top of page
Search

Predictive Values (PPV, NPV) & Prevalence — MRCP Part 1

TL;DR

Predictive values tell you how useful a test result is in real patients, and they change with disease prevalence. In MRCP Part 1, most candidates lose marks by forgetting that PPV rises with higher prevalence, while NPV rises with lower prevalence. If you can identify the population and reason directionally, these questions become straightforward.


Why this topic matters in MRCP Part 1

Predictive values sit at the intersection of epidemiology, screening, and clinical reasoning. They are frequently tested in MRCP Part 1 because they assess understanding rather than rote calculation. Examiners commonly frame questions around screening programmes, primary care, or specialist clinics to see whether you appreciate how prevalence alters the meaning of a test result.

This article supports the main MRCP hub on Crack Medicine:


Core concepts you must know

Key definitions (learn these precisely)

  • Sensitivity: Probability that a test is positive given the disease is present.

  • Specificity: Probability that a test is negative given the disease is absent.

  • Positive Predictive Value (PPV): Probability that disease is present given a positive test.

  • Negative Predictive Value (NPV): Probability that disease is absent given a negative test.

  • Prevalence: Proportion of the population that has the disease.

The single most important rule

PPV and NPV depend on prevalence; sensitivity and specificity do not.

If you remember only one sentence for the exam, make it this one.

How prevalence changes predictive values

Think clinically rather than mathematically.

  • Low prevalence population (e.g. population screening):

    • Many healthy people

    • False positives accumulate

    • PPV is low, NPV is high

  • High prevalence population (e.g. tertiary referral clinic):

    • More true disease

    • Positive tests are more likely to be true positives

    • PPV is high, NPV is lower


High-yield comparison table

Clinical context

Prevalence

PPV

NPV

Population screening

Low

Low

Very high

Primary care

Moderate

Moderate

High

Specialist referral clinic

High

High

Lower

Exam insight: Sensitivity and specificity stay the same in all three rows. Only prevalence changes.


Comparison of screening and diagnostic testing contexts demonstrating low PPV in screening and higher PPV in specialist clinics.

The 5 most tested subtopics in MRCP questions

1. Screening programmes

Screening is performed in largely healthy populations. Even excellent tests generate false positives when disease prevalence is low. This principle underpins national screening policy.

Authoritative reference:

2. Diagnostic testing in secondary care

In hospital clinics, patients are pre-selected by symptoms or referral criteria. Prevalence is higher, so PPV increases. This is why confirmatory tests are often reserved for specialist settings.

3. False positives in rare diseases

A classic MRCP trap. A test with 99% specificity can still produce more false positives than true positives if prevalence is extremely low.

4. Directional reasoning over calculation

MRCP Part 1 rarely requires full Bayes’ calculations. Instead, you are asked what happens to PPV or NPV when prevalence increases or decreases.

5. Sensitivity/specificity vs predictive values

Sensitivity answers: “How good is the test? ”PPV answers: “How much can I trust this result in this patient? ”Examiners expect you to distinguish these clearly.


Mini-case (exam style)

Scenario A screening blood test for Disease Y has:

  • Sensitivity: 95%

  • Specificity: 95%

Disease Y has a prevalence of 0.2% in the screened population.

Question Which statement is most accurate?

A. PPV is high because sensitivity is highB. PPV is low because prevalence is lowC. Specificity decreases in screening populationsD. NPV is low due to false negativesE. Sensitivity depends on prevalence

Correct answer: B

Explanation Despite high sensitivity and specificity, the disease is rare. Most positive tests will therefore be false positives, giving a low PPV. Sensitivity and specificity do not change with prevalence, and NPV will be very high.


The 5 most common exam traps

  1. Confusing PPV with sensitivity.

  2. Assuming high specificity automatically means high PPV.

  3. Ignoring the population described in the stem.

  4. Over-calculating when only direction of change is required.

  5. Forgetting that prevalence is often implied, not stated.


Practical study-tip checklist

Before answering any predictive value question in MRCP Part 1, run through this checklist:

  1. Identify the setting (screening, GP, hospital clinic).

  2. Estimate prevalence (low vs high).

  3. Ask: “What happens to PPV and NPV in this context?”

  4. Eliminate options that confuse predictive values with sensitivity/specificity.

  5. Choose the answer consistent with Bayes’ logic.

For active practice, use:


FAQs

Does PPV depend on prevalence?

Yes. PPV increases as disease prevalence increases and falls when prevalence is low, even if the test itself is excellent.

Why is NPV usually high in screening programmes?

Because screening targets low-prevalence populations, most negative results are true negatives.

Do sensitivity and specificity change with prevalence?

No. They are intrinsic properties of the test and remain constant across populations.

Are calculations required in MRCP Part 1?

Rarely. Most questions test conceptual understanding rather than numerical computation.


Ready to start?

Consolidate this topic by pairing conceptual revision with question practice. Review the MRCP Part 1 overview, practise targeted items in the Free MRCP MCQs, and benchmark readiness with a full mock test.


Sources

 
 
 

Comments

bottom of page