How experts think: Making clinical reasoning visible in Oncology

Expert oncologists do more than follow guidelines. They weigh tradeoffs, interpret uncertainty and adapt decisions to individual patients. Living Algorithms make this clinical reasoning visible, helping clinicians understand not just what to do, but how experienced physicians think through complex cases.

The hidden layer of oncology care

When they first enter oncology, many trainees assume experts make decisions by simply following guidelines.

But in practice, experienced clinicians do something much more nuanced. They:

• Interpret data in context

• Weigh competing priorities

• Adapt to real-world patients

• Navigate uncertainty
  

This process is often invisible.

Why this matters

Two clinicians may review the same patient and arrive at different, yet reasonable, treatment decisions. Why?

Because oncology is not about memorizing pathways, it's about clinical reasoning.

What expert reasoning actually looks like

Experienced oncologists are constantly balancing questions like:

• Is this patient fit enough for intensification?

• Does the potential benefit justify the toxicity?

• How much weight should I give this biomarker?

• Does this patient resemble the trial population?

• What are the patient's goals and priorities?
  

These decisions rarely come from a single rule. They come from judgment.

The problem with traditional algorithms

Most traditional algorithms show:

• Recommendations

• Treatment pathways

• Evidence summaries
  

But they often don't show:

• Why one option is favored

• How experts think through tradeoffs

• What factors drive the decision
  

The reasoning layer is missing.

Why this gap matters for trainees

For trainees, this creates a major challenge. They may understand:

• The guideline

• The trial data

• The available therapies
  

But still struggle with: "how do I choose?"

This is where many learning gaps emerge.

Why this matters for community oncologists

The same issue affects practicing clinicians, especially when they're:

• Managing less familiar diseases

• Navigating rapidly evolving therapies

• Treating patients who don't fit standard pathways
  

In these scenarios, understanding expert reasoning becomes just as important as understanding the evidence itself.

Clinical decisions are rarely binary

Many oncology decisions involve multiple reasonable options. For example:

• Hormonal therapy vs chemotherapy

• Triplet vs doublet therapy

• Aggressive treatment vs tolerability-focused care
  

The right answer depends on:

• Disease burden

• Patient fitness

• Toxicity tolerance

• Goals of care

• Real-world context
  

This is where expert thinking matters most.

The role of uncertainty

Expert clinicians are also comfortable with uncertainty. They recognize:

• Data-free zones

• Evolving evidence

• Areas lacking consensus
  

Instead of pretending uncertainty doesn't exist, they:

• Frame the tradeoffs

• Prioritize key factors

• Move toward the best practical decision
  

This reasoning process is rarely captured in standard tools.

How Living Algorithms make reasoning visible

Living Algorithms are designed not just to show decisions, but to expose the thinking behind them.

What this looks like in practice

Expandable clinical rationale

With Living Algorithms, clinicians can explore:

• Why a treatment is recommended

• Which factors influence the choice

• What tradeoffs exist
  

Real-world context

Algorithms incorporate practical considerations like:

• Performance status

• Comorbidities

• Prior therapies

• Patient preferences
  

Multiple acceptable pathways

Instead of forcing a single rigid answer, Living Algorithms can show:

• Several reasonable options

• Where expert opinions may differ

• How clinicians think through those differences
  

Transparency around uncertainty

Where evidence is limited or evolving, that is made explicit. This reflects how experienced clinicians actually practice.

From memorization to reasoning

The goal of oncology education shouldn't be memorizing pathways. It should be learning how to think through decisions.

Living Algorithms help support this transition.

Why this becomes more important over time

As oncology grows more complex, no clinician can memorize everything. There are too many:

• Biomarkers

• Therapies

• Sequencing decisions

• Edge cases
  

Future decision support tools must do more than provide information. They must help clinicians reason.

A more human model of decision support

Many tools focus on delivering answers, but expert care is not just about answers.

It's about:

• Judgment

• Context

• Interpretation

• Adaptation
  

Living Algorithms are designed around this more realistic model of clinical practice.

Bottom line

Expert oncologists do more than follow pathways. They interpret, prioritize and adapt decisions to the patient in front of them.

Making this reasoning visible helps clinicians:

• Learn faster

• Navigate complexity

• Make more confident decisions
  

Living Algorithms are built to expose this layer of clinical thinking and help to bridge the gap between evidence and real-world care.

Try it with your next case

The next time you review a treatment pathway, ask yourself: do I understand why this decision is being made?

If not, you're seeing the recommendation, but not the reasoning. That's where deeper decision support begins.