Relative risk vs odds ratio: the 2x2 table explained
One 2x2 table can give a relative risk of 4 and an odds ratio of 4.75. Learn how to calculate each, why they differ, and which one a cohort or case-control study can report.
Chapters (10)
Calculate relative risk and odds ratio from a 2x2 table and choose the measure appropriate to cohort or case-control designs.
The short version
- Relative risk is a ratio of risks; odds ratio is a ratio of odds, calculated as ad / bc.
- A value of 1 means no association, above 1 means more outcome with exposure, below 1 means less.
- The odds ratio approximates the relative risk only when the outcome is rare, with risk below about 10%.
- When the outcome is common, the odds ratio sits further from 1 than the relative risk and overstates the effect.
- Cohort studies give relative risk; case-control studies give only the odds ratio.
Relative risk vs odds ratio: same table, two answers
Three hundred workers handle a solvent and three hundred do not. Years later, 60 of the exposed workers have the disease, and 15 of the unexposed.
From that one table, the relative risk is 4 and the odds ratio is 4.75. This lesson explains relative risk vs odds ratio: how to calculate each from a 2x2 table, why they differ, and which one fits a cohort or a case-control study.[1][2][3]
The 2x2 table and cells a, b, c, d
Rows are exposed and unexposed. Columns are disease and no disease. The four cells are named a, b, c and d.[1][2]
| Disease | No disease | |
|---|---|---|
| Exposed | a | b |
| Unexposed | c | d |
- a: exposed with disease
- b: exposed without disease
- c: unexposed with disease
- d: unexposed without disease
Risk vs odds
Risk is the people with the outcome divided by everyone in the group. Odds is the people with the outcome divided by the people without it.[1][3]
A risk of 20%, one in five, is odds of 1 to 4. As numbers, a risk of 0.20 is odds of 0.25.
Relative risk: a ratio of risks
risk in the exposed divided by risk in the unexposed
- Exposed risk: 60 / 300 = 0.2
- Unexposed risk: 15 / 300 = 0.05
- RR = 0.2 / 0.05 = 4
The exposed workers have four times the risk of the unexposed.[1][2]
Odds ratio: a ratio of odds, and how to read both
odds in the exposed divided by odds in the unexposed; ad / bc is the cross-product shortcut
- Exposed odds: 60 / 240 = 0.25
- Unexposed odds: 15 / 285 ≈ 0.053
- OR = 0.25 / (15 / 285) = 4.75
- Shortcut: (60 × 285) / (240 × 15) = 17,100 / 3,600 = 4.75
Both measures read alike. A value of 1 means no association, above 1 means more outcome with exposure, and below 1 means less.[1][2][3]
Why the odds ratio is bigger: common vs rare outcomes
When the outcome is common, odds run well above risk, so in this table the odds ratio (4.75) is bigger than the relative risk (4). When the outcome is rare, the gap closes.[1][3]
- Exposed: 12 cases in 3,000. Unexposed: 4 cases in 3,000
- RR = (12 / 3000) / (4 / 3000) = 3
- OR = (12 × 2996) / (2988 × 4) = 35,952 / 11,952 ≈ 3.01
When the outcome is rare in both groups, below about 10%, the odds ratio is close to the relative risk. When the outcome is common, the odds ratio sits further from 1 than the relative risk, whichever direction the effect goes, so it overstates the effect.[1][3]
Cohort vs case-control: which measure you can use
Which measure you can use depends on how the study sampled people.[1][2]
| Design | How people are sampled | Measure |
|---|---|---|
| Cohort | Exposed and unexposed people are followed forward and outcomes are counted | Risk, so relative risk, can be calculated |
| Case-control | Cases and controls are picked, then exposure is looked back at | Odds ratio only |
In a case-control study the researchers chose how many cases and how many controls to recruit. So the risk of disease in each exposure group cannot be calculated from these data, and relative risk is not available.[1][2]
The odds ratio survives. The odds of exposure in cases against controls has the same cross-product as the odds of disease in exposed against unexposed. So case-control studies report the odds ratio, which approximates the relative risk when the disease is rare in the population.[1][3]
Practice question
Original · USMLE styleResearchers investigate whether past occupational exposure to a chemical solvent is associated with a rare adult cancer. They identify 80 patients with the cancer and 160 healthy adults without it. Review of employment records shows prior exposure to the solvent in 48 of the patients and 40 of the controls.
Common questions
What is the difference between relative risk and odds ratio?
Relative risk is the risk in the exposed divided by the risk in the unexposed. Odds ratio is the odds in the exposed divided by the odds in the unexposed, which equals ad / bc.
How do you calculate an odds ratio from a 2x2 table?
Multiply a by d and divide by b times c. For the table with 60, 240, 15 and 285, that is 17,100 / 3,600 = 4.75.
When does the odds ratio approximate the relative risk?
When the outcome is rare in both groups, with risk below about 10%. For a common outcome the odds ratio sits further from 1 and overstates the effect.
Why can a case-control study not give relative risk?
The researchers chose how many cases and controls to recruit, so risk in each exposure group cannot be calculated. The odds ratio can still be calculated.
What does an odds ratio of 1 mean?
No association. Above 1 means more outcome with exposure, and below 1 means less.
