The Use of Research to Discuss the Benefits of Doula Support

Catie Mehl For Doulas, Labor & Birth, Medical Professionals 3 Comments

I often see doulas citing statistics about the benefits of doula support. And why not? There’s a fantastic 118 page Cochrane Review on the benefits of continuous labor support and a terrific summary of the review can be found on Evidence Based Birth’s blog about doulas.

It seems pretty clear from the research that doulas are awesome, so why not use that research to sell our services?

First, it’s important to note that the Cochrane Review is about continuous labor support in general, not doula support specifically. The review found continuous support in labor resulted in:

  • 34% decrease in the risk of being dissatisfied with the labor and birth experience*
  • 31% decrease in the use of synthetic oxytocin (also known as Pitocin)*
  • 28% decrease in the risk of cesarean birth*
  • 14% decrease in the risk of newborns being admitted to a special care nursery
  • 12% increase in the likelihood of a spontaneous vaginal birth*
  • 9% decrease in the use of any medications for pain relief

Again, most of these stats are for continuous labor support in general. Only the results with an * showed a difference between doula support and continuous labor support provided by a laboring person’s partner, family member, friend, loved one, nurse, nurse midwife, or someone else.

Now, don’t get me wrong, having data showing that doula support can decrease the cesarean rate and the risk of being dissatisfied with the birth experience and the use of synthetic oxytocin as well as increase the likelihood of a spontaneous vaginal birth are fantastic and should absolutely be considered by policy makers and hospital administrators when making decisions about their L&D units.

But when it comes to the marketing of doula services to consumers, we doulas need to back off on the use of studies and these stats and focus on the importance of support. Here’s why:

First, the above Cochrane Review on continuous labor support looked at over 15,000 women and 22 different studies. “The trials were conducted in Australia, Belgium, Botswana, Brazil, Canada, Chile, Finland, France, Greece, Guatemala, Mexico, Nigeria, South Africa, Sweden, Thailand, and the United States, under widely disparate hospital conditions, regulations, and routines.”

So right off the bat, we’re looking at the role of support in countries that are very different from the US in terms of population (Sweden and Nigeria are very homogeneous, for example, where the US is very heterogeneous), their culture around birth, support, hospital culture and abilities, etc. The population differences alone can impact all sorts of birth outcomes due to pelvic shape and muscular structure (the pelvis is also responsible for thermoregulation, therefore we often see one predominant pelvis shape in homogeneous countries w/ warmer climates vs. homogeneous countries w/ colder climates vs. countries like the US which is a relatively young country on an evolutionary level and is heterogeneous in population).

In other words, there are other possible things going on aside from a doula.

Second, the full review includes over 15,000 women and 22 studies, only 5 of the included studies were specific to doulas (a total of 1,501 women). All of the doula studies were from 1986-1991 except one, and while that one was published in 2008, the data is from 1988-1992.

So, if we look at just the five studies on doulas, we’re dealing with a very small sample size and studies that were conducted more than 2 decades ago when doulas and hospitals didn’t have the issues that they do today. Also, at least one of the studies was done in Guatemala (refer back to point one on why this is an issue). 

These are important things to understand in general as the studies most certainly suggest a correlation but the small number of women doesn’t clearly show causation, either. There could be other factors at play that are not being considered.

Lastly, research is based on populations and we work with individuals.

When the research on doulas says that doula support decreases the cesarean rate by 28%, that doesn’t mean we decrease a single person’s cesarean rate by 28%, rather the overall cesarean rate.

Example: Riverside Methodist Hospital (RMH), a large and local to me hospital, had 1,980 nulliparous term singleton vertex births last year. Nulliparous term singleton vertex (NTSV) means the data is only for a person who is giving birth for the first time, they were at term (defined as 38 weeks) at the onset of labor, they were pregnant with one baby, and the baby was head down at the onset of labor. Their NTSV cesarean birth rate was 24.8485% (492 births were cesarean births).

Now, according to the Cochrane Review, doula support showed a 28% decrease in the cesarean rate.

So let’s say RMH decided to start a doula program and required all NTSV patients to have a doula (this is what makes this example an extreme and not possible) in an effort to decrease their cesarean rate. The hospital should expect their NTSV cesarean rate to drop from 24.8485% to 17.89092%. But that’s as a hospital with 1,980 NTSV births a year. Not an individual person (because 354 people would have both a doula and a cesarean). But people don’t understand this.

People tend to hear “a doula decreases the risk of cesarean”, but they think doula = no cesarean.

So, yes, technically doulas can improve outcomes overall, but it would take a huge number of women to have the support of a doula to make a difference. Also, and remember that this is what makes this example so extreme, this example is assuming all NTSV patients at RMH would have a doula. If we do something smaller, let’s say 50% of all NTSV patients, it only decreases the cesarean rate to 21.36966%. (Special thank you to Alex Weinberger of Toronto Family Doulas and fellow ProDoula Trainer for helping me figure this stat out!)

Additionally, remember that the number I’m using is only the number of NTSV births and this number does not include anyone who has

  • previously given birth OR
  • has started labor before 38 weeks OR
  • has multiples OR
  • the baby is anything but head down.

To give a frame of reference, RMH averages about 5,000-6,000 births a year (5,005 for the year I have the data for). NTSV births make up around 40% of their births every year. It would take a huge number of doulas at that one hospital alone, and it’s only 1 of 12 in my area. We would need a minimum of 13 doulas to support the 1,980 NTSV births at RMH alone (assuming each doula works 3 12-hour shifts a week and no one takes a week off or ever gets sick).

Then you have to figure out how much those doulas should get paid and compare it to how much cost savings the lower cesarean rate would bring to determine if it’s even something the hospital would be willing to do. Then you have to add in the fact that the most significant decrease in the cesarean rate is specific to doulas who do not work for the hospital so the reality will likely be something less than 28% decrease.

So the point of all of this is to say that research and statistics are far more complex and nuanced than many doulas realize and it is because of this that we need to stop using these stats to sell our services.

Yes, doulas are awesome. But we’re not awesome because we have the potential to lower the cesarean rate, but because we provide something no one else can: nonjudgmental emotional support.


Comments 3

Leave a Reply

Your email address will not be published. Required fields are marked *