Home births

A reader recently asked for my comments on a Vancouver Sun article on the safety of home births, published on February 20, 2008. This is not the first time I've been forwarded home birth-related material since this post, so rather than replying with a private email, I'm publishing my response here.

The main point of the Sun article is:

[A] growing body of research - including some in B.C. - suggests, when done properly and attended by a midwife, [home births] can be just as safe as delivering in a hospital.

In 2002, Patricia Janssen, a perinatal epidemiologist at UBC, published a study comparing 800 planned home births in B.C. to about 1,300 births in hospital.

The two groups of women were carefully matched, to ensure both involved low-risk pregnancies and women of similar economic backgrounds.

What Janssen found was home births were no more risky, either for women or their babies, than delivering in a hospital.

I'm not sure why the article is deemed newsworthy, when, despite mention of a "growing body of research", the only published study referred to is six years old, and did not go without criticism when it was published in 2002. You can read the study by Janssen et al here.

The study authors' conclusions were:

There was no increased maternal or neonatal risk associated with planned home birth under the care of a regulated midwife.

But the Sun article makes no mention of these other statements by the authors, published in the same study:

Five babies in the home birth group required assisted ventilation for more than 24 hours, compared with none in either comparison group.

In the home birth group there were 2 cases of obstetric shock, and there were none in either comparison group. Three women in the home birth group required a blood transfusion, as did one in the midwife comparison group and none in the physician comparison group.

There were 3 cases of perinatal death in the home birth group: 2 stillbirths and one neonatal death. There was one stillbirth in the physician comparison group . . . The relative risk of perinatal death comparing midwife-attended births at home with physician-attended, planned hospital births was 2.5 (95% confidence intervals 0.27–24.5) . . . Although our perinatal death rate lies well within the rates described in other population-based studies, it deserves ongoing monitoring in view of the small numbers of births it is based on to date. A study large enough to compare perinatal death rates accurately, if the annual rate of home births with regulated midwives in BC were to remain the same as it is today, would require 7–8 years of data collection.

Of the babies who were exposed to thick meconium and whose Apgar score at one minute was less than 7, 45% in the home birth group received tracheal suction compared with 75% in each comparison group. Babies exposed to thick meconium who are not vigorous at birth may be disadvantaged in the home birth group, particularly in view of the trend toward increased need for assisted ventilation in this group.

[T]he consequences of some of the expected complications (thick meconium in the amniotic fluid and hemorrhage) may be more serious for women and their babies when women deliver at home than when they deliver in hospital. These comparisons are based on small numbers and warrant ongoing evaluation.

The rates of some adverse outcomes were too low for us to draw statistical comparisons, and ongoing evaluation of home birth is warranted.

Clearly, reading the study itself is more sobering than the enthusiastic Sun article.

There were many well-written letters to the CMAJ editor in response to the publication of this study. Here are some excerpts:

Jim Petzold Family Physician, Gibsons, BC:

As a family physician who has provided obstetric services in rural British Columbia for over 20 years, I am upset by the implied safety attached to home births. Statistically significant or not, in the study group involving 862 home births there were 3 times as many perinatal deaths compared with the cohort group involving 1314 in-hospital births. As well, 5 infants in the study group required prolonged ventilatory support versus none in the cohort group, and the only 2 cases of hemorrhagic shock occurred in the study group. 

If, as the authors state, 7 to 8 years of data collection are required to compare perinatal death rates accurately, why did they then feel compelled to state that "there are no indications of increased risk associated with planned home births attended by regulated midwives"? The lay press has concluded that home births have been shown to be as safe as, if not safer than, in-hospital births. If we look at serious complications, this is clearly not the case. 

If nothing else, the study should raise legitimate concerns regarding the safety of home births. Unfortunately, these concerns have not been conveyed to expectant mothers trying to make an informed choice.

Lawrence Fawcett Family Physician, Listowel, ON:

The study stated: "The median total time from a 911 call to arrival at hospital was 37 minutes, with a range of 15–93 minutes." Oh, how this vapid statement glosses over an incredible amount of needless suffering. If it takes 90 minutes from the decision to call 911 (and that is only after the immediate attempts at resuscitation have been recognized as insufficient) until arrival at the hospital, then presumably it takes at least 40 minutes for the ambulance to arrive. What do the midwife and patient talk about while the hemorrhaging uterus is being massaged and the blood pressure drops? . . . Do they reassure themselves that the pleasure of delivering at home is worth the agony of waiting for an ambulance while the blood continues to gush, or the gasping/flat/blue baby continues to be bagged? What if the ambulance personnel are also unable to perform neonatal intubation, so that inadequate assisted bag and mask ventilation continues for another 30 or 40 minutes in the ambulance? . . . 

