although having read the first part of the rather massive PDF files linked to on the methodology...
they used data from a 'stratified sample' of CLUs - that is to say they will have chosen a variety of types in order to represent the general picture of care in the UK. the data quality isn't quite as good as that for HB or MLU - particularly as 5 trusts did drop out to be resampled (though this may not be for my postulated reason above: the study went over two years and in that time some trusts changed/merged/closed their CLUs).
However any potential for inaccuracy between CLU data and HB data would still not answer the question: why are MLU outcomes for babies in this study better than for first time HB?
I didn't even consider a MLU as a potential birthplace, precisely because what was on offer there held no benefit over home, and some drawbacks. Due to the fact that our local MLU is an hours drive away also, i don't know anyone IRL who did give birth in a MLU.
could it be that the typical MLU - is attached to a better-than-average CLU and therefore transfer would offer less disadvantage? there doesn't seem to be a 'typical' MLU - the freestanding ones the study notes have quite a range of characteristics...
is that the disinclusion of babies born on the way to chosen birthplace (something associated with poor outcomes) has been done for MLU and CLU figures, but not for HB mothers whose MW didn't turn up (or only got a paramedic)?
anyone shed some light on this?