Diagnostic tests are an extension of clinical examination for signs. They are not 100% accurate or reliable. Tests are not necessarily more accurate or more objective than physical signs. Test results can not stand alone; they are interpreted in light of other relevant clinical data.

Validity is when a test measures what it is supposed to measure. Sensitivity, specificity, and predictive value are measures of validity (accuracy). Sensitivity is a measure of the strength of association. Specificity measures the uniqueness of association.

The figure below shows the relation between test results and diagnosis.

The following parameters can be read from the figure:

True positives (TP) = a;

True negatives             (TN)=d;

False negative  (FN)=b;

False positives             (FP)=c;

Sensitivity =     a/(a+b);

Specificity =    d(/c+d).

Both FP and FN have serious ethico-legal consequences. FP cases can be hospitalized and undergo expensive and risky diagnostic confirmatory procedures that are not necessary. FN cases

have a disease but think they are healthy and will not pay attention to early warning symptoms in order to seek early medical attention.

There is a trade-off between specificity and sensitivity. High sensitivity is associated with low specificity & vice versa. High specificity is associated with low sensitivity & vice versa.

Interest focuses on how well a particular test predicts the true diagnosis. This includes the positive predictive value (detection of disease) and the negative predictive value (correct indication of absence of disease). The relationships between test results and the true diagnosis can be seen in figure below

Predictive value of positive test (PV+ve)= a/ (a+b).