Chapter 10 — Exercises

C1 — the diet-history autopsy — is the most uncomfortable exercise in this book and the one readers most often say changed something.

Difficulty legend: ⭐ basic recall · ⭐⭐ applied · ⭐⭐⭐ synthesis and judgment · ⭐⭐⭐⭐ extension

Selected answers in Appendix J.

Part A — Conceptual Understanding ⭐

A1. State this chapter's threshold concept, and then state the qualifier that stops it being a cop-out.

A2. What did DIETFITS compare, over what period, in how many people, and what did it find?

A3. What two personalization predictors did DIETFITS pre-specify, and what happened to them?

A4. Which was larger in DIETFITS: the difference between diets, or the variation within each diet?

A5. Name four reasons head-to-head diet trials converge.

A6. List the features that essentially all successful dietary patterns share. Which one do they not share?

A7. What does the paleo premise get wrong, anthropologically? Give three specific points.

A8. Why did carnivore receive ⚗️ Untested rather than ❌ Not supported?

A9. Describe the typical weight trajectory across 24 months on any diet. At which point is the between-diet separation widest, and why does that matter commercially?

A10. What is a "reset-on-failure rule," and why does the chapter say to reject any regime with one?

A11. Give the three reasons the diet argument is unusually bitter.

A12. Roughly what is the annual cost difference between a beans-and-grains plant-based pattern and a carnivore diet?

Part B — Applied Analysis ⭐⭐

B1. For each of Theo's five diets, identify the specific mismatch that ended it, and propose one design change that would have addressed it without abandoning the diet.

B2. A friend says: "Keto is objectively better — I lost 18 lb in four months and only 6 lb on Weight Watchers." Respond using §10.8's trajectory and Chapter 4 §4.9.

B3. Someone offers you a DNA test that will identify whether you're a "low-carb responder." Evaluate the offer using §10.7. What exactly has been tested, and what hasn't?

B4. A trial finds no difference between two diets at 12 months. List four reasons this could occur even if a real compositional difference exists, and say which is most likely.

B5. Rank these five patterns by strength of long-term randomized evidence, and justify: keto, Mediterranean, low-fat, carnivore, DASH.

B6. A patient with type 2 diabetes on metformin and a sulfonylurea wants to start keto. What are the specific clinical concerns, and what must happen first?

B7. Apply §10.12's decision framework to a specific person: a 45-year-old shift worker, cooks for three, $70/week food budget, hates cooking, wants to lose 10 kg. What do you recommend and why?

B8. The chapter says the diets with the best evidence are among the cheapest. Does that make the cost argument an argument for those diets, or a coincidence? Distinguish carefully.

Part C — Skills & Practice ⭐⭐–⭐⭐⭐

C1. Do the diet-history autopsy. The Project Checkpoint. Build the table, then answer the four questions — especially the last one about the language you used.

C2. Circle the moral vocabulary. Go through what you wrote in C1 and circle every instance of failed, cheated, fell off, was good, was bad, gave in, slipped. Count them. Then rewrite three of those entries in neutral, descriptive language and notice how differently they read.

C3. Find your recurring constraint. From your table: which constraint ended the most attempts? Hunger, other people, cost, time, energy, boredom, social eating? Design one intervention around that constraint specifically, rather than around a macronutrient.

C4. Write your reversibility plan. For whatever you're currently doing or considering, write the answer to "what is my plan for the day this ends?" in two sentences. If you can't, that's the finding.

C5. Audit a diet book. Pick any popular diet book. Find: (a) what it eliminates; (b) what mechanism it claims; (c) what the elimination actually removes from a typical diet; (d) what it says about day one after. Then ask whether (c) explains its results better than (b).

C6. Price four diets. Build a week's shopping list for one adult on each of: Mediterranean, low-carb, plant-based, and your current pattern. Price them in a real shop. Compare to §10.11's table.

C7. The steelman exercise. Pick the diet in §10.6 you find least plausible. Write its strongest case in 200 words — the version its smartest advocate would recognize. Then write what would have to be true for it to be right.

