A highly sensitive test that yields a negative result is most useful for ruling out disease because it is associated with which likelihood ratio?

• A. A low negative likelihood ratio.
• B. A high negative likelihood ratio.
• C. A high positive likelihood ratio.
• D. No change in likelihood ratio.

Answer: (A)

When you want to rule out disease, a negative result needs to be highly informative. The negative likelihood ratio (LR-) measures exactly how much a negative test lowers the probability of disease. The smaller this value, the more a negative result can rule out disease. A highly sensitive test minimizes false negatives, so if the test comes back negative, the odds that the patient has the disease drop significantly. This is because LR- = (1 − sensitivity) ÷ specificity, and with very high sensitivity, (1 − sensitivity) becomes very small; even with reasonable specificity, the ratio stays small. Thus, a low LR- is the best indicator for effectively ruling out disease after a negative result. In contrast, a high LR- would make a negative result less reassuring, a high LR+ is about ruling in with a positive result, and no change in LR wouldn’t change the interpretation.