Instructor Notes — Chapter 36

Chapter 36: What's Actually Coming — Peptides in the Next Decade, and How to Read Anyone's Prediction Including This One · Part VI · intermediate · estimated 6–8 hours · prerequisites 35, 9, 5

What this chapter is for

Chapter 36 is the book's methods chapter wearing a topical costume. Students will arrive expecting a preview of coming attractions and the chapter refuses to give them one, on the explicit ground that no evidence can exist for a claim about the future. The teaching target is §36.2 — the six questions — and everything else is worked application. If a session runs short, cut §36.7 and §36.8 before touching §36.2 or §36.10.

The three sections every student must read in full, regardless of track: §36.1, §36.2, §36.10.

The single sentence to get across: a prediction says what will happen; a forecast says what will happen and tells you what would prove it wrong. Almost every claim the course has taught students to distrust fails on that property alone, before any evidence is examined.

Where this chapter sits

It closes the argumentative arc. Chapters 5, 6, 9, 10, and 16 built the rating machinery; Part III applied it to a gray market with no evidence behind it; Part VI turns it on the future, which is the hardest case because the absence of evidence there is structural rather than culpable. Chapter 37 assembles every rating in the book into one table, Chapter 39 handles the clinician conversation, and Chapter 40 closes the dossier. Chapter 36 is the last chapter that teaches a new skill.

Common misconceptions, in the order they appear

  • "🔬 means promising." It does not. It means a serious question is being asked seriously, with no timeline and no probability attached, and most 🔬 becomes ❌. This is the single most likely misreading in the chapter and it is worth stating twice in the first ten minutes.
  • "❌ on the claim form means the drugs will disappoint." No. §36.2's ❌ describes the evidence available for the sentence as constructed. Some next-generation drugs will be better; the sentence still tells you nothing. Students conflate a rating of a claim with a prediction about the world.
  • "It's in clinical trials" as reassurance. Read against the base rate it should lower confidence. Expect resistance here; the intuition runs the other way.
  • "Phase 2 optimism means companies cheat." It does not. Small samples, selected populations, sensitive endpoints, and ordinary reporting statistics produce it without anyone lying.
  • "Statistically significant" as a synonym for "large" or "important." Reliably present in every cohort. Address it during item o, not before.
  • "The drug enters the brain." Students who have absorbed that GLP-1 drugs affect appetite routinely infer bulk penetration. §36.8's two routes are the correction, and note that the opposite error — dismissing the central effects as imagined — is equally wrong.
  • "Preserved lean mass on a scan means preserved function." The central error of §36.6. The chain has three links and the trial measured the first one.
  • Mixing the price ledger with the evidence ledger. Both directions appear in class: students who distrust industry pricing discounting the trial data, and students defending the data waving off access concerns. Name both as the same error.

Suggested sequencing

Session 1 (§36.1–§36.2). Open by asking students to write one sentence they believe about the future of this field, before reading. Collect them. Teach §36.1's framing problem, then the six questions. Return to their sentences at the end and have them run their own claim through all six. This is the highest-yield forty minutes in the chapter and it works because the specimen is theirs.

Session 2 (§36.2 applied — exercises n, o, p, q, r). Constructed announcements only. Item r's side-by-side table is the deliverable; put it on a board. The realization to engineer is that n, p, and q are different categories of knowledge that would be covered in nearly identical language.

Session 3 (§36.3–§36.5). Oral delivery, duration, multi-agonists. The pedagogical spine is the one-percent arithmetic in §36.3 running forward into §36.9's supply constraint. §36.4's counterweight is where clinically minded students engage most.

Session 4 (§36.6 + Case Study 2). The muscle question. Best run as a trial-design workshop: give groups the question and have them specify population, endpoint, and comparator before reading the case study, then compare.

Session 5 (§36.7–§36.10 + Dossier). Conjugates and the barrier can be brisk. Reserve at least half the session for §36.10 and the three 🔬 dossier rules, with in-class writing of the four-line entries.

Assessment notes

  • Exercises n, q, and kk grade cleanly and are good for a quiz or a short written check.
  • Items d†, m†, and i† are the best discriminators between students who have memorized the six questions and students who understand what a rating is. Item m† in particular — the defense of "mechanism does not move the base rate" against its strongest objection — separates the top of a cohort reliably.
  • Items ll† and mm† are the only ones that cannot be graded on content, since they ask students to forecast in their own domain and to audit their own sources. Grade on structure: are the items events rather than trends, are they dated, could a stranger mark them in ten years?
  • Item ii† requires students to find real coverage. Ask them to submit the link with the analysis; otherwise the assessment becomes an argument about a specimen you cannot see.
  • Case Study 2's discussion questions 3 and 5 are the strongest essay prompts in the chapter.
  • Beware over-rewarding cynicism. A student who concludes "nothing can be known and all coverage is worthless" has failed the chapter as completely as one who believes every press release. §36.2 is a discrimination procedure, not a rejection procedure — item q exists precisely so that students have to recognize a genuinely strong claim when they see one.

A note on tone

This chapter is deliberately unexciting and some students will read that as evasion. It is worth naming the design decision out loud: the author had the option of a thrilling future chapter and declined it, because a hype-heavy future chapter would retroactively discredit the thirty-five chapters before it. Students who see that as a choice rather than as a limitation get considerably more out of §36.10.

Practical items

  • Have a trial registry open in class. Look up one compound live. The gap between what the registry record says and what students believed they knew is the fastest demonstration available, and it takes two minutes.
  • Track material — 💊 GLP-1: §36.3–§36.6, with §36.6 read twice. 🏋️ Performance: §36.6, and note this is the one place in the book where a muscle question is genuinely open rather than settled against the marketing. 🔬 Science: straight through. 💄 Cosmetic: a light chapter, but §36.2 applies unmodified to ingredient claims. 🏥 Clinical: §36.2 is the section to hand a colleague; §36.4's safety counterweight and §36.6's endpoint discipline are the ones most likely to change a real conversation.
  • No compound named in this chapter should be discussed in terms of use. The chapter contains no dosing, protocols, or sourcing, and class discussion should not supply any. Redirect to Chapter 39.