Chapter 10 — Quiz

Twenty questions. Attempt before opening.

Multiple Choice

1. DIETFITS randomized 609 adults to healthy low-fat or healthy low-carbohydrate for 12 months. The primary finding was:

  • a) Low-carb produced significantly more weight loss
  • b) Low-fat produced significantly more weight loss
  • c) No significant difference between groups
  • d) Both groups gained weight
Answer **c.** Both lost roughly 5–6 kg. And the variation *within* each group was far larger than the difference *between* groups — which is the single most important number in the chapter.

2. DIETFITS pre-specified tests of two personalization predictors. They were:

  • a) Age and sex
  • b) Genotype patterns and insulin secretion
  • c) Microbiome composition and sleep
  • d) Baseline weight and activity level
Answer **b.** **Neither predicted who did better.** This is a pre-specified negative result in a well-designed trial, and it's why the DNA-diet-match claim gets a ❌.

3. POUNDS LOST compared four diets varying macronutrient ratio over two years in 811 adults. It found:

  • a) A clear winner
  • b) No meaningful difference in weight loss
  • c) That higher protein was dramatically superior
  • d) That all participants regained everything
Answer **b.** Together with DIETFITS and A TO Z, it forms the backbone of the "composition matters less than you think" conclusion.

4. The threshold concept of this chapter is:

  • a) All diets are equally healthy
  • b) Adherence beats composition — the differences between diets are small, and within-diet variation is far larger
  • c) Low-carb is superior for insulin-resistant people
  • d) Calories don't matter
Answer **b.** With the necessary qualifier: **composition sets the floor** — protein, fiber, micronutrients, ultra-processed food — and *among nutritionally reasonable diets*, differences are small.

5. Which is NOT one of the four reasons head-to-head trials converge?

  • a) The assigned diet is not the eaten diet
  • b) Dropout is large and non-random
  • c) Every reasonable diet removes the same foods
  • d) Researchers deliberately design trials to find no difference
Answer **d.** The other three are real and mechanical, plus the fourth: everything runs through energy balance.

6. What do essentially all successful dietary patterns share?

  • a) A specific macronutrient ratio
  • b) Low ultra-processed food, lots of vegetables, adequate protein, high fiber, home cooking
  • c) Elimination of an entire food group
  • d) Carbohydrate restriction
Answer **b.** And the striking thing is the negative: **no two of them agree on a macronutrient ratio.** The one universal feature is low ultra-processed food.

7. The paleo premise is criticized primarily because:

  • a) Agriculture never happened
  • b) There was no single Paleolithic diet; modern plants and animals differ from ancestral ones; and humans have demonstrably continued evolving (lactase persistence, amylase copy number)
  • c) Hunter-gatherers ate only meat
  • d) The diet doesn't produce weight loss
Answer **b.** Note the chapter's distinction: **the diet performs comparably to other reasonable diets; the rationale given for it is what fails.** Being able to separate those is the transferable skill.

8. Carnivore received ⚗️ Untested rather than ❌ Not supported because:

  • a) There is strong evidence it works
  • b) There are no long-term controlled trials in either direction — only testimonials and mechanism
  • c) It has been proven safe
  • d) The chapter is sympathetic to it
Answer **b.** The verdict scale distinguishes "evidence against" from "no evidence." What *can* be said is that people reporting improvement are reporting something real and attributing it to the wrong variable — the bundled-intervention problem from Chapter 3.

9. The between-diet separation in weight loss is widest at approximately:

  • a) 1 month
  • b) 3–6 months
  • c) 12 months
  • d) 24 months
Answer **b.** Which is **exactly when most diet books stop reporting.** "Lose more weight on X" is almost always a statement about month three.

10. Which pattern has the strongest long-term randomized evidence in this chapter?

  • a) Ketogenic
  • b) Carnivore
  • c) Mediterranean (PREDIMED)
  • d) Paleo
Answer **c.** And note that it is **not low-fat** — PREDIMED's Mediterranean arms were higher in total fat than the control arm advised to reduce fat.

11. A "reset-on-failure rule" — where one lapse restarts the count — is criticized because:

  • a) It's too easy
  • b) It converts one bad day into total failure, guaranteeing eventual termination
  • c) It doesn't produce weight loss
  • d) It's expensive
Answer **b.** It writes Chapter 4 §4.10's what-the-hell effect into the terms and conditions. Theo completed 31 consecutive days of two daily workouts and called it "failing."

12. Roughly what is the annual food cost range for a carnivore diet versus a beans-and-grains plant-based pattern?

  • a) They're about the same
  • b) Carnivore ~$5,200–$8,300 vs. plant-based ~$1,800–$2,600
  • c) Plant-based is more expensive
  • d) Cost has never been estimated
Answer **b.** A factor of two to four. **The patterns with the best long-term evidence are among the cheapest** — worth remembering before the advice reaches a household at $6.43 per person per day.

13. §10.10 argues the diet argument is unusually bitter because:

  • a) The effect sizes are enormous
  • b) A diet is an identity, everyone has a genuine personal success story, and conversion costs are high
  • c) Researchers are corrupt
  • d) People don't care about health
Answer **b.** And the chapter's own conclusion "offers the least identity of any position in the argument" — no in-group, no enemy, nothing to post.

