Low Primary Cesarean Rate and High VBAC Rate With Good Outcomes in an Amish Birthing Center

Thank you for your thoughts regarding our recent publication and your description of the excellent results being obtained in accredited birth centers across the country. It is heartening that the opportunity for TOLAC is expanding in this setting. Hopefully we will see continued increase in the hospital setting as well. Susan Stapleton is correct to point out that many aspects of our birth center practice are less generalizable to non-Amish populations, particularly the inclusion of higher risk patients such as multiple gestation, breech deliveries, prolonged ROM, and transfers from unsuccessful home birth. The management of these more complex patients at the birth center gives the opportunity for a safer birth experience for these women who would otherwise be delivering at home.

Competing interests:   None declared

In response to the comments from Dr Tuteur regarding the neonatal death rate, many higher risk patients were delivered at the birthing center (see also the recent letter of December 17, 2012 which addressed the issue of patient selection). The following higher risk patients were allowed to deliver at the birth center: TOLAC (10%); past history of postpartum hemorrhage; postdates inductions with oligohydramnios; patients of advanced maternal age (25% of our patients were over 35); twin gestation if twin #1 was vertex; prolonged ROM (without clinical evidence of infection); gestational hypertension; unstable lie; frank breech planning vaginal delivery; patients transferring in from home after unsuccessful home birth. In addition patients with marked preterm status were delivered as well as infants with multiple anomalies not felt to be compatible with life.

A review of Supplemental Table 2 describes neonatal deaths. There were actually five such deaths for a rate of 5.4/1000 live births. These included one infant at term with multiple anomalies; two infants with severely preterm status (23 & 24 wk gestation); one infant with a lethal genetic neurological syndrome; one infant with macrosomia without dystocia, encephalopathy, and seizures. The infant with lethal genetic syndrome had had two siblings who died before one year of age with the same syndrome in spite of extensive medical evaluation at Mayo Clinic and University of Wisconsin.

Many families seek out our guidance and deliver at the birthing center rather than at home after such abnormalities are discovered, intrauterine fetal death has occurred, or onset of labor occurs early in the second trimester. We consider this part of our mission and assist such patients though it may adversely affect perinatal mortality statistics.

Competing interests:   None declared

The authors have not accurately represented their outcome because they compared their patients to the wrong control group, thereby making their outcomes look much better than they were.

The women who delivered at the birth center were white (race is a risk factor), low risk women at term with babies of normal weight, with no pre-existing medical illnesses and no serious pregnancy complications.

According to the CDC, the neonatal death rate for low risk white women at term with non-IUGR babies is 0.4/1000. The authors of this study report 3 neonatal deaths at term for a neonatal death rate of 3.2/1000. That's eight times higher than the expected rate.

The authors erroneously compared the death rate at the birthing center with the overall US neonatal death rate that includes women of all races, at all gestational ages, with all possible pre-existing medical conditions and all possible pregnancy complications.

The death rate at the birthing center is not low, it is appallingly high.

Competing interests:   None declared

The authors are to be commended for their efforts to meet the needs of childbearing families in their community, reporting outcome data, and excellent outcomes in this unusual birth center. Although the outcomes achieved in this unique population are very good, they are not dissimilar from those reported by researchers from American Association of Birth Centers (AABC) from 79 birth centers across the US in which clinical practice and risk criteria are consistent with the American Association of Birth Centers Standards of Care in Birth Centers.[1] These Standards have been periodically reviewed, judged against the best available scientific evidence, and reaffirmed by AABC members since 1978. The AABC Standards are also used by the Commission for the Accreditation of Birth Centers (CABC), the only national organization devoted exclusively to evaluating and accrediting birth centers. Although not universally accepted by all U.S. birth centers, the AABC Standards have been supported by evidence in several large prospective studies of birth center care in the U.S.[2,3,4]

While the authors are correct in reporting that AABC initially advised member birth centers against offering Trial of labor after Cesarean TOLAC) in 2005 after the publication of its Results of the national study of vaginal birth after cesarean in birth centers, 5 the organization revised this position based on other studies in 2008 to advise members that each birth center should make a decision regarding whether to offer TOLAC based upon its own unique situation, including level of access to this service in its community. Likewise, CABC initially prohibited TOLAC in accredited birth centers, but in July, 2008, issued a revised VABC policy permitting repeat TOLAC in any accredited birth center and primary TOLAC in accredited birth centers meeting certain criteria. Since that time, many CABC accredited birth centers have resumed offering TOLAC to women in their communities. AABC continues to collect outcome data for these women via its online Perinatal Data Registry(TM) in order to evaluate the safety of this policy. TOLAC

