statement
Rebecca Dekker’s ARRIVE Commentary: https://www.stitcher.com/podcast/rebecca-dekker/evidence-based-birth
Marie-Celine Farver RN, BSN, IBCLC © 2018