MENTORSHIP IN PRACTICE

Practical strategies for developing better clinicians

A Highbar Clinical Education Series

Your Learner Made a Mistake. Don’t Correct It Yet.

The visible error tells you what happened. Better mentorship starts by figuring out why.

A resident forgets the neurological screen.

You notice immediately.

So you say:

“You missed the neuro screen. Make sure you do that next time.”

Clear feedback. Easy correction.

Except you may have corrected the behavior without identifying the problem.

Maybe the resident does not know when a neurological screen is indicated. Maybe they know perfectly well but become overloaded during complex evaluations. Maybe they recognize the need for it but cannot perform the screen efficiently. Maybe they were uncertain and waited for you to tell them what to do.

Same mistake. Completely different developmental problems.

Some of the most useful clinical mentorship strategies start here: if the developmental problem is different, the mentor response should be different too.

The visible mistake is data. It is not the diagnosis.

Before you judge the gap, know the target

Before deciding that a learner is struggling, ask two questions:

• What should competent performance look like here?
• How independently should this learner reasonably perform at this stage?

The same mistake can carry very different meaning depending on when it occurs.

A resident early in training may reasonably need substantial help organizing a complex evaluation. A resident approaching graduation should be functioning with much greater independence.

The behavior may look identical.

The developmental significance is not.

This is also where individualized mentorship can get confused with individualized standards.

They are not the same thing.

We can change how we teach. We can provide more structure. We can temporarily simplify complexity. We can step back as capability develops.

But the expected standard for safe, competent practice still needs to remain clear.

Adapt the support. Keep the standard clear.

The mistake is only the starting point

In the first article in this series, The Best Mentors Don’t Always Give You the Answer, we centered on one question:

What does this learner need from me right now?

To answer that well, we first have to understand what is actually limiting the learner.

That means moving beyond the reflex to see a mistake and immediately correct it.

Observe: What actually happened?

Not your interpretation. Not your judgment about the learner. What did you actually see or hear?

Locate: Where is the performance gap showing up?

Clinical reasoning? Communication? Examination? Intervention? Professionalism?

Hypothesize: What is most likely limiting performance?

Respond: What mentor action best matches that problem?

Reassess: Did performance change?

If it did not, reconsider your hypothesis.

The goal is not to perfectly diagnose a learner after one interaction. It is to create a useful working hypothesis that leads to a better teaching decision.

Where the problem shows up is not why it is happening

This distinction is the heart of educational diagnosis.

Suppose a resident struggles during the examination.

That tells us where the problem is showing up.

It does not tell us why.

Where the problem shows up is not the same as why it is happening.

If we confuse the two, mentorship gets imprecise very quickly.

A reasoning problem gets treated with more information.

An execution problem gets talked about instead of practiced.

A capable but dependent learner gets even more guidance.

Then we wonder why performance does not change.

Match the mentor move to the problem

Once you have a reasonable hypothesis, the mentor response becomes much clearer.

The goal is not to become the mentor who always questions, always teaches, or always lets the learner struggle.

The goal is to choose the mentor move that matches the developmental problem.

Your mentor move is also a test

There is another reason to be deliberate about your response.

The learner’s response gives you more information.

Suppose you think the problem is knowledge. You teach the concept and performance immediately improves.

That supports your hypothesis.

But what if you teach it clearly and nothing changes?

Knowledge may not have been the real constraint.

Or suppose you think a learner lacks confidence. Instead of reassuring them, you step back and require an independent decision.

They make an excellent decision.

You just learned something important.

They may not have needed more instruction. They may have needed less help.

If the learner responds as predicted, your hypothesis gains support.

If they do not, revise it.

We are comfortable doing this clinically: the patient did not respond as expected, so our initial hypothesis may have been wrong.

Mentorship deserves the same humility.

One bad performance is not a personality trait

A learner struggles once, and the performance can quickly become a description of the person.

• “They’re weak at examination.”
• “They don’t have good clinical reasoning.”
• “They aren’t confident.”
• “They’re not prepared.”

Be careful.

