The polarized history of birth politics is neatly contained within my family. I was born in a second-floor bedroom, with candles burning and The Chieftains on the turntable. My wife, Beth, was born at a hospital via scheduled cesarean section, without musical assistance. My parents were hippies. Hers were Republicans. In our respective rebellions, we met somewhere in the middle. Both of us lacked the fervor that had divided the previous generation during the culture wars of the ’70s. She, a nurse, was aware of the occasional excesses of modern medicine, and I, having spent my childhood testing the beneficence of nature, no longer believed that “natural” was synonymous with “healthy.” Nonetheless, before we went to our first prenatal appointment in December, Beth asked me not to say anything crazy.

“What do you mean?” I asked, aghast.

“I just don’t want you to go into investigative-reporter mode on the doctor,” she said.

For the last few years pregnancy and birth has been my beat as a journalist. It began when a source showed me the findings of an unreleased report from the California Department of Public Health revealing that the number of women dying from causes directly related to pregnancy had more than doubled in the previous decade. The researchers immersed in this data said that this statistic was probably the most visible sign of much more widespread problems haunting maternal health care.

The U.S. is now the most dangerous developed country in which to give birth, and by far the most expensive. Usually money is the fulcrum of the health-care debate–we argue about how many we can afford to cover with insurance, and about who should get the benefit of expensive new tests and drugs. But as I reported a series of stories on maternal health it became clear to me that access to health care is beneficial only if the quality of that health care is good. And quality, it turns out, is highly variable. Even the best hospitals regularly perform procedures with no basis in scientific evidence.

My reporting, done from an abstract height, took on a new immediacy once Beth and I entered the health system ourselves. It was pretty hard to question the standard treatment at all without violating Beth’s directive against sounding like a crazy person. Hospitals, both for better and for worse, are industrial systems. They are optimized for efficient treatment, and not for patients who want to examine the science backing each routine procedure. They run most smoothly when patients have faith in their efficacy. It’s rough going for patients like me, who would rather leave faith out of medical decisions entirely.

It hadn’t taken much research to see why medical authorities tended to think home-birthers like my parents were crazy. Maternal mortality peaked in this country around the turn of the last century, when more than 800 women were dying for every 100,000 live births. Since that time, it has fallen more than 99 percent, reaching its nadir in 1997 at 7.7 deaths per 100,000 live births. Infant mortality fell a little over 90 percent in that time period. Clearly, obstetrics has been wondrously effective–so effective that some practitioners feel they can legitimately dismiss any complaints about modern childbirth as frivolous.

These improvements have coincided with a shift in obstetrics toward more technology and greater standardization. Rather than adapting techniques to the vagaries of each birth, doctors have sought out methods that yield reliable results regardless of context. In pondering this trend, the physician-journalist Atul Gawande showed that the cesarean surgery was one of the best methods for bringing standardization to birth. It’s easy to learn, and it replaces the atavistic complexities of mammalian reproduction with a few legible, controllable steps.

In a 2006 article for The New Yorker (titled “The Score”) Gawande asked if there was anything wrong with this: “The question facing obstetrics was this: Is medicine a craft or an industry? If medicine is a craft, then you focus on teaching obstetricians to acquire a set of artisanal skills–the Woods corkscrew maneuver for the baby with a shoulder stuck, the Lovset maneuver for the breech baby, the feel of a forceps for a baby whose head is too big. You do research to find new techniques. You accept that things will not always work out in everyone’s hands. “But if medicine is an industry, responsible for the safest possible delivery of millions of babies each year, then the focus shifts. You seek reliability. You begin to wonder whether 42,000 obstetricians in the U.S. could really master all these techniques.”

Gawande concluded that, notwithstanding a romantic twinge at the thought of losing human connection to yet another natural process, it was the quantifiable results of improving infant health that had driven this industrial revolution in hospitals–and who could argue with that? Though several years have passed since the publication of this article, it’s worth considering here because it epitomizes both the way most medical institutions frame the issue of C-sections to this day, and the error that undermines that framing. This error is a willingness to allow the logic of industrialization to drive medicine (or the reasoning in a magazine story), while failing to use that same logic to check the results. And checking isn’t so hard.

Medical researchers, thanks in large part to the work of Donald Berwick (now running Medicaid and Medicare for the Obama administration), have increasingly embraced the discipline of quality improvement, a set of techniques borrowed from industrial engineering. Quality improvement seeks out measurable benchmarks to show if reforms make things better or worse. These data allow administrators to see problems that are otherwise invisible–either because cause is separated from effect, or because they defy conventional wisdom. “The Score” focused on one such measurement, the Apgar score, which quantifies the health of a newborn on a scale of 0 (limp and blue) to 10 (pink and wriggling). And while Gawande showed how attention to this measurement could drive doctors to do more cesareans (on this point he is typically brilliant), he didn’t mention that as the cesarean rate rose from 15 percent in 1978 to 32 percent today, overall Apgar scores did not budge. In other words, quantification may have caused more surgeries, but those surgeries didn’t produce a quantifiable improvement. There are also other measurements used to determine the success of obstetrics. One is maternal mortality.

This article came up when I met with Aaron Caughey, head of obstetrics at the Oregon Health and Sciences University (at the time Caughey was at UC San Francisco). Gawande and Caughey had gone to Harvard Medical School together where they had become friends. “So did he call you when he was writing ‘The Score?'” I asked. Caughey shook his head. I wondered if he thought there was anything more to be said on the subject. “I think there is,” Caughey said. “I mean it was a great article. But most people have been missing something when it comes to C-sections.”

