Decision-Making Models

Modeling decisions and treatment options in dentistry are particularly difficult since they involve risk that is continuous over time, and timing in dental care is important.  For instance, caries and periodontal diseases can both occur at a specific time in the past and 're-occur' at another time in the future.  Additionally, these disease entities can happen more than once, and they can occur in multiple sites.  Humans have more than one tooth that is vulnerable to disease, and teeth can be restored more than once.  Under these 'real world' conditions, the use of conventional decision trees may require unrealistic simplifying assumptions.

Here, the Markov Model can be used to model prognoses for clinical problems with ongoing risk and changing transition probabilities over time.⁴  In prosthodontic practice it is now possible to calculate the risk of a patient becoming edentulous based upon the patient's actual state within the prosthodontic cycle.²  Algorithms adapted from Hollenberg enable analysts to determine cost-effectiveness as well as the financial cost of being in a certain state for a single or multiple cycle.⁵  Since clinical data are needed to perform such calculations, the need to develop a large database becomes apparent.  The probabilities for the resulting causal network are learned and adjusted from stored findings in a corresponding relational database.  Initial versions of the Dental Clinical Advisory System, which is in progress, are described in a 2000 paper by Umar, et al.³

Click Here To Navigate Within Article >> Abstract Introduction Decision-Making Models Expert Systems For Clinical Decision-Making The EOHR and Capturing Patient Care Data Diagnostic Technology Treatment Planning, Pretreatment, Treatment Dental Implications Recall and Outlook Conclusion References About the Author
	Page 3 of 16
Citation Number: Vol. 3, No. 1, Page 029