Bayesian Reasoning in Medical Contexts — Practice

Practice bank for UCAT Decision Making questions involving Bayesian/diagnostic reasoning.

---

Quick-Reference: Diagnostic Test Measures

Key Definitions

Measure	Definition	Formula
Sensitivity	P(positive test | disease)	True positives / (True positives + False negatives)
Specificity	P(negative test | no disease)	True negatives / (True negatives + False positives)
PPV	P(disease | positive test)	True positives / (True positives + False positives)
NPV	P(no disease | negative test)	True negatives / (True negatives + False negatives)
Prevalence	P(disease)	Total with disease / Total population

The 2×2 Table Method

	Disease +	Disease −	Total
Test +	True Positive (TP)	False Positive (FP)	TP + FP
Test −	False Negative (FN)	True Negative (TN)	FN + TN
Total	TP + FN	FP + TN	N

Step-by-Step for UCAT

1. Start with a convenient population (e.g., 10,000 or 1,000)
2. Apply the prevalence to split into disease+ and disease−
3. Apply sensitivity to the disease+ group → TP and FN
4. Apply specificity to the disease− group → TN and FP
5. Answer the question using the completed table

The Base Rate Trap

When prevalence is low, even a highly specific test produces many false positives relative to true positives. This makes PPV low despite high sensitivity and specificity.

---

Strategy

• Always build the 2×2 table — do not try to do Bayesian reasoning in your head
• Use round numbers for the hypothetical population
• Target: 60–75 seconds (these questions are calculation-heavy)

---

Practice

Complete the 10 assessment questions.