Why oncology algorithms fail trainees (and how to fix them)
- May 3
- 3 min read
Most oncology algorithms are written by experts, for experts.
But trainees need something different: clear structure, definitions and guidance on how to think through decisions.
Living Algorithms are designed to make pathways easier to understand, navigate and apply in real clinical scenarios.
The gap between expertise and learning
Oncology algorithms are often created by leading specialists. They reflect:
Deep knowledge
Years of experience
Familiarity with complex decision making
But many of the people using algorithms are:
Fellows
Residents
Early-career clinicians
Generalists covering multiple diseases
There's a gap between how experts think and how trainees learn.
What trainees actually experience
When trainees open a traditional algorithm, they often encounter:
Abbreviations they don't fully recognize
Pathways that assume prior knowledge
Decision points without explanation
Dense or hard-to-follow structure
The information may be correct, but it's not always accessible.
A common scenario
A trainee is preparing for clinic. They open an algorithm to review a case.
Instead of clarity, they find themselves asking:
What does this abbreviation mean?
Why is this the recommended step?
How do I choose between these options?
Am I even looking at the right pathway?
This slows them down and limits learning. Even worse, it makes some trainees feel stupid when nothing could be further from the truth.
Why traditional algorithms fall short for trainees
Assumed knowledge
Experts naturally skip steps in their thinking, and algorithms often reflect that.
But trainees need those steps made explicit.
Lack of context
Decision points are presented without explanation. Trainees are left to infer:
Why a treatment is chosen
What factors matter
How alternatives compare
Unclear structure
Some algorithms are difficult to navigate.
Too many branches
Poor labeling
Hidden details
This makes it harder to build a mental model.
Overuse of abbreviations
Short forms may be efficient for experts. But for trainees, they can create confusion.
Even common terms can vary by disease.
What trainees actually need
Trainees are not just looking for answers. They are trying to learn how to think.
They need:
Clear definitions
Logical structure
Brief explanations at key steps
Visibility into clinical reasoning
In other words: not just the pathway, but the thinking behind it.
How Living Algorithms are designed differently
Living Algorithms are built with both experts and learners in mind.
What this looks like in practice
Clear, patient-centered structure
Algorithms start with:
Disease type
Stage
Biomarkers
Prior treatments
This helps trainees orient quickly.
Expandable details
Key steps include:
Rationale
Supporting data
Clinical context
Trainees can go deeper when needed, without overwhelming the main pathway.
Reduced ambiguity
Nodes are designed to be:
Concise
Specific
Actionable
With clear thresholds and criteria.
More intuitive navigation
Interactive elements allow users to:
Move step by step
Focus on relevant branches
Avoid unnecessary complexity
Opportunity for clarification
Where needed, algorithms can include:
Rationale
Definitions
Explanations
Supporting visuals
This helps trainees build their understanding and level of knowledge over time.
Making algorithms more educational
A strong algorithm should do two things:
Help you make a decision
Help you understand why
For trainees, the second step is just as important as the first.
Learning through real decisions
Trainees learn best in context. Not by memorizing lists, but by:
Seeing how decisions are made
Understanding tradeoffs
Connecting data to action
Algorithms that reflect real clinical reasoning accelerate this process.
Supporting confidence in clinic
One of the biggest challenges for trainees is confidence.
Am I choosing the right option?
Did I miss something?
Do I understand the reasoning?
Clear, structured algorithms help answer these questions quickly.
Better tools, better training
As oncology becomes more complex, training needs to evolve. There is too much:
Information
Data
Variation
...to rely on memorization alone.
Trainees need tools that help you:
Guide decisions
Support learning
Reduce cognitive load
Bottom line
Most oncology algorithms are built for those who already understand the system.
Trainees need something different: clarity, context and structure.
Living Algorithms are designed to bridge that gap, helping clinicians not only decide what to do, but understand why.
Try it as a trainee
The next time you review a case, ask: do I understand how this decision was made?
If not, the algorithm is incomplete.
If yes, you're learning in the way that matters most. And your patients will thank you.