I would imagine that it is possible for thyroid problems to completely overwhelm and suppress benefits from promotion of breastfeeding.  Thyroid problems would not necessarily be a "confounding factor" in that both the treatment and the control would most likely have the same amount of radiation exposure and hence risk of thyroid abnormalities.

Epidemiologically, if the study population was exposed to radiation this would be an "interaction" effect.  This means that the impact of the intervention alone and the "interactive factor" are not additive.  For instance,  if your normal state is to breastfeed and have a latrine and childhood mortality is your outcome:

and you might hypothetically see  following results:

a) the childhood mortality rates among those who have a latrine and were NOT breastfed is 20 more deaths per 1000 children
b) the childhood mortality rates among those who don't have a latrine and were breastfed is 10 more deaths per 1000 children
c) the childhood mortality rates among those who don't have a latrine and were NOT breastfed is 50 more deaths per 1000 children

 and conclude that the impact of a) not breastfeeding and b) not having a latrine is MULTIPLICATIVE, not additive.

Applying this to radiation, if you compared a trial with an EFFECTIVELY IMPLEMENTED intervention to increase breastfeeding rates in areas where there was exposure to radiation at levels that reduced the milk supply to the SAME intervention in areas where the exposure to radiation was negligable, to give a HYPOTHETICAL example --- if you considered negligible radiation and an effectively implemented intervention to be your norm:

a) the breastfeeding rates among those who were exposed to the intervention and lived where the exposure to radiation was substantial might see a decrease of 20 fewer babies who breastfed to 3 months per 100 babies
b) the breastfeeding rates among those who were NOT exposed to the intervention and lived where the exposure to radiation was negligable might see a decrease of 10 fewer babies who breastfed to 3 months per 100 babies
c) the breastfeeding rates among those who were NOT exposed to the intervention and lived where the exposure to radiation was negibable might see a decrease of 21 fewer babies who breastfed to 3 months per 100 babies.

In this case -- you would hypothetically see a difference of 10 between those that received the intervention and those that did not when the intervention was implemented in a situation of negligable radiation exposure
You would hypothetically see a difference of only 10 between those that received the intervention and those that did not when the intervention was implemented in a situation of substantial radiation exposure.

This again is an interaction effect.  If you are missing any one of the four boxes (which includes the defined normal state) you would not be able to determine if that condition (e.g. radiation exposure) had an interactive impact on the intervention.

This is DIFFERENT than confounding.

Best, Susan Burger, MHS, PhD, IBCLC

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

Archives: http://community.lsoft.com/archives/LACTNET.html
To reach list owners: [log in to unmask]
Mail all list management commands to: [log in to unmask]
COMMANDS:
1. To temporarily stop your subscription write in the body of an email: set lactnet nomail
2. To start it again: set lactnet mail
3. To unsubscribe: unsubscribe lactnet
4. To get a comprehensive list of rules and directions: get lactnet welcome