At the end of cardiac surgery, Great Ormond Street Hospital for Children had what looked like a communication problem.

But communication was only one thing moving.

The child, ventilation, monitoring, medications, equipment, accumulated knowledge, risk, and responsibility all had to move from the operating theatre to intensive care without the system losing stability.

Calling it a communication problem was not exactly wrong.

It was too flat.

The label did not show where the transition began or ended. It did not reveal which tasks depended on one another, when authority changed hands, or how one failure could propagate into the next.

The team was trying to solve a cloud.

It needed to give the problem a shape.

A Hospital Looks at a Pit Stop

Physicians Martin Elliott and Allan Goldman recognized a similar underlying structure while watching a Formula 1 race.

A pit crew also had to bring multiple specialists into a constrained space, perform interdependent tasks under time pressure, anticipate failure, and move a complex system through a brief, high-risk transition without leaving anything unresolved.

The useful comparison was not child to car.

It was transition to transition.

The Great Ormond Street team visited Ferrari and worked with human-factors specialists and aviation training captains. They filmed clinical handovers and invited Formula 1 observers to review the footage. A dance choreographer helped the team reconsider where people stood and how they moved around one another.

Each observer made a different part of the process visible.

The resulting protocol separated the transfer into distinct phases: preparation before arrival, equipment and technology transfer, an uninterrupted information handover, and then group discussion and planning. The anesthetist coordinated the process until responsibility was explicitly transferred to the intensivist. Equipment stability was confirmed before the information briefing began.

This separation mattered. The receiving team no longer had to absorb the operative story while simultaneously connecting equipment, resolving alarms, and determining who was directing the room.

The intervention did not add another reminder to communicate better.

It redesigned the conditions under which communication had to survive.

In a prospective study of 50 surgical handovers—23 before the intervention and 27 after it—the mean number of technical errors fell from 5.42 to 3.15. Mean information omissions fell from 2.09 to 1.07. Average handover time decreased from 10.8 minutes to 9.4 minutes.

Before the new protocol, nine of 23 observed handovers had more than one technical error and more than one information omission. Afterward, three of 27 did. Regression analysis found a statistically significant reduction in technical errors.

This was a small, single-center before-and-after study. It did not prove that a pit-crew model would improve every clinical process.

The narrower conclusion is more useful.

The handover became safer not because the team looked harder at the same description of the problem. It became safer because another system exposed relationships that the description had hidden.

The problem had not become simple.

It had become legible.

The Observer Is Part of the Model

Why could a Formula 1 team and a choreographer contribute to a pediatric cardiac handover?

Not because medicine lacked expertise.

Because expertise is also a vantage point.

The sending team sees what occurred during surgery. The receiving team sees what must be stabilized and understood next. A Formula 1 observer is trained to notice task sequence, interference, role boundaries, and recovery from failure. A choreographer attends to position, movement, timing, and the use of shared space.

Each view preserves some relationships and compresses others.

That is true of every representation. A clinical note, dashboard, map, workflow diagram, org chart, and garden plan are useful because they reduce reality. The danger begins when we forget that reduction occurred and mistake the projection for the object.

Giving a problem shape is therefore not the same as declaring its final form.

It means making one model explicit enough that we can inspect it, turn it, and compare it with another.

Once the model can be rotated, the observer becomes part of what is being examined.

When a Yellow Leaf Became a Finding

The hospital changed its model by looking outward.

I made a smaller version of the same change in my garden.

For a long time, I understood a garden primarily as an arrangement of plants. I thought about placement, color, sunlight, watering, and whether the whole composition looked healthy.

I became a better gardener when a yellow leaf stopped being a visual defect and became a finding.

Was the pattern isolated or systemic? Sudden or gradual? What had changed before it appeared? Was it spreading? Had the last intervention improved the condition or created another one?

The leaf was no longer an answer.

It was evidence inside a trajectory.

The plant had not changed.

My model had.

Tool hygiene changed in the same way. Cleaning pruners had once felt like fussy gardening etiquette. Once I understood that a blade could carry pathogens from infected tissue to another cut, the practice acquired a clearer structure.

