Internal Medicine · Clinical Reasoning · Clinical Pearl
A diagnostic test only helps when it meaningfully changes probability
Lede: A test result is clinically useful only if it changes what you believe enough to alter what you do next.
Why it matters
Every diagnostic test is interpreted against a pretest probability. The same positive result may be persuasive in one patient and misleading in another. Likelihood ratios provide a practical framework for estimating how strongly a result shifts diagnostic odds, but the key bedside question is whether the post-test probability crosses a decision threshold for additional testing, treatment, or exclusion.
The pearl
Do not ask only, “Is this test accurate?” Ask, “Will this result move the probability enough to change management?”