As long as statistical twisting can be used to advantage, the obvious will be ignored, and the health of mothers and their babies will be sacrificed at the altar of personal choice.

Claudio De Lorenzi Plastic Surgeon, Kitchener, ON:

My interpretation of these results would be: the study showed a similar outcome for home births because the poorer outcomes expected of the more vigorous mothers who gave birth at home matched the better outcomes of the less healthy mothers who opted for hospital birth. 

Caroline McIntyre Meite Moser Division of Emergency Medicine, Vancouver General Hospital, Vancouver, BC:

The interpretation section states that "there are no indications of increased risk associated with planned home birth." This is dangerously misleading: rates for perinatal mortality and assisted ventilation were both higher in the home-birth population. 

Although the authors acknowledge that the rates of some adverse outcomes were too low to provide statistical comparisons, they still suggest no difference in adverse outcomes. Clearly, one preventable episode of perinatal mortality or requirement for assisted ventilation is one too many. Given that this study is not large enough to detect a clinically relevant difference in these major outcomes, the authors have no basis to make this claim. 

Unfortunately, the claims have already made it into the popular press, with the CBC stating: "Home births with a midwife are as safe as births in a hospital with a doctor." Once again, a medical publication has played a hand in misinforming the public. 

Liz Okon Registered Nurse, Registered Midwife, Clarenville, Nfld. M. A. Okon Physician, Clarenville, Nfld.:

This study contained significant biases. The groups were not like for like because members of the hospital specialist group were shorter, more likely to have had a previous cesarean section, weighed more and were less likely to be multiparous. Hence, they were more prone to dystocia than members of the home-birth group. 

As well, comparisons were made for induction of labour and epidural/spinal analgesia, but these interventions are usually unavailable during home births.  Are the authors implying that they are available at home in British Columbia?

This article is too biased to allow us to draw any meaningful comparisons between home and hospital births. Moreover, the conclusions are not justified by the evidence presented. The first step would be to compare like for like — a randomized selection of appropriate patients for home or hospital birth. 

Don Farine Professor of Obstetrics and Gynecology, University of Toronto, Toronto, ON:

Hospital deliveries and births are safer, and this is why there is a selection process for assigning patients to home birth. The issue is how small the risk is to women delivering at home . . .

Janssen and colleagues showed that the risks of home birth are quite low but possibly significant. An analogy may be that keeping patients in hospital for the full 9 months of pregnancy would be the safest thing to do. However, neither patients nor caregivers would consider the risks worthy of such a drastic measure. The still unanswered question is if home delivery carries a similar low risk in selected patients.

So, what are my thoughts on the Vancouver Sun article? This is a perfect example of how the media inaccurately presents medical studies to appeal to their subscribers. The article mentions only the study findings that support home birth. It completely ignores the limitations of the study and the cautions of the authors themselves: that more rigorous studies are needed to conclude that home birth is as safe as hospital birth.

And my thoughts on home birth? For starters, that parents should make informed choices. And that the popular media is probably not the best place to gather that information.

 

Comments

[The Other Pete]

Our attempt at a homebirth didn't work out and we're glad it didn't. The labour for Jonathan, who's now 4 and as sharp as a tack, wasn't progressing and so we asked to move to the hospital. Jonathan got his shoulders stuck on the way out (he was an unexpected 11.5 pounds) and he needed resuscitation. Thankfully, the midwife knew what she was doing and got him out quickly and the full support team was there, it seemed, within seconds of his delivery.

The response of the midwives was interesting. The younger midwife expressed her thanks that we had asked for the move to the hospital. The older midwife went into damage control, assuring us that things would have worked out fine if we had been at home. We're just thankful we weren't.

[Matthew Collinge]

We had our 3 at the hospital with a doctor. I think a doola that comes to the hosptial with you sounds like an interesting option, but I think the best option for us is to not have anymore kids.

[Sheri]

I had three doctors miss that Adam was breech even after an ultrasound at 36 weeks showed he was (I had him at 37 weeks) I was fully dialated and pushing when the doctor figured out his head was actually his bum. After a nasty T incision, emergency c-section, I am VERY HAPPY I was at the hospital and not attempting a home birth.

[Liana]

Hi,

I found your blog via Dr. J. My take on the Janssen study was essentially the same as yours and the doctors' comments you quoted. I wrote a very similar post on my blog (can't seem to find it now).

I like your blog... will be back often :)

Liana