Surprise! Banning deliveries before 39 weeks isn't such a great idea after all.

I've written repeatedly about the foolishness of a "hard stop" policy banning elective deliveries (by induction or scheduled C-section) before 39 weeks. In Oops, reducing early elective delivery leads to more deaths, I reported on a paper that demonstrated that:

... The reduction in early elective delivery reduced NICU admissions, reduced both the induction rate and the C-section rate, and ... presumably reduced costs. However, these benefits were achieved at a very steep price. The stillbirth rate increased from 2.5 to 9.1 per 10,000 term pregnancies. Instead of 3 stillbirths between 37-39 weeks among 12,000 patients, there were 11 stillbirths between 37-39 weeks among a similar number of patients after reduction in early elective deliveries.

You didn't need a crystal ball to realize beforehand that a reduction in early elective delivery will INEVITABLY lead to an increase in perinatal deaths. Based on everything we know about stillbirth at term, any reduction in morbidity due to delivery before 39 weeks will INEVITABLY be accompanied by increasing mortality.

It appears that even the prime movers behind a "hard stop" policy banning early elective deliveries have finally realized their mistake. In a paper aptly entitled Oversight of elective early term deliveries: avoiding unintended consequences, authors Clark, Meyers and Perlin recognize that they embraced a "hard stop" policy without thinking through the consequences. Of note, the authors are employees of the Hospital Corporation of America, and a major motivation behind the policy was an effort to reduce short term costs.

... The position of the Hospital Corporation of America, as a primary driving force behind these changes, and our experience with such care improvement efforts in well over 1 million deliveries during the past 5 years suggest some areas of significant concern in the implementation of this policy and a number of practical remedies. A review of these issues may assist interested parties in helping patients reap the benefits of this practice change, while avoiding the associated perils.

The first problem with the policy is that many people have deliberately misinterpreted it as a ban on any deliveries before 39 weeks. The authors are at pains to clarify that the policy ONLY applies to a subset of early deliveries, those undertaken for purely social purposes:

It cannot be over emphasized that the campaign to reduce early term deliveries applies only to purely elective procedures. In this discussion, elective refers only to those scheduled deliveries that are performed without a valid medical indication. Many clinical conditions exist in which the well-described risks of early term or even preterm delivery are outweighed by the benefits of delivery to mother or child... Our concern is that a misinterpretation both of our policies and of the nature of our specialty's opposition to purely elective early term deliveries may result in inappropriate reluctance to deliver women who are at risk for serious complications...

Bans on early elective delivery are meant to reduce perinatal morbidity, but NOT at the expense of increasing perinatal mortality:

... Accepting the risk of such morbidity in select individual cases in which the dangers of continuing the pregnancy because of valid medical complications is significant is often the best choice. Thus, it is incumbent on any entity that promotes a reduction in early term delivery to make it clear that the target practice is early term delivery without medical indication, not generic early term delivery, and that occasional indicated early term or preterm delivery remains an important part of good obstetric care (my emphasis).

The authors had previously recognized the importance of early delivery in a variety of situations, and had attempted to create criteria to differentiate between indicated and unindicated early deliveries. They now recognize that their criteria are not based on hard data.

... How close must the blood pressure be to 160/110 mm Hg level to justify delivery at 37 weeks gestation or even before? How poorly controlled must the diabetes mellitus of a noncompliant patient be to justify delivery at 38 weeks' gestation? In the absence of hard data to guide the clinician, physician judgment and informed consent will continue to play a major role in such cases. Any facility that uses the "hard stop" approach must have in place the availability of an easy-to-access chain of command 24 hours a day to resolve such issues.

And:

... [I]t is critical to realize that, because the Joint Commission definitions for indicated early term deliveries are based on diagnosis-related group (DRG) codes and because many valid indications for early term delivery exist that do not have such a code, the rate of “elective” early term delivery for any institution will never be and should never be consistently zero. There is no code for a multiparous woman whose most recent labor lasted 10 minutes and who lives 1 hour from the hospital. Yet, when that patient is seen at 37-38 weeks' gestation with a cervix that is dilated 4 cm, delivery is clearly indicated, not elective. Similarly, there is no DRG code for "history of a classic cesarean section delivery," yet such women should be delivered routinely at <39 weeks' gestation. Numerous other examples exist.

In other words, "soft" indications, comprising borderline cases and based on the clinical judgment of the obstetrician, are real indications. I'm glad that the authors acknowledge this, but their recommendations for addressing it are poor. There should no barriers to clinicians exercising their judgment, and calling for permission is a barrier. Rather, clinicians should have to justify their decisions retrospectively, presenting evidence to other obstetricians as to why they felt the early delivery was indicated.

Finally, the authors address the "elephant in the room," the insurance companies who push for short term savings at the expense of infant lives:

... Although a retrospective review of early term deliveries with nonpayment for those without a valid indication has been proposed, the aforementioned discussion suggests that this is a particularly bad idea, with the potential to promote bad practice and catastrophic outcomes... [N]o evidence exists to validate the appropriateness of off-site, post-hoc reviews to determine payment when dealing with deliveries that possibly were elective... Knowledge of the potential for such oversight error and its associated financial penalties establishes for the clinician and facility a perverse incentive to delay delivery when delivery may be in the best interest of the mother and baby, with potentially catastrophic results...

The authors insist that they still support a "hard stop" policy:

... The observations presented here do not represent any weakening of our commitment to the elimination of elective deliveries at <39 weeks' gestation.

However, their caveats belie their claim and they ought to change their terminology to reflect that. They should make it clear that they are talking about purely social indications by explicitly naming the policy a ban on purely social indications for early deliveries. Elective early deliveries for ANY other indication, including soft indications, are necessary, often life saving, and should not require permission.