On Fri, 19 May 2000, Eric Armstrong wrote:

> One problem with epidemiological studies, of course, is correlations.
> People who eat more nuts have what else in common? Do they eat more
> healthily in general? Or less sugar (because they like salty foods).
> Or do they exercise more and are more weight conscious? I saw one
> epidemiological study in which it was shown that sunlight causes
> weight loss, so you should get sun every day. Of course, what the
> study showed was that people in sunnier areas of the country weighed
> less. Well, duh, they also exercise more because its warmer outside!

Well constructed epidemiological studies do attempt multivariate
analysis and identification of confounding variables.  This by no
means makes them perfect.  In this instance, one of the nut
studies was part of the Seventh Day Adventist study.  The SDAs
already enjoy lower heart disease rates than the general
population; the purpose of the study was to learn what dietary
factors within this group were correlated with the lowest heart
disease rates, and nut consumption turned out to be an important
one.  In this and other nut studies, incidentally, peanuts are
included as "nuts" and are in fact more frequently consumed than
real nuts.

It is a fact that most nuts are roasted, so their PUFAs have been
exposed to significant heating.  If this were a significant
contributor to heart disease, as some contend, then one would
expect that people who eat more roasted nuts would have more
problems, not less.  It is reasonable to infer that either the
heated PUFAs in the nuts are not as dangerous as we may think or
there is something else in the nuts that prevents them from
having a deleterious effect.  Or both.

> But an epidemiological study only shows a correlation, not a
> cause/effect relationship. I always use such studies as a guide, and
> then look for the deeper biological explanation of why. If a
> reasonable explanation can be found, I tend to accept it. If not, I
> tend to look for other causative factors that are equally correlated.

And this is what the epidemiologists try to do, too.  It can be
difficult, though.

> Frankly, though, I'm *still* looking for the definitive word on nuts.
> There are all these anti-germination enzymes that only go away after
> soaking. So you have to soak nuts overnight before you eat them?

So they say, although it is hard to believe that paleo people did
this.  After all, you don't *have* to do it for the nuts to be
edible.  Why would they have bothered?

> But then, are those enzymes on the surface only? If not, how does the
> water get inside? And if you heat them, doesn't that make trans fats?

No.  You don't make trans fats just by heating.

> It seems to me that when it comes to nuts and seeds, there is something
> major that we do not understand.

There's probably a lot that we don't understand.  There are
probably nutrients in them that we haven't even identified yet.

Todd Moody
[log in to unmask]