東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

A data-driven approach to optimizing...

Holmes, Ashley.

FindBook

Google Book

Amazon

博客來

A data-driven approach to optimizing dis-enrollment for diabetes patients in care management.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	A data-driven approach to optimizing dis-enrollment for diabetes patients in care management./
作者:	Holmes, Ashley.
面頁冊數:	143 p.
附註:	Source: Masters Abstracts International, Volume: 55-05.
Contained By:	Masters Abstracts International55-05(E).
標題:	Systems science. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10133956
ISBN:	9781339924588

A data-driven approach to optimizing dis-enrollment for diabetes patients in care management.
Holmes, Ashley.

A data-driven approach to optimizing dis-enrollment for diabetes patients in care management. - 143 p.

Source: Masters Abstracts International, Volume: 55-05.

Thesis (M.S.)--State University of New York at Binghamton, 2016.

Care management organizations help improve overall population health and wellness by offering specialized programs that target specific chronic conditions with evidence-based interventions, in addition to providing care coordination, wellness promotion, medical management, and other services. Chronic care management programs target common diseases and conditions with several important characteristics. These diseases typically have a significant effect on population wellness, and if patients improperly manage them, they develop deadly complications that are often costly to the industry. The ideal diseases targeted by chronic care management have a plethora of evidence-based interventions that to prevent or delay such complications. As the care management industry becomes responsible for an increasing portion of the population, care management programs need to place an increased emphasis on the enrollment of the most appropriate members, and the dis-enrollment of these members when their health is no longer benefitting from receiving care management services.

ISBN: 9781339924588Subjects--Topical Terms:

3168411
Systems science.

A data-driven approach to optimizing dis-enrollment for diabetes patients in care management.
LDR:04504nmm a2200325 4500 001 2079187
005 20161210083247.5
008 170521s2016 ||||||||||||||||| ||eng d
020 $a 9781339924588
035 $a (MiAaPQ)AAI10133956
035 $a AAI10133956
040 $a MiAaPQ $c MiAaPQ
100 1 $a Holmes, Ashley. $3 3194850
245 1 2 $a A data-driven approach to optimizing dis-enrollment for diabetes patients in care management.
300 $a 143 p.
500 $a Source: Masters Abstracts International, Volume: 55-05.
500 $a Adviser: Mohammad T. Khasawneh.
502 $a Thesis (M.S.)--State University of New York at Binghamton, 2016.
520 $a Care management organizations help improve overall population health and wellness by offering specialized programs that target specific chronic conditions with evidence-based interventions, in addition to providing care coordination, wellness promotion, medical management, and other services. Chronic care management programs target common diseases and conditions with several important characteristics. These diseases typically have a significant effect on population wellness, and if patients improperly manage them, they develop deadly complications that are often costly to the industry. The ideal diseases targeted by chronic care management have a plethora of evidence-based interventions that to prevent or delay such complications. As the care management industry becomes responsible for an increasing portion of the population, care management programs need to place an increased emphasis on the enrollment of the most appropriate members, and the dis-enrollment of these members when their health is no longer benefitting from receiving care management services.
520 $a One of the most targeted diseases for chronic care management is diabetes, a group of diseases characterized by the body's inability to produce a sufficient amount of insulin. Currently, many diabetes management programs determine enrollment and dis-enrollment eligibility based upon a singular clinical criterion, the (glycated hemoglobin) HbA1c level. However, an abundance of literature indicates that there are many other contributors to the health of a person with diabetes, such as diet quality, blood pressure, and body mass index. Although HbA1c is useful as a current indicator of health status, by itself it is a poor predictor of future complications.
520 $a This thesis proposes the development of a clinical decision support system that uses multiple clinical indicators to help aid clinician's in determining member dis-enrollment eligibility. Data from a care management company was collected, and four different data mining models (decision tree, logistic regression, support vector machine, and artificial neural network) were developed and compared to determine the best one for use in a decision support system. A number of measures (accuracy, sensitivity, specificity, and area under the curve) were used to evaluate the different models. The artificial neural network model achieved the highest accuracy (87.07%) for the identification of members eligible for dis-enrollment.
520 $a Furthermore, a systematic methodology is used to develop a data-driven Diabetes-specific Healthy Eating Index (DS-HEI) as a way to quantitatively measure diet quality of people with diabetes. This diet quality index identifies the food components (whole grain and total dairy) that significantly affect HbA1c. Neither the HEI-2010 or the DS-HEI was found to have a significant correlation with HbA1c score nor participants of the 2007--2008 National Health and Nutrition Examination Survey (NHANES) that have 24 hour food recall data and HbA1c level available. To the best of the authors' knowledge, this is likely because of the small cohort of available data to build the DS-HEI.
520 $a For future research, dis-enrollment criteria for other chronic care management programs and populations can be included, as well as larger cohort of people with diabetes and dietary data to verify the findings from this study. Additionally, future work could include an analysis of members in diabetes care management who were dis-enrolled and then re-enrolled as a result of their poor health returning.
590 $a School code: 0792.
650 4 $a Systems science. $3 3168411
650 4 $a Health care management. $3 2122906
690 $a 0790
690 $a 0769
710 2 $a State University of New York at Binghamton. $b Systems Science Industrial Engineering. $3 2104041
773 0 $t Masters Abstracts International $g 55-05(E).
790 $a 0792
791 $a M.S.
792 $a 2016
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10133956