What is Pre-Test and Post-Test Probability?


In the medical field, a diagnostic test is used to determine whether or not an individual has a particular disease.

Whenever a diagnostic test is performed, there are always two probabilities of interest:

1. Pre-Test Probability: The probability that an individual has the disease before the diagnostic test is even performed.

  • This is calculated as the proportion of individuals who are known to have the disease in the population of interest.
  • This can be calculated using data that has been collected in prior studies or it can be roughly estimated by professionals in the field.

2. Post-Test Probability: The probability that an individual has the disease after testing positive in the diagnostic test.

  • This is calculated using pre-test probability and the known sensitivity and specificity of the diagnostic test being used.
  • Sensitivity is the “true positive rate” – the percentage of positive cases the model is able to detect.
  • Specificity is the “true negative rate” – the percentage of negative cases the model is able to detect. 
  • Both sensitivity and specificity can be calculated using data from prior studies.

The following example shows how to calculate pre-test and post-test probability in practice.

Example: Calculating Pre-Test and Post-Test Probabilities

Suppose it is known that about 7 in 100 individuals in a certain population have disease X.

If we selected an individual from this population at random and performed a diagnostic test to determine if they have disease X, the pre-test probability that they have the disease would be 0.7 or 7%.

Now suppose it’s also known that the sensitivity of the diagnostic test is 0.74 and the specificity is 0.92.

We can use the following formulas to calculate the post-test probability:

  • Likelihood ratio positive = sensitivity / (1−specificity) = .92 / (1−.92) = 11.5
  • Likelihood ratio negative = (1−sensitivity) / specificity = (1−.74) / .92 = .2826
  • Pre-test odds =pre-test prob. / (1−pre-test prob.) = .07 / (1−.07) = .0752
  • Positive post-test odds = .0752 * 11.5 = 0.8648
  • Positive post-test probability = .8648 / (.8648+1) = .4637

Here is how to interpret these results:

The pre-test probability is 7%

  • This means the probability that a randomly selected individual has disease X is 7%, even before any diagnostic test is performed.

The post-test probability is 46.37%.

  • For an individual who tests positive on this diagnostic test, the probability that they actually have disease X is 46.37%.

You might be thinking to yourself that a positive test result on the diagnostic test should indicate that an individual definitely has the disease, but keep two things in mind:

1. The probability that a randomly selected individual from the population has the disease (7%) is very low to start.

2. The diagnostic test is known to not be perfect at detecting true positive cases and true negative cases.

Keeping both these facts in mind, it’s a little easier to understand how a positive result on the diagnostic test doesn’t necessarily mean that the individual actually has disease X.

Additional Resources

The following tutorials provide additional information about topics related to probability:

What is a Probability Distribution Table?
What is the Law of Total Probability?
How to Find the Probability of “At Least One” Success

Leave a Reply

Your email address will not be published.