Recently I wrote about the natural childbirth website Science and Sensibility, detailing how it is neither scientific, nor makes much sense. That’s probably because every discussion has to be jammed into the same pre approved narrative arc: evil obstetricians, whose raison d’être is ruining birth “experiences” create a theory/practice/procedure which ignores scientific evidence, the evil obstetricians persist in using this theory/practice/procedure even though it doesn’t work, but now we’ve learned that they are utterly wrong and still they continue what they have been doing. Since the pre approved narrative arc has nothing to do with the truth, the post misstates or misinterprets the science in critical ways.

Henci Goer’s recent post, Iatrogenic Norms: How Fast Do First-Time Mothers Beginning Labor Spontaneously Actually Dilate, is a perfect example of natural childbirth as an unscientific smear. The first principle of the NCB is smear is to start with a gratuitous swipe combined with a little made-up “medicine.”

Iatrogenic norm: a defined range of normal values for a biological process that, rather than describing actual normal physiology, instead measures the consequences of a health care provider’s beliefs, actions, or therapies or the effects of exposure to a health care facility.

Oooh, sounds fancy and scientific. Too bad Henci Goer just made up that “definition,” which exists nowhere else.

The body of the post is an attempt to smear “the famous ‘Friedman curve’.” Recent research, looking at the ways in which epidurals influence labor, suggest new norms. Goer is outraged that anyone would presume to define any “arbitrary” norms to distinguish normal from abnormal labor.

Nevertheless, while revising norms to match reality would take a big step in the right direction, I would argue it doesn’t go nearly far enough because it still sticks us with the assumption that active first-stage dilation progresses smoothly. Anyone who has spent time with laboring women knows that this is often not the case. Neat graphical lines (or curves) come from averaging many highly variable individual labors, so the very expectation of how labors progress, at whatever pace, is itself an iatrogenic norm.

That sounds fancy and scientific, too. What a shame, then, that is nothing more than made up baloney centered around a made up “definition.” Then there is this witless gem:

Moreover, the published review points out that both the old and the proposed new threshold for “abnormal” are statistically derived (e.g. two standard deviations beyond the mean). No study links a cut point for “abnormally slow” with an increase in perinatal morbidity, but averting adverse outcomes should form the basis for intervening medically because of the risks of intervention. In fact, even if a study tried to establish an outcome-based threshold, it would be hard to determine whether the increase was due to labor duration per se or to the interventions used to treat slow labor…

Statistically derived? Two standard deviations beyond the mean? Well, duh. That’s not some nefarious plot; that’s the entire point of statistical analysis.

And finally there’s this:

… No study links a cut point for “abnormally slow” with an increase in perinatal morbidity, but averting adverse outcomes should form the basis for intervening medically because of the risks of intervention. In fact, even if a study tried to establish an outcome-based threshold, it would be hard to determine whether the increase was due to labor duration per se or to the interventions used to treat slow labor. So we have yet another iatrogenic norm, this one having to do with a definition of “abnormal” with no clinical significance.

No link with perinatal morbidity? No fooling! That’s because the curve has nothing to do with perinatal morbidity and no one ever claimed that it did.

There are so many mistakes and misinterpretations in this piece that it’s hard to know where to begin. I’ll confine my discussion to the three most egregious mistakes, one historical, one statistical, and the third a serious misrepresentation of the purpose of the curve itself.

I know a bit about the Friedman curve because I trained with Dr. Friedman himself. He was the chief of my department at Beth Israel in Boston for the four years of my residency. He was an extremely difficult man to work with, but he was brilliant and a strong advocate for women.

How and why did Dr. Friedman define the curve?

Dr. Friedman did his residency in the 1950s. He was not a man to suffer fools gladly and he considered a lot of his superiors to be fools. He felt that they made medical judgments based on their intuition and not on science, and he set out to accumulate the research data necessary to give the profession a firm scientific foundation.

During his residency, when he was on call every other night, he used his “spare” time to compile detailed observations about every laboring woman who came through the hospital. The goal was no less than to find out what normal labor looked like. Using observations from tens of thousand of women, he created a curve. Women who followed the curve were almost certain to have a vaginal delivery. Women who fell off the curve were more likely to need a C-section.