Transcript
Read the full narration, by chapter648 words
0:00The hook: relative risk 4, odds ratio 4.75, same table
Three hundred workers handle a solvent. Three hundred do not. Years later, 60 exposed workers have the disease, and 15 unexposed. Same table, two answers: the relative risk is four, and the odds ratio is 4.75. Why do they differ? Today: calculate both from a two by two table, and choose the right one for cohort and case-control studies.
0:26The 2x2 table and cells a to d
First, the table. Rows are exposed and unexposed. Columns are disease and no disease. Call the cells a, b, c and d: exposed with disease is a, exposed without is b, unexposed with disease is c, unexposed without is d. Risk is the people with the outcome divided
0:47Risk versus odds
by everyone in the group. Odds is the people with the outcome divided by the people without it. A risk of 20%, one in five, is odds of one to four.
1:01Relative risk calculated
Relative risk is the risk in the exposed divided by the risk in the unexposed. Our workers: 60 out of 300 is a risk of 0.2. Fifteen out of 300 is 0.05. 0.2 over 0.05: the relative risk is four.
1:21Odds ratio calculated, the ad / bc shortcut and how to read RR and OR
The odds ratio is the odds in the exposed divided by the odds in the unexposed. Exposed odds: 60 to 240, 0.25. Unexposed: 15 to 285, about 0.053. Divide: 4.75. The shortcut, a times d over b times c, gives the same 4.75. Both read alike: one means no association, above one more outcome with exposure, below one less.
1:57Common vs rare outcomes: why they differ
So why is the odds ratio bigger? When the outcome is common, odds run well above risk. When the outcome is rare, the gap closes: 3,000 exposed with 12 cases, 3,000 unexposed with four. The relative risk is exactly three. The odds ratio is about 3.01. When the outcome is rare in both groups, below about 10%, the odds ratio is close to the relative risk. When the outcome is common, the odds ratio sits further from one than the relative risk, whichever direction the effect goes, so it overstates the effect.
2:37Cohort vs case-control: which measure you can use
Which one you can use depends on how the study sampled people. A cohort study follows exposed and unexposed people forward and counts outcomes, so risk, and relative risk, can be calculated. A case-control study starts with outcome: pick the cases, pick the controls, look back at exposure. The catch: the researchers chose how many cases and how many controls to recruit. So risk in each exposure group cannot be calculated from these data, and relative risk is not available. But the odds ratio survives: the odds of exposure in cases against controls has the same cross-product as the odds of disease in exposed against unexposed. So case-control studies report the odds ratio, which approximates the relative risk when the disease is rare in the population.
3:30Practice question
Your turn. Researchers study a rare cancer, recruiting 80 patients and 160 people without it. Past chemical exposure is found in 48 of the patients and 40 of the controls. Which is the odds ratio? A, 0.5. B, 1.5. C, 2.4. D, 2.6. E, 4.5. Pause here and decide.
4:01Answer and explanation
A is out: that pairs the wrong cells, a times b over c times d. B is out: that is only the odds of exposure among the cases. C is out: that divides the proportion of cases exposed by the proportion of controls exposed. D is out: that is a relative risk, and this design cannot calculate risk. So the answer is E. Forty-eight times 120 is 5,760. Thirty-two times 40 is 1,280. The odds ratio is 4.5.
4:39Recap and next lesson: absolute risk reduction, relative risk reduction and NNT
Three things to keep. One: relative risk is a ratio of risks; odds ratio is a ratio of odds, a d over b c. Two: cohort studies give relative risk; case-control studies give only the odds ratio. Three: the odds ratio approximates relative risk only when the outcome is rare, risk below about 10%. Next, absolute risk reduction, relative risk reduction and number needed to treat.
Sources
Numbers in the text, e.g. [1], point here- Kim H-Y. Statistical notes for clinical researchers: risk difference, risk ratio, and odds ratio. Restor Dent Endod. 2017;42(1):72-76. pmc.ncbi.nlm.nih.gov/articles/PMC5300861/
- Relative risk (Wikipedia, secondary). en.wikipedia.org/wiki/Relative_risk
- Odds ratio (Wikipedia, secondary). en.wikipedia.org/wiki/Odds_ratio
All numbers on screen are teaching values: 300 exposed and 300 unexposed workers; a rare-outcome example with 3,000 per group; the practice question uses its own values.
The practice question is original and its patient is synthetic. Every fact is checked against the sources above in independent accuracy reviews. Always follow your local guidelines and your own laboratory’s ranges.
Image credits. Illustrations: NIAID NIH BioArt Source (bioart.niaid.nih.gov/bioart/331), public domain: adult man upper-body figure, adapted (recoloured, outlined). The tables, dot grids, number line, study-design cards, question card and all handwriting are drawn in code (original). No AI-generated teaching images.
Corrections. None so far. Spotted an error? Email info@medservatory.com with the lesson code MS-S1-BIOS-003.
For exam study only, not medical advice. Read the medical disclaimer.
In this series
Playlist on YouTube- EP 01Sensitivity vs specificity: the 2×2 table, SnNout and SpPin5:34
- EP 02PPV, NPV and prevalence explained5:27
- EP 03Relative risk vs odds ratio: the 2x2 table explained5:15
- EP 04Number needed to treat: ARR, RRR and NNH from trial data5:00
- EP 05Study designs explained: case-control, cohort, cross-sectional and RCT5:22
- EP 06Types of bias: selection, recall, lead-time and confounding5:30
- EP 07Type I and type II errors: alpha, beta, power and the p-value5:16
- EP 08Confidence intervals and significance: is the null value inside?5:12