The gentlebirth.org website is provided courtesy of
Ronnie Falcao, LM MS, a homebirth midwife in Mountain View, CA


Evidence Regarding Place of Birth

Easy Steps to a Safer Pregnancy - View e-book or Download PDF - FREE!
An interactive resource for moms on easy steps they can take to reduce exposure to chemical toxins during pregnancy.

Other excellent resources about avoiding toxins during pregnancy

These are easy to read and understand and are beautifully presented.

The following international committees maintain recommendations on evidence-based maternity care:

World Health Organization


World Health Organization's Summary of Research on Place of Birth

[excerpted from Care in Normal Birth: A Practical Guide Report of a Technical Working Group

General aspects of care in labour Part 1

2.4  Place of Birth

Does the place of birth have an impact on the course of labour and delivery?  This question has been abundantly researched in the past two decades (Campbell and Macfarlane, 1994).  When in many developed countries labour went from a natural process to a controlled procedure, the place of birth changed from home to hospital.  At the same time much of the human touch was taken out.  Pain was alleviated pharmacologically and women were left alone for long periods of time as they were in a light sleep anyway; they were monitored closely from afar.  This was the opposite end of the spectrum of those parts of the world where fewer than 20% of women have access to any type of formal birth facility.  For them, home birth is not an option, it is virtually inevitable, for reasons ranging from the economic to the cultural, and including the geographical (Mbizvo et al 1993, Onwudiego 1993, Smith 1993).  The call for a return to the natural process in many parts of the developed world opened up delivery rooms to fathers and to other family members, but the location stayed the same: the hospital.  Some hospitals have made an effort by installing a home-like birth room and this was found to increase maternal satisfaction and reduce the rate of perineal trauma, as well as reducing the desire for a different setting for the next birth, but randomised trials found no effect on the use of epidural analgesia, forceps delivery and caesarean section (Klein et al 1984, Chapman et al 1986).  These trials were primarily concerned with a more attractive labour ward setting without a fundamental change in care; apparently this is not enough to improve the quality of care and the obstetric outcome.

Other studies found that a woman with a low-risk delivery giving birth to her first child in a teaching hospital could be attended by as many as 16 people during 6 hours of labour and still be left alone for most of the time (Hodnett and Osborn 1989b).  Routine, though unfamiliar, procedures, the presence of strangers and being left alone during labour and/or delivery caused stress, and stress can interfere with the course of birth by prolonging it and setting off what has been described as a "cascade of intervention".

Home birth is a practice which is unevenly spread across the world.  With the widespread institutionalisation of childbirth since the 1930s the option of a home birth in most developed countries disappeared, even where it was not banned.  The system of obstetric care in the Netherlands, where still more than 30% of pregnant women deliver at home, is exceptional among developed countries (Van Alten et al 1989, Treffers et al 1990).  On the other hand, in many developing countries, great distances between women and the health facilities restrict options and make home birth the only choice.

Although risk assessment may be appropriately performed by trained birth attendants their advice about the place of birth, made on the basis of such assessment, is not always followed.  Many factors keep women away from higher level health facilities.  These include the cost of a hospital delivery, unfamiliar practices, inappropriate staff attitudes, restrictions with regard to the attendance of family members at the birth and the frequent need to obtain permission from other (usually  male) family members before seeking institutional care (Brieger et al 1994, Paolisso and Leslie 1995).  Often, high and very high risk women do not feel ill or show signs of disease, so they give birth at home, attended by a family member, by a neighbour or by a TBA (Kwast 1995a).  [TBA = Traditional Birth Attendant]

However, a properly attended home birth does require a few essential preparations.  The birth attendant must make sure that there is clean water at hand and that the room in which birth takes place is warm.  There is a need for careful  handwashing.  Warm cloths or towels must be ready to wrap the baby to keep it warm.  There must also be at least some form of clean delivery kit as recommended by WHO in order to create as clean a field as possible for birth and to give adequate treatment to the umbilical cord.  Furthermore, transport facilities to a referral centre must be available if needed.  In practical terms this means that community participation and revolving funds are necessary to enable transport to be arranged for emergencies in areas where transportation is a problem.

In some developed countries birth centres in and outside hospitals have been established where low-risk women can give birth in a home-like atmosphere, under primary care, usually attended by  midwives.  In most such centres electronic fetal monitoring and augmentation of labour are not used and there is a minimum use of analgesics.  An extensive report about birth centre care in the USA described care in alternative birth centres in and outside hospitals (Rooks et al 1989).  Experiments with midwife-managed care in hospitals in Britain, Australia and Sweden showed that women's satisfaction with such care was much higher than with standard care.  The number of interventions was generally lower, especially obstetric analgesia, induction and augmentation of labour.  The obstetric outcome did not significantly differ from consultant-led care, though in some trials perinatal mortality tended to be slightly higher in the midwife-led models of care (Flint et al 1989, MacVicar et al 19983, Waldenstrom and Nilsson 1993, Hundley et al 1994, Rowley et al 1995, Waldenstrom et al 1996).

