Synonym: detection bias
Surveillance bias occurs when one group is believed to be at greater risk for a disorder or complication and so is examined more closely than the general population.
Issues that need to be considered:
(1) overdiagnosis (false positives), especially in borderline cases
(2) detection of subclinical disease which may be irrelevant
(3) detection of a condition at an earlier stage of evolution (which is often beneficial but not always)
(4) underdiagnosis in a low surveillance group, since resources have been allocated to the high surveillance group
This can result in problems for an epidemiological study, especially one done for a fixed time interval. It may appear that a condition is more common in the high surveillance group than a group not so closely scrutinized.
Why a specific group is scrutinized more closely than another may reflect other forms of bias.
