Workplace-Based Assessments Decoded

Two registrars finish a list. One approaches the consultant in the corridor: "Could you just tick a PBA for that hemiarthroplasty?" Thirty seconds later it is done, rated at the top of every domain, free text reading "good case, well done."

The other asks the same consultant for ten minutes at the end of the list to go through the bits they found difficult — the reduction that took three attempts, the moment they were not sure about version. They come away with a form rated lower, and with three specific things to work on.

The first registrar has a better-looking portfolio. The second is going to be a better surgeon. The entire problem with workplace-based assessments is contained in that gap.

This post covers what WBAs are for, why they fail, and how to get value from them anyway. It sits within our training systems series.

The contradiction at the centre

WBAs were designed as formative tools — structured opportunities for observation and feedback, where the value is the conversation and the score is almost incidental.

They are used as summative evidence — proof of progression, counted at annual review, contributing to whether you advance.

Those two purposes pull in opposite directions. A formative tool rewards choosing the case you struggled with. A summative one rewards choosing the case that went well. Ask trainees to do both with the same form and they will optimise for the one with consequences.

Nobody is behaving badly here. The system asks for a demonstration of competence and then expects candour about incompetence, on the same document, assessed by the same person who decides your progression.

Being clear about that tension is the only way to get value from the exercise: you have to decide, deliberately, to use them formatively despite the summative pressure.

The main types

Names vary by country and college; the underlying instruments are common.

The PBA is the one that matters most in orthopaedic training. It assesses the entire operative episode — pre-operative planning, consent, positioning, the procedure itself, and the post-operative plan — not just whether the implant went in.

Its most useful feature is the level of supervision it records: whether the trainer had to intervene, guide, talk you through, or simply observe. That progression, tracked across cases, is a far more meaningful picture than any global rating.

Entrustment is the direction of travel

The concept underneath all of this is increasingly explicit: can this trainee be entrusted to do this unsupervised?

North American programmes have formalised it as Entrustable Professional Activities, and the entrustment framing is spreading. It is a better question than "was this performance satisfactory?" because it is what a supervisor actually needs to know and what a patient actually depends on.

If your system uses supervision levels rather than EPAs, the same question applies. When you look at your PBAs across a year, the meaningful pattern is not the ratings — it is whether the level of supervision required is falling.

Why they fail

Predictable and consistent failure modes:

Tick and run. Completed in a corridor, minutes after the case, with no discussion. Produces a record with no learning attached.

Everything maxed. A portfolio of top ratings tells an assessor nothing except that the trainee chose easy cases or the trainer avoided difficulty. Uniformly perfect scores are read as an absence of evidence, not proof of excellence.

All at year end. Thirty assessments in the fortnight before review. Transparent, and it demonstrates the opposite of continuous engagement.

Empty free text. "Good case, well done." The free text is the entire value of the form and it is routinely left blank.

Cherry-picked cases. Only straightforward cases you knew would go well.

Assessor shopping. Always the same soft-touch consultant, never the demanding one.

Every one of these is a rational response to summative pressure, and every one destroys the thing that makes the exercise worth doing.

How to get real value

If you accept that the score matters less than you fear — and it does; reviewers look at patterns, not individual ratings — you can use these properly.

• Pick the case you found hard. That is where the feedback is. A lower-rated PBA on a difficult revision is worth more to your development, and reads better to a thoughtful reviewer, than a perfect one on a routine case.
• Ask before the case, not after. "I'd like a PBA on this one — can you watch the approach particularly?" Directed observation produces specific feedback.
• Ask for one thing to work on, explicitly, and write it in yourself.
• Close the loop. Reference the previous feedback in the next assessment. Demonstrating that you acted on it is the single most persuasive thing in a portfolio.
• Spread them across the year and across assessors.
• Include some that did not go well. A portfolio containing genuine difficulty, addressed, is stronger evidence of insight than a flawless one.
• Vary your assessors deliberately, including the ones who will be hard on you.

The evidence a reviewer is really looking for is trajectory — see Surviving ARCP. A flat line of perfect scores demonstrates nothing. A rising line with specific feedback acted upon demonstrates everything.

If you are the assessor

Consultants complete these badly at least as often as trainees game them, and the fix costs ten minutes.

• Do it with them, not about them. The form is a prompt for a conversation, not a record of a verdict.
• Be honest about the rating. Inflated scores deny the trainee the feedback they came for, and reviewers discount them.
• Write specific free text. "Struggled to maintain reduction while inserting the guidewire; discussed using the ball-spike differently, to try next case" is worth more than any tick.
• Name one thing done well and one to change. Both.
• Do it promptly, while the case is fresh.
• Say what level of supervision you actually provided. Recording independence you did not observe is a patient-safety issue as much as an educational one.
• Don't sign for what you did not see. It happens, and it is a form of dishonesty.

If a trainee only ever brings you straightforward cases, say so — kindly, and early.

Quantity, and its limits

Every programme sets a minimum number. Treat it as a floor, not a target — and understand that a reviewer can distinguish thirty meaningful assessments from thirty ticks in about ninety seconds.

The pattern that reads well: spread through the year, across multiple assessors, spanning a range of case complexity, with substantive free text, showing decreasing supervision and evidence that earlier feedback was acted on.

The pattern that reads badly: clustered, single assessor, uniformly maximal, free text absent or generic.

The honest summary

WBAs are a compromise between two purposes that do not sit comfortably together, and most of what is wrong with them follows from that. They will not be fixed by another form.

What you can control is how you use them. The score on any individual assessment is close to meaningless. The conversation it prompts is where the value is, and the pattern across a year is what a reviewer actually reads.

So take the difficult case to the demanding consultant, ask them to watch the part you are worst at, and write down what they tell you. That is worth more than a portfolio of perfect scores, and it is also — not coincidentally — how you get better at operating.