Proof of concept for the role of glycemic control in the early detection of infections in diabetics

Taxiarchis Botsis, Albert M. Lai, George Hripcsak, Walter Palmas, Justin B. Starren, Gunnar Hartvigsen

Research output: Contribution to journalArticle

Abstract

The relationship of infections and glycemic control in diabetes has been previously investigated but no solid findings have been described. Meanwhile, the detection of any infection at the early stages of disease progression, i.e. during the incubation period, is critical. In order to study this topic, we used the infection evidence and the daily glycemic control data of 248 type-2 diabetics who participated in a large telemedicine study. The results showed that morning blood glucose was significantly elevated and that diabetics performed the measurements at a later time when infected. A simple model for predicting the occurrence of infection based on the glycemic control variables showed good performance (sensitivity: 56%, specificity: 92%). A set of variables that synthesize a diabetic's profile could be included in a dedicated model and facilitate the early detection of infections; other aspects, such as continuous self-monitoring and personalized medical records, should be examined in this direction.

Original languageEnglish (US)
Pages (from-to)26-35
Number of pages10
JournalHealth Informatics Journal
Volume18
Issue number1
DOIs
StatePublished - Mar 1 2012
Externally publishedYes

Fingerprint

Infection
Telemedicine
Infection Control
Medical Records
Disease Progression
Blood Glucose
Sensitivity and Specificity
Critical Period (Psychology)
Direction compound

Keywords

  • blood glucose
  • diabetes mellitus
  • early detection of disease
  • incubation period
  • infection

ASJC Scopus subject areas

  • Health Informatics

Cite this

Proof of concept for the role of glycemic control in the early detection of infections in diabetics. / Botsis, Taxiarchis; Lai, Albert M.; Hripcsak, George; Palmas, Walter; Starren, Justin B.; Hartvigsen, Gunnar.

In: Health Informatics Journal, Vol. 18, No. 1, 01.03.2012, p. 26-35.

Research output: Contribution to journalArticle

Botsis, Taxiarchis ; Lai, Albert M. ; Hripcsak, George ; Palmas, Walter ; Starren, Justin B. ; Hartvigsen, Gunnar. / Proof of concept for the role of glycemic control in the early detection of infections in diabetics. In: Health Informatics Journal. 2012 ; Vol. 18, No. 1. pp. 26-35.
@article{69c1b573273d4105a76df83bef021f2c,
title = "Proof of concept for the role of glycemic control in the early detection of infections in diabetics",
abstract = "The relationship of infections and glycemic control in diabetes has been previously investigated but no solid findings have been described. Meanwhile, the detection of any infection at the early stages of disease progression, i.e. during the incubation period, is critical. In order to study this topic, we used the infection evidence and the daily glycemic control data of 248 type-2 diabetics who participated in a large telemedicine study. The results showed that morning blood glucose was significantly elevated and that diabetics performed the measurements at a later time when infected. A simple model for predicting the occurrence of infection based on the glycemic control variables showed good performance (sensitivity: 56{\%}, specificity: 92{\%}). A set of variables that synthesize a diabetic's profile could be included in a dedicated model and facilitate the early detection of infections; other aspects, such as continuous self-monitoring and personalized medical records, should be examined in this direction.",
keywords = "blood glucose, diabetes mellitus, early detection of disease, incubation period, infection",
author = "Taxiarchis Botsis and Lai, {Albert M.} and George Hripcsak and Walter Palmas and Starren, {Justin B.} and Gunnar Hartvigsen",
year = "2012",
month = "3",
day = "1",
doi = "10.1177/1460458211428427",
language = "English (US)",
volume = "18",
pages = "26--35",
journal = "Health Informatics Journal",
issn = "1460-4582",
publisher = "SAGE Publications Ltd",
number = "1",

}

TY - JOUR

T1 - Proof of concept for the role of glycemic control in the early detection of infections in diabetics

AU - Botsis, Taxiarchis

AU - Lai, Albert M.

AU - Hripcsak, George

AU - Palmas, Walter

AU - Starren, Justin B.

AU - Hartvigsen, Gunnar

PY - 2012/3/1

Y1 - 2012/3/1

N2 - The relationship of infections and glycemic control in diabetes has been previously investigated but no solid findings have been described. Meanwhile, the detection of any infection at the early stages of disease progression, i.e. during the incubation period, is critical. In order to study this topic, we used the infection evidence and the daily glycemic control data of 248 type-2 diabetics who participated in a large telemedicine study. The results showed that morning blood glucose was significantly elevated and that diabetics performed the measurements at a later time when infected. A simple model for predicting the occurrence of infection based on the glycemic control variables showed good performance (sensitivity: 56%, specificity: 92%). A set of variables that synthesize a diabetic's profile could be included in a dedicated model and facilitate the early detection of infections; other aspects, such as continuous self-monitoring and personalized medical records, should be examined in this direction.

AB - The relationship of infections and glycemic control in diabetes has been previously investigated but no solid findings have been described. Meanwhile, the detection of any infection at the early stages of disease progression, i.e. during the incubation period, is critical. In order to study this topic, we used the infection evidence and the daily glycemic control data of 248 type-2 diabetics who participated in a large telemedicine study. The results showed that morning blood glucose was significantly elevated and that diabetics performed the measurements at a later time when infected. A simple model for predicting the occurrence of infection based on the glycemic control variables showed good performance (sensitivity: 56%, specificity: 92%). A set of variables that synthesize a diabetic's profile could be included in a dedicated model and facilitate the early detection of infections; other aspects, such as continuous self-monitoring and personalized medical records, should be examined in this direction.

KW - blood glucose

KW - diabetes mellitus

KW - early detection of disease

KW - incubation period

KW - infection

UR - http://www.scopus.com/inward/record.url?scp=84859132585&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84859132585&partnerID=8YFLogxK

U2 - 10.1177/1460458211428427

DO - 10.1177/1460458211428427

M3 - Article

VL - 18

SP - 26

EP - 35

JO - Health Informatics Journal

JF - Health Informatics Journal

SN - 1460-4582

IS - 1

ER -