This past Friday, the American Journal of Obstetrics and Gynecology released an article online about a study–more accurately, a meta-analysis–by Dr. Joseph R. Wax of the obstetrics and gynecology department at Maine Medical Center and his colleagues. The analysis reached the loud conclusion that homebirth, while having significantly better intervention and morbidity rates than planned hospital birth, had a threefold neonatal mortality rate. This is in sharp contrast to previous studies that have been done on the subject, which reached the conclusion that homebirth is as safe as a hospital birth for low-risk women. Bad news for homebirth supporters, right?
Firstly, the advantage of a meta-analysis is that it provides impressive numbers. Which is why this study is going to be flaunted and relished by anti-homebirth OBGYNs for years to come–the sheer scale of it is impressive. But the disadvantage is that it is synthesizing data from many different studies, it is not controlled and the population is not uniform. That means the conclusions are less reliable than a controlled study.
Also, a closer look at the study reveals some major differences between this analysis and the pro-homebirth studies. For one thing, those studies’ conclusion was not that all homebirths–attended or unattended by a professional–are safe. It was that low-risk births, attended by a professional and well-equipped midwife, are as safe as hospital births with significantly lower morbidity rates.
And actually, this analysis reaches the exact same conclusion (emphases and brackets mine):
“The analysis excluding studies that included home births attended by other than certified or certified nurse midwives had findings similar to the original study [meaning, significantly lower intervention and morbidity rates], except that the ORs [odds ratios] for neonatal deaths among all (OR, 1.57; 95% CI, 0.62–3.98) and nonanomalous (OR, 3.00; 95% CI, 0.61–14.88) newborns were not statistically significant.”
Hmm. They failed to mention that in the official conclusion.
Morever, as in this report on CBC, other epidemiologists are questioning the political motivation of the analysis as well as the methodology. According to UBC epidemiologist Patricia Janssen, quoted in the article in the link, the analysis includes studies that are highly questionable–some with as few as five recorded births, some dating back to the 1970s, and some in which the qualifications of the caregiver is unknown.
Another glance at the statistical analysis will raise a few eyebrows among those who know a thing or two about the US national birth rates. That’s right–the rate of C-section in the planned hospital births. 9.3%, huh? Extremely impressive, considering the national rate is higher than 25%. How optimistic. What other statistics listed in the analysis are so far removed from reality?
Basically, Dr. Wax… nice try, but no cigar.