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 26 — Further Reading
⚠️ This is the chapter where the primary sources are most worth reading directly, because the effect sizes are specific and the summaries lose them.
⚠️ The three trials that carry the chapter
DiRECT — Lean MEJ et al., The Lancet, 2018 (12-month) and 2019 (24-month).
⚠️ Read the 12-month paper's figure showing remission by weight-loss category. That single figure is §26.3, and it is the most clinically useful image in this book. Then read the 24-month follow-up for what happens to durability.
The SSaSS salt substitute trial — Neal B et al., New England Journal of Medicine, 2021. Search "salt substitute stroke cardiovascular events rural China trial."
⚠️ Randomized evidence for hard outcomes from a dietary substitution. There is almost nothing else like it in nutrition, which is exactly why it's worth an hour. Read the eligibility criteria — the population was at elevated stroke risk — because generalization is the open question.
DASH and DASH-Sodium — search "DASH trial blood pressure controlled feeding" and "DASH-Sodium trial." ⚠️ Note that these are controlled feeding studies, which is why the effect sizes are clean and why translating them to free-living advice loses some of it.
On LDL-C and cardiovascular risk
The EAS Consensus Panel statement on LDL causality — search "LDL cholesterol causal atherosclerotic cardiovascular disease consensus statement." ⚠️ This document assembles the genetic, trial and mechanistic evidence in one place, and it is the best available demonstration of what a settled causal question looks like. Read it deliberately against Part IV.
On Mendelian randomization: Chapter 12's references, ⚠️ and note that the same method that settled alcohol also settled LDL-C.
On ApoB: search "apoB discordance LDL-C cardiovascular risk" and your national lipid guideline's position on ApoB measurement. ⚠️ Several now recommend it, particularly with raised triglycerides or diabetes.
On Lp(a): search "lipoprotein(a) once in a lifetime measurement consensus." ⚠️ The "measure once" recommendation is increasingly standard and is still not widely implemented.
On the portfolio diet: David Jenkins's work — search "portfolio diet LDL cholesterol reduction." ⚠️ This is where §26.5's 20–30% figure comes from, and it is the best case for what diet alone can achieve.
PREDIMED — search "PREDIMED Mediterranean diet cardiovascular prevention 2018 republication." ⚠️ Read about the retraction and republication as well as the result. How you weight a republished trial is a genuine methodological question and this book has taken a position on it.
On type 2 diabetes
The ADA Standards of Care (diabetesjournals.org, free) — ⚠️ the nutrition section, which is
notable for declining to specify a single eating pattern.
On remission definitions: search "consensus report definition and interpretation of remission in type 2 diabetes." ⚠️ Knowing the definition matters, because "remission" has a specific meaning and monitoring requirement.
On the twin-cycle mechanism: Roy Taylor's work — search "twin cycle hypothesis liver pancreatic fat type 2 diabetes." ⚠️ This is §26.1b and §26.3's mechanism, and it explains why Walt's glucose moved in week one.
On low-carbohydrate approaches: search for recent systematic reviews on carbohydrate restriction and glycaemic control, ⚠️ and read your national guidance on medication adjustment when starting one. The safety material is the part that's usually missing.
On fatty liver disease
Search "MASLD nomenclature change 2023" for the terminology, then ⚠️ "weight loss histological improvement NASH dose response" for the 3–5 / 7–10 / 10%+ gradient.
On coffee: search "coffee liver fibrosis cohort meta-analysis." ⚠️ Consistent, plausible, and one of the few genuinely enjoyable findings in this chapter.
⚠️ On cancer — read the reference document, not the internet
The World Cancer Research Fund / AICR Continuous Update Project and Cancer Prevention
Recommendations (wcrf.org) — ⚠️ free, graded by strength of evidence, and the reference standard.
Read the grading, which is where the honesty lives.
On nutrition DURING treatment: search "ESPEN guideline nutrition in cancer patients" ⚠️ and your national oncology dietetic guidance. This is §26.10's inversion, written by people who do it.
On cachexia: search "cancer cachexia definition management." ⚠️ It is not reversed by eating more alone, and under-nutrition on top of it is avoidable and common — both halves matter.
⚠️ And if you have been given an "anti-cancer diet" during treatment, or given one to someone, read Case Study 2 and then the ESPEN guideline. In that order.
⚠️ On kidney disease
KDIGO guidelines (kdigo.org, free) ⚠️ and your renal unit's dietetic guidance. CKD nutrition
is stage-specific and individualized, and this is one of the few places in this book where I'd say
plainly: do not self-manage from a textbook.
On the protein question: search "MDRD study protein restriction CKD progression" ⚠️ and then the debate about it. The effect sizes are more modest and more contested than the confidence of the advice suggests.
On phosphate: search "phosphate additives bioavailability CKD." ⚠️ This is the practical, high-yield instruction that is rarely given, and it is free to act on.
On sodium, if you want to go deeper
Read the Cochrane review on sodium reduction and blood pressure, then ⚠️ PURE's sodium papers, then a methodological critique of spot-urine estimation — search "spot urine sodium estimation Kawasaki bias."
⚠️ Read them in that order and form your own view. §26.8 states where I land and why; the disagreement is real and you should be able to reconstruct both sides.
A note on what to be careful with
⚠️ This chapter has two opposite failure modes and they map onto its two halves.
1. ⚠️ Under-claiming on the responsive conditions. Type 2 diabetes remission is real, quantified and reproducible, and a large number of people with the condition have never been told it exists. Fatty liver is the most reversible thing on most people's charts and is frequently flagged and never explained. ⚠️ If this chapter makes you fatalistic about chronic disease, you have read it backwards.
2. ⚠️ Over-claiming on the unresponsive ones — which does more damage. Diet does not treat cancer. It does not move Lp(a). It is not a substitute for a statin in someone at high risk, and it is not a substitute for urate-lowering therapy. ⚠️ And "I'll manage it with diet" is a decision that should be made with the magnitudes in front of you, not instead of them.
Three things I'd most want carried out:
⚠️ Ask which of your numbers respond to food, and by how much. Nobody volunteers it.
⚠️ If you have type 2 diabetes, ask about remission and ask for the number. §26.3's gradient turns an aspiration into a target.
⚠️ And if someone you love is having cancer treatment, the useful question is not what they should cut out. It is whether they are eating enough, and of what.