St Emlyn’s Lesson Plans
Diagnostic testing: probability and clinical decisions
Use a simple test table and a PE case to connect pre-test probability, test results and the next clinical decision.
Medical Students · 45-minute session
For learners
Before: Allow about 15 minutes in total. With the reading alternative: about 11 minutes.
When we meet: We will calculate test performance from a simple table, then use a PE case to decide when a test is useful and what should follow its result.
For facilitators
Send: the learner preparation link before teaching.
Bring: Board, paper and optional calculators; no patient images are required.
Before the session
If you are attending this session, complete the core preparation below before you come. Allow about 15 minutes in total. With the reading alternative: about 11 minutes.
Core preparation
Core listening · 11 min 15 sec
St Emlyn’s Podcast: Ep 238 - Positive and Negative Predictive Values: Critical Appraisal Nugget
Understand how prevalence changes the meaning of a positive or negative test result.
Search for this episode by title in MedPod Learn if you want to record your listening there.
Prefer to read? Use Oxford CEBM: diagnostic accuracy studies (PDF) instead of listening. Allow about 8 minutes.
Read the measure definitions and explanations on pages 3–4. Leave the worked table exercise for the session; focus on sensitivity, predictive values and prevalence.
https://www.cebm.ox.ac.uk/files/ebm-tools/diagnosticaccuracystudies1.pdf
Listen for the distinction between sensitivity and predictive value, and the effect of prevalence on a test result. The current NICE PE pathway will be used for the supervised clinical case.
https://www.stemlynspodcast.org/e/ep-238-positive-and-negative-predictive-values-critical-appraisal-nugget/
Prepare (allow 3 minutes): Write one sentence explaining the difference between sensitivity and positive predictive value.
Think about: How would a positive D-dimer change your plan for a patient with suspected pulmonary embolism?
Picture the scene
You suggest a D-dimer for chest pain. Your supervisor asks what you will do if it is positive. You know the test is sensitive, but have not yet decided what question you are asking.
By the end of the session
- Explain sensitivity, specificity and predictive values using a two-by-two table.
- Show why a negative result has different meaning at different pre-test probabilities.
- Choose a test only after deciding how its possible results would change management.
After the session
- Before your next supervised investigation request, say aloud what a positive and a negative result would change.
Facilitator guide
Send learners the preparation link several days before teaching. Allow about 15 minutes in total. With the reading alternative: about 11 minutes.
Before learners arrive
- Draw the first numerical table and check the calculations before the session.
- Open your department’s PE pathway and identify the D-dimer assay and threshold; do not substitute a universal cutoff.
Equipment: Board, paper and optional calculators; no patient images are required.
Setup: Pairs calculate together, then explain a result to a neighbouring pair. Invite uncertainty about maths without grading speed.
If the preparation has not been completed
Replace the five-minute “Establish the essentials” slot with this shared preparation. Use a board and the first case’s two-by-two table: spend two minutes labelling disease and test results, two identifying the denominators for sensitivity and positive predictive value, and one checking the answers. Keep the calculations and PE pathway application for the cases.
Suggested session plan
- 0–5 min — Explore. Ask what happens after a positive D-dimer.
- 5–10 min — Establish the essentials. Label the test table and distinguish the four denominators.
- 10–32 min — Discuss & Apply. Spend 12 minutes on the numerical exercise and 10 on the changing PE case.
- 32–40 min — Consolidate. Ask pairs to explain the change in predictive values between the two populations, then explain a positive D-dimer to a patient.
- 40–45 min — Reflect & Transfer. Ask each learner to choose an upcoming supervised investigation request and state what they will clarify before ordering it.
Start the discussion
Read the opening scenario together. Give everyone a moment to think, then ask what concerns them and what they need to know next.
Cases and discussion
Use these cases, or invite an anonymised case from the group that addresses the same questions.
The fictional FRB test
FRB is a fictional test used for this exercise. Among 200 people with heart failure, it is positive in 80 and negative in 120; among 200 without heart failure, it is positive in 20 and negative in 180.
Add this information as the discussion develops:
- Ask each pair to calculate one quantity, then explain its denominator to another pair.
