Overview
Post-test probability is the chance a patient has a condition after a positive test, taking into account how likely it was beforehand. It's how clinicians interpret diagnostic tests.
Enter the pre-test probability, the test's sensitivity and its false positive rate (100% − specificity). This is for learning and isn't medical advice.
Worked examples
| Inputs | Result |
|---|---|
| Prior Probability of Condition (%)2Sensitivity (%)92False Positive Rate (%)8 | Posterior probability P(ConditionPositive): 19.01%False-alarm share among positives: 80.99% |
Show 2 more examples
| Prior Probability of Condition (%)30Sensitivity (%)92False Positive Rate (%)8 | Posterior probability P(ConditionPositive): 83.13%False-alarm share among positives: 16.87% |
| Prior Probability of Condition (%)10Sensitivity (%)85False Positive Rate (%)3 | Posterior probability P(ConditionPositive): 75.89%False-alarm share among positives: 24.11% |
How it works
The pre-test probability is combined with the test's accuracy using Bayes' theorem. A positive result raises the probability, but how far depends heavily on where it started. The same result means much more in a high-risk patient than in a low-risk one.
The formula
| pre | Pre-test probability |
| FPR | 100% − specificity |
Practical tips
- A 2% pre-test probability with 92% sensitivity and 92% specificity gives a post-test probability of 19.01%.
- With a 30% pre-test probability, the same test gives about 83%.
- Likelihood ratios are another common way to do the same calculation.
Frequently asked questions
Where does pre-test probability come from?
Prevalence in similar patients, adjusted for symptoms and risk factors, often from clinical scores.
What about a negative result?
That needs the negative version of the formula, using 1 − sensitivity and specificity.
Is this the same as positive predictive value?
Yes, when the pre-test probability equals the prevalence in the tested group.