In a number of developed countries dissatisfaction with hospital care led small groups of women and caregivers to the practice of home birth in an alternative setting, often more or less in confrontation with the official system of care.  Statistical data about these home births are scarce.  In an Australian study data were collected which suggested that the selection of low-risk pregnancies was only moderately successful.  In planned home deliveries the number of transfers to hospital and the rate of obstetric interventions was low.  Perinatal mortality and neonatal morbidity figures were also relatively low, but data about preventable factors were not provided (Bastian and Lancaster 1992).

The Netherlands is a developed country with an official home birth system.  The incidence of home deliveries differs considerably between regions, and even between large cities.  A study of perinatal mortality showed no correlation between regional hospitalisation at delivery and regional perinatal mortality (Treffers and Laan 1986).  A study conducted in the province of Gelderland, compared the "obstetric result" of home births and hospital births.   The result suggested that for primiparous women with a low-risk pregnancy a home birth was as safe as a hospital birth.  For low-risk multiparous women the result of a home birth was significantly better than the result of a hospital birth (Wiegers et al 1996).  There was no evidence that the system of care for pregnant women can be improved by increasing medicalization of birth (Buitendijk 1993).

In Nepal the decentralization approach of maternity care has been adapted to the special needs of urban areas in a developing country, where a hospital's capacity to deliver the specialist obstetric services needed by women with childbirth complications was being swamped by the sheer numbers of low-risk women experiencing normal birth - a common scenario in many countries.  The development of a "low-technology" birthing unit in the vicinity of the main hospital not only took the pressure off the specialist unit but made it much easier to deliver appropriate care to women in normal labour.  A similar, larger-scale project took place in Lusaka, Zambia, where a University teaching hospital, serving as a specialist referral centre for the entire country, was overcrowded by large numbers of low-risk pregnant women.  The extension of the capacity of the peripheral delivery centres and the opening of new centres for low-risk births reduced the number of deliveries in the hospital from around 22,000 to around 12,000 and at the same time the total number of births in the dozen satellite clinics rose from just over 2000 in 1982 to 15,298 in 1988.  The care of high-risk women in the hospital was improved by the reduction in numbers of low-risk women, while in the peripheral units time ws available to ensure that the low-risk women received the care and attention they needed (Nasah and Tyndall 1994).

So where then should a woman give birth?  It is safe to say that a woman should give birth in a place she feels is safe, and at the most peripheral level at which appropriate care is feasible and safe (FIGO 1992).  For a low-risk pregnant woman this can be at home, at a small maternity clinic or birth centre in town or perhaps at the maternity unit of a larger hospital.  However, it must be a place where all the attention and care are focused on her needs and safety, as close to home and her own culture as possible.  if birth does take place at home or in a small peripheral birth centre, contingency plans for access to a properly-staffed referral centre should form part of the antenatal preparations.


Most doctors and many other health professionals strongly believe that hospital births are safer than home births.  This opinion, which is shared by many childbearing women, may in part stem from the poor perinatal outcomes of unplanned, precipitate home births, which include a high proportion of preterm and low-birthweight babies.  These unfortunate statistics do not, however, apply to planned homebirth for eligible women attended by caregivers experienced in home birth, backed up by a modern hospital system.

Several methodologically sound observational studies have compared the outcomes of planned home-births (irrespective of the eventual place of birth) with planned hospital-births for women with similar characteristics.  A meta-analysis of these studies showed no maternal mortality, and no statistically significant differences in perinatal mortality between the groups.  The number of births included in the studies was sufficiently large to rule out any major difference in perinatal mortality risk in either direction.  Significantly fewer medical interventions occurred in the home-birth groups (including women transferred to hospital), and there were significantly fewer low Apgar scores, neonatal respiratory problems, and instances of birth trauma among the babies.

Only one small randomized trial, involving 11 women, has been mounted to compare home with hospital birth.  This was done more to demonstrate the feasibility of randomizing women to home or hospital than concern about outcomes.  The majority of the women in the hospital group were disappointed by the allocation.  This finding was not surprising.  Choosing a home (or hospital) birth is a very individual and personal choices for a woman based on her own priorities and values.

Maternal and perinatal mortality are so low in low-risk pregnancies that these cannot be the primary outcome measures for a trial.  Yet they are the outcomes of real interest and the source of the polarized concerns.  A study looking at issues of less importance would provide data that are relevant to those who wish to make a choice based on considerations of safety.

Women who have no factors that contra-indicate a home birth, and who prefer a planned, attended home-birth with facilities for prompt transfer to hospital if necessary, should not be advised against this.  [Ed. - This wording is interesting; it addresses the serious problem that many hospital-based practitioners intentionally remain ignorant or misrepresent their knowledge about homebirth safety when asked.  They thus violate generally accepted informed consent guidelines by not presenting information about reasonable options.]

Home versus hospital birth (Cochrane Review) - the online abstract

Search other topics from the The Cochrane Library Issue 1, 2002

This Web page is referenced from other pages containing related information about Homebirth Safety/Advocacy and For Parents - How to Get the Best Care


SEARCH gentlebirth.org

Main Index Page of the Midwife Archives

Main page of gentlebirth.org         Mirror site

Please e-mail feedback about errors of fact, spelling, grammar or semantics. Thank you.

Permission to link to this page is hereby granted.
About the Midwife Archives / Midwife Archives Disclaimer