Handoffs That Don’t Drop Patients: I-PASS — IM Bootcamp

Why this session

Every transfer of responsibility is a place where information can die quietly. The evidence that structure fixes this is unusually strong for medical education: in the landmark nine-site study — run in pediatric residency programs, and since adopted far beyond them — implementing a structured handoff program built around the I-PASS framework cut medical errors by 23% and preventable adverse events by 30%, from 4.7 to 3.3 per hundred admissions.¹ Few teaching interventions anywhere in residency have numbers like that.

The failure mode this session attacks is specific. A new intern’s sign-out says “he’s fine, nothing to do” — a status report, true at 5 p.m. and useless at 2 a.m. A trained sign-out says “if this number does that, do this, and call them” — a contingency plan, which is a different genre of document. The difference is teachable in an hour, and the case below is built to make interns feel it from the receiving end first.

What interns leave able to do

Label every patient by illness severity — stable, watcher, or unstable — and defend the label.
Write if/then contingency plans a covering intern could execute at 2 a.m. without calling anyone.
Run receiver synthesis: read the plan back, ask until it is executable, and own what they accept.
Execute a contingency overnight and close the loop at morning sign-out.

The case

Disclosed one part at a time; interns commit before each reveal. This case deliberately starts on the receiving end — the fastest way to learn what a good sign-out contains is to inherit a bad one.

Part 1 — Receiving

5:10 p.m. You are the night intern. The day intern signs out twelve patients in eleven minutes. Four of them, verbatim:

Bed 6: “82-year-old woman, urosepsis day 3, doing great, on her antibiotics, nothing to do.”
Bed 9: “58-year-old man, rule-out — troponin negative times two, third one pending around 8, probably nothing. Chest-pain free.”
Bed 11: “74-year-old man, pneumonia, pressure’s been a little soft all day but okay — keep an eye on it, I guess?”
Bed 12: “90-year-old woman, failure to thrive. Family wants an update tonight — they’re anxious, they might page you.”

Your pager goes off mid-sign-out — Bed 2 wants a sleep aid. The day intern is visibly rushing to leave. Two co-interns are chatting at the desk behind you.

What do you do — in this moment, and about each of these four sign-outs?

Part 2 — The contingency fires

8:40 p.m. Bed 9’s third troponin posts: elevated, and rising. You open the sign-out in the EHR. Under Bed 9 it says, in full: “r/o MI. trops neg ×2.”

Now what? Walk your next thirty minutes, in order.

Part 3 — The morning bounce

6:45 a.m. You sign out to the day team: Bed 9’s positive troponin, the ECG, the cardiology consult, the heparin. The day intern asks: “What was his potassium and creatinine? Did anyone recheck the magnesium? And what’s the plan for the drip this morning?” You realize you never looked — the night got away from you after Bed 11’s pressure dropped at midnight. The “keep an eye on it” patient, it turns out, was septic; you gave two liters and called the rapid response at 1 a.m.

How do you handle this moment — and what should have been different at 5:10 p.m. yesterday?

The framework: I-PASS — and where SBAR fits

I-PASS is the structured handoff mnemonic developed by the I-PASS Study Group² and validated in the nine-site trial above.¹ It is widely used in internal medicine, and it names the five things a safe transfer of responsibility contains:

I — Illness severity. One word per patient: stable, watcher, or unstable. The word does real work — it tells the receiver where their night’s attention lives before any details arrive.
P — Patient summary. The one-liner, the working diagnoses, the treatment plan, and what actually happened today.
A — Action list. The to-dos, with times and owners: what is pending, what must be checked, what must be done.
S — Situation awareness & contingency plans. The if/thens: “if the third troponin is positive → ECG, repeat in three hours, page cardiology, start heparin per protocol once they weigh in.” “If the hemoglobin drops below 7 → transfuse one unit.” Trigger, action, who to call.
S — Synthesis by receiver. The receiver reads it back, asks questions, and confirms the plan. The handoff is not over when the giver stops talking; it is over when the receiver could run the night alone.