As noted by the authors, the outcomes reported in this retrospective analysis in the birth center may be to some extent related to both clinician experience and cultural attitudes surrounding birth in the Amish community. AABC researchers reported a cesarean section rate of 6% in birth centers in which twins, previous cesarean section, breech presentation, and indications for induction or augmentation of labor or vacuum-assisted delivery were risk factors precluding birth center birth; and in a population consisting of almost half nulliparous women.[6] The model of birth center care limited to physiologic labor and birth for women who are at low risk of obstetric complications seems likely to result in good outcomes across broader patient populations and a variety of provider backgrounds than the model described in this unique birth center. While clearly meeting the needs of this community, the significant role of the unique population in avoiding adverse outcomes should not be underestimated by those tempted to replicate the model in other communities.

1. American Association of Birth Centers. Standards for Birth Centers. Perkiomenville, PA. 2007. Available at http://www.birthcenters.org/open-a-birth-center/birth-center-standards

2. Rooks, J., Weatherby, N., Ernst, E., Stapleton, S., Rosen, D., Rosenfield, A. Outcomes of care in birth centers: The National Birth Center Study. New England Journal of Medicine. 1989: 321 (26)1804-11.

3. Jackson, D. J., Lang, J. M., Swartz, W. H., Ganiats, T. G., Fullerton, J., Ecker, J., et al. Outcomes, safety, and resource utilization in a collaborative care birth center program compared with traditional physician-based perinatal care. American Journal of Public Health. 2003: 93(6), 999-1006. Available at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1447883/pdf/0930999.pdf

4. Stapleton, S., Osborne, C., Illuzzi, J. Outcomes of care in birth centers: demonstration of a durable model. (in press) Journal of Midwifery & Women's Health.

5. Lieberman, E., Ernst, E., Rooks, J., Stapleton, S., Flamm, B. Results of the National Study of Vaginal Birth After Cesarean in Birth Centers. Obstetrics & Gynecology: 104(5), 933-42.

6. Stapleton, S. & Osborne, C. 2011. American Public Health Association presentation, Washington, DC.

Competing interests:   None declared

The issue of selection bias has been raised regarding the Amish patients delivered in our out of hospital birth center. Further clarification of selection criteria may be of interest to readers. The Amish of southwest Wisconsin traditionally deliver at home and avoid hospital delivery. Our five year substudy (see paragraph 1 in RESULTS) documented 1178 Amish births in Vernon and Monroe counties. Of these, only 5 patients self referred for hospital birth. Of the 448 patients who were considered for delivery in the birthing center, 17 (3.7%) were referred by us antenatally for a hospital birth. 36 patients were transferred intrapartum and included in statistics for transfer and/or cesarean birth. The above statistics support our premise that Amish patients with high risk status or prior adverse obstetrical events tended to self-refer to the birthing center, not to the hospital.

In the early years of our birth center experience, considerable efforts were made to refer higher risk patients for hospital delivery. Acceptance of hospital birth remained poor with such patients delivering then at home. With time we accepted such patients for delivery in the birth center. Though we were 20 minutes from the hospital, at the birth center we had experienced staff, IV fluids running when appropriate, on site ultrasound (during the last ten years), and ambulance service within minutes from the birth center. The following higher risk patients were allowed to deliver at the birth center: TOLAC (10%); past history of postpartum hemorrhage; postdates inductions with oligohydramnios; patients of advanced maternal age (25% of our patients were over 35); twin gestation if twin #1 was vertex; prolonged ROM (without clinical evidence of infection); gestational hypertension; unstable lie; frank breech planning vaginal delivery; patients transferring in from home after unsuccessful home birth. Excluded were patients with known placenta previa; third trimester bleeding; twins with twin #1 not vertex. In addition, patients with fixed malpresentation in whom external version was unsuccessful or declined or frank breech with patient preference for cesarean were referred to the hospital for cesarean birth. As noted above, during the 5 year substudy, only 17 patients (3.7%) were referred antepartum for these or other indications. All others were allowed a trial with clinic birth - and are therefore included in our data.

Competing interests:   None declared

The experience of Dr Longenecker in his work in west-central Ohio is similar to ours here in southwest Wisconsin. Most of the small hospitals in our region have also discontinued offering TOLAC after the restrictive policies advocated by ACOG in 1999. While our community hospital has continued to allow TOLAC, there is an atmosphere of fear and anxiety in younger staff out of proportion to the risks associated with TOLAC. The data regarding uterine rupture discussed in ACOG's 1999 report could have be used to give guidance to practitioners regarding factors associated with increased risk of uterine rupture rather than advocating policies which essentially eliminated TOLAC as an option except in urban centers. Mandating that surgeons, anesthesia, and surgical teams remain "in house" during TOLAC is unrealistic in the rural setting and sets up an adversarial relationship between physicians performing TOLAC deliveries and their colleagues. Our hospital has been able to continue to perform TOLAC deliveries by simply mandating that such staff have an appropriate response time. Dr Longenecker's comments regarding driving Amish patients desiring TOLAC into home birth due to lack of a hospital option are very appropriate. Hospital policies which do not allow TOLAC or the use of oxytocin with TOLAC put mothers and their babies at greater risk.

