What is a diagnostic likelihood ratio for a test, and how is it used in clinical decision making?

• A. It quantifies how much a test result changes the odds of disease and is used with pretest probability to compute the post-test probability.
• B. It measures how accurate a test is in the population and is used to determine sample size.
• C. It represents the cost-effectiveness of a test.
• D. It indicates the time required to perform a test.

Answer: (A)

A diagnostic likelihood ratio is a statistic that shows how much a test result changes the odds that a patient has the disease. It combines sensitivity and specificity into a single number and comes in two forms: one for a positive result and one for a negative result. The idea is to take your pretest probability (your estimate of how likely the disease is before testing) and update it to a post-test probability using Bayes’ rule: post-test odds = pretest odds × likelihood ratio.

For example, if your pretest probability is 30% (odds about 0.43) and the test yields a positive result with an LR of 5, the post-test odds become about 2.15, which corresponds to a post-test probability around 68%. If the test result is negative and the negative LR is 0.2, the post-test odds drop to about 0.086, giving a post-test probability around 8%. This updating helps guide clinical decisions about diagnosis, treatment, or further testing.

The other options describe population-level accuracy, sample size implications, cost-effectiveness, or time to perform, none of which capture how a diagnostic likelihood ratio changes disease probability with a given test result.