The Johnson and Daviss Study

As I discussed in the last post, we can only use studies that use home birth data from the United States to determine whether home birth in the United States is safe.

And when you look at that data, you find that every study on planned, midwife-attended home birth in the United States shows that significantly more babies die at home birth than at hospital birth:

Other than those listed, there is one more major study on home birth in the United States. It is called "Outcomes of planned home births with certified professional midwives: large prospective study in North America," it was published in 2005 by the authors Johnson and Daviss, and it claims to have found that planned home birth had "lower rates of medical intervention but similar intrapartum and neonatal mortality to that of low risk hospital births in the United States."

The problem is, the data didn't actually show that.

When analyzing the number of interventions for their study, Johnson and Daviss compared numbers from home birth in the year 2000 to hospital births in the year 2000. But then when looking at mortality (death) rates, they compared the home birth numbers from 2000 to hospital numbers stretching all the way back to 1969, when death rates in the hospital were much higher.  After doing this, they said that found a neonatal death rate of 2.7 per 1000 for home births and 2.6 per 1000 for hospital births.

When it was pointed out to them that the data actually showed a hospital death rate of 0.9 per 1000 in the year 2000, meaning that their study really showed home birth had 3 times the death rate as hospital birth, the authors admitted that the hospital death rate in 2000 was indeed 0.9 per 1000. They claimed that the data wasn't available to make the correct comparison at the time of their study, even though it was available since 2002 and their study was published in 2005.

After admitting the correct hospital neonatal mortality rate in 2000 was 0.9 per 1000, they then tried to claim that home birth in 2000 *actually* had that same death rate. They did this by excluding groups of deaths from the home birth data that they did not exclude from the hospital data. If those same groups were excluded from the hospital data, the hospital death rate would drop down to 0.34 per 1000 - still showing the same conclusion: three times as many babies died at home birth than at the hospital.

Why would Johnson and Daviss go to all these lengths to claim that the data showed home birth is safe, even when it didn't? Well, the reader might be surprised to know (since they did not disclose it in their paper) that Johnson used to work for the Midwives Alliance of North America, and Daviss is a homebirth midwife.