What You Made of It

Why a brief can include every detail and still leave the next person working from the wrong picture.

A man with light skin, hair tied back and round glasses holds out a tall stack of chart pages marked every fact, while his other hand keeps a small blue folded note against his chest. A Black woman with deep brown skin, locs and a shirt and tie reaches toward the stack with open hands, and the blue words the verdict float in the gap between them.

"OK, this is our sickest patient, and he's full code."

A four-year-old boy has pneumonia. The next doctor needs to check on him around midnight.

This is a teaching sample from a paper in Pediatrics, not a recorded exchange. The authors have put the patient's condition and the next check into a short handoff.

The opening sentence tells the receiver how to read everything that follows. The boy's age cannot do that. Neither can the diagnosis on its own.

You can pass on everything you wrote down and still leave someone working from the wrong picture. That does not always mean you forgot a detail. Sometimes you left out what you made of the details.

What was missing

A handoff passes information along. From there, it is easy to assume that more information makes a better handoff.

But the designers of I-PASS, a handoff method for hospital residents, began with particular absences. In their 2012 Pediatrics paper, they described what faculty observers found "most commonly absent" from resident handoffs: an assessment of illness severity, contingency planning, and a read-back by the receiving resident.

Those are different from missing facts about the patient.

Duty-hour restrictions had increased the number of handoffs. The Joint Commission required a standardized approach. The problem needed a structure people could use while passing responsibility to someone else.

I-PASS gave that structure five parts. I stands for illness severity. P is patient summary. A is action list. The first S is situation awareness and contingency planning. The second is synthesis by receiver.

Four kinds of information

The elements observers found most often missing sit outside the record.

The record holds the patient summary and action list. The boy has a left lower lobe pneumonia and hypoxia. The incoming doctor should look in around midnight, check that his vital signs and oxygen saturation remain stable, and check his blood culture.

That gives the receiver a condition to understand and work to do. It also gives that work a time. Those details matter; the other categories do not replace them.

But a chart can hold every fact and the sender's judgment can still be missing.

I call the sender's assessment of illness severity the verdict. I-PASS asks for stable, watcher, or unstable. In the sample, the sender begins by identifying the sickest patient. That judgment tells the next doctor how much attention the record deserves.

The forecast names what could change and what to do if it does. In the sample, worsening breathing should prompt another chest radiograph to check for an effusion. The sender has passed along a response to a possible change, rather than leaving the next doctor to infer it.

The proof comes from the receiver. The receiver repeats the patient summary, the midnight check, and the response to worsening breathing. Only after that does the sample end with "I think I have it."

Those words alone would prove little. The summary before them lets the sender hear what arrived, while there is still time to correct it.

What the results support

A hospital handoff is not a business brief. And the later study cannot tell us that these three missing elements produced the improvement.

The 2014 study by Starmer and colleagues in the New England Journal of Medicine tested a bundle across nine pediatric residency programs. It combined the mnemonic with communication training, faculty development and observation, and a sustainability campaign.

It compared outcomes before and after implementation. It did not separate the parts.

Across 10,740 admissions, medical errors fell from 24.5 to 18.8 per 100 admissions, a 23% reduction. Preventable adverse events fell from 4.7 to 3.3 per 100 admissions, a 30% reduction.

Nonpreventable adverse events showed no significant change: 3.0 before and 2.8 after. Errors fell significantly at six of the nine sites.

Oral handoffs took 2.4 minutes per patient before and 2.5 after. That difference was not significant.

The harm that changed was the preventable kind. The study found no significant added cost in handoff time. That fits better communication, but it does not identify which part of the program mattered most.

My expectation is that the verdict, forecast, and proof account for much of the benefit. I would be wrong if handoffs containing only the record, under the same training and support, cut errors just as much.

For now, I see a reason to test those missing parts, though the study does not promise the same results elsewhere.

More leads

A brief saying the client wants more leads leaves the SEO team without a clear optimization goal and the PPC team without a CPA target.

It also leaves the verdict unstated: how sure are we that more leads is the real goal? The forecast is missing too: what would make us change course? Until the person doing the work says it back, the proof is missing.

Writing the intent down keeps it available. The say-back check against that written intent tests whether it survived the handoff.

Keep the judgment visible. Keep the conditions for changing course visible. Do not let the work move on until the receiver can explain it accurately.

If they cannot, the work goes back upstream.

The sample handoff begins by naming the sickest patient.

When I find what the sender knew and the page left out, I write it down and email the notes worth sharing.