Abstract
We estimated the effects of confounder adjustment as a part of the underlying quantitative risk assessments on the performance of a hypothetical example of a risk-based surveillance system, in which a single risk factor would be used to identify high risk sampling units for testing.
The differences between estimates of surveillance system performance with and without unwarranted confounder adjustment were shown to be of both numerical and economical significance.
Analytical procedures applied to multiple risk factor datasets which yield confounder-adjusted risk estimates should be carefully considered for their appropriateness, if the risk estimates are to be used for informing risk-based surveillance systems.
The differences between estimates of surveillance system performance with and without unwarranted confounder adjustment were shown to be of both numerical and economical significance.
Analytical procedures applied to multiple risk factor datasets which yield confounder-adjusted risk estimates should be carefully considered for their appropriateness, if the risk estimates are to be used for informing risk-based surveillance systems.
Original language | English |
---|---|
Publication date | 2011 |
Number of pages | 1 |
Publication status | Published - 2011 |
Keywords
- Former LIFE faculty