Instructor Notes — Chapter 19

What this chapter is doing

Chapter 19 closes Part III and is the book's central safety chapter. It is also the chapter most likely to change what a student actually does — and, for the same reason, the one most likely to fail if it is taught in the wrong register.

The single most important thing to understand before teaching it: a meaningful fraction of any class taking a course on peptides has already bought something. Not hypothetically. Some of them are using something this week. The chapter is written on that assumption, and its stated position is pro-knowing-what-you-have, not anti-peptide. If your teaching drifts into disapproval, those students disengage and return to sources with worse information and better manners. That is a worse outcome than the one you were trying to prevent, and it is the failure mode to guard against for the whole unit.

The practical version: teach §19.7 as the payoff, not §19.3 as the warning. The six failure modes are the technical core, but they are not what persuades. What persuades is the concrete list of what a clinician does — baselines that cannot be recovered later, a differential diagnosis instead of an attribution, the escalating-dose trap — because it is specific, non-moralizing, and unavailable elsewhere.

Prerequisites and dependencies

Formally [6, 17, 5]. In practice students also need:

• Chapter 2 §2.4 (amplification) for the "a concentration error is not a proportional error" argument in §19.3.
• Chapter 2 §2.8 (desensitization) for §19.7 item 3, which is the chapter's most valuable single claim.
• Chapter 4 §4.7 (aggregation) for the seventh problem.
• Chapter 10 CS2 (observational data asymmetry) for §19.8. If students have forgotten it, §19.8 will not land — spend two minutes restating it before you get there.
• Chapter 3 (cadaver growth hormone) — used three times. Worth a one-slide refresher.

Timing

Suggested allocation for a standard unit:

If you only have two sessions: teach §19.3, §19.7, and §19.10, and assign the rest. Those three carry the chapter.

Teaching the hard parts

Endotoxin (§19.3). This is new to essentially everyone, including students with science backgrounds. Do not lead with the biochemistry. Lead with the sentence — sterility asks whether anything is alive in the vial; endotoxin asks whether anything ever was — and only then explain why it survives sterilization and passes filters. The mechanism sticks once the distinction is in place; it does not stick the other way round.

The nonlinearity argument (§19.3, content). Students reliably assume that being off by a factor of two means a twofold difference in effect. Draw the sigmoid on the board. Ask where a twofold error is invisible and where it is total. This takes four minutes and permanently fixes an intuition the rest of the book depends on.

Aggregation (§19.3). The correction here is counterintuitive and easy to under-teach. The intuitive model — mishandled vial equals weaker vial — is not merely imprecise, it is wrong in kind. Say explicitly: mishandling changes the risk profile, not the potency. Watch for students who nod and then, two questions later, describe a temperature excursion as "so it just wouldn't work as well."

§19.6. This section is deliberately written so that it cannot be followed, and it says so in the text. Do not supplement it. Students will ask; the honest answer is that the class is not the place and the answer belongs with a clinician. Treat the question itself as a teaching moment about why the chapter is structured this way.

§19.9 legality. Resist giving jurisdiction-specific answers, even if you know them, and even if asked directly. The teaching goal is the structure — six separable questions answered by different authorities that need not agree. If you supply an answer, students will remember the answer and forget the structure, and the answer will be out of date before they need it.

Anti-doping (§19.9). If any student competes at any level, this is the highest-stakes passage in the chapter for them personally. The two facts that matter: the non-approved substances category prohibits a class without naming compounds, and strict liability means intent is irrelevant. Make sure both are stated aloud.

Common student errors

1. Collapsing the two axes. The most common and most consequential. Students argue about whether a compound is "safe" while meaning two different things — evidence for a claim, and quality of a preparation. The §19.10 diagram exists for this; use it early, not only at the end.
2. Treating a certificate of analysis as a property of the vial. Expect this. The corrective is the logical form: a report of tests on a sample.
3. Assuming "legal" implies "reviewed." Especially for compounded products. Say the sentence: neither 503A nor 503B products are FDA-approved or reviewed for safety and efficacy.
4. Wanting a failure-rate number. Students find §19.4's refusal unsatisfying and will go looking for a percentage online. Pre-empt it: explain that the number they will find is not supported by any sampling frame, and that the ability to say so is the skill being taught.
5. Hearing "peptides are dangerous." If a student leaves with that summary, the chapter has failed. Check for it explicitly.
6. Treating self-testing as a substitute for supervision. The chapter forbids this framing and so should you. Analysis tells you what is in a vial; it says nothing about what it will do in a person.

Assessment guidance

• exercises.md ships without answers; model responses and marking notes are in _scratch/answers/ch19.md. Items marked † are the extended ones — roughly ten of thirty-four, and they carry most of the discriminating power.
• quiz.md has 22 items with a collapsed key. Items 6, 11, 16, 21, and 22 are short answer and are where partial understanding shows.
• Best single summative task: exercise F33 (write a four-line Evidence Rating for "material from a facility that also supplies pharmaceutical customers is pharmaceutical-grade"). The fourth line — what would change the rating — separates students who have understood batch release from those who have memorized vocabulary.
• Best single formative task: exercise F34 (the 200-word reply to a friend). It assesses tone, which is the thing this chapter is actually teaching and the thing a multiple-choice item cannot reach.