Death rate of new-born babies study
From Angry Mob it's our old friend the Daily Mail scaring people with their headlines again. In this case the news of a study that shows Babies born at night in hospital 'have higher risk of dying' which is what the study shows. Of course you have to read the article to discover that this study took place in the Netherlands, and that the absolute risk of death is between 0.05% and 0.09%. The risk increase is given as between 32% and 47%.
So what does that mean? Time for the mathematics, don't worry it's really simple.
Let's divide the deaths into two categories - those born during the day and those born at night. Now if we're told that risk increase for night-born babies is 32% that's compared to the day-born babies we set the absolute risk for the day-born at 1 and the risk for the night-born at 1.32. If we use the risk increase value of 47% then it'd be 1.47. As we know the overall absolute risk is 0.05% or 0.09% we can create some equations. Let x stand for the day-born death rate and we get our highest and lowest figures from:
1x+1.32x=0.09
2.32x=0.09
x=0.09/2.32
x=0.039
1x+1.47x=0.05
2.47x=0.05
x=0.05/2.47
x=0.020
So the absolute death rate for the day-born is between 0.020% and 0.039% and the night-born is 0.029% and 0.051% (multiply by the risk factor). Take the mean and we get 0.030% and 0.040% that is we 'expect' there to be 7 deaths per 10,000 births of which 3 will be those born during the day and 4 will be those born at night.
Is this worth the scary headline - no.
2 comments:
Correlation ≠ causality.
How many babies are born in hospital because the doctors know they're at risk of dying?
Think about that for a moment.
Now, what's this? A study shows that more of the babies born in hospital died? Wow, what a shocker! (Although, as you rightly point out, the probabilities either way are so pifflingly tiny as to not be worth worrying about anyway…)
The silly thing is the Mail quotes the reasoning behind the statistics which is that late at night more of the staff aren't there, they're tired, etc.
In other words the statistic goes up at night pretty much like every accident statistic, big shock.
As a study it's important and it highlights some of the differences between the larger and smaller hospitals as well as how they're run; but as a headline... no it's nothing.
Post a Comment