True / False

One-line justification.

14. "Adherence beats composition" means all diets are equally good and you can eat anything.

Answer **False.** Composition sets the floor: adequate protein, adequate fiber, adequate micronutrients, low ultra-processed food. *Among diets meeting that floor*, differences are small. **Composition sets the floor; adherence determines whether you ever stand on it.**

15. Theo's diets ended because they stopped producing weight loss.

Answer **False, and it's the chapter's opening point.** Not one stopped working. All five stopped being *executable* — hunger, a spouse, a holiday, exhaustion, a rule designed to end.

16. A DNA test can currently identify whether you'll do better on low-fat or low-carb.

Answer **False.** DIETFITS tested exactly this, prospectively and pre-specified, and the genotype patterns did not predict outcome. The premise — that people vary — is right; the product doesn't follow from it.

17. The typical two-year outcome across diets is complete regain to baseline.

Answer **False, and the nuance matters.** Average weight at two years is commonly a *fraction* of peak loss — not zero. And modest sustained loss carries real benefits in blood pressure, glycemia, lipids, and function that persist even when weight partially returns.

18. The Mediterranean pattern is a low-fat diet.

Answer **False.** PREDIMED's Mediterranean arms were supplemented with extra-virgin olive oil or nuts and were **higher** in total fat than the control arm. The variable that mattered was food quality and pattern, not ratio.

Short Answer

19. Explain why the trials converge, and why that is a finding rather than a methodological failure.

Answer **Four mechanisms.** (1) **Assigned ≠ eaten** — adherence drifts, so by twelve months the groups' actual intakes are much closer than their assignments. (2) **Dropout is large (20–40%) and non-random** — the people doing worst leave, inflating every arm and compressing differences. (3) **Every reasonable diet removes the same foods** — whatever the ratio, people stop drinking soda, stop snacking, and start cooking. (4) **Everything runs through energy balance.** **Why it's a finding:** because the question a real person is asking is *"what happens if I attempt this?"* — and drift, dropout, and food-overlap are part of the answer, not noise obscuring it. A metabolic-ward trial that eliminated all four would answer a cleaner question that is less relevant to anyone's life.

20. Someone tells you they failed at four diets. Reframe it, and say what you'd ask them.

Answer **The reframe:** they didn't fail at four diets. Four diets each produced results for as long as they could be executed, and each had a design feature that made indefinite execution impossible in their particular life. *"The diet failed" and "I failed the diet" are both wrong* — the accurate framing is **mismatch**. **What I'd ask — and only this:** *"Why did each one end?"* Not what they ate, not how much they lost. The last column. Then look for the **recurring constraint**, because there almost always is one — the same thing ending three different attempts. Hunger, or other people, or cost, or energy, or a rule that guaranteed termination. **That constraint, not the diet, is the actual problem**, and it's the thing to design around. And finally, notice their language. If they say "cheated," "fell off," "was good" — that vocabulary is doing damage, and naming it is often more useful than any dietary advice.

Applied Scenario

21. A 38-year-old asks you to settle an argument: their brother says keto is scientifically proven superior; their colleague says whole-food plant-based reverses heart disease; their trainer says it's all just calories. They want to know who's right so they can start on Monday.

Answer them.

Answer **All three are partly right and none of them is answering the question.** **The brother** is right that keto works, that it suppresses appetite in many people, and that it has genuine evidence for glycemic control in type 2 diabetes. He's wrong that it's *superior* — head-to-head trials at 12 and 24 months don't show it, and the early advantage is partly glycogen and water. **The colleague** is right that plant-based patterns have good evidence and that some intensive programmes have shown improvements in cardiovascular risk factors. They're overstating "reverses," and the cohort evidence carries healthy-user bias. **The trainer** is right that it runs through energy balance and wrong that this is useful advice — Chapter 4's threshold, both halves. **What I'd actually say:** > *"Nobody's going to win that argument because they're arguing about the wrong variable. The trials > that compare these directly find differences too small to matter by two years, and the differences > between individuals on the same diet are bigger than the differences between diets. So the question > isn't which is best — it's which one you'll still be doing next October."* **Then the four questions from §10.12:** Would you still be doing this in a year? Does it survive a bad week? Can you eat with other people? Can you afford it, in money and time? **And one piece of concrete advice, because "it depends" is not a recommendation:** if they have no strong preference, start Mediterranean — best long-term randomized evidence, additive rather than eliminative, nutritionally complete, mid-range cost, and it doesn't require them to give up a food category or explain themselves at dinner. **Then, before Monday, one thing in writing:** *what is my plan for the day this ends?* Because it will, and that's where the outcome is actually determined.

Scoring

Score Reading
18–21 Strong. You can now defuse most diet arguments without joining one.
14–17 Good. Reread §10.3 and §10.5.
10–13 Reread §10.2–§10.5 together — the trials and the threshold have to land as one thing.
Under 10 Reread the chapter. Chapters 21, 24, and 37 all build on it.

And do the autopsy (Exercise C1). Reading this chapter changes what you think; the autopsy changes what you notice about yourself.