Which item indicates a design issue where the time between index tests is not short enough?

Which item indicates a design issue where the time between index tests is not short enough?

• A. Blinding of examiners
• B. Standardized procedures
• C. Gold standard comparison
• D. Time between index tests not short enough

Answer: (D)

Time interval bias in diagnostic accuracy studies is tested here. The idea is that the index test and the reference standard should be performed within a short time of each other so the patient’s disease status hasn’t changed in the interim. If the time between tests is not short enough, the underlying condition could evolve, treatment could start, or symptoms could worsen or improve, making the reference standard result no longer reflect the patient’s state at the time the index test was done. That drift creates bias in estimates of how well the index test performs. So, the option that describes the time between index tests not being short enough directly points to this design flaw and is the best choice. The other items—blinding examiners, standardized procedures, and using a gold standard comparison—address different sources of bias or validity concerns but do not capture the issue of a long interval between tests.

Time interval bias in diagnostic accuracy studies is tested here. The idea is that the index test and the reference standard should be performed within a short time of each other so the patient’s disease status hasn’t changed in the interim. If the time between tests is not short enough, the underlying condition could evolve, treatment could start, or symptoms could worsen or improve, making the reference standard result no longer reflect the patient’s state at the time the index test was done. That drift creates bias in estimates of how well the index test performs.

So, the option that describes the time between index tests not being short enough directly points to this design flaw and is the best choice. The other items—blinding examiners, standardized procedures, and using a gold standard comparison—address different sources of bias or validity concerns but do not capture the issue of a long interval between tests.