This gaper presents the combined use of decision trees and cost-effect
iveness analysis to determine on which patients new diagnostic tests c
an be useful, and whether they can change the dentist's decision to pr
ovide different services and thus improve patient outcomes in 5 differ
ent categories of patients. A decision tree is constructed in which th
e primary decision branches are ''test'' and ''no test''. The treatmen
t and outcome branches in the test branch are affected by the results
of the test which is then either negative or positive. In contrast, th
e treatment and outcome branches in the no-test branch represent pathw
ays for patient therapy when the new (or additional) test is not used.
Results show that even with excellent test efficacy parameters (sensi
tivity 0.87 and specificity 0.84) it is not cost effective to use a di
agnostic test in the general population or for adults under age 45 yea
rs. However for adults > 45 years, patients referred to a periodontist
, and patients with a history of periodontal disease, the test would c
hange treatment decisions and improve patient care outcomes.