RT Journal
A1 Zilberberg MD
T1 Bayesian thinking: A safer tool for reassurance
JF Archives of Internal Medicine
JO Archives of Internal Medicine
YR 2011
FD April 11
VO 171
IS 7
SP 702
OP 710
DO 10.1001/archinternmed.2011.111
UL http://dx.doi.org/10.1001/archinternmed.2011.111
AB 
Ideally, a test is ordered to augment the accuracy of decision making, given the particular patient's clinical presentation. A perfect test, while identifying the disease 100% of the time when present (100% sensitivity), would never misidentify those without the disease as having it (100% specificity). Unfortunately, no such test exists, and for this reason a clinician must consider the value of a positive (positive predictive value) or a negative (negative predictive value) result in the context of the given patient (pretest probability of the disease).