A multi-part series originally posted on HealthCare_2.0 (to read entire document)
Part 4 ~:~ Safety & the maternity-care continuum in an essentially healthy population
Two kinds of data are required to intelligently determine if the current configuration of medical and midwifery care is making the kind of cost-effective contribution that economists refer to as ‘value added’. In order to identify “value-added”, we must first identify the innate riskiness of normal childbearing and directly quantify the relative benefits of each category of maternity care. This requires baseline data for “no care” in an essentially healthy US population, as well as outcome statistics for each type of birth attendant. Having done that, the essential qualities of maternity care can be distinguished from customary practices not directly associated with better outcomes for mothers and babies.
The ultimate goal is to test the validity of all current maternity care policies, protocols and practices in regard to safety, cost-effectiveness and patient satisfaction and to arrive at ‘standard care’ characteristics. This would integrate the principles of physiological management with best advances in obstetrical medicine to create a single, evidence-based standard for all healthy women with normal pregnancies, with obstetric interventions reserved for those with complications or if requested by the mother. Having scientifically identified ‘standard care’ characteristics, this model of ‘best practices’ would apply to all birth settings and be used universally by all categories of birth attendants when providing care to healthy women.
Childbirth risks in healthy women & policies that reduce them
In searching for the essential qualities of safe and cost-effective maternity care, I have identified 5 useful sources – 4 published studies in combination with a consensus of the research literature for hospital-based maternity services in the US. These included:
(1) a contemporary study of purposefully unattended births & rejection of necessary emergent care
(2) a retrospective study of lay midwife-attended OOH w/ access to comprehensive medical service
(3) a prospective study of professional midwife-attended OOH w/ access to comprehensive medical services
(4) a prospective Canadian study that contrasted outcomes for a select groups of low-risk women that gave birth under three different circumstances (a) OOH setting under the care of direct-entry midwives, (b) in hospital under the same category of midwives and (c) in hospital by MDs
(5) a configuration of hospital-based obstetrical studies of medically-managed hospital births in low and moderate risk women by obstetricians, FP physicians and certified nurse- midwives.
CONTROL GROUP: The study of unattended birth functions as a ‘control group’ that allows comparison of ‘care’ versus ‘no care’. Then we can compare the 3 major groups of birth attendants to one another and to the ‘no care’ cohort.
Taken together these studies provide information on the biological background rate of maternal and perinatal mortality and morbidity when all the benefits of modern biological sciences are absent, inaccessible or rejected by a childbearing population due to cultural traditions or religious beliefs.
The poor outcomes for purposefully unattended birth in the US in an educated and essentially healthy population are consistent with available statistics for maternal-infant mortality in the late 19th and early 20th century in the US, and the current high maternal mortality rates in developing countries such as Afghanistan, Ethiopia and the Gambia. The major cause of MM in this first-world cohort was hemorrhage and infection and establishes a background rate of biological risk that is independent from poverty, malnutrition and other factors specific to deprivation.
CONVENTIONAL OBSTETRICS: At the other end of the scientific continuum, this collection of studies also helps us to distinguish between maternity care policies and practices that benefit healthy women and those non-productive traditions, customs and provider preferences that increase the economic cost, but do not directly contribute to improved maternal-infant outcomes.
Five Models, Five Perspectives, Five Insights
Study #1 Perinatal & maternal mortality in a religious group avoiding obstetric care – Am Jour Obst Gyne 1984 Dec 1: 150(7):926-31:
This control group consists of women with the same general health and demographic characteristics that are seen in the CDC birth registration data. This is predominately healthy, white, middle-class women who had economic access to all categories of maternity care providers and settings, but in this case, purposefully choose unattended births. Data on this group of unattended home births came from Indiana state mortality statistics for a fundamentalist religious group that rejected all forms of medical care under all circumstances – no prior diagnosis or treatment of chronic medical problems, no risk-screening of mothers during pregnancy, no prenatal care, no trained attendant during childbirth and no emergency transfer of mother or baby with life-threatening complications to a medical facility – a situation similar to rural parts of the developing world.
Out of 344 births, the unattended group had 6 maternal deaths and 21 perinatal losses. The baseline mortality rate for unattended childbirth was one maternal death per 57 mothers or MMR of 872 per 100,000 live births (92 times higher than Indiana’s MMR for the same period) and one perinatal loss for every 16 births or PNM rate of approximately 45 per 1,000.
Study #2: “Home Delivery & Neonatal Mortality in North Carolina”, Claude Burnett, Judith Rooks; JAMA, Dec 19, 1980, Vol. 244, No. 24, p. 2741-2745:
Planned home birth (PHB) in an impoverished and medically-indigent minority population attended by experienced lay midwives. These demographically high-risk maternity patients were risk-screened one time by a public health officer prior to being approved for PHB under the care of a lay midwife. However, state laws did not authorized non-nurse midwives to carry oxygen or emergency anti-hemorrhagic drugs (Pitocin) or to suture perineal tears. These county-registered midwives were required to transfer patients with complications to a local hospital in an appropriate and timely manner. The lay midwife-attended group had no maternal deaths and 4 neonatal losses per 1,000 (including 2 fatal birth defects).
