A critical care physician built this so the clinician stays sharper.
I'm Mike Hansen, MD — an ICU intensivist who teaches 1.1M+ on YouTube as Dr. Mike Hansen. Operation ADH is clinical reasoning reps for clinicians who want to use AI without surrendering judgment. Commit first, compare second, teach back.
Founding access. Be first in when seats open — no spam, unsubscribe anytime.
MH
Mike Hansen, MD
Critical care intensivist · 1.1M+ on YouTube
⚛
A real doctor built it
Designed at the bedside by a practicing intensivist, not a content mill.
⟳
You commit first
You lock your answer before any explanation opens. The reasoning stays yours.
📚
Cited, then compared
After you commit, you get cited reasoning to compare against your own.
Try it yourself — no sign-up
Do a real rep. Commit first — then the reveal opens.
This is an actual Operation ADH case. Read it, lock your answer, and the cited reasoning unlocks — exactly like the product. The explanation stays sealed until you commit.
Demo rep · liveAcid–base · high anion gap metabolic acidosis
🎧 Hear Jon read the case
Press play — Jon reads the vignette and the answer choices, highlighting each word as he speaks.
A 70-year-old female presents with several days of poor appetite, fatigue, and mild confusion. History: osteoarthritis treated with daily over-the-counter acetaminophen for months; she has been eating very little lately; mild chronic kidney disease.
What is the single most likely cause of this high anion gap metabolic acidosis?
Lock your choice — the reasoning stays sealed until you commit. That order is the whole point.
🎧 Hear Jon explain the answer
Jon walks the reasoning below — each word highlights as he says it.
Why D
The Plant (the routine detail that is the answer): chronic daily therapeutic acetaminophen + poor nutrition + female + mild CKD. That cluster — not an overdose — is the classic acquired 5-oxoproline picture. The sub-toxic acetaminophen level (18 mcg/mL) is the tell people dismiss.
acetaminophen depletes hepatic glutathione → loss of glutathione's feedback inhibition on γ-glutamylcysteine synthetase → the γ-glutamyl cycle overproduces 5-oxoproline.
Why the others are wrong
A — D-lactic acidosis: the seductive one. A normal lactate doesn't exclude it, because the standard lactate assay reads only L-lactate — D-lactate is invisible to it. But D-lactic acidosis needs a substrate-delivery setting (short bowel / malabsorption / bacterial overgrowth), and there is no such gut history here. Excluded on history, not on the lactate number.
B — Methanol / C — Ethylene glycol: toxic alcohols. Here the osmolar gap is normal and screens are negative. Caveat (teach it): in late toxic-alcohol presentations the parent alcohol may already be metabolized to acids, so the osmolar gap can normalize — a normal osmolar gap does not by itself fully exclude a late toxic-alcohol ingestion. The clinical picture (chronic APAP, malnourished woman, CKD), not the osmolar gap alone, points away from these.
Cited evidence
Fenves AZ et al. Increased anion gap metabolic acidosis as a result of 5-oxoproline: a role for acetaminophen. Clin J Am Soc Nephrol 2006 (22 patients, 82% women, most malnourished with kidney dysfunction). DOI
Liss DB et al. Clinical significance of 5-oxoproline following paracetamol exposure. Clin Toxicol 2013. DOI
Tailor P et al. Recurrent high anion gap metabolic acidosis secondary to 5-oxoproline. Am J Kidney Dis 2005. DOI
Educational content only. Not medical advice. Clinician verification required. No PHI. Synthetic, de-identified vignette.
As Dr. Mike Hansen, I've spent years teaching clinical medicine to over a million people — board-certified in pulmonary & critical care. Operation ADH turns that teaching into reps that keep clinicians sharp. See how I teach:
The worry I hear from clinicians isn't that AI is wrong. It's that leaning on it quietly makes us duller. So I built the opposite tool — one that keeps you the clinician who still thinks.
In the ICU, the cost of a missed pattern is real — and when AI hands you the answer before you've reasoned, the rep never happens. Operation ADH flips the order: you read the case and commit first, then the cited reasoning opens. You still get AI's speed; you just earn the thinking first.
Mike Hansen, MDCritical care intensivist · Founder, Operation ADH
The science
Studies show…
In critical care, the cost of a missed pattern is real. A growing body of research keeps pointing to one theme: how you use AI shapes whether it sharpens your judgment — or quietly erodes it. Operation ADH is built on the Cognitive Autonomy Framework — active recall, case-based reasoning, spaced repetition, teach-back.
Lancet Gastro. & Hep.2025 · observational study
−6.0%detection
Lean on it, and the skill can fade. Endoscopists detected fewer pre-cancers when working without AI after routine AI-assisted use — adenoma detection fell 28.4% → 22.4%.
Answers-first backfires. When the answer-first AI was taken away, learners scored 17% worse than peers who never used it — while a version built to make them reason first protected the learning.
