On paper, shadowing sounds reasonable.
A new APP follows experienced people, watches enough cases, picks up the flow, and gradually starts doing more. No one has to build a formal curriculum. No one has to stop and define every standard. The team just teaches in real time.
That can work for exposure. It does not work well as a full training system.
Because shadowing alone creates a hidden problem: the learner sees a lot, but the team never has to agree on what “good” actually looks like.
Why shadowing feels better than it is
Shadowing survives because it gives everyone the feeling that training is happening.
The new APP is present. They are hearing the language. They are observing workflow. They are slowly taking on pieces of the job. From the outside, that looks like progress.
But observation is not the same as a defined progression path.
In high-trust environments like cardiac surgery, there are too many subtle decisions for people to infer correctly just by being nearby. Without an explicit framework, the learner starts building a picture from fragments.
What shadowing alone quietly creates
1. Trainer drift
One trainer emphasizes anticipation. Another emphasizes speed. Another explains almost nothing and expects the learner to adapt. Another corrects only when something goes obviously wrong.
If the team has not agreed on the standard, the learner is not being trained into one system. They are being pulled through multiple personal styles.
2. Subjective readiness
At some point someone asks, “Do we think they are ready?”
That question becomes hard to answer when the training path has been mostly informal. People fall back on impressions instead of criteria. Confidence, personality, and communication style start to weigh too heavily.
The result is predictable: one person thinks the APP is ahead, another thinks they are risky, and the APP is stuck in the middle trying to decode what is actually expected.
3. Slower trust with surgeons
Surgeon trust is built on consistency. Not just effort. Not just potential. Consistency.
If a first assist looks sharp with one trainer and uncertain with another, surgeons do not experience that as “normal learning.” They experience it as variable reliability.
That slows down the trust curve, even when the learner is capable.
4. Hidden friction for the team
Informal shadowing also taxes the senior team more than leaders realize. Trainers keep reteaching the same points in different ways. Expectations drift. Frustration builds because people think they are handing off a learner at one level, but the next trainer receives someone at another.
That is not just inefficient. It creates avoidable strain inside the team culture.
What a stronger model looks like
The answer is not to eliminate shadowing. Shadowing is useful. The answer is to stop pretending it is enough on its own.
A stronger first-assist training system keeps observation, but adds structure around it:
- clear milestones for what the APP should be able to do by stage
- shared trainer standards for what “acceptable” looks like
- specific feedback loops instead of vague post-case impressions
- defined signs of readiness, not just gut feel
- repeatable checklists or protocols that reduce schedule-dependent variation
That shift matters because it moves training from personality-dependent to system-dependent.
The deeper issue
When teams rely on shadowing alone, they are usually carrying one hidden assumption: that strong people will piece the job together if you expose them long enough.
Sometimes they do. But that is not a training strategy. That is a survivor bias strategy.
It rewards the people who can tolerate ambiguity longest, not necessarily the system that develops people best.
A better question for leaders
Instead of asking, “Who should they shadow next?” ask this:
What exactly are we trying to help them become more reliable at this week?
That question forces specificity. And specificity is what makes training faster, safer, and easier to scale.
Want a cleaner first-assist training starting point?
Start with the Cardiac Surgery First-Assist Protocol Template if you want a lightweight asset for standardizing preparation, anticipation, and OR-readiness expectations.