Turn on the Lights Podcast

The episode surfaces a handful of genuinely useful observations - notably that healthcare punishes errors while tolerating bad behaviour, and that improving care continuously generates new safety problems - but the first third is dominated by biography and career origin stories, and most of the conceptual ground (latent conditions, systems vs. blame, aviation analogies) is decades-old patient safety canon rather than fresh insight.

“I think healthcare is very odd in responding very badly and oppressively punitively, if you like, to people making mistakes and being far too tolerant of people behaving badly or disrespectfully or being excessively hierarchical and dominant, which is in itself dangerous”

“Sepsis was not a safety problem 15, 20 years ago. Now it is. But that's because people have got better at managing sepsis. So now it's a safety problem”

The inversion - healthcare is intolerant of error yet tolerant of genuinely dangerous behaviour - is a crisp and underused framing, and the 'moving target' argument about improvement creating new hazards is non-trivial; but the core frameworks (Swiss cheese, latent conditions, psychological safety, aviation analogy) are well-worn, and the episode never really argues against any mainstream patient safety orthodoxy.

“Very tolerant of poor behavior, as you might say, and intolerant of error. That's a rather. I'm caricaturing it. Yeah. Rather curious situation”

“as healthcare improves, we acquire new safety problems for a variety of reasons. We have new technologies, we have new risks”

Vincent is a genuine practitioner-researcher with decades of empirical fieldwork - he ran fracture-miss studies, embedded observers inside operating theatres for a decade, and co-authored with Amalberti - making him a credible primary source rather than a recycled conference speaker; the interview does not fully exploit his depth, leaving his most technical contributions largely undescribed.

“we became quite unpopular. My friend Peter Driscoll and I was an emergency doctor for discovering that in the emergency department in central London, junior doctors missed a third of serious fractures”

“we would have observers in the operating theater. And the observers often would be a psychologist and also often a young surgeon, very experienced”

There are some concrete anchors - the one-in-three missed fracture rate, named researchers (Reason, Amalberti, Leape), the Harvard Medical Practice Study, the WHO surgical checklist - but the episode is predominantly qualitative and anecdotal, with no outcome data, timelines, or scale metrics attached to any intervention or improvement claim.

“in the emergency department in central London, junior doctors missed a third of serious fractures. I broken hip and things like that”

“This was before Lucian Leap and everybody's, before the Harvard Medical Practice study came out”

The host lands a few good follow-up moves - pressing Vincent to name a specific bad behaviour, asking what pre-Reason mental models looked like - but frequently injects his own opinions and career anecdotes, frames questions in admiring rather than probing terms, and relies on a pre-scripted closing question the show uses for every guest, leaving several substantive claims (e.g. on payment incentives, on curriculum change) unchallenged and underdeveloped.

“What was the pre existing theory you had in your head about errors? Let's talk about reasons. Approach. But what before reason?”

“Are you willing to say what you mean by a bad behavior? An example of that. And then what is its connection to errors and safety?”