|front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |19 |20 |21 |22 |23 |24 |25 |26 |27 |28 |29 |30 |review|
A number of parameters may be used to define a testís performance. The performance characteristics of assays may be assessed at a point in time to provide baseline data for future quality assurance.
Specificity and sensitivity are traditionally the most important of the performance characteristics.
The sensitivity of an assay is defined as its ability to detect antibody in samples that actually contain antibody . Therefore, one must know the HIV status of the individuals from whom the samples were drawn or results obtained in other assays before the assay's sensitivity can be assessed.
The specificity of an assay is defined as its ability to identify samples that do not contain antibody as negative. Thus, the specificity of an assay is determined in large populations known to be anti-HIV negative. Australian blood donors make up such a population, where only 46 in 5.4 million donations have been HIV positive from 1985-1990.
Evaluations must be scientifically rigorous and statistically valid.