東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Modeling individual healthcare expen...

Gao, Jie.

Linked to FindBook

Google Book

Amazon

博客來

Modeling individual healthcare expenditures by extending the two-part model.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Modeling individual healthcare expenditures by extending the two-part model./
Author:	Gao, Jie.
Description:	151 p.
Notes:	Advisers: Edward W. Frees; Marjorie A. Rosenberg.
Contained By:	Dissertation Abstracts International68-12A.
Subject:	Business Administration, Management. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3294044
ISBN:	9780549383437

Modeling individual healthcare expenditures by extending the two-part model.
Gao, Jie.

Modeling individual healthcare expenditures by extending the two-part model. - 151 p.

Advisers: Edward W. Frees; Marjorie A. Rosenberg.

Thesis (Ph.D.)--The University of Wisconsin - Madison, 2007.

Frequency or the number of utilization is generated from a negative binomial process for both the inpatient and outpatient care. Severity or expenditures per event is assumed to be log-normal or gamma.

ISBN: 9780549383437Subjects--Topical Terms:

626628
Business Administration, Management.

Modeling individual healthcare expenditures by extending the two-part model.
LDR:03567nam 2200373 a 45 001 940709
005 20110518
008 110518s2007 ||||||||||||||||| ||eng d
020 $a 9780549383437
035 $a (UMI)AAI3294044
035 $a AAI3294044
040 $a UMI $c UMI
100 1 $a Gao, Jie. $3 1264840
245 1 0 $a Modeling individual healthcare expenditures by extending the two-part model.
300 $a 151 p.
500 $a Advisers: Edward W. Frees; Marjorie A. Rosenberg.
500 $a Source: Dissertation Abstracts International, Volume: 68-12, Section: A, page: 5130.
502 $a Thesis (Ph.D.)--The University of Wisconsin - Madison, 2007.
520 $a Frequency or the number of utilization is generated from a negative binomial process for both the inpatient and outpatient care. Severity or expenditures per event is assumed to be log-normal or gamma.
520 $a Methods for estimating healthcare expenditures link the financial risks of providing health services to individuals' characteristics. This dissertation developed statistical methods based on extensions of the two-part model (TPM).
520 $a The emphasis of this dissertation is on two extensions of the TPM. The first extension is an aggregate loss approach that models frequency of utilization and severity for both inpatient and outpatient care. Total individual healthcare expenditures are expressed as a combination between frequency and severity. The second extension is based on the interdependency between inpatient and outpatient care. I developed a joint aggregate loss model where I conditioned counts of outpatient visits on inpatient admissions. A random effects specification as well as a copula function were introduced to explain correlations among expenditures by subject.
520 $a Results indicate that modeling the frequency of utilization is an important step in modeling total healthcare expenditures as many individual characteristics are significant in explaining the number of inpatient and outpatient care events. Inpatient and outpatient care are complements and the joint dependency is largely due to inpatient admissions inducing more frequent outpatient care.
520 $a The log-normal distribution provides support for the principal-agent relationship and better fit than the gamma distribution for both inpatient and outpatient severity. With the gamma assumption, more personal characteristics become significant in determining the severity. The differences in estimates from the two distribution assumptions on severity can be due to log-scale errors that are heterogeneous.
520 $a In out-of-sample predictions, the aggregate loss approach outperforms existing methods when medians of the gamma distribution are used to predict inpatient severity. The joint aggregate loss models improve on existing methods when means and medians of the log-normal distribution are used as predictors for inpatient severity. Predictions of outpatient expenditures from the developed methods are comparable to competing methods. Results from predictions suggest that medians of distributions are meaningful predictors because of the large mass at zero and the skewness observed in healthcare data.
590 $a School code: 0262.
650 4 $a Business Administration, Management. $3 626628
650 4 $a Economics, General. $3 1017424
650 4 $a Health Sciences, Health Care Management. $3 1017922
650 4 $a Statistics. $3 517247
690 $a 0454
690 $a 0463
690 $a 0501
690 $a 0769
710 2 $a The University of Wisconsin - Madison. $3 626640
773 0 $t Dissertation Abstracts International $g 68-12A.
790 $a 0262
790 1 0 $a Frees, Edward W., $e advisor
790 1 0 $a Rosenberg, Marjorie A., $e advisor
791 $a Ph.D.
792 $a 2007
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3294044