- Reveal a second fictional population with 100 affected and 900 unaffected people, assuming the same sensitivity and specificity: 40 true positives, 60 false negatives, 90 false positives and 810 true negatives.
Ask the group:
- Which denominator answers ‘how many people with disease test positive’?
- If the result is positive, what proportion actually have the condition in each population?
- Would a negative FRB safely exclude heart failure?
Discussion notes
Points to draw out:
- For the first table: sensitivity 40%, specificity 90%, PPV 80%, NPV 60%. LR+ = 0.4/(1−0.9) = 4; LR− = (1−0.4)/0.9 ≈ 0.67.
- In the second fictional population PPV is 40/130 ≈31% and NPV is 810/870 ≈93%. A high NPV can coexist with poor sensitivity when disease is uncommon.
- A test result changes probability; it does not abolish uncertainty. The clinical decision also depends on consequences of missing disease and harms of investigation.
Common misconceptions:
- Confusing sensitivity with the chance that a positive result is true.
- Calling these invented numbers evidence that FRB exists or is clinically useful.
For a more experienced group: Use odds to show that a 10% pre-test probability becomes about 31% after LR+4; then explain that result without mathematical terminology.
A breathless patient and a D-dimer request
A 51-year-old has mild pleuritic pain and breathlessness. They are stable. The supervisor's full assessment gives a PE-unlikely two-level Wells score.
Add this information as the discussion develops:
- The D-dimer result is above the applicable laboratory threshold.
- Change the history: there are now convincing DVT signs and PE is clinically likely.
Ask the group:
- What decision follows the positive result?
- Why is a positive D-dimer not a diagnosis?
- What changed in the second version before any blood was taken?
Discussion notes
Points to draw out:
- Use the validated PE pathway and local assay; a positive D-dimer in the appropriate branch leads to further investigation.
- In the likely branch, arrange imaging rather than using a negative D-dimer alone to dismiss concern.
- Continue to assess other explanations and discuss any interim treatment with the supervising clinician.
Common misconceptions:
- Using PERC in every patient with chest pain.
- Calling a low Wells score ‘negative’ as though it were a laboratory result.
For a more experienced group: Ask which groups fall outside NG158, including pregnancy and children, and why an adult score cannot simply be exported.
Important clinical points
- Students should use the locally adopted pathway with a supervisor rather than independently ordering indiscriminate D-dimer tests.
- A negative test cannot be interpreted safely without its timing, limitations and pre-test probability.
Closing the session
Use these questions to check what the group will take away. Go back over any clinical points that remain unclear.
- Explain why FRB’s NPV rises in the second population even though sensitivity is unchanged.
- Name one harm from ordering a test without a next-step plan.
- Explain a positive D-dimer to a patient without saying they have a clot.
Key teaching points:
- Interpret the result alongside the clinical history and the question the test is meant to answer.
- Predictive values change with disease prevalence.
- Before ordering, decide what each possible result would change.
Before leaving, ask each person to choose one action from After the session and say when they will try it. They complete that follow-up in practice, with supervision where needed.
Adapting the discussion
Uncertainty and local practice
Accuracy depends on the assay, threshold, population and reference standard. Check those conditions before using a test to support exclusion of disease.
If the discussion drifts
What would you do differently for this patient after either result?
For a more experienced group
Compare test thresholds or consequences of false positives without drifting into a general statistics lecture.
A 30-minute version
30 minutes: opening 3, table 7, PE case 12, consolidation 5, reflection 3. Omit likelihood-ratio calculations.
Evidence and guidance
- NICE NG158 — Current adult VTE pathway; source accessed via indexed recommendations.https://www.nice.org.uk/guidance/ng158/chapter/recommendations
- CEBM: SpPin and SnNout — Retained conceptual resource at its current Oxford URL; used with explicit limits.https://www.cebm.ox.ac.uk/resources/ebm-tools/sppin-and-snnout
Suggested curriculum links
GMC Outcomes for graduates / MLA 2026: Clinical reasoning; selecting and interpreting investigations
Learners justify an investigation and communicate uncertainty.
Check these links against the learner’s stage and programme. Attendance at a teaching session does not establish competence.
First published 2021-03-17 · Iain Beardsell.