One performance is information.

It is not necessarily a pattern.

People have difficult days. Some patients create unusually high cognitive demand. Performance can change because someone knows they are being observed. A learner can understand something and still make an isolated mistake.

Before turning one performance into a stable judgment:

• Ask what they were thinking.
• Observe again.
• Invite self-assessment.
• Change the context.
• Look for recurrence.

And describe the behavior instead of labeling the person.

One is a judgment.

The other is observable, testable, and developable.

Confidence is not competence

Mentors can also be fooled by presentation.

A learner sounds certain. They speak quickly. They answer without hesitation. They appear decisive.

None of that guarantees good reasoning.

Likewise, hesitation does not automatically indicate poor reasoning.

A cautious learner may be considering several competing possibilities before committing. A confident learner may have prematurely locked onto the first explanation that came to mind.

Confidence is not competence. Hesitation is not incompetence.

Evaluate the thinking, not simply how confidently the answer is delivered.

• Why?
• What evidence supports that?
• What alternatives did you consider?
• What would change your mind?

That is where capability becomes visible.

Stop correcting everything

Experienced clinicians usually see more than the learner does.

During one evaluation, you may notice eight things that could be better.

The temptation is to correct all eight.

Don’t.

If everything becomes the priority, nothing is.

Find the developmental target with the highest value right now.

Maybe the learner performed several examination techniques imperfectly. But perhaps the larger problem is that none of the tests they selected were likely to change management.

Technique matters.

But examination prioritization may matter more.

Selective feedback is not ignoring problems. It is prioritizing learning.

Match the teaching method to the learning problem

Learners differ, but we do not need to reduce clinicians to “visual,” “auditory,” or “kinesthetic” categories to individualize mentorship.

A more useful question is:

What teaching method best fits this learning problem?

• Can’t conceptualize it? Map or diagram it.
• Can’t execute it? Demonstrate and practice.
• Can’t organize the reasoning? Talk through the case and require prioritization.
• Can’t generalize? Change the case and test transfer.
• Can’t evaluate their own performance? Require self-assessment before feedback.

Individualization should improve the pathway to learning.

It should not change the expected destination.

Ask questions that reveal the level of thinking

Even good questions can give mentors false confidence about what a learner actually understands.

Take cervical radiculopathy.

Same clinical topic.

Very different level of thinking.

If you only ask recall questions, you cannot conclude that the learner can reason.

“Improve your clinical reasoning” is not a development plan

Educational diagnosis only matters if it changes what happens next.

Suppose a resident struggles with examination prioritization.

You could tell them:

“You need to improve your clinical reasoning.”

Maybe that is accurate.

But what should they actually do differently tomorrow?

Instead, make the target observable.

Now you have something you can develop.

You can watch for change:

• Can they explain why they are choosing each test?
• Are they prioritizing findings that could change the plan?
• Are they eliminating lower-value testing?

After three evaluations, reassess.

If performance improves, progress the challenge.

If it does not, reconsider your educational diagnosis or change the mentor strategy.

“Improve your clinical reasoning” is an expectation. It is not a development plan.

A useful development plan needs four things:

• a focused target
• a mentor strategy matched to the problem
• an observable behavior
• a reassessment point

That is the difference between giving feedback and creating development.

The better sequence

The next time a learner struggles, resist the reflex to immediately correct what you see.

Start with the target:

What should competent performance look like at this stage?

Then observe:

What actually happened?

Diagnose:

Where is the gap showing up—and what is most likely limiting performance?

Choose the mentor move:

What response best matches that problem?

Then reassess:

Did performance change?

If not, revise your hypothesis.

The visible error is not always the real learning problem.

The better we become at understanding why someone is struggling, the less likely we are to overteach, rescue too quickly, correct the wrong problem, or give feedback that never changes performance.

Eventually, though, there is a bigger goal.

We do not simply want mentors to become better at diagnosing the learner.

We want learners to become better at diagnosing themselves—to recognize what they were thinking, where their reasoning broke down, and what they need to change next.

That is when mentorship starts becoming self-directed development.