The garden was not a sterile field, and cleaning a pruning tool was not operating-room technique. But a useful relationship survived the transfer between domains: when a tool crosses a protective boundary, what is on that tool matters.

Grafting also became easier to understand. A successful graft depends on complementary cuts, contact between the vascular cambium of the scion and rootstock, stabilization of the union, protection from moisture loss, and enough time for the tissues to heal into a functioning vascular system.

It is not surgery.

But it has a procedural shape that clinical training makes easy to recognize:

Prepare. Align. Secure. Protect. Reassess.

Nursing made me a better gardener not because it supplied horticultural answers, but because it changed what counted as information. It trained me to see an individual condition inside an environment, distinguish a symptom from a cause, intervene with purpose, and reassess rather than assume that completing a task produced the intended result.

How to Give a Problem Shape

When a problem remains stubbornly abstract, I now try to define five things.

  1. The boundary. Where does the problem begin and end? What belongs inside the system, and what is merely adjacent to it? At Great Ormond Street, the relevant boundary extended from preparation before the patient arrived in intensive care through the explicit transfer of responsibility.
  2. The moving parts. What is actually being transferred or transformed? In the handover, the moving parts included the child, physiological support, equipment, medications, information, risk, attention, and authority.
  3. The relationships. Which parts depend on one another? What must occur in sequence, what can happen in parallel, and where can one failure propagate into another? Equipment needed to be stable before the briefing could receive undivided attention. Discussion needed to follow the structured transfer rather than interrupt it.
  4. The observer. Where am I standing, and what does that position make easy to see or ignore? The operating team, ICU team, Formula 1 observers, and choreographer did not see competing realities. They saw different projections of the same one.
  5. The analogous system. Where else does the same structure appear? The useful match is not shared vocabulary or surface appearance. It is a shared pattern of boundaries, dependencies, transitions, constraints, or failure modes.

The contents changed in my garden, but the structure of the method did not.

The boundary became an individual plant and its environment. The moving parts became water, nutrients, pathogens, tissue, and time. The relationships became exposure, response, recovery, and recurrence. The observer shifted from someone arranging a composition to someone following a living trajectory.

Once those elements are visible, generic labels begin turning into actionable descriptions.

“Better communication” becomes: protect the briefing from competing technical tasks and name the moment responsibility changes hands.

“A plant that is not doing well” becomes: a progressing pattern that appeared after a specific environmental change and did not improve after the last intervention.

Definition does not solve the problem.

It determines what kind of solution could belong.

Return Before the Analogy Becomes an Identity

Analogy can reveal structure.

It can also smuggle assumptions across a boundary where they do not belong.

A hospital is not a racetrack. A child is not a machine. A garden is not a clinical unit. A plant is not a person.

The value lies in transferring a relationship that holds, not forcing every feature of one domain onto another.

The Great Ormond Street team could borrow role clarity, choreography, sequencing, workspace design, and anticipatory planning from Formula 1. It could not borrow unlimited rehearsal, interchangeable equipment, or the assumption that every contingency could be enumerated.

The clinical system contained irreducible biological variability and far fewer opportunities for repetition.

The same restraint applies in the garden. Clinical attention can sharpen questions about trajectory, contamination, intervention, and reassessment. It does not turn horticulture into medicine.

A good analogy produces a question that can be tested.

It does not produce a conclusion that must be obeyed.

Definition Before Solution

We often rush from naming a problem to prescribing an intervention.

More staffing. Better communication. More training. A new tool. Another policy. More water. Less water.

Those may be answers, but they arrive before we have established what kind of object we are trying to change.

Sometimes the most productive first intervention is definition.

Draw the boundary. Name what moves through it. Map the dependencies. Locate the observer. Then look for another system in which the same relationships are easier to see.

Look from the receiving side instead of the sending side. Look at the handover as choreography instead of conversation. Look at the garden as a collection of individual trajectories instead of a single arrangement.

Formula 1 did not make a pediatric handover less complex. Nursing did not make a garden less biological.

Each supplied an angle from which consequential relationships became visible.

A change in perspective does not change reality.

It changes which parts of reality become actionable.

Related: Technology Doesn’t Replace the Work. It Moves the Bottleneck. — on what must survive when work crosses a boundary.

Sources and further reading

Paul Peck