Why Homebirth is 1,000 Times Safer Than Hospital Birth for Low Risk US Women

Oft quoted research studies state 3X to 10X more babies die in the first week after low risk homebirth than hospital birth. In order for low risk homebirth to have higher perinatal mortality rates there would have to be a theory to explain this. There would have to be one or more complications of low risk homebirths that result in death in the first week that can be prevented by being in hospital, and death from these complications would have to occur more often than low risk deaths at planned hospital births.

The problem is... there is no explanation for why low risk newborns would have a higher rate of death in the first week after Planned Attended Homebirth than after Planned Hospital birth. All the studies that make this conclusion pretend to count low risk perinatal deaths but actually count other things. The rate of preventable perinatal deaths is no more than 1 per 10,000 low risk births for both home and hospital low risk births. Cord Prolapse and attendants not expert in resuscitation are the only complications that cause preventable low risk homebirth deaths in the first week. Some of these deaths might have been prevented by the woman planning a hospital birth or hiring a better trained midwife. These deaths are balanced by deaths in hospital, for examples: caused by the doctor not being present to do resuscitation, nurse busy doing paperwork and didn't notice baby stopped breathing, and hospital-caused: infections, cord prolapses, placenta abruptions, uterine ruptures, and Amniotic Fluid Embolisms (AFE) at low risk births. These hospital-caused deaths of the newborn in the first week likely could have been prevented by the woman having a homebirth with a well-trained attendant. Although it appears that the preventable newborn deaths at home and hospital birth balance out, homebirth is clearly safer when you take into consideration the risk of maternal death that 20% of low risk U.S. women face as a result of avoidable cesareans which became necessary because they went to hospital. When this 20% risk of death is compared to the 0.02% rate of cord prolapse during labor at homebirth that might have a better outcome if it happened in hospital, this means that a low risk woman has a 1000 times higher chance of having a life threatening complication either to her life or her fetus/newborns life at planned hospital birth, than if she plans to have an attended homebirth with a well-trained practitioner. High risk births such as premature, breech and twins ARE at high risk of perinatal death and overall have superior outcomes when delivered in hospital. About 10% of breech delivered at homebirths suffer damage or death.

An In Depth Explanation

Most low risk babies that are stillborn or die at birth, die as a result of congenital defects incompatible with life or unexplained stillbirths and would die no matter where the birth takes place. The rate of unexplained stillbirth at term for low risk pregnancy is 1/1000 (Smith Lancet 2003). No more than 1 in 10,000 low risk babies, whose mothers are healthy, non-smokers, normal blood pressure, with adequate protein and micronutrients in their diet, die in the first week from plausibly preventable causes at home or hospital. While this has never been directly published, it can be extrapolated from Pasupathy JAMA 2009, which reported 432 deaths due to intrapartum anoxia in the first FOUR weeks among 1,012,266 (or 4 deaths/9372 births ) term, singleton, head down births including all types of high risk pregnancy known to have high perinatal mortality rates such as Type 1&2 diabetics, gestational diabetics, cholestasis, toxemia, preeclampsia, IUGR, drug addicts, alcoholics, heavy smokers, etc. Logically at least 3 out of 4 of those deaths occurred in women who were unsuitable for homebirth or occurred after the first week, which would amount to a maximum of 1 in 10,000 preventable low risk deaths in the first week after birth. A 3/10,000 intrapartum death rate for same criteria- head down, term, singleton including all high risk was found in Ireland. (Walsh. 2008. AJOG)

The three recent papers published in American Journal of ObGyn: Wax metaanalysis (2010), Chervenak (2013), Grunebaum****(see note at bottom) (Apgar 0, 2013) and the U.K. Birth Place study (2013) report perinatal death rates from homebirth as 3 times or 10 times higher than perinatal death rates in the first week than hospital birth. All of these studies profess to count how many babies supposedly die at planned attended low risk homebirth, but none of them do. Wax et al includes unattended, unplanned, preterm, breech, hypertensive, gestational diabetics and other high risk homebirths. Chervenak quotes the UK Daily Mail newspaper as his source of data. Grunebaum et al count the fetuses who died before labor started (antepartum stillbirths). The U.K. BirthPlace study admits that perinatal death is too rare to count so counts transient events of no relevance to long term outcomes. None of these studies suggest a theory to explain their findings. It is possible the bias of these authors originate from never having attended a homebirth and extrapolating from the horrendous emergencies that happen at hospital births, thinking that they also happen at homebirths, when they don't.

