Performance here means. In a hospital, performance is patient outcomes and safety, the clinicians who don't burn out, and the work done under heavy regulatory load — not a generic engagement score. And the setting subdivides: what 'good teamwork' means in the OR isn't what it means in the clinic.

Here is a fact that should bother anyone who sells a one-size-fits-all team survey. Of every field that takes teamwork seriously, the one that takes it most seriously — where a bad handoff kills someone by lunchtime — looked at the standard organizational-behavior instruments, the engagement scores and the climate batteries, and quietly refused them. Then it built its own, by the dozen: the Mayo High Performance Teamwork Scale, the Ottawa crisis-resource-management rating, and a long shelf of others. Medicine did not adopt the generic tool. It decided the generic tool was measuring the wrong room.

That's the whole argument of this guide, and a hospital is where it's least deniable. Take its sharpest room, the operating theater: good teamwork there isn't morale, it's whether someone caught the thing that was about to go wrong — and the outcome data is unusually clean, the patient recovered or didn't. A survey built for a software team or a sales floor doesn't even ask about the conditions that produce that. But here's the part that should really unsettle the one-size-fits-all instinct: "hospital" is already too coarse. The right tool differs between the OR and the ward and the outpatient clinic — the setting keeps subdividing under you, and the generic battery was at the wrong altitude from the start.

So what does bind performance here, and why is it invisible to the off-the-shelf battery?

Capability in the OR is not credentials on a wall, and it isn't years. It's two layers — the technical skill, and the non-technical skills the surgical world had to name explicitly because the technical ones weren't enough: situational awareness, decision-making, communication under load. These get assessed the only valid way, on representative tasks, in simulation, against what actually arrives — not by tenure, which the expertise research already told us predicts almost nothing.

Alignment is where the OR gets its reputation, and it's the condition the generic survey is least equipped to see. It lives in shared mental models and closed-loop communication, and most of all in whether a nurse will speak up to a surgeon when something looks wrong. That is not a personality trait; it's a condition of the team, and it has a name — psychological safety — and it is not an accident that the research establishing it was done in hospitals, watching which nursing units logged more medication errors and discovering, against every intuition, that the units with the best climate logged the most — not because they erred more, but because they were safe enough to record what went wrong. The surgical safety checklist is, underneath the clipboard, an alignment device: it forces the shared model and licenses the junior person to talk. The board's engagement score will never tell you whether the most important sentence in the room — "stop, I think this is wrong" — can actually be said.

Support is the condition medicine learned the hard way and named for everyone else. When the field finally confronted its error rate, the conclusion was not that it had bad people; it was that good people were working inside systems that manufactured the errors — fatigue, broken handoffs, thin staffing, equipment that invited the mistake. Look at the environment first. The search for a bad apple is usually a way of not fixing the system that grew the mistake.

Motivation is the strange one. These are some of the most driven people in any workforce — the problem isn't a deficit of will, it's the overdraft. Burnout is the hospital's defining condition crisis, and not only in the OR — and the instrument the whole world now uses to measure it came out of the helping professions in the first place. The demand is relentless and the resources don't always refill; measure the will and you'll miss the depletion.

Not one of those four is captured by a generic engagement number, and every one of them is something medicine had to build a specific instrument to see. Which returns us to the fact we started with, now pointing the other way. If the most measured, most consequential teamwork on earth required medicine to throw out the universal survey and build its own — a different one for the OR than for the clinic — then the quiet belief every other leader carries, that their setting is different and the off-the-shelf tool doesn't quite fit, isn't a bias to be corrected. It's the hospital's lesson, arriving everywhere else. The tools miss what the setting makes specific. Medicine just had the least room to pretend otherwise.

---

This guide is grounded in the publish-worthy findings feed and the cited sources below. Where the underlying literature reports specific effect sizes that haven't been verified against the primary text, this guide states the direction of the finding and not the number.

Sources

1. Marlow et al. (2018), Small Group Research; Kash et al. (2018), Health Services Insights — healthcare built its own team measures (Mayo High Performance Teamwork Scale, Ottawa CRM rating); the right tool differs OR vs. clinic
2. Edmondson — psychological safety; the foundational studies were in hospital teams (units with better climate logged more detected errors, because it was safe to record them)
3. Haynes et al. (2009), NEJM — WHO Surgical Safety Checklist (a shared-model / speak-up alignment device; associated with fewer complications)
4. Institute of Medicine, To Err Is Human (2000) — error as a system property, not a bad apple; environment first
5. Maslach — Maslach Burnout Inventory (originated in the helping professions; clinician burnout)