Dr. Friedman was the first to say that you should not section a woman in latent phase because a long latent phase was not a sign that the baby doesn’t fit. He insisted that you should not section a woman in the active phase of labor unless she failed to make a certain amount of progress in a certain amount of time. Dr. Friedman used to express the utmost disgust for doctors who would say, “she looks like a C-section to me”, instead of adhering to established criteria.

So Goer has thoroughly misrepresented the Friedman curve. It was created precisely to AVOID unnecessary C-sections, not to justify them. And it is hardly “arbitrary.” It reflects the observation of thousands of labors, both normal and abnormal and graphically represents those observations.

But perhaps Goer is confused into thinking that the curve is arbitrary since she completely misunderstands and misrepresents standard deviation.

NCB advocates like to claim that medical definitions of “normal” are utterly arbitrary and exist merely for the convenience of doctors. Nothing could be further from the truth. Often, “normal” is based on knowing the outcomes from previous experience. We can confidently say that having an Apgar score of 1 at 5 minutes of life is not normal, because babies who have Apgar scores of 1 at 5 minutes always have serious medical problems of one kind or another.

Sometimes “normal” is defined as a range. That is not an accident, and it does not mean that a range was chosen arbitrarily. A normal range in medicine is almost always based on a basic and widely accepted form of statistical analysis, the standard deviation.

There is an excellent simple explanation of standard deviation on SensibleTalk.com. It is written for journalists who have no background in statistics:

Let’s say you are writing a story about nutrition. You need to look at people’s typical daily calorie consumption. Like most data, the numbers for people’s typical consumption probably will turn out to be normally distributed. That is, for most people, their consumption will be close to the mean, while fewer people eat a lot more or a lot less than the mean.

When you think about it, that’s just common sense. Not that many people are getting by on a single serving of kelp and rice. Or on eight meals of steak and milkshakes. Most people lie somewhere in between.

When you graph the data with calories on the x-axis and numbers of people on the y-axis, you will get a bell shaped curve. The curve is a graphical representation of all the possible things that can happen. The important point, though, is that every possible thing that can happen is not necessarily normal. How do we tell the difference between normal and abnormal? We start by calculating the standard deviation. The formula for calculating the standard deviation is complicated, but the result is relatively simple to understand. The standard deviation is a reflection of distribution of all possible outcomes.

Mathematically, one standard deviation on each side of the mean (the average) encompasses 68% of individuals. Two standard deviations encompasses 95% of individuals. Therefore, only 5% of individuals will be outside of two standard deviations from the mean. This is always true, regardless of whether the bell curve is tall and narrow or short and extended. “Normal” is usual defined as within two standard deviations. That means that “normal” is a range, but the range is hardly arbitrary. It reflects the actual distribution of results among large populations of human beings.

So when we look at how long a first labor lasts, for example, we can graph the labors of large numbers of women and we will get a bell curve. Ninety-five percent of women will fall within two standard deviations of the mean. It is only those women who are outside of two standard deviations that are considered abnormal. That does not mean that a woman whose labor is lasting longer than two standard deviations from the mean cannot possibly have a vaginal delivery, but it does mean that a woman whose labor is lasting longer than two standard deviations from the mean is far less likely to have a vaginal delivery.

The bottom line is this: defining normal as a range is not arbitrary. It is a reflection of what we know about human variation. The range of normal accounts for most of human variation. Anything that lies outside the range of normal is very unlikely to be normal.

Finally, the swipe at the curve for not being related to perinatal morbidity and therefore being clinically irrelevant is just plain bizarre. The Friedman curve has NOTHING to do with morbidity and mortality. That wasn’t its purpose when it was developed and it is not its purpose today. It is, however, quite important clinically because it tells us the likelihood that woman will deliver vaginally.

Oh dear, it seems that the story of the Friedman curve does not fit the predetermine arc of the NCB smear. The Friedman curve was NOT created to ruin women’s birth experiences; it was created to reduce unnecessary C-sections. The Friedman curve is NOT arbitrary; it is simply a graphical representation of thousands of labors. Standard deviation is NOT arbitrary; it is at the foundation of statistical analysis. No matter. Who cares about the truth? Certainly not Henci Goer.