Preventable causes of death at low risk birth attended by a trained practitioner:

Shoulder Dystocia : Big shoulders getting stuck after the head is out occurs in 1/200 births and is the most life threatening event of low risk birth. In the event of asphyxia, the newborn almost never dies in the first week of life so does not contribute to death rates in the first week. No death rate from shoulder dystocia has ever been published, possibly partly because it is often months after the birth. 

Cord prolapse occurs at about 1/400 low risk hospital births and about 1/5,000 (0.02%) homebirths and only where rupturing membranes is not restricted. Cord prolapse definitely has better outcomes when it happens in hospital but when it occurs during labor it is usually caused by the routine of breaking the water. Artificially rupturing membranes is routine at hospital birth. Hospital caused cord prolapse has nephariously rarely been documented in the literature, but a recent paper documents 23 hospital-caused cord prolapses among 33,000 intended vaginal deliveries (Gabbay-Benziv et al J Matern Fetal Neonatal Med. 2013). About 1 life threatening cord prolapse occurs as a result of every 300 artificial rupture of membranes in low risk women. (Cohain J Matern Fetal Neonatal Med.2013) This can be eliminated wherever birth takes place by severely restricting breaking the sac as well as vaginal exams.

Amniotic fluid embolism (AFE) appears in 70% of cases to be caused by/associated with a combination of artificial rupture of membranes and zealous induction. The increasing rate of inductions and cesareans may account for all or most AFE in the USA. What else would explain the increase in AFE from 1/120,000 during the years 1950 to 1990 (Clark SL Am J Obstet Gynecol 1995) to 1/13,000 for the years 2000-2008 (Abenhaim HA. Am J Obstet Gynecol 2008)? It is unknown whether it has ever happened at low risk homebirth with a well-trained attendant present.

Hospital Malpractice is poorly documented due to liability. An educated guess would be that in 1/50,000 live births it is responsible for perinatal death. Examples I personally can document: a plague of deadly bacteria in the newborn nursery killing 3 previously healthy newborns, exploding the fetal lungs with too forceful dose of oxygen after birth by inexperienced doctor, crushing the skull during forceps extraction, overdose of adrenalin to newborn by a nurse, slow paging or slow response to call to resuscitate newborn, exploding the uterus(uterine rupture) and/or placental abruption as a result of high IV dose of oxytocin in labor in a low risk women.

Home birth Malpractice- attendant unable to properly perform needed resuscitation. Can be eliminated with proper training, which could be taught on YouTube.

Deaths in low risk pregnancy as a result of hospital routine: Inductions with prostaglandin and Pitocin, epidurals, rupturing membranes, forceps and anesthesia have been documented to cause rare but serious complications including death or near death of the fetus. These deaths are completely preventable by restricting the frequent use of hospital interventions that cause them: inductions and augmentations (currently 50% of low risk births), forceps & vacuum (5% of low risk births), rupturing membranes (85% of low risk births), epidurals (50% of low risk births), frequent vaginal exams (98% of low risk births), general anesthesia at cesareans (5% of low risk births). Today 1/3000 low risk (Landon 2004 NEJM) women undergoing cesareans die from complications of anesthesia, puerperal infection, and venous thromboembolism (Deneux-Tharaux 2006 OBGYN). Epidurals are well known to dramatically lower the mother's blood pressure which can put the fetus into stress. In an attempt to prevent this, the woman receives a liter of IV fluids before administration. Inductions cause unnaturally strong contractions that cut off oxygen to the fetus. For this reason, the woman is put on a monitor so oxytocin can be shut off if the fetal heart drops. The largest studies of a million term births failed to conclusively show that high cesarean rates are saving the lives of term babies (Pasupathy JAMA 2009). Nutrition and exercise are still the most important keys to good pregnancy outcomes. The key to good birth outcomes among low risk women is patience and a motivated well trained attendant. If there is a good connection between the mom and the birth attendant, when the midwife tells the mother to get into hands knees on the floor and push real hard to get the shoulders out, she does so quickly. Or if the mother is taking a long time to birth, the midwife is committed to waiting patiently with her until she does, avoiding unnecessary artificial augmentation of the labor or an unnecessary cesarean. In hospitals, patients are frequently referred to by their room number, for example "room number 5". There is financial motivation to process women out of the labor and delivery ward.

