AtomicHipster:

Can you please specify which book and which website you're speaking of?

JS

Without buying the book, I can't know exactly how they calculate their number — but from the description and the numbers given in their table (which seem to be the same numbers used by nutritiondata.self.com), it seems to be a combination of "reasonable" and "bogus".

Reasonable: omega-3s are anti-inflammatory, high-GI and heavily carb-laden foods are pro-inflammatory, certain antinutrients are pro-inflammatory (example: adzuki beans get a negative rating.)

Not reasonable: gluten is non-inflammatory, insoluble fiber is anti-inflammatory, saturated fats are inflammatory.

Again, without reading their book I can't be sure whether my take on this is strictly correct.  However, given the numbers I see for various foods (game meat is more inflammatory than store-bought hamburger? All-Bran whole wheat cereal is less inflammatory than beef liver?) I can't possibly take their "index" seriously.  

Just like the ANDI, this is a way to cloak disproven conventional wisdom and subjective judgments with a layer of indirection that makes them look like science.  Hey, it's an Index, and numbers are objective measurements, right?

That's perhaps a bit harsh, so let me explain.  The problem with trying to distill many different pieces of information into a single "index" is that if you screw up even one of the components, the resulting "index" is worthless -- i.e. you're only as strong as your weakest link.

JS

AtomicHipster:

Sometimes there's no quick numeric shortcut to understanding the issues.  I'm doing my best to explain them!

JS