prev next 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 |31 |32 |review
Sensitivity and specificity values do have some shortcomings. First, they describe the rates of true tests given that the condition is either present or absent. In the clinical situation, however, we know the test result only. Second, specificity or sensitivity can be artificially inflated at the expense of the other. An answer to these difficulties can be found in the calculation of likelihood ratios.