Surprise! There were homebirth deaths in the Dutch study that claimed to show that homebirth has lower risks.

Surprised girl

On June 14, I wrote about Ank de Jonge’s latest attempt to show that homebirth is safe (No, new Dutch study does NOT show that homebirth is safe). As I mentioned at the time, de Jonge continues to slice and dice the Dutch homebirth data is an effort to somehow prove that homebirth is safe, when the data suggests that it is not.

In the latest paper discussed in that post, Severe adverse maternal outcomes among low risk women with planned home versus hospital births in the Netherlands: nationwide cohort study,  de Jonge concluded:

Low risk women in primary care at the onset of labour with planned home birth had lower rates of severe acute maternal morbidity, postpartum haemorrhage, and manual removal of placenta than those with planned hospital birth. For parous women these differences were statistically significant…

In other words, there was no difference in severe acute maternal morbidity (SAMM) between home and hospital among nulliparous women and a slightly lower rate of SAMM for parous women at homebirth.

There was just one teensy, weensy problem. de Jonge left out the mortality rates. Severe maternal morbidity is an appropriate measure of safely ONLY when death rate is zero or nearly zero. If the death rate is not zero, that MUST be taken into account in assessing safety. My Letter to the Editor of the BMJ regarding this inexplicable oversight was published the same day. de Jonge and colleagues have finally responded, and what do you know, the maternal mortality was NOT zero.

The reply appears to continue the trend of apparent obfuscation of the results.

The authors claim:

We did not mention maternal deaths in our study, but they were included among the women with severe acute maternal morbidity (SAMM). There were two maternal deaths in the planned home birth group (2 per 100,000) and three in the planned hospital birth group (6 per 100,000). The differences between these rates were not statistically significant (Fisher’s exact test, P=0.367).

They described 1 homebirth death due to cerebral hemorrhage possibly secondary to pre-eclampsia. The authors try to blame the doctors who evaluated the woman at 37 weeks, at which time she was felt to fine. A lot can and does happen in the last week of pregnancy. To blame the doctors who saw the woman a week before her collapse and absolve the midwife who cared for her at the time of birth is bizarre.

What about the other homebirth death? Funny you should mention that. The authors did not say. They lumped the second homebirth death in with the hospital deaths and reported:

The other four women were referred during labour from primary to secondary care because of meconium stained liquor. One woman suffered from sudden collapse during labour, when she was already in secondary care, and died. Although no definite diagnosis was made at postmortem examination, a cardiac cause appeared to be most likely.

A woman who gave birth spontaneously was discharged after one day. On the fourth day postpartum she was readmitted because of profuse vaginal bleeding and shortness of breath. She had a sudden collapse and died. Postmortem examination showed sinus sagittalis superior thrombosis.

Two women died a few weeks after they gave birth from causes not related to the delivery; one from a severe asthma attack, the other one fell down the stairs, had a skull fracture and died of a subarachnoid haemorrhage.

Since the authors did not specify that either of the woman who died of causes unrelated to delivery was in the homebirth group, it seems safe to assume that they were both in the hospital group.

Therefore, as far as I can determine, there were 3 maternal deaths attributable to pregnancy in the entire study, 2 in the homebirth group and one in the hospital group, for a death rate of 2/100,000 in each group. The only one that appears to have been potentially preventable was the one that occurred in the homebirth group. Therefore, the homebirth group had one death that was potentially preventable in the hospital, while the hospital group had none.

The study is underpowered to determine whether there is a statistically significant difference in the death rate between the two groups, but the fact that even one woman in the homebirth group died of a potentially preventable cause means that there is no basis for concluding that homebirth is as safer or safer than hospital birth among the women in this study.

Simply put, the death rate was not zero and until the difference (if any) between maternal deaths at home and in the hospital is determined, we cannot draw any conclusions about the safety of homebirth for Dutch mothers.

A more appropriate conclusions for the study would be:

Low risk women in primary care at the onset of labor with planned home birth had lower rates of severe acute maternal morbidity, but this difference was statistically significant only for parous women. However, there was a potentially preventable death in the homebirth group, while there were no potentially preventable deaths in the hospital group. The study is underpowered to detect a difference in maternal mortality between home and hospital, therefore, no conclusion can be drawn about the safety of homebirth.

Yes, fewer women in the homebirth group experienced severe acute maternal morbidity, but that’s nothing to crow about if one of them died and might have been saved in the hospital.