Affiliate disclosure
Book titles on this page link to Amazon. As an Amazon Associate, DataField.Dev earns from qualifying purchases — at no additional cost to you.
Chapter 7 — Further Reading
The reference documents
Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids (Institute of Medicine / National Academies, 2005). Freely readable through the National Academies Press.
This is where the 130 g/day RDA and the 45–65% AMDR come from, and the reasoning behind both is laid out in full. Reading the committee's actual justification for the RDA — that it's based on average minimum brain glucose use — resolves the apparent contradiction in §7.6 better than any summary, including mine. For: students, clinicians, and anyone who wants to see how a recommendation gets made.
USDA FoodData Central — fdc.nal.usda.gov. Free. Look up any food's carbohydrate, fiber, and sugar
content from the reference database that every app and label is downstream of.
FDA guidance on the Nutrition Facts label, particularly the 2016 update that introduced the Added
Sugars line (fda.gov). Worth reading the rationale — it's a rare case of a labelling change
designed around a genuine evidence-based distinction, and it's the single most useful line on the
panel.
For: everyone. Ten minutes, permanent payoff.
On glycemic index
The original Jenkins work. David Jenkins and colleagues introduced the glycemic index in the early 1980s in the American Journal of Clinical Nutrition. Read the original if you can access it — the authors were considerably more modest about its scope than the subsequent marketing has been, and the gap is instructive.
The University of Sydney glycemic index database (glycemicindex.com) — the main published source
of GI values. Use it and notice what §7.4 describes: the same food appearing with different values, and
the methodological caveats stated up front by the people who maintain it.
On individual variation: the Weizmann Institute work by Zeevi, Segal, and colleagues on personalized glycemic responses, published in Cell in 2015. Genuinely interesting science — people diverging widely in response to identical standardized meals. Read at least one critique alongside it. Chapter 35 revisits it. For: anyone considering a CGM or a personalized nutrition product.
On whole grains
The whole grain cohort and trial literature. Search PubMed for "whole grain intake systematic review meta-analysis" and read a recent one. Pay attention to how the authors handle healthy-user bias — the good reviews are explicit about it, which is itself a quality signal.
The Whole Grains Council (wholegrainscouncil.org) is a useful reference for definitions and for
what the various stamps and claims legally mean — with the standing caveat that it is an industry
organization, so use it for the definitions and go elsewhere for the health claims. Chapter 1 §1.5's
rule applies in every direction.
For: label decoding.
On resistant starch and sugar alcohols
The resistant starch literature. Search for reviews on resistant starch types and glycemic response. Compare the measured magnitudes to how the cooking-and-cooling finding is presented online — this is one of the cleanest examples in the book of a real, reproducible, small effect being inflated into a strategy.
On sugar alcohols: the NIH Office of Dietary Supplements and FDA materials on sugar alcohols and non-nutritive sweeteners. For the erythritol question specifically, read the 2023 Nature Medicine paper by Witkowski and colleagues and at least two published responses to it. The disagreement is about confounding and endogenous production rather than about the measurements, and seeing where the argument actually sits is more useful than any summary. For: anyone using sugar substitutes regularly, especially with cardiovascular risk factors.
On food safety with cooled rice: national food safety agencies (the UK Food Standards Agency and the US FDA) both have plain-language guidance on Bacillus cereus and rice. Short, practical, and the one genuinely safety-relevant item in this chapter.
On the history
Gary Taubes, Good Calories, Bad Calories (Knopf, 2007) and Nina Teicholz, The Big Fat Surprise (Simon & Schuster, 2014).
These are the two most influential books making the low-carbohydrate historical case, and I'm recommending them deliberately. They are well written, they document real problems with how the low-fat era unfolded, and the historical critique in both is worth engaging with seriously.
They are also arguments, and they select their evidence. Read them after Chapter 2 and with the Claim Filter in hand, and apply the test from Chapter 2's further reading: does the author ever describe a finding that complicates their own position? For: anyone who wants to understand the strongest version of the case this chapter argues against.
For the counterweight: the published responses and reviews of both books, which are easy to find and which are often as selective in the other direction. Reading both sides of this particular argument is one of the better available exercises in evidence literacy, precisely because both sides are intelligent and neither is disinterested.
On the cost dimension
USDA Economic Research Service (ers.usda.gov) publishes data on food prices, food spending, and
the cost of dietary patterns. If §7.10's cost check interested you, this is where the actual numbers
live — and Chapter 32 uses this material properly.
For: anyone advising people with real budgets, which is everyone advising anyone.
A note on what to be careful with
The carbohydrate corner of the internet has two loud camps and both are selective.
One tells you carbohydrate is uniquely fattening. Watch for: controlled feeding studies being ignored, traditional high-carbohydrate populations being explained away, and mechanism (insulin) being weighted above outcomes.
One tells you fat is the problem and carbohydrate is exonerated. Watch for: refined and intact carbohydrate being merged into one defence, and sugar-sweetened beverage evidence being minimized.
The tell in both directions is the same: does the source distinguish refined from intact? A source that treats "carbohydrate" as one thing — in either direction — has not yet reached the first useful distinction in this chapter, whatever else it knows.