Dear all:

I now know why my major dissertation advisor, a very brilliant nutritional epidemiologist who sometimes even confounded the analysts and the World Bank, went on rants about Linus Pauling.  His point was that if you are a mathmetician -- you only need to understand math.  A physicist needs to understand math and physics.  A chemist needs to understand math, physics and chemistry.  A molecular biologist like Linus Pauling needs to understand math, physics, chemistry, biochemistry and molecular biology.  An epidemiologist, however, needs to understand math, physics, chemistry, biochemistry, molecular biology, physiology, human behavior and epidemiology.  Linus stopped extrapolated from molecular biology without fully taking into account physiology, human behavior and epidemiology and many of the hypotheses he derived from looking at the molecular level have been disproved.

Now as to rats -- there are merely a model from which to test a few ideas.  What happens in rats doesn't necessarily happen in humans.  In fact, sometimes what happens in higher primates doesn't happen in humans.

The NIH study on SSRI cannot be extrapolated to autism in humans for three key reasons:

1) From what I read in that article, the researchers did not appear to test the SSRI in DEPRESSED rats.  Therefore, this does not apply to depressed humans who are taking SSRIs -- it would only apply to humans who were not depressed who were given SSRIs.  To give a different example, one study in Scandinavia gave beta-carotene to smokers to see if the beta-carotene reduced cancer risk.  The study was done in a fish eating population of smokers.  This means that they had high stores of retinol and would not absorb the beta-carotene.  Had they done the study in a population that was deficient in retinol, they may have found results.

2) The biochemical response in rats to a medication is not always the same as the response in humans

3)  The behavioral response in rats may not translate to a specific behavioral trait in humans.

As for the ASSOCIATIONAL study of autism and SSRI use -- they did not provide enough information about whether they completely controlled for the subtle effects of the severity of the depression.  Hospitalization rates may be more a factor of socioeconomic status (or perhaps access to health insurance that covers out patient services) than severity of depression.  We KNOW maternal depression affects brain development.  So even if the extrapolation from rat to human turns out to have some basis, you are still in a balancing of risks situation.  

Autism is most likely a multi-causal phenomenon and so it makes for very challenging research.  We've already had one pseudoscientist take an association that never existed to begin with (but was published as such because he completely changed the real data) and turn it into a widespread public health scare.  In this case at least, the researchers themselves are cautioning against extrapolating too far from associations and rats.

Best, 
Susan Burger 

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

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