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 21 — Further Reading
The three trials to read
⚠️ These do most of the work in §21.5 and §21.7, and all three are more interesting in the original.
Lowe DA et al., the TREAT trial — JAMA Internal Medicine, 2020. 16:8 versus three structured meals, 12 weeks. ⚠️ Read the lean-mass secondary outcome AND the published correspondence about it. The methodological argument about how body composition was measured is the whole reason §21.8 is 🟡 rather than a finding.
The 12-month time-restricted eating trial — New England Journal of Medicine, 2022 (search "time restricted eating calorie restriction 12 months NEJM"). ⚠️ This is the cleanest design available: both arms restricted calories, only one restricted the window. Read the methods for how adherence was measured — it's the hardest part of any free-living fasting trial.
Sutton EF et al. — Cell Metabolism, 2018. Early time-restricted feeding, eucaloric, crossover. ⚠️ Read the sample size before the results, and then read the results, and hold both. This is the most interesting finding in the chapter and it rests on very few people.
And read a meta-analysis: search "intermittent fasting versus continuous energy restriction meta-analysis weight." ⚠️ Note how consistently the conclusion is "comparable" and how consistently it gets reported as "as good as" or "better than."
On autophagy
Start with the Nobel Assembly's own summary of the 2016 prize (nobelprize.org) — free, short, and
⚠️ it will show you immediately what the prize was actually for. Ohsumi's work is genuinely
beautiful and the summary is written for non-specialists.
Then search "autophagy human fasting measurement markers" and see how thin the human literature is compared with the confidence of the claims. ⚠️ Look specifically for anything establishing a dose-response between fasting hours and autophagic flux in human tissue. I have not found it; if you do, I'd want to know.
On the translation problem: search "metabolic scaling mouse human fasting equivalence". ⚠️ This is a general problem in translational research and it is not specific to fasting — but fasting protocols make the error unusually directly, because they translate rodent hours into human hours.
On circadian nutrition
Search "diurnal variation glucose tolerance insulin sensitivity" for the underlying rhythm, and "melatonin glucose tolerance evening meal" for the mechanism behind late-eating effects.
Satchin Panda's research and his book The Circadian Code — ⚠️ read it as the strongest popular statement of the timing case, and note where it moves from established circadian biology to specific protocol recommendations. The first is much better supported than the second.
On shift work: search "shift work metabolic syndrome cohort" and the IARC evaluation of shift work involving circadian disruption. ⚠️ This is the strongest population-level evidence that timing matters at all — and Chapter 20 §20.11's hazard-versus-risk caution applies to the IARC classification here too.
On hunger and ghrelin
Search "ghrelin anticipatory meal entrainment humans." ⚠️ The meal-schedule-shifting studies are the interesting ones and they explain the subjective experience better than anything in the fasting literature does.
On appetite regulation generally: Chapter 3 §3.6's references, plus search "appetite hormones energy restriction compensation." ⚠️ The compensation literature is what predicts Theo's curve, and it is not specific to fasting.
⚠️ On the contraindications — read these if any apply
On diabetes and fasting: ⚠️ search your national diabetes organization's guidance on fasting (including Ramadan-specific guidance, which is excellent and detailed). Diabetes UK, the ADA and the IDF all publish on this. If you take insulin or a sulfonylurea, read it before you read anything else on this page.
On RED-S and low energy availability: ⚠️ search "IOC consensus statement relative energy deficiency in sport." Free, authoritative, and it is the single best document for anyone who trains seriously or coaches someone who does. Case Study 2 is built on it.
On eating disorders and IF: search "intermittent fasting disordered eating" and "fasting eating disorder risk young adults." ⚠️ The literature is growing and the association is consistent. Chapter 34 handles this properly and is worth reading out of sequence if §21.12's subsection described someone.
On older adults: search "protein distribution sarcopenia older adults anabolic resistance" (Chapter 8's references) ⚠️ and then consider what a narrow window does to that.
On Ramadan
Search "Ramadan fasting metabolic effects systematic review" for the general literature, and ⚠️ "Ramadan diabetes risk stratification" for the clinical guidance — which is more sophisticated than most secular fasting advice and worth reading regardless of your religion.
⚠️ And a note: this is a spiritual practice, not a health protocol. Reading it primarily as metabolic data is scientifically limited and somewhat discourteous. If you want to understand the practice, read about the practice.
The commercial edge
On "fasting-mimicking": search for the published trials, and ⚠️ then look at the price of the retail kit. Note that the intervention is a defined low-calorie diet and the product is a packaged version of it. Chapter 1 §1.5's question, which finds nothing elsewhere in this chapter, finds something here.
On fasting apps: ⚠️ if you use one, look at what its streak mechanic rewards. §21.12's subsection is not hypothetical.
A note on what to be careful with
⚠️ This chapter's failure mode is unusual: the risk is not that you'll believe too much, it's that you'll take away "it's just calories" and miss two things.
1. ⚠️ The circadian findings are real and the weight trials weren't designed to detect them. Dismissing timing entirely goes further than the evidence does — §21.7 is 🟡 and I hold the mechanism at moderate confidence, not zero.
2. ⚠️ The adherence case is not a consolation prize. §21.10. If a rule about when is one you'll keep and a rule about how much isn't, that is a real and sufficient reason to use it — Chapter 10 §10.4's finding is the best-supported thing in this book, and it applies here.
And the symmetric warning, for the reader who came in enthusiastic:
⚠️ If you finish this chapter still believing 16 hours triggers a threshold, go and look for the human dose-response yourself. It's the most concrete request in this chapter's further reading and the whole claim rests on it. If you can't find it, that's not a gap in your searching.
And ⚠️ read §21.12 properly, even if you're sure it doesn't apply. It is the section with the largest gap between how much attention it gets and how much harm it prevents — and unlike everything else in this chapter, getting it wrong is not merely a waste of money.