© John Wiley & Sons, Inc.
FIGURE 13-4: Results from a study of a new experimental home pregnancy test.
The structure of the table in Figure 13-4 is important because if your results are arranged in that way,
you can easily calculate at least five important characteristics of the experimental test (in our case, the
home pregnancy test) from this table: accuracy, sensitivity, specificity, positive predictive value
(PPV), and negative predictive value (NPV). We explain how in the following sections.
Overall accuracy
Overall accuracy measures how often a test result comes out the same as the gold standard result
(meaning the test is correct). A perfectly accurate test never produces false positive or false negative
results. In Figure 13-4, cells a and d represent correct test results, so the overall accuracy of the home
pregnancy test is
. Using the data in Figure 13-4,
, which is 0.84,
or 84 percent.
Sensitivity and specificity
A perfectly sensitive test never produces a false negative result for an individual with the condition.
Conversely, a perfectly specific test never produces a false positive result for an individual negative
for the condition. The goal of developing a screening test is to balance the sensitivity against the
specificity of the test based on the context of the condition, to optimize the accuracy and get the best of
both worlds while minimizing both false positive and false negative results.
In a screening test with a sensitivity of 100 percent, the test result is always positive whenever the
condition is truly present. In other words, the test will identify all individuals who truly have the
condition. When a perfectly sensitive test comes out negative, you can be sure the person doesn’t have
the condition. You calculate sensitivity by dividing the number of true positive cases by the total
number of cases where the condition was truly present: a/c1 (that is, true positive/all present). Using
the data in Figure 13-4,
, which is 0.89. This means that the home test comes out
positive in only 89 percent of truly pregnant women, and the other 11 percent were really pregnant, but
had a false positive result in the test.
A perfectly specific test never produces a false positive result for an individual without the condition.
In a test that has a specificity of 100 percent, whenever the condition is truly absent, the test always
has a negative result. In other words, the test will identify all individuals who truly do not have the
condition. When a perfectly specific test comes out positive, you can be sure the person has the
condition. You calculate specificity by dividing the number of true negative cases by the total number
of cases where the condition was truly absent:
(that is, true negative/all not present). Using the