Hi Gonneke,

Thanks for the opportunity to discuss one of MY favourite topics!! The rate of 
SIDS in breastfeeding bed-sharing infants is indeed extremely low (for Europe 
and UK anyway, which are the data I know the best). So low that case 
control studies cannot calculate relative risks separately for this group 
because they have so few cases of SIDS in this category the stats are 
meaningless (I explored this with Pete Blair (statistician CESDI study) several 
years ago, and he couldn't do it with the CESDI data alone for this reason).

Prior to the special session on bed-sharing and SIDS (organised by Aurore 
Cote) at the SIDS international conf in Edmonton a couple of years ago I had 
asked Prof. Carpenter (lead author of controversial ECAS study report, and 
statistician) if he could devise a way to separate breastfeeding related bed-
sharers from non-breastfeeders in his SIDS-risk calculations using the ECAS 
(European Concerted Action on SIDS) data (which basically pools all the data 
from a dozen or more national case-control studies conducted across Europe 
in the nineties). To his credit, he accepted the challenge.

He reported at this conference that his data showed that when 'type of 
feeding in last 7 days' was added to a multivariate model of 17 significant 
variables, the risk of SIDS was halved for fully breastfed infants. He also 
confirmed our previous reports that 'bed-sharing was associated with a 
substantial increase in the duration of breastfeeding'. 

He devised a 'cumulative mortality model' for breastfeeding and bed-sharing 
that extrapolated from the ECAS data-set (i.e. were modelled mathematically, 
not actual odds ratios). The outcome was a prediction that the cumulative 
rate of SIDS among bed-sharing breastfeeders up to age 6 months was 2 in 
10,000 infants. In comparison the predicted cumulative rate for non-bed-
sharing breastfeeders was 1 in 10,000. For what he termed 'bottle-fed' infants 
the predicted cumulative rate was 4 in 10,000 for non-bed-sharers and 11 in 
10,000 for bed-sharers. This is against a backdrop of a general SIDS rate in 
the UK of 1 in 2000. 

For me this confirms that SIDS is extremely low for breastfed babies whether 
they bed-share or not, and frightening breastfeeding mothers about SIDS in 
the context of bed-sharing is unwarranted (I am not confident that a 
mathematical model based on rather disparate data (with inconsistent 
definitions of what constitues bed-sharing) is able to accurately predict 
differences in mortality to the level of 0.0001 -- so to me 1 or 2 per 10,000 
does not signify an important difference). 

What it told Carpenter however (his published conclusion) was that 'bed-
sharing doubled the risk of SIDS for both breastfed and bottle-fed infants'. 

You are all intelligent women on this list, so you can draw your own 
conclusions...

[All of our papers in this conference were published in a special issue of the 
Canadian Paediatrics Society Journal edited by Dr Cote. The above data can 
be found in: Carpenter, R.G. (2006) The Hazards of Bed-sharing. Paediatric 
and Child Health 11 (Suppl A) 24A-28A].

Regards, Helen

-------------
Helen Ball
Anthropology Dept & Parent-Infant Sleep Lab,
Durham University, UK

             ***********************************************

Archives: http://community.lsoft.com/archives/LACTNET.html
Mail all commands to [log in to unmask]
To temporarily stop your subscription: set lactnet nomail
To start it again: set lactnet mail (or [log in to unmask])
To unsubscribe: unsubscribe lactnet or ([log in to unmask])
To reach list owners: [log in to unmask]