Caughey’s research predicts that if we brought the national C-section rate down to 20.6 percent by 2020, every year we’d be saving the lives of 110 moms who would otherwise die as the trajectory continued upward. These deaths are almost nearly impossible for doctors to see because most complications and injuries stemming from cesareans do not appear until the next birth, or the one after that. By performing the surgery an obstetrician can lower the risk that anything going terribly wrong, but she passes an increased risk on to the next doctor–not to mention the patient.

Caughey said it would be “really, really easy” to bring the national cesarean rate back down to 20 percent–meaning that doctors and hospitals have the tools they need to safely perform more natural births. Other quality improvement reformers have come to similar conclusions.

Intermountain Healthcare, a nonprofit hospital system based in Utah, started on a new course more than a decade ago, after the industrial audit recommended more “low-tech, high-touch” births. Intermountain focused most of its efforts on reducing the number of babies unnecessarily delivered early (either via cesarean or induction). Bryan Oshiro, who worked for Intermountain as a neonatologist in the early days of this project, told me a story of the defining experience that had made him an evangelist for this issue. The head of the neonatal intensive care asked him to look in on a group of babies with problems.

“He said, ‘You doctors are doing this. These babies are here because you allowed them to be delivered early,'” Oshiro said. “That just kind of stopped me dead. It was really clear–we’re hurting babies. And we can stop it.”

The effects of the changes were wholly positive. Mothers, on average, were able to go home sooner. Complications and admissions to the neonatal intensive care unit decreased. C-section rates declined. But there was also a hitch: The reforms have reduced the amount Intermountain charged its patients by more than $250 million since 1999. “Intermountain is nonprofit and that makes it easier for us to take such steps, but the incentive for most organizations is to provide more care, not less,” Intermountain representative Daron Crowley told me.

My own analysis showed that a pregnant woman walking into a for-profit hospital in California was significantly more likely to have a cesarean surgery than a woman walking into a hospital without an incentive for profit. I doubt that doctors are the mechanism driving this disparity (they usually receive about the same reimbursement regardless). But it is clear that nonprofit hospitals are the ones adopting these reforms. I imagine it would be hard to convince any hospital administrator to adopt a low-tech birth program that costs money, sometimes alienates doctors (no one likes to be told they are doing things wrong), and (oh by the way) reduces revenues by millions of dollars. But it would be especially hard if that administrator’s performance was measured by hospital profits.

As my wife and I began talking about where she would give birth, all this was troubling. Of course, we could have simply opted out of a hospital birth entirely. That’s not as foolhardy as those statistics at the beginning suggest. One point usually left out of the improvement-of-obstetrics story is that obstetric practice itself was responsible for a large part of those shocking turn-of-the-century statistics. In a 1999 review of the century’s improvements, the Centers for Disease Control concluded that “Poor obstetric education and delivery practices were mainly responsible for the high numbers of maternal deaths, most of which were preventable.”

Uneducated midwives were certainly part of the problem in those days when the majority of births occurred at home. But historical examples suggest home births weren’t the crux of the problem, as in the well-documented case of Eastern Kentucky’s Frontier Nursing Service, which, starting in the 1920s, made it safer to give birth in a log cabin, attended by well-trained horseback midwives, than in the best hospitals in the city.

Even today the evidence is perplexingly mixed on home birth. The people who do meta-analyses of all the various studies (like the Cochrane Systematic Reviews) basically say that–for a low-risk pregnancy–it’s a wash. There’s insufficient evidence to advise patients that it’s safer to give birth in a hospital or home.

Beth and I were left to sort out the choice on our own. It made intuitive sense that a hospital would be safer in an emergency–I could think of several conditions (an undetected placental abruption for instance) where you would want to be seconds, rather than minutes, from an operating room. On the other hand, I’d also interviewed the families of women who had died because of mistakes made in the bustle of hospitals. I found I was weighing Gawande’s rhetorical question: Do we go with craft, or industry?

It’s a question that reformers now are asking about the entire medical system. Ideally health care would be post-industrial, assuming either form depending on the circumstance; it would be nice if medical record-keeping worked with the effortless efficiency of industry, while office visits emphasized the personal touch of an artisan. Family medicine and psychiatry–craft; surgery-industry. But if we treat all medicine as an industry, as is increasingly the case, it would behoove the country to also adopt the best practices of industry–to make decisions based on evidence–rather than simply using the metaphor to justify a reliance on technology. Any industrial system is by necessity reductive, it boils complexity down into measurable numbers. At the very least we should pay attention to what those numbers are telling us.

One of those numbers is cost. In the United States maternal mortality is the most salient example of our health care paradox–the more we spend the sicker we get. The costs of maternal and infant health have grown to absolutely astronomical levels, while the outcomes have stagnated or gotten worse. Troubling as they are, the rising maternal mortality statistics are also an opportunity. If we can figure out what is happening in reproductive health then we will have a set of tools to make health care better and more affordable.

But this is a challenge for policy makers and hospital administrators. It’s not easily undertaken by individual patients. There’ no place for nuanced discussions about quality improvement during prenatal checkups. The only thing that seemed within our power to control was our choice of systems–or of faiths, if you will. Neither my wife nor I wanted to place our trust in the power of The Chieftains, but we also didn’t want to be nudged into a cesarean surgery by a health system with policies based on industrial logic rather than science. In the end, the idea of giving birth in a hospital just seemed a little more comfortable–a little more safely mainstream. It’s frustrating, but despite all the statistics and studies, we ended up making the one choice available to us based on a gut feeling.

Nathanael Johnson’s reporting on maternal death rates for the non-profit investigative reporting outlet California Watch appeared last year on the front pages of newspapers across the state, from the Sacramento Bee to The Press-Enterprise of Riverside.