Evidence-appraisal glossary
Likelihood ratio
A likelihood ratio tells you how much a diagnostic test result shifts the odds that a condition is present. The positive LR compares the chance of that result in people with the condition versus without it. Values far above 1 raise probability; values near 0 lower it.
Also called: LR, positive likelihood ratio, negative likelihood ratio, LR+, LR-.
A likelihood ratio (LR) summarizes how strongly a test result changes the probability of a condition. The positive LR is the true-positive rate divided by the false-positive rate; the negative LR is the false-negative rate divided by the true-negative rate. An LR of 1 is useless because the result is equally likely with or without disease. As a rough guide, a positive LR above 10 or a negative LR below 0.1 produces large, often decisive shifts in probability, while values near 1 barely move it. When reading a diagnostic study, look for reported LRs (or the sensitivity and specificity needed to compute them) rather than accuracy alone, and ask how much the test moves probability from the starting (pre-test) estimate. Example: a test result with a positive LR of 10 applied to a pre-test probability of 20 percent raises the post-test probability to roughly 70 percent, using a probability-to-odds conversion.
This is a plain-language methodology definition for reading research. It is general education, not medical advice.