Can low accuracy disease risk predictor models improve health care using decision support systems?

Douglas Benn, D. D. Dankel, S. H. Kostewicz

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

A prototype decision support system has been designed for managing dental caries using a risk assessment model. Caries is a multifactorial disease with risk prediction models having low sensitivity (65%) and moderate specificity (80%) for 2 or more new lesions. These models are inaccurate for targeting resources at high risk people. However, low risk individuals can be more accurately identified. If the activity of early tooth decay lesions, in low risk people, are monitored over time and only lesions beyond 1/3 of the dentin depth are filled, the number of annual fillings may be reduced by 50%. Currently, most US dental schools do not teach risk assessment for caries and encourage early treatment of lesions leading to a repair destruction cycle. The combination of a decision support system with a moderate accuracy specificity risk model for predicting low risk individuals may produce a significant improvement in caries management.

Original languageEnglish
Pages (from-to)577-581
Number of pages5
JournalProceedings / AMIA ... Annual Symposium. AMIA Symposium
StatePublished - 1998
Externally publishedYes

Fingerprint

Delivery of Health Care
Dental Schools
Dental Caries
Dentin
Tooth

Cite this

Can low accuracy disease risk predictor models improve health care using decision support systems? / Benn, Douglas; Dankel, D. D.; Kostewicz, S. H.

In: Proceedings / AMIA ... Annual Symposium. AMIA Symposium, 1998, p. 577-581.

Research output: Contribution to journalArticle

@article{2068d7c07b7343088a1c739a57b0420a,
title = "Can low accuracy disease risk predictor models improve health care using decision support systems?",
abstract = "A prototype decision support system has been designed for managing dental caries using a risk assessment model. Caries is a multifactorial disease with risk prediction models having low sensitivity (65{\%}) and moderate specificity (80{\%}) for 2 or more new lesions. These models are inaccurate for targeting resources at high risk people. However, low risk individuals can be more accurately identified. If the activity of early tooth decay lesions, in low risk people, are monitored over time and only lesions beyond 1/3 of the dentin depth are filled, the number of annual fillings may be reduced by 50{\%}. Currently, most US dental schools do not teach risk assessment for caries and encourage early treatment of lesions leading to a repair destruction cycle. The combination of a decision support system with a moderate accuracy specificity risk model for predicting low risk individuals may produce a significant improvement in caries management.",
author = "Douglas Benn and Dankel, {D. D.} and Kostewicz, {S. H.}",
year = "1998",
language = "English",
pages = "577--581",
journal = "Proceedings / AMIA . Annual Symposium. AMIA Symposium",
issn = "1531-605X",
publisher = "Hanley & Belfus",

}

TY - JOUR

T1 - Can low accuracy disease risk predictor models improve health care using decision support systems?

AU - Benn, Douglas

AU - Dankel, D. D.

AU - Kostewicz, S. H.

PY - 1998

Y1 - 1998

N2 - A prototype decision support system has been designed for managing dental caries using a risk assessment model. Caries is a multifactorial disease with risk prediction models having low sensitivity (65%) and moderate specificity (80%) for 2 or more new lesions. These models are inaccurate for targeting resources at high risk people. However, low risk individuals can be more accurately identified. If the activity of early tooth decay lesions, in low risk people, are monitored over time and only lesions beyond 1/3 of the dentin depth are filled, the number of annual fillings may be reduced by 50%. Currently, most US dental schools do not teach risk assessment for caries and encourage early treatment of lesions leading to a repair destruction cycle. The combination of a decision support system with a moderate accuracy specificity risk model for predicting low risk individuals may produce a significant improvement in caries management.

AB - A prototype decision support system has been designed for managing dental caries using a risk assessment model. Caries is a multifactorial disease with risk prediction models having low sensitivity (65%) and moderate specificity (80%) for 2 or more new lesions. These models are inaccurate for targeting resources at high risk people. However, low risk individuals can be more accurately identified. If the activity of early tooth decay lesions, in low risk people, are monitored over time and only lesions beyond 1/3 of the dentin depth are filled, the number of annual fillings may be reduced by 50%. Currently, most US dental schools do not teach risk assessment for caries and encourage early treatment of lesions leading to a repair destruction cycle. The combination of a decision support system with a moderate accuracy specificity risk model for predicting low risk individuals may produce a significant improvement in caries management.

UR - http://www.scopus.com/inward/record.url?scp=0032239217&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032239217&partnerID=8YFLogxK

M3 - Article

SP - 577

EP - 581

JO - Proceedings / AMIA . Annual Symposium. AMIA Symposium

JF - Proceedings / AMIA . Annual Symposium. AMIA Symposium

SN - 1531-605X

ER -