A debate: Should obstetricians participate in homebirth?

The editorial section of this month’s issue of the journal Obstetrics and Gynecology addresses the issue of obstetrician participation in homebirth. Although it it not presented in this way, it appears that the journal asked the question “is obstetrician participation in homebirth ethical?” and commissioned one group to argue “yes” and the other to argue “no.”

The first group argues that medical ethics allows obstetricians to participate; the second asserts that obstetrician participation in homebirth violates their ethical responsibilities. Both raise important ethical issues and both, in my judgment, fail to address critical points.

What Are Physicians’ Ethical Obligations When Patient Choices May Carry Increased Risk? is written by Jeffrey Ecker and Howard Minkoff. The authors succinctly frame the issues:

… 1) When does respect for patient choice and autonomy become support for poor decision-making? 2) When is participation not respectful but enabling? …

Their answer to the questions is succinct, too:

… Weighing benefits and burdens should focus not on RR [relative risk] … but absolute risk … It is the absolute risk of adverse outcome (eg, 1/1,000 risk of neonatal mortality) that can be balanced against the benefits valued by a patient in choosing a particular choice (the comfort and control of delivery at home; a decrease in obstetrical interventions). As noted for home birth, the magnitude of the RR in comparison to a hospital birth remains unsettled but even in those studies that show a difference, the absolute risk remains low…

Ecker and Minkoff acknowledge that we don’t really know the absolute risk of homebirth in the US; they simply assume it is comparable to the risk in other first world countries. It’s a completely unjustified assumption for reasons they fail to address.

First, they appear to assume that homebirth in the US occurs under the supervision of midwives trained to the same standards as European, Canadian and Australian midwives. They don’t seem to realize that there is an entire class of self proclaimed homebirth midwives (certified professional midwives or CPMs) who have nothing more than a post high school “certificate.” They have less education and training than midwives in any other first world country, and, indeed, would be ineligible for licensure in any other industrialized nation. Their education and training is grossly substandard.

Second, they appear to be unaware of the fact that a huge database of American homebirths exists, the data on 18,000 homebirth attended by CPMs, collected by MANA (the Midwives Alliance of North America), the organization that represents CPMs. Throughout the years that the data was collected, MANA repeatedly proclaimed that the data would be used to prove the safety of homebirth. Now that the data have been analyzed, MANA is refusing to release the death rates. The MANA data almost certainly shows that homebirth has an unacceptably high rate of neonatal death.

Third, they appear to be unaware of the fact that individual states such as Colorado have collected data on planned homebirth with a licensed midwife, and the death rates are nothing short of appalling. Moreover, US data on homebirth shows that homebirth with a CPM has a higher death rate than homebirth with a certified nurse midwife (CNM), the type of midwife with training equivalent to European, Canadian and Australian midwives.

So the absolute risk of American homebirth is known, but being hidden by homebirth midwives, and the absolute risk varies widely depending on attendant.

The authors also fail to address a more important issue. Homebirth practitioners are not honest about homebirth risks. Indeed, in the homebirth community, it is axiomatic that homebirths are “as safe or safer” than hospital births. The entire argument of Ecker and Minkoff is predicated on patients’ willingly accepting an increase in absolute risk, but they don’t seem to realize that most women choosing homebirth don’t believe there is an increased risk of death associated with homebirth.

To use the language with which they framed the initial questions, obstetricians who participate in homebirth are merely enabling poor patient decision making. Most patients who choose homebirth have not given informed consent since they lack information on the real risks of homebirth, and have been told by their homebirth practitioners to ignore the information supplied by the obstetrician.

For me, the most remarkable thing about the Ecker and Minkoff piece is the stunning naivete regarding legal liability. The entire argument rests on the belief that a small increase in absolute risk of perinatal death is “acceptable.” But if the American legal system has taught obstetricians anything, it is that ANY increased risk of perinatal death, no matter how small, is utterly unacceptable.

The liability issue is further complicated by the fact that patients also think that any increase in the risk of perinatal death is unacceptable. When choosing homebirth, they aren’t choosing an increased absolute risk of death; they are pretending that there is no absolute risk of death. That’s why they often turn around and sue the obstetrician even if he or she warned them against homebirth, arguing that they didn’t “understand” that there was any increased risk of perinatal death.

Despite all this, Ecker and Minkoff conclude:

In sum, physicians are obliged to use their skills to minimize risks, even for women who have shunned physician’ recommendations and advice..

As far as I’m concerned, that’s an inexplicable non-sequitur. There’s a big difference between something being ethically acceptable and being ethical mandated. Ecker and Minkoff have made an argument (in my judgment a poor argument) that it is ethically acceptable for obstetricians to support and participate in homebirth. But obstetricians are not obliged to use their skills for any particular patient, unless they have agreed to care for her, and they are certainly not obligated to practice medicine in ways that violate their understanding of appropriate medical care.

Obstetricians are free to support and participate in homebirth if they want to do so. And it is certainly ethically acceptable to suggest that a small absolute increase in perinatal mortality MIGHT be acceptable in certain circumstances. However those circumstances include an accurate assessment of the absolute increase in risk, high level education and training for midwives, transparency on the part of CPMs in revealing their death rates from past cases and on an ongoing basis, and a legal system that agrees that a small absolute increase in perinatal mortality at homebirth is acceptable.

None of those conditions obtain at the moment and their advent is not even on the horizon. If Drs. Ecker and Minkoff want to support and participate in homebirths, they are free to do so, but they are not ethically obligated to do so, and neither is any other obstetrician.

Tomorrow, I’ll look at the paper presenting the opposite point of view.