Where SBAR fits. Interns meet SBAR³ in this bootcamp’s cross-cover session, and the two tools are easy to conflate. The distinction is scope: SBAR is for escalating one problem, right now — a focused call about a single patient to someone who must act — while I-PASS transfers a whole service from one team to another, every patient, for hours. Some programs run their handoffs on SBAR or a local variant, and your program’s structure wins; but whatever the letters, the five I-PASS elements name what any handoff must contain — severity, summary, actions, if/thens, and a read-back. A handoff missing one of them is missing it regardless of the mnemonic.

One more element belongs in the sign-out document itself — the shared electronic list, whichever EHR module your program keeps it in: the flags that are social rather than physiologic. The family expecting a call tonight, the pain-management boundaries already negotiated, the patient who has been asking to leave. The night team inherits those conversations too, and Bed 12 is in this case on purpose.

Running the room

Minutes	Block
0–5	Frame: “sign-out is a procedure — today you learn the technique from both ends”
5–20	Part 1 — receiving sign-out
20–40	Part 2 — the contingency fires; board the I-PASS skeleton mid-discussion
40–55	Part 3 — the morning bounce + a live sign-out rep
55–60	Pocket card

Part 1 — what good looks like

The receiver owns the handoff as much as the giver. “Keep an eye on it, I guess” is not acceptable, and the intern who accepts it has co-signed it. The move is kind and relentless: “Soft how soft? What’s the number where I act, and what’s the action?” A sign-out you do not understand is a sign-out you have not accepted.

Severity labels for all twelve. Bed 6 is stable. Bed 9 is stable with a pending test — which means it needs an if/then, not reassurance. Bed 11 is a watcher: a pneumonia patient with a soft pressure all day is how 1 a.m. rapid responses get made, and he needs a vitals frequency, a fluid-and-antibiotics plan, and a number that triggers the call. Bed 12 is stable — and carries a communication contingency instead of a clinical one.

If/then plans — in the sign-out, executable by a stranger. The test for every contingency: could a covering intern who has never met this patient execute it at 2 a.m. without calling anyone? “If troponin #3 is positive → ECG, repeat troponin in three hours, page the cardiology fellow, heparin per protocol after they weigh in, update the sign-out.” If the plan is secretly “call me,” it is not a plan.

Protect the space. Sign-out is no-pager, no-chatter time: hand Bed 2’s sleep-aid page to a co-intern or defer it ten minutes — and note, from the cross-cover session, that a sleep-aid request gets evaluated, not reflexed, when its turn comes. Punctuality and protected attention at sign-out are not etiquette; they are the safety mechanism.

Read back the watchers. “Bed 11 — if the systolic drops under 100 I give 500 and call you or the rapid response; antibiotics due at 9. Did I get it?” That sentence is receiver synthesis, and it is the part new interns skip.

Part 2 — the 8:40 troponin, in order

See the patient — the cross-cover rule holds here too. Chest pain now? Vitals? Exam.
ECG, compared against the admission tracing.
Open the contingency plan — and let the room feel the payoff moment this case is built for: a good sign-out has already done your thinking. The one on file (“r/o MI. trops neg ×2”) has not, which is the lesson from the other direction.
Page cardiology with SBAR and a specific ask: “rising troponin in a chest-pain admission — do you want heparin started, and is telemetry enough or do you want him upgraded?”
Call your senior. A new positive troponin is a call, not a text.
Orders and loose ends: confirm telemetry, heparin per cardiology, repeat troponin timed — and the labs that travel with the drip: potassium, creatinine, magnesium. (Part 3 is waiting for whoever skips this line.)
Document, and update the sign-out in the EHR so the 6:45 a.m. version of you has it in one place.

The teaching point to say out loud: a contingency plan converts a surprise into a checklist. The adrenaline still arrives; you just don’t have to improvise inside it.

Part 3 — the morning bounce

Own it without theater. “I don’t have the potassium — I should have pulled it when the drip started. I’m checking it now and it’s on the list.” No groveling, no defending; fix it and move on. The intern who hides overnight uncertainty to look competent at sign-out is rehearsing the exact failure this session exists to prevent.

Synthesis runs both directions. The morning sign-out needs the same structure demanded at 5:10 p.m. — severity labels, what happened, new if/thens. An overnight event the day team never hears about did not, operationally, happen; the cross-cover session made the same point from the note’s side.

The systemic read. Bed 11’s 1 a.m. rapid response was predicted at 5:10 p.m. by a soft-pressure mention with no plan attached. The fix was never “be smarter at 1 a.m.” It was “demand the contingency at 5:10.” You cannot control the night; you can control the sign-out.

