Reverse causation in cross-sectional studies is most likely when which scenario occurs?

• A. Exposure is fixed over time.
• B. Exposure is time-variant and the direction of causality cannot be determined.
• C. The study uses randomization.
• D. The outcome clearly precedes exposure.

Answer: (B)

Cross-sectional studies capture exposure and outcome at roughly the same time, so we can’t confirm which came first. Reverse causation happens when the outcome influences the exposure rather than the other way around. This problem is most likely when the exposure can change over time (time-variant). If exposure can vary, and we don’t have data to pin down the sequence, the direction of causality becomes ambiguous, making reverse causation a real concern. For example, activity level or stress can fluctuate, and measuring them at one moment doesn’t reveal whether the exposure led to the outcome or the outcome led to changes in exposure.

If the exposure were fixed over time, the temporal ambiguity would be less pronounced, and randomization would further reduce concerns about reverse causation. If there were clear evidence that the outcome preceded the exposure, that would point to reverse causation too, but in typical cross-sectional settings that timing isn’t reliably known. The scenario where the exposure is time-variant and the direction of causality can’t be determined best captures the risk of reverse causation.