Answer all 12 scored questions, then submit. Several questions (4, 7, 8, 10, 12) draw directly on the transcript-and-draft-note exercise from Handout Section C — if you haven't worked through it yet, do that first. Question 13 is a short reflection and isn't scored.
Question 1
An "ambient documentation" tool primarily —
Correct answer: drafts a note from the encounter for the clinician to review — nothing more.
Question 2
Why does this course introduce documentation tools before diagnostic AI tools (Module 3)?
Correct answer: lower relative risk plus a fast visible win — deliberate sequencing, not a claim that documentation tools are risk-free.
Question 3
Which Module 1 failure mode explains an AI-drafted note stating a blood pressure reading that was never actually taken or spoken aloud?
Correct answer: hallucination / plausible-value filling — generating a realistic-sounding value where the actual data was missing.
Question 4 · Applied — from the worked exercise
In the Handout's exercise, the patient stated the blood pressure reading as 138/86, but the AI-drafted note recorded 128/82.
What category of error is this?
Correct answer: a fabricated value close enough to be plausible but factually wrong — the most dangerous kind, because it doesn't look like an error.
Question 5
Between a fabricated detail and a silently omitted true detail, which is generally more dangerous in a clinical note?
Correct answer: the omission is often more dangerous precisely because it doesn't look wrong — it looks like a normal, unremarkable note.
Question 6
Which of the following is NOT part of the verification checklist in this module?
Correct answer: writing style is not a safety-relevant checklist item — the checklist focuses on clinical accuracy, not prose preference.
Question 7 · Applied — from the worked exercise
In the transcript, the patient confirmed a penicillin allergy at the end of the visit. The AI-drafted note stated "no known drug allergies."
What should happen before this note is signed?
Correct answer: correct it before signing. This is the single highest-stakes error type covered in the module.
Question 8 · Applied
A colleague tells you: "I don't really read the AI's note anymore — it's usually right, so I just sign it." What is the clearest risk in this habit?
Correct answer: the second option — this is precisely the "never trust-and-paste" rule from Handout Section C.
Question 9
According to the "documentation feedback loop" idea, why might a sloppy, unverified note matter beyond this one patient's chart?
Correct answer: the feedback-loop effect — depending on the vendor's data practices, today's sloppy note can shape tomorrow's tool.
Question 10 · Applied — from the worked exercise
The patient said "10 milligrams" of lisinopril; the AI draft recorded "20 mg."
What single habit from this module would have caught this before signing?
Correct answer: the medication cross-check step — exactly why it's on the checklist.
Question 11
True or false: once a physician has signed off on an AI-drafted note, responsibility for its accuracy shifts to the software vendor.
Correct answer: False. The signing clinician remains accountable for the note's accuracy — the tool assists, it doesn't absorb responsibility.
Question 12 · Applied
You're evaluating a documentation tool for a clinic with unreliable internet and patients who speak Tetum, Portuguese, and Indonesian. Which two questions from Section E's worksheet matter most before cost even enters the conversation?
Correct answer: connectivity tolerance and language accuracy — a feature-rich, well-known tool that fails on either is simply unusable in that setting.
Question 13 · Reflection (not scored)
In your own words: what would "trust-and-paste" look like in your own practice, and what's one concrete habit from this module you'll use to avoid it?