Home birth risk remains unclear

Wednesday April 2 2008

The risks of home births compared with hospital births are addressed in today’s newspapers. The Guardian says that “home births are generally considered safe” but that there is “a significant rise in death rate of babies when mothers have to be transferred to hospital”. The Daily Telegraph also reports this study. It says: “Women who opt for a home birth are more likely to lose their baby than those who have them in a hospital.”

The reports are based on a large UK study which has calculated national death rates following home births over a 10-year period. The study found that home births were generally safe and not associated with an increased risk of death. However, it also revealed that the risk of mortality increased if the mother required emergency transfer to hospital due to complications.

This study is one of the first to attempt to quantify the risks associated with home births. Home births currently make up a small proportion of births in the UK but are increasing in popularity. However, the conclusions that can be drawn from the study are limited as there were gaps in the data. In particular, the definition of “home births transferred to hospital” included not just those that occurred during labour as a result of complications, but also those who transferred during the pregnancy (which may have been due to personal choice). For home births, transfer could be a proxy indicator of complications, and as such it is unsurprising that intended home births that transferred had higher risks. In hospital, pregnancy associated with complications probably leads to higher-risk births too.

Further research and improved data collection is required to clarify the issue. It would be better to compare women that have complications at home with others that have the same complication in hospital. For now, expectant parents should be fully supported and informed so that they can make the correct decision for themselves on where they would like their baby to be born.

Where did the story come from?

Rintaro Mori and colleagues at the Osaka Medical Center and Research Institute for Maternal and Child Health, Japan, carried out this research. The study was funded by the National Institute for Health and Clinical Excellence. It was published in the British Journal Obstetrics and Gynaecology , a peer-reviewed medical journal.

What kind of scientific study was this?

This was a cross-sectional study. It aimed to estimate the rate of death of babies in the time around labour and birth (the intra-partum related perinatal mortality rate, or IPPM) for booked home births in England and Wales.

The researchers used the Confidential Enquiry into Maternal and Child Health (CEMACH) to examine the outcomes of all women who gave birth at home, intentionally or not, between 1994 and 2003. The CEMACH collects data on death rates and records whether women had booked hospital or home for delivery. The IPPM rate included all still births or deaths within the first week of life from asphyxia, lack of oxygen or trauma. The researchers looked at actual home births (those births that were booked to and did take place at home, and those that occurred at home unintentionally) and at booked home births (which may not have been actual home births if women either chose later to move to hospital or were transferred for emergency reasons). Within these two groups, the researchers also looked at whether there were differences in IPPM rate between women who chose to have a home birth and did, and those who had an unintended home birth.

Some of the information they needed was available through national data sets (such as the Office for National Statistics and CEMACH). However, data on how many home births were unintended and how many intended home births transfer to a hospital booking were collected through a systematic review where researchers pooled the results of studies that had previously considered these measures.

What were the results of the study?

Between 1994 and 2003, 4,991 infant deaths occurred out of a total of 6,314,315 births in England and Wales (0.08%). IPPM tended to decrease with time. Among the 130,700 actual home births (which includes intended and unintended), there were 120 infant deaths (0.09%).

The researchers used two ways to determine the unintended home birth rate, which gave widely differing figures, ranging from 0.31% to 56%. Their systematic review suggested that the transfer rate of births that had originally been planned to take place at home was an average of 14.3%. The researchers used the unintended home birth rate and the transfer rate to calculate the IPPM rate. They found that in women who intended to have a home birth and did, IPPM rates were either 0.48/1000 or 0.28/1000, depending on which value of “unintended” birth rate they used (both results are lower than the overall IPPM rate of 0.79/1000).

Women in the “transferred group” (i.e. those who had intended to have a home birth but then transferred to hospital for whatever reason) tended towards a higher IPPM rate, either 6.05/1000 or 3.53/1000. There was also a higher IPPM rate in women who had not intended to have home births but did (either 1.42/1000 or 4.65/1000).

What interpretations did the researchers draw from these results?

The authors conclude that the findings from their study “need to be interpreted with caution due to inconsistencies occurring in the recorded data”. However, they note that infant mortality rates around the time of home birth do not appear to have improved much over the study period, even though overall rates did. They also note that the mortality rate for babies delivered at home appeared to be low, while rates were higher for women who transferred to hospital.

What does the NHS Knowledge Service make of this study?

This is a large study which attempted to quantify the risks associated with home births. The majority of births take place in hospital, but home births are increasing in popularity, so their safety is paramount. However, the authors openly acknowledge that this study has important limitations because of the data available for analysis.

  • This study does not provide any evidence that there is any more risk of infant death associated with home birth than with hospital birth when it considers women who elect for a home birth and actually have one. In fact, the mortality rate in these women was lower than the overall rate.
  • The highest risk was found for transferred birth that was originally planned to take place at home. There is no information available on the reasons for the transfers, but a higher rate is unsurprising if transfer occurs because of an emergency. “Transfer to hospital” could be a proxy indicator of complications during a home birth. 
  • No information was available on the numerous factors that may have a significant effect on infant mortality rates, such as the mother’s medical and obstetric history, lifestyle, ethnicity and socioeconomic status.
  • The transfer rates and unintentional home births were obtained from a selection of regional studies. It is not possible to comment on the accuracy of these studies or the methods or definitions that they used, which may have been different. It would also be helpful to look at demographic data and characteristics of the women in the combined studies as these factors may influence death rates. 
  • Due to reliance upon national databases, there may be errors introduced from miscoding of the different study variables that were investigated.

Further research and improved data collection will be needed to clarify the safety of home births. For now, expectant parents should be fully supported and informed so that they can make the correct decision for themselves about where they would like their baby to be born.

Sir Muir Gray adds...

As is so often the case in medicine the key question is not ‘is treatment A better than treatment B?’ but ‘which people do best with A and which with B?’ and ‘how can we best distinguish the two groups?’.

Analysis by Bazian
Edited by NHS Choices