A surgeon is about to make the first incision. The patient is anaesthetised. The team is assembled. The lights are bright, the instruments laid out, and everyone in the room is highly trained, highly focused, and highly confident.

​

Only no one has confirmed which leg they are operating on.

​

This is not a hypothetical. Wrong-site surgery (operating on the wrong limb, the wrong organ, the wrong patient) happens in real hospitals, performed by competent surgeons, more often than anyone in the profession would like to admit. It is not caused by incompetence. Rather by the fact that skilled professionals, working under pressure, routinely skip steps they consider obvious.

​

In 2007, surgeon Atul Gawande and a team at the World Health Organisation set out to measure how often things went wrong in surgery. Not dramatic, headline-grabbing errors, just the routine kind that could have been easily prevented. The base rate they found was startling: across eight hospitals in eight countries, major complications occurred in 11% of surgical procedures. One in nine.

​

Why had this been happening? Up to then, most individual surgeons estimated their own complication rate as well below that number. Afterall, the vast majority felt certain that they would not make the kind of simple errors their average colleagues do. This gap between what actually happens and what the individual in charge believes happens is one of the most dangerous features of overconfidence. It is not that surgeons lack knowledge. It is that the feeling of expertise creates a sense of reliability that the evidence does not support. And that certainty is precisely what allows the error to happen unchecked.

​

Gawande's team also noticed a pattern in the errors. They were not failures of skill or judgment but failures of omission: steps skipped, checks not performed, information not shared, antibiotics not given before incision, allergies not confirmed, the surgical team not introduced to each other by name. Each omission was individually trivial. But collectively they were very damaging. And they occurred in one every nine surgeries.

​

The surgeons' instinctive response to these findings was revealing. Most were framing the problem as a question about themselves: am I skilled enough to not need a checklist? Framed that way, the answer was obviously yes.

​

But Gawande's team reframed the question. The issue was not whether any individual surgeon was good enough, but whether the process surrounding the surgery was reliable enough to catch the errors that any individual, however skilled, would inevitably make under pressure. This was not a question about competence. It was a question about system design.

​

This reframe meant that the checklist Gawande's team introduced was no longer a crutch for inadequate surgeons. It was a process tool, the same kind of tool that aviation had been using for decades, not because pilots were incompetent but because the consequences of omission were too high to leave to individual memory. Resistance faded. Surgeons started using it.

​

The checklist itself was almost absurdly simple. A single sheet of paper with 19 items, completed in a few minutes at three points during every operation: before anaesthesia, before the first incision, and before the patient leaves the operating room. Confirm the patient's identity. Confirm the site. Confirm the procedure. Has the anaesthesia equipment been checked? Are antibiotics on board? Does anyone in the room have concerns they have not voiced?

​

Each item functions as a tripwire: a predefined condition that must be met before the process continues. If the patient's identity has not been verified, you stop. If antibiotics have not been administered, you stop. The stopping decision is built into the process, not left to individual judgment under pressure.

​

The results, published in the New England Journal of Medicine in 2009, were unambiguous. Across 7,688 patients in hospitals in 8 countries (Toronto, New Delhi, Amman, Auckland, Manila, Ifakara, London, Seattle), complications fell from 11% to 7%. Deaths fell from 1.5% to 0.8%, a reduction of more than 40%. The results held across both high-income and low-income settings. Four countries (UK, Ireland, Jordan, Philippines) established nationwide programmes immediately. The checklist has since been implemented on six continents.

​

Later research revealed that the checklist's effectiveness depends heavily on how it is used. When teams treat it as a bureaucratic formality rather than a genuine pause for communication, it loses its power. Gawande himself has acknowledged this: the checklist is not a solution, it is a catalyst for the behavioural and cultural changes that produce the solution.

​

 

 

What to notice

​

Several of the traps from earlier in this masterclass are at work here simultaneously, which is what happens in real life. The surgeons' overconfidence made the problem invisible: each one believed their own complication rate was well below the 11% base rate, and that certainty was precisely what prevented the problem from being addressed. The question itself was narrowly framed "am I skilled enough to not need a checklist?", which excluded the more useful question about whether the process was reliable enough to catch inevitable human errors. And the normalisation of deviance meant that skipping routine checks had become standard practice, measured against recent behaviour rather than the original standard, so nobody noticed the drift.

​

The tools that addressed these traps didn't just target single traps. The outside view made the gap between belief and reality undeniable. The reframe shifted the question from individual competence to system reliability. The checklist created structural stops that removed the decision to check from the person most likely to skip it. And the explicit prompt "does anyone have concerns they have not voiced?" built psychological safety into the process rather than hoping individuals would overcome the hierarchy on their own.