Note: This study also reported the perinatal mortality rate for medically indigent women in the same rural regions of North Carolina who delivered unattended, often because local hospitals turned away laboring women who did not have the prescribed ‘cash in hand’. These unattended births had a dramatically increased perinatal mortality rate ranging from 30 to 120 stillbirth and neonatal deaths per 1,000, a perinatal mortality rate consistent with 3rd world countries and unattended births among the religious group in Indian.
This highlights the preventive value of physiologically-based pregnancy and childbirth services and the equally important access to medicalized maternity care during pregnancy as indicated and the ability to call on comprehensive medical services during the intrapartum and immediate postpartum-neonatal period whenever necessary. Compared to the combined mortality statistics for the control group, the care of these lay midwives saved the lives of 14 mothers and 58 babies. If their care were a drug or medical device, it would be illegal for every childbearing women not to have one of them.
Study #3: Outcomes of planned home births with certified professional midwives: large prospective study in North America; Kenneth C Johnson, senior epidemiologist; BMJ 2005;330:1416 (18 June 2005)
Planned home birth (PHB) in a generally healthy population as attended by nationally-certified direct-entry (non-nurse) midwives in the year 2000. All clients were risked-screened and received prenatal care and those with medical or pregnancy complications were referred to medical services. Professional midwives monitored maternal vital signs and fetal heart tones during labor and were authorized to carry emergency supplies such oxytocin (Pitocin + Methergine), IV fluids, oxygen, neonatal resuscitation equipment and also to suture perineal lacerations. Twelve percent of PHB patients were transferred to the hospital during labor or after birth, the majority of whom were first-time mothers. Cesarean rate was < 4% for PHB women hospitalized during labor. This group had no maternal deaths and 2.6 neonatal losses per 1,000 (including lethal birth defects).
Study #4: Outcomes of planned home birth with midwives versus planned hospital birth with midwife or physician; Janssen PA, Saxell L, Page LA, et al. CMAJ 2009;181:377-383:
A 5-year Canadian prospective study published in 2009 compared the outcomes of PHB in British Columbia attended by professional direct-entry midwives btw 2000 and 2004. It compared planned hospital births also attended by this same category of professional midwives and a matched low-risk cohort of physician-attended hospital births. They found that the risk of perinatal death associated with PHB attended by midwives did not differ significantly from the low rate associated with planned hospital birth. The study also found that women who planned a home birth had a reduced number of obstetric interventions and adverse maternal outcomes.
The neonatal death rates per 1,000 births were 0.35 for midwife-attended planned OOH birth, 0.57 for midwife-attended hospital births, and 0.64 for physician-attended hospital births. Maternal mortality for all three groups was zero. Inclusion in the two hospital categories required the childbearing women to have the same low risk-based characteristics as those who were planning to labor at home. These finding echoed a Dutch study published in July that also found a planned home birth to be as safe as a planned hospital birth, provided that a well-trained midwife is available, transportation and medical referral system is in place, and the mother is at low risk of developing any complications.
The authors concluded: “… (the) study showed that planned home birth attended by a registered midwife was associated with very low and comparable rates of perinatal death and reduced rates of obstetric interventions and adverse maternal outcomes compared with planned hospital birth attended by a midwife or physician”.
#5 Neonatal mortality rates for planned hospital birth as reflected in a consensus of scientific literature, plusCDC birth registration stats for birth after 37 completed wks and data on obstetrical intervention levels in general population from the “Listening To Mothers” survey, Childbirth Connection; 2002 and 2006:
Planned hospital services for low and moderate risk women — labor attended by a professional nursing staff, routine intrapartum use of continuous electronic fetal monitoring (93%), IVs (86%) and epidurals (63%); birth conducted as a surgical procedure by a physician or certified nurse midwife. Medical intervention rate for this group was 99%; aggregate surgical intervention rate was 70% (episiotomy, forceps, vacuum extraction and Cesarean section). The CS rate was approximately 25% in 2002 (now 32%). The scientific literature reported neonatal mortality for obstetrically-managed hospital birth for low-risk women to range from a low of 0.79 to 4.1, with an average NNM rate of 1.5 per 1,000.
The patient-choice issue of genetic testing & termination
The routine use of ultrasound and prenatal genetic screening in the hospital cohort, in conjunction with termination of affected pregnancies during the pre-viable state, slightly lowers the rate of perinatal and neonatal mortality when compared to the sub-set of families who choose OOH birth. This is due to a reduced number of babies in the hospital cohort with lethal anomalies who are carried to term.
Families that choose non-medical maternity care are statistically less likely to utilize prenatal genetic and ultrasound screening or to terminate affected pregnancies when indicated. One study in PHB in Washington State (1996) documented a disproportionate increase NNM due to congenital anomalies, not all of which were incompatible with life. Among this specific sub-set of non-testing parents, prenatal diagnosis and planned hospital care would have reduced (but not eliminated) the incidence of neonatal mortality. However, this is a patient choice and is not a provider or place-of-birth issue.
In regard to the great debate about safety, it is useful to realize that birth-related morbidity and mortality can be time-shifted, place-shifted and practitioner-shifted, but they cannot be eliminated. In other words, increasing rates of pregnancy termination reduces neonatal mortality rates but obviously does not reduce over all perinatal mortality. There is nothing that birth attendants can do or not do that reliably, and with economically sustainability, can create a condition of zero risk for both mother and baby 100% of the time.
Continue to part 5: The risk-benefit continuum among the 4 responses to normal childbirth and the 3 types of birth attendants
{ 0 comments… add one now }