Part D — Synthesis & Critical Thinking ⭐⭐⭐

D1. "Adherence beats composition" could be read as "nothing matters, do whatever." The chapter denies this. Draw the line precisely: what does composition determine, and where does adherence take over?

D2. DIETFITS tested two personalization hypotheses and both failed. Does that mean personalization is a dead end, or that we tested the wrong things? What would a genuinely good test of personalization look like — and would it be feasible?

D3. The chapter gives carnivore ⚗️ Untested rather than ❌. Is that too generous? Construct the case that absence of evidence, combined with a strong prior from the fiber literature, justifies a negative verdict. Then say whether it survives the book's own verdict-scale rules.

D4. §10.10 argues that a diet is an identity and that the chapter's own conclusion "offers the least identity." If true, what follows for public health communication? Is there an ethical way to give the boring answer an identity, or would that be manipulation?

D5. Look AHEAD achieved weight loss and improved many secondary outcomes but did not reduce its primary cardiovascular endpoint. How should that trial be weighted? Consider what it implies about weight loss as a goal versus as a proxy.

Part M — Mixed & Interleaved Practice ⭐⭐–⭐⭐⭐

M1. (With Chapter 4.) Explain how each of the six diets in §10.1 produces an energy deficit. Name the specific mechanism for each — appetite, elimination, satiety, decision reduction.

M2. (With Chapters 4 and 6.) A friend regains 25 lb after keto and blames a wrecked metabolism. Give the four-part answer from §10.8's retrieval prompt without looking.

M3. (With Chapter 9.) PREDIMED's Mediterranean arms were higher in fat than the control. Explain why this is a problem for the framing of the diet wars, not just for low-fat advocates.

M4. (With Chapter 7.) The chapter says every reasonable diet removes the same foods. Using Chapter 7's intact/refined axis, name the specific foods, and explain why the overlap is larger than the prescribed difference.

M5. (With Chapter 8.) Which of the six diets is most likely to leave someone below their protein target? Which is most likely to overshoot? What does that predict about lean mass outcomes in a deficit?

M6. (With Chapters 1 and 2.) §10.10 says everyone in the argument has a genuine personal success story. Place that evidence on the ladder, and explain why it feels more compelling than a trial.

M7. (With Chapter 3.) The carnivore verdict invokes the bundled-intervention problem from Chapter 3's Case Study 1. Explain the parallel, and name the eight things a carnivore diet changes at once.

M8. (Revisit at Chapter 24.) Predict now: what proportion of people who lose 10% of body weight maintain it at five years? Write your number, seal it.

Part E — Research & Extension ⭐⭐⭐⭐

E1. Read the DIETFITS paper (Gardner et al., JAMA, 2018) in full, including the pre-specified secondary analyses of genotype and insulin secretion. Then find at least two critiques. Where do the critics say the trial fell short, and are they right?

E2. Find POUNDS LOST (Sacks et al., NEJM, 2009) and a recent network meta-analysis of named diets. Compare their conclusions. Then look at how each was reported in the popular press.

E3. Read about Look AHEAD — the intensive lifestyle intervention trial in type 2 diabetes. Find its primary outcome result and its secondary outcomes. Write a page on how a trial can "fail" and still be informative, and on what its early termination means.

E4. Investigate the anthropological literature on hunter-gatherer diets — the Hadza, the Tsimané, Arctic populations. How much variation is there? Then reassess §10.6's paleo verdict, and consider what a genuinely evidence-based "evolutionary" diet would even look like.

E5. Look into the Blue Zones literature and the recent methodological critiques concerning the quality of birth and death records in some of the identified regions. This is a live and instructive dispute about whether a widely-cited body of evidence rests on reliable data at all — and it's a good test of whether you can hold a favourite finding loosely.

E6. Research dropout rates in dietary intervention trials, and how different analytical approaches (intention-to-treat, per-protocol, completers-only) change reported results. Then reread any diet study's headline result with that in mind.