top of page
Search

Odds Ratios vs Relative Risk Explained — MRCP Part 1

TL;DR;

Odds ratio (OR) and relative risk (RR) are measures of association commonly tested in MRCP Part 1. RR is intuitive and used in cohort studies and randomised controlled trials, while OR is standard in case–control studies and logistic regression. OR approximates RR only when the outcome is rare; misinterpreting this is a frequent exam trap.


Why this topic matters for MRCP Part 1

Medical statistics questions in MRCP Part 1 are designed to test interpretation, not advanced calculation. OR and RR appear across epidemiology, cardiology, endocrinology, and public health questions. Candidates who can quickly identify the study design and choose the correct measure gain easy marks.

This article supports systematic revision alongside the official MRCP syllabus and evidence-based medicine principles used in UK clinical practice.


Core concepts you must know

Risk vs odds (the foundation)

  • Risk (probability):= Number with outcome ÷ total number at risk

  • Odds:= Probability of outcome ÷ probability of no outcome

Risk and odds are related but not interchangeable.

Relative risk (RR)

Definition:

Risk of outcome in exposed group ÷ risk of outcome in unexposed group

Where it is used:

  • Cohort studies

  • Randomised controlled trials

Interpretation:

  • RR = 1 → no association

  • RR > 1 → increased risk with exposure

  • RR < 1 → protective effect

RR answers the clinically intuitive question:

“How many times more likely is the outcome in the exposed group?”

Odds ratio (OR)

Definition:

Odds of outcome in exposed group ÷ odds of outcome in unexposed group

Where it is used:

  • Case–control studies

  • Logistic regression (adjusted analyses)

Why MRCP examiners like it:In case–control studies, incidence is unknown, so RR cannot be calculated. OR is therefore the only valid measure.


Odds ratio vs relative risk — exam comparison

Feature

Relative Risk

Odds Ratio

Compares

Risk

Odds

Study designs

Cohort, RCT

Case–control

Intuitive

Yes

No

Logistic regression output

No

Yes

Approximates the other

—

Yes, if outcome is rare

Key exam rule:

OR ≈ RR only when the outcome is rare (<10%).

The 5 most tested subtopics

1. Study design first

Always identify the design before choosing a measure.

  • Case–control → OR

  • Cohort/RCT → RR

2. Rare disease assumption

When the outcome is rare, odds and risk are similar.When the outcome is common, OR overestimates the strength of association.

3. Logistic regression

If a question mentions adjusted analysis, multivariable model, or logistic regression, the answer will involve odds ratios.

4. Interpretation over calculation

MRCP Part 1 rarely requires arithmetic. Focus on what the number means clinically.

5. Association is not causation

Neither OR nor RR proves causality. This distinction is frequently tested indirectly.


Mini-MCQ (MRCP Part 1 style)

Question A case–control study evaluates the association between smoking and pancreatic cancer. Which is the most appropriate measure of association?

A. Absolute riskB. Relative riskC. Odds ratioD. Number needed to treatE. Hazard ratio

Correct answer: C. Odds ratio

Explanation: Case–control studies begin with outcome status, so incidence cannot be calculated. Relative risk is therefore invalid; odds ratio must be used.


Medical student revision setup for MRCP Part 1 statistics and epidemiology

Common exam pitfalls (high-yield)

  • Using relative risk in a case–control study

  • Interpreting odds ratio as risk when the outcome is common

  • Assuming OR or RR implies causation

  • Forgetting logistic regression outputs ORs

  • Ignoring confidence intervals when judging significance


Practical revision checklist

  • Identify the study design first

  • Ask whether incidence is known

  • Case–control → OR

  • Cohort/RCT → RR

  • Rare outcome → OR ≈ RR

  • Look for keywords: adjusted, logistic regression, multivariable

To practise exam-style interpretation, use timed MCQs from a reliable MRCP question bank.


How this fits into your MRCP Part 1 preparation

Odds ratios and relative risk should be revised alongside confidence intervals, p-values, bias, and screening test performance. A structured approach is outlined in the official MRCP syllabus and reinforced through repeated question practice.

For a broader framework, review the MRCP Part 1 overview:https://www.mrcpuk.org/mrcpuk-examinations/part-1


FAQs (People Also Ask)

What is the main difference between odds ratio and relative risk?

RR compares probabilities, while OR compares odds. RR is used when incidence is known; OR is used when it is not.

Why is odds ratio used in case–control studies?

Because participants are selected by outcome, incidence cannot be calculated, making RR invalid.

When does odds ratio approximate relative risk?

When the outcome is rare, typically less than 10%.

Does an odds ratio greater than 1 mean causation?

No. It indicates association only; causation requires additional criteria.


Ready to start?

Consolidate this topic by attempting mixed statistics questions under exam conditions. Start with Practise statistics MCQs in our Qbank, then progress to timed papers once interpretation becomes automatic.


Sources

 
 
 

Comments

bottom of page