Sunday, January 17, 2010

Homebirth statistics

The Medical Journal of Australia will release a paper tomorrow claiming increased death rates from homebirths in South Australia over a 16-year period (1991-2006).
[Link to the Editorial, and the Paper]

Midwives and homebirth activists have reviewed the paper titled Planned home and hospital births in South Australia, 1991-2006: differences in outcomes (Authors Kennare RM, Keirse MJNC, Tucker GR and Chan AC).

Although this paper is presented in the AMJ as research, it is just a report on what has actually happened. It is not a research study, and should not be held out as such. There is no ability to control for biases. This report has no statistical power, and any conclusions drawn from it are no more than opinion.

South Australia's high profile midwife Lisa Barrett has provided comments at her blog.

Click here for Homebirth Australia's comments; here for further informed discussion.


The following comments and response have been attributed to NSW midwives Hannah Dahlen and Caroline Homer. (Source midwives yahoo group)

Summary of key issues regarding the MJA homebirth paper


• There is no way to tell if these planned homebirths were under the care of a registered midwife between 1991-2006 (sixteen years). This is also a retrospective population based study (low level evidence)
• One of the problems is that the planned home birth group includes women who planned homebirth at booking but then developed risk factors and had their babies in hospital. There are probably only two women whose babies died who started labour at home planning a homebirth and one of these was a twin pregnancy (high risk). The others had all transferred before the onset of labour. The authors admit they ‘could not differentiate all planned homebirths according to whether transfer to hospital had occurred before or during labour.’ So for low risk women who start labour at home the risk is very low - 1 death in 16 years.
• There was a high rate of post-term pregnancy (3.8% vs 1.2%); twins (five sets of twins); VBAC 8.8% - so not a low risk population.
• Significantly less intervention in homebirth group: C/S 9.2% vs 27.1(one third); instrumental birth 4.4% vs12.8% (one third); Episiotomy 3.6% vs 21.7% (one sixth). More than three times the rate of intact perineum in homebirth group (seven years missing data on episiotomy and perineal injury).
• No difference in major maternal morbidity measures of severe perineal trauma 1% vs 1.8% and postpartum haemorrhage 5.5% vs 4.4%
• The numbers of planned homebirths are small (1141) (birthed at home n=792; in hospital n=349). You cannot look at the rare outcome of intrapartum death in such a small sample (CI 1.53-35.87)(there is 1 intrapartum death at home and 1 in hospital). You also can’t look at intrapartum asphyxia due to the low numbers (CI -8.02-88.83) (1 at home and 2 in hospital). You would need about 10,000 births at home to show this. Also they don‚t say how they define this and bias is possible. The authors state the ‘small numbers with large confidence intervals limit the interpretation of these data.’
• There is no difference in perinatal mortality (stillbirths and neonatal deaths within 28 days of birth) between home and hospital (7.9 vs 8.2/1000). Perhaps the authors were surprised by this and looked further trying to find something. For those born at home the perinatal mortality rate is 2.5/1000.
• The numbers of perinatal deaths are also small (9 deaths). There were two deaths actually occurring at home and 7 in hospital. Of these, only 3 are related to perinatal asphyxia.
• The deaths in hospital were: (1) one had lethal congenital abnormalities (known beforehand and a decision made to be born at home), (2) 1 had hydrops and (born in hospital), (3) 1 was unexplained with a cord entanglement (born in hospital), (4) 1 had pulmonary hypoplasia after a early rupture of membranes and (born in hospital), (5) 1 was growth restricted with an abnormal karotype (born in hospital) and (6) 1 'seriously' postdates, induction in hospital without fetal monitoring (the woman refused) and eventuated in a stillbirth and (7) one was a woman with known haematological risk factors whose baby had a lethal abnormality.
• Of the 2 other deaths at home (8) one was at home after a waterbirth which was not found to be the cause of death but increased monitoring may have identified the baby was in distress and (9) one was a second twin (first twin born at home and second twin born after a delay in transfer). Three therefore are potentially preventable and related to the model of care:1 - waterbirth at home; 2 - second twin; 3 - postdates.
• Therefore 3 deaths in 16 years - two of which had risk factors.
• Therefore, 1 death in 16 years where there were no risk factors.
• There were no differences in Apgar scores or NICU admissions but infants born at home were half as likely to receive specialised neonatal care compared to planned hospital birth
• The paper also highlights that the system must be so terrible for some women that they would choose to give birth outside of it than in it even with risk factors. Therefore, this is an indictment on the current system - that needs fixing - not removing homebirth.
• The conclusion of the paper is actually very sensible about risk assessment, transfer and fetal monitoring.

Responses
• Despite a malfunctioning system where midwives are uninsured and have no visiting rights the perinatal mortality rate is no different. This is remarkable.
• The intervention rates are to be commended
• Risk assessment, transfer and fetal monitoring will be improved when private midwives are no longer excluded from mainstream services so we should be aiming for this not continuing the witch hunt against private midwives.
• Some women will always choose homebirth so we should support this choice with safe responsive systems of care. The authors state that ‘women's autonomy in choosing reproductive behaviour is a fundamental human right enshrined in Australian law'.
• The excess mortality continues to be found in high-risk women and women need to be informed of this risk.
• Freebirth is rising and this is a concerning outcome of restrictions on options like homebirth and trauma from hospital births

2 comments:

Lisa Barrett said...

I dispute one of the conclusions, there is excess mortality in high risk. You can't even begin to say that unless you look at hospital twin birth outcomes. The waterbirth was not the cause of the death of the baby and further investigation reveals much more about what really happened as the case was reviewed by the Coroner.

This paper because it is written by those who are credited with respect, even though they are not homebirthing midwives seems to be trying to justify outcomes when justification is just not necessary. The figures speak for themselves. There is such a big problem in midwifery with our own academics playing into the hands of the AMA by marginalising women who take responsibility for their own births and midwives who support them. How sad for women. The press releases from the College et al should really be far more positive. Lets look at the thousands of hospital deaths and pick each one apart. I'm sure we will find many faults. The assumption that closer monitoring could save the one baby that apparently died at home for no reason is purely speculation by the authors of the paper and should be given absolutely NO comment.

Joy Johnston said...

Lisa I agree with your assertion that this study cannot be used to claim that there "The excess mortality continues to be found in high-risk women", which is in the comments by our academic midwifery colleagues. Anyone with knowledge of statistics knows this.

The conclusion in the Kennare etal MJA paper that "Perinatal stfety of home births may be improved substantially by better adherence to risk assessment, timely transfer to hospital when needed, and closer fetal surveillance" is a statment that has no bearing on the data that was reviewed by the study. When there is no investigation of risk assessment, time of transfer, and status of fetal surveillance, the 'researchers' cannot conclude that outcomes may be "substantially improved" if it was done better.

In my opinion this paper may well discredit the authors and the organisation that published it.