And the grace note: the rushing day intern is not a villain — every intern in the room will be that person some Friday. The norm to install is not shame; it is “two more minutes on the watchers, then go.”

The live rep (do not skip it): one intern re-signs-out Bed 11 properly against the boarded skeleton; another receives it and reads it back. Ninety seconds, and it encodes more than the whole discussion.

Common pitfalls to surface

“Doing great, nothing to do” sign-outs on anyone with an active process.
If/thens that are actually “call me” in disguise.
Accepting sign-out while half-listening to a pager.
Updates that stay verbal overnight — the sign-out in the EHR is the team’s shared brain; update it or lose it.
Hiding uncertainty at morning sign-out to look competent.

Key teaching points

Every patient gets a severity label — stable, watcher, unstable — and the watchers get plans, not vibes.
If/then, executable by a stranger: trigger, action, who to call. If it needs you, it isn’t finished.
The receiver owns the handoff too. Ask until you could run the night alone, then read it back.
The sign-out in the EHR is the team’s shared brain — keep it current in real time, clinical flags and social ones.
Overnight events only count if the day team hears them.

Pocket card

Carry this

Stable / watcher / unstable — label every patient.
Watcher = trigger + action + who to call. In the sign-out.
“Could a stranger run this at 2 a.m.?” If not, rewrite it.
Receiver synthesis: read it back before you accept it.
SBAR escalates one problem; I-PASS hands over a service.
If it happened overnight, it’s in the morning sign-out.

Variations

The social contingency. Add “watch out for Bed 4 — the family is angry” to Part 1 and make the room write it as a real plan: who calls, what has been promised, where the boundary is. Handoffs carry communication plans too — a bridge to Day 2’s de-escalation session.
The final handoff. For a stronger group: the discharge summary is a handoff to a doctor you will never meet, and it fails the same ways — status report instead of contingency plan, pending results unowned. Have interns name the if/thens a discharge summary should carry (pending cultures, titration plans, who follows the incidental finding).
The SBAR-shop version. If your program hands off on SBAR or a local variant, run the same case and map each beat onto your structure — then ask which of the five I-PASS elements your format carries implicitly and which one someone must remember to add. (It is almost always the contingency plans or the read-back.)
Second iteration (the reunion session): interns bring a real if/then that fired, de-identified, and the group critiques it against the “stranger at 2 a.m.” test.

Notes

The case above is a fictional composite; no patient in it is real. The evidence for why handoffs sit at the center of an intern bootcamp is laid out in the bootcamp introduction.

Sources

Starmer, A. J., Spector, N. D., Srivastava, R., West, D. C., Rosenbluth, G., Allen, A. D., Noble, E. L., Tse, L. L., Dalal, A. K., Keohane, C. A., Lipsitz, S. R., Rothschild, J. M., Wien, M. F., Yoon, C. S., Zigmont, K. R., Wilson, K. M., O’Toole, J. K., Solan, L. G., Aylor, M., … Landrigan, C. P., for the I-PASS Study Group. (2014). Changes in medical errors after implementation of a handoff program. New England Journal of Medicine, 371(19), 1803–1812. https://pubmed.ncbi.nlm.nih.gov/25372088/ ↩₁ ↩₂
Starmer, A. J., Spector, N. D., Srivastava, R., Allen, A. D., Landrigan, C. P., Sectish, T. C., & the I-PASS Study Group. (2012). I-PASS, a mnemonic to standardize verbal handoffs. Pediatrics, 129(2), 201–204. https://pubmed.ncbi.nlm.nih.gov/22232313/ ↩
Haig, K. M., Sutton, S., & Whittington, J. (2006). SBAR: A shared mental model for improving communication between clinicians. The Joint Commission Journal on Quality and Patient Safety, 32(3), 167–175. https://pubmed.ncbi.nlm.nih.gov/16617948/ ↩

This page is a teaching scenario for facilitated small-group education. Every patient in it is fictional. It is not clinical guidance for the care of any actual patient, and it does not replace your hospital’s handoff policies or your program’s sign-out structure — localize before teaching. Last reviewed July 2026.

Handoffs that
don’t drop patients