Amal, you said:  "What is a bit complicating in the Coutsoudis study (and it
pointed out) is
the fact that the randomisation was on the basis of vitamin A
supplementation and not on feeding mode. This means that there was
self-selection into feeding category and makes the interpretation a bit
complex. FOr example, could it be that the HIV positive mothers who were in
better state of health chose to breasfeed exclusively whilst the others who
were worse off chose the other options and hence the findings? It would not
be possible to randomise feeding option, neither ethically or practically,
so how does one come around this?"

If one were in the position fo setting up a new study (!), I suppose one
could ask women to choose feeding method (breastfeeding vs replacment from
birth).  Within the  breastfeeding cohort one could randomise women to a
control group to do whatever they decide to do (and monitor this) and an
intervention group to receive support for exclusive breastfeeding.  This
technique would be a bit like the PROBIT study -- looking at the differences
within a group of women who choose to breastfeed with different conditions.
This would not eliminate any differences between women who choose to
breastfeed or choose not to breastfeed, but it could randomise behaviour
around exclusivity.

Magda Sachs
Breastfeedign Supporter, BfN

             ***********************************************
The LACTNET mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software together with L-Soft's LSMTP(TM)
mailer for lightning fast mail delivery. For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html