Differential outcome misclassification can bias study results in which ways?

• A. It always biases toward the null.
• B. It has no effect on the estimate.
• C. It can bias toward or away from the null depending on how accuracy differs by exposure.
• D. It always biases away from the null.

Answer: (C)

When outcome misclassification varies by exposure, the accuracy of outcome measurement is different between exposed and unexposed groups. This differential misclassification distorts the observed counts of who truly has the outcome, because errors occur more in one exposure group than the other. As a result, the association you observe can be shifted in either direction depending on whether the exposed group has higher or lower accuracy (more true cases and fewer false positives, or vice versa). Therefore, the bias can pull the estimate toward the null or away from it. This contrasts with nondifferential misclassification, which more commonly biases toward the null. The key point is that the direction of bias with differential outcome misclassification depends on how accuracy differs by exposure.