Properly managed, most patients with a single prior cesarean birth and a low transverse uterine incision can have a safe vaginal birth if desired. This is true even in the more challenging patients in whom the initial cesarean was performed for failure to progress in spite of an adequate trial of labor. Patients undergoing a primary VBAC are likely to have a labor pattern similar to a nulliparous patient. A prolonged first or second stage of labor should be anticipated and is not a cause for undue concern. Misoprostol for cervical ripening and labor induction should be avoided. Oxytocin should be used only when there is a strong indication and dosing for induction and augmentation should be conservative. Patients should be advised of the increased risk of uterine rupture with oxytocin and given the option to decline. Approached in this way, TOLAC is a safe and reasonable option for interested patients with prior cesarean delivery.

Competing interests:   None declared

Obesity is common in the Amish community of southwest Wisconsin. We did record the weight of our obstetrical patients. However as we often did not see the patients until late in pregnancy, our weight measurements are essentially "at term." For this reason, BMI calculations are not possible. 26% of our patients were greater than 90kg at term. The graph below displays weight at term in our population.

Competing interests:   None declared

After 30 years of clinical practice among the Amish of west-central Ohio, I applaud the authors in documenting their experience for the benefit of the rest of us and share their respect for the Amish people. Our local hospital does not "allow" TOL/VBAC, yet our Amish women continue to choose this. Most of them now birth at home, rather than face coercion to accept repeat Cesarean in the hospital setting. To me, this represents the worst of all options and flies in the face of good evidence. The best approach in my experience, and I believe is supported by evidence, is low- intervention care and a low primary cesarean rate (8-10% for our practice over the past 15 years; only 15% of our patients were Amish patients for whom it was even lower). The Amish, in discussing the risks of a trial of labor during their prenatal visits with us (both group and individual visits) consider the risks of both VBAC and repeat CS, based on an informed discussion of the evidence and a decision-making process unencumbered by threat of litigation. They take ownership of their decision, and when they choose other than what I would have chosen for them, they do so with autonomy in a relationship of trust and mutual respect.

The numbers, of course, will never be great enough nor the evidence strong enough for those who the Amish say "wouldn't believe it if it were true!"

Competing interests:   None declared

Thanks for your interest in our birthing work with the Amish. Our "birth center" is really just a modified exam room equiped with a basic birthing bed and infant warmer. It measures 17' X 11' (including a 5' X 7' bathroom - no shower or tub). See Figure 1 in the electronic version. If we have 2 moms in labor, the other uses a similar exam room next door (delivering on the exam table if we are unable to move her to the birthing bed). While the setting is basic, our team tries to make it as comfortable as possible (warm blankets; low level lighting when desired; one on one support during active labor). Many of our moms have IVs running or saline wells given high risk status (VBAC, Hx PPH, etc).

Generally moms come into the center when in active labor but this is variable. Occasionally a mom will renain for as long as 24 hours though this is a stress on staff who are limited in number. They generally stay two hours postpartum, occasionally longer if needed for stabilization.

Competing interests: None declared

Thank you for this very interesting and carefully presented analysis of the outcomes of a unique birth center. Our Cochrane Review of Alternative versus Conventional Institutional Settings for Birth suggests that the physical characteristics of the birth setting may also play important roles in labor and birth outcomes, by influencing the behavior and emotional responses of the women and their care providers. For example, when "home-like" is interpreted as "bedroom-like," the bed is the central feature, and laboring women may spend most of labor in bed. I'm curious about the physical characteristics of the birth center, as well as whether women usually arrived for birth center care in early or active labor.

Competing interests:   I am the first author on two related Cochrane Reviews, one of which is cited in the paper.

May I ask whether the authors have data on what percentage of the Amish women in their study were overweight or obese?

The results of this research are being compared to what's happening in the wider American population of pregnant women and their birth experiences - and in particular, media headlines have suggested that this research may offer clues about how to reduce the nation's cesarean rate.

It's important that any comparisons are made in the context of important obstetric risk factors (of which high maternal weight is one), and therefore it would be useful to understand the maternal weight characteristics of the Amish women in this study.

Thank you.

Competing interests:   Co-author of Choosing Cesarean, A Natural Birth Plan