Infection: THE EASIEST WAY TO GET AN INFECTION IS HOSPITALIZATION. HOSPITALS ARE THE GERM RESERVOIR OF THE WORLD. If it were financially possible to do open heart surgery at home, it would obviously be safer for the patient to avoid the reservoir of germs in hospital. Infection may be dealt with better in hospital, but infection in low risk newborns might actually be largely a product of hospital birth. As with adults, infection occurs in stressed or vulnerable newborns- largely premature and high risk babies. There is no research on the rate of infection at low risk planned attended homebirths, but in theory homebirth prevents infection and hospital birth increases the risk of infection. One way to prevent newborn infection is to eliminate artificial rupture of membranes. This results in 5% of babies being born in the sac, protected from all bacteria until after birth, as in the picture above, and 85% having ruptured membranes minutes before the birth.

The CDC begins all its report on Group B Streptococcus (GBS) proclaiming that "GBS is the most common infectious cause of neonatal mortality."(https://www.cdc.gov/mmwr/PDF/RR/RR5111.pdf). A more balanced statement would be: GBS was part of the normal harmless intestinal flora present in about 25% of full term pregnancies for centuries only being documented as virulent and deadly after hospital birth became common in the 1940s. GBS may be the most common infectious cause of neonatal mortality but it is extremely rare in low risk birth. A total of 80 deaths due to GBS in premature and full term newborns occur per year among the 4 million low and high risk births in the US or 1/50,000. It has recently been shown that Streptococcus turns from harmless to deadly only when the person releases stress hormones like norepinephrine. (Marks et al MBio. 2013 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3735180/) Most women justify going to hospital to give birth because they fear a bad outcome at birth. This fear may be the factor at hospital birth that causes GBS to turn from harmless to virulent and deadly. Fetal distress complicates about 30% of hospital low risk births and may be related to this fear as well. Hospitals and drug companies/CDC feed into this fear with carefully sculpted and fear-promoting statements like: "GBS is the most common infectious cause of mortality." 

Today 30% of laboring women in the US receive IV antibiotics. It took about 40 years for MRSA to evolve anti-biotic resistance to Penicillin (1945-1985). Giving millions of women and newborns Prophylactic Penicillin is short sighted. Already strains of GBS are resistant to erythromycin and clindamycin. The protocol to give a million women prophylactic antibiotics in labor began in late 1990s. By 2040, many strains of GBS will most likely be resistant to Penicillin. As GBS deaths go down, babies dying of other bacteria increases, such as antibiotic resistant Clostridium difficile. Therefore, the total infection rate of low risk HOSPITAL births, which has always been very low, is not likely to decrease overall. In addition, since GBS is normally present in about 25% of term pregnant women, when a baby dies for unknown reasons, and a culture is done, GBS is found in 25% of autopsy cultures and is sometimes incorrectly assigned to be cause of death. Cord problems are also another convenient scapegoat for unknown causes of death or malpractice. 35% of full term babies have wrapped the cord around themselves and 1% have a knot in the cord.  When a newborn dies and the cord is around the neck or there is a true knot, the cord is often blamed for the death, when it might not have been the cause of death.

The summary of my findings appear in a 5 minute YouTube video entitled "Home birth is 1000 times safer than Hospital birth for low risk women".

A second version, more geared for doctors is also available: https://www.youtube.com/watch?v=O-OqnqfHQ2Q

**** Dr A. Grunebaum ([email protected]) did not answer my emails requesting a public debate. In his latest Letter to the Editor (Grünebaum A, Chervenak FA, McCullough LB. Am J Obstet Gynecol. 2013 Nov 7) he was reduced to Amy Tuteur-style name- calling seemingly in lieu of admitting to the lack of science or math behind his Apgar 0 paper. More on this in a future post!