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

Surrogate endpoints in clinical tria...

Xu, Jane (Jing).

Linked to FindBook

Google Book

Amazon

博客來

Surrogate endpoints in clinical trials: A Markov chain Monte Carlo approach.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Surrogate endpoints in clinical trials: A Markov chain Monte Carlo approach./
Author:	Xu, Jane (Jing).
Description:	116 p.
Notes:	Adviser: Scott Zeger.
Contained By:	Dissertation Abstracts International60-11B.
Subject:	Biology, Biostatistics. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9950618
ISBN:	0599528117

Surrogate endpoints in clinical trials: A Markov chain Monte Carlo approach.
Xu, Jane (Jing).

Surrogate endpoints in clinical trials: A Markov chain Monte Carlo approach. - 116 p.

Adviser: Scott Zeger.

Thesis (Ph.D.)--The Johns Hopkins University, 2000.

Longitudinal data in biomedical and public health research usually comprise repeated measures of markers of health or disease data as well as times to key clinical events. In many longitudinal studies, particularly randomized trials, the main question is whether the time to a clinical event <italic> T</italic> depends upon a treatment <italic>X</italic>. When the disease under study is chronic or with long latency periods, it can take many years and be very expensive to finish the study if we only use time-to-event data. One alternative is to use biological marker data <italic>Y</italic> as a “surrogate endpoint” for <italic>T</italic>. In this situation, models for the joint distribution of <italic>T</italic> and <italic>Y</italic> given <italic> X</italic> are valuable in assessing whether <italic>Y</italic> is a useful surrogate for <italic>T</italic> in studying <italic>X</italic>. In this thesis, I present a general model for the joint analysis of [<italic>T</italic>, <italic> Y</italic>|<italic>X</italic>] and apply the model to estimate the marginal distribution [<italic>T</italic>|<italic>X</italic>] using the relevant information in <italic>Y</italic>. I adopt a latent variable formulation like that of Fawcett and Thomas (1996) and extend it to the problem of surrogate endpoints. A Markov chain Monte Carlo algorithm is implemented for estimating parameters of interest. Several quantities of clinical relevance are proposed to determine the efficacy of a treatment. I illustrate the joint model for time to event and repeated measures with analysis of data from a randomized trial comparing risperidone to placebo for treatment of persons with schizophrenia. I extend the work to handle the possibility of multiple surrogate endpoints. The use of this methodology in validating surrogate endpoints from multi-trial study is discussed and sensitivity analyses are carried out.

ISBN: 0599528117Subjects--Topical Terms:

1018416
Biology, Biostatistics.

Surrogate endpoints in clinical trials: A Markov chain Monte Carlo approach.
LDR:02795nam 2200277 a 45 001 929281
005 20110427
008 110427s2000 eng d
020 $a 0599528117
035 $a (UnM)AAI9950618
035 $a AAI9950618
040 $a UnM $c UnM
100 1 $a Xu, Jane (Jing). $3 1252765
245 1 0 $a Surrogate endpoints in clinical trials: A Markov chain Monte Carlo approach.
300 $a 116 p.
500 $a Adviser: Scott Zeger.
500 $a Source: Dissertation Abstracts International, Volume: 60-11, Section: B, page: 5291.
502 $a Thesis (Ph.D.)--The Johns Hopkins University, 2000.
520 $a Longitudinal data in biomedical and public health research usually comprise repeated measures of markers of health or disease data as well as times to key clinical events. In many longitudinal studies, particularly randomized trials, the main question is whether the time to a clinical event <italic> T</italic> depends upon a treatment <italic>X</italic>. When the disease under study is chronic or with long latency periods, it can take many years and be very expensive to finish the study if we only use time-to-event data. One alternative is to use biological marker data <italic>Y</italic> as a “surrogate endpoint” for <italic>T</italic>. In this situation, models for the joint distribution of <italic>T</italic> and <italic>Y</italic> given <italic> X</italic> are valuable in assessing whether <italic>Y</italic> is a useful surrogate for <italic>T</italic> in studying <italic>X</italic>. In this thesis, I present a general model for the joint analysis of [<italic>T</italic>, <italic> Y</italic>|<italic>X</italic>] and apply the model to estimate the marginal distribution [<italic>T</italic>|<italic>X</italic>] using the relevant information in <italic>Y</italic>. I adopt a latent variable formulation like that of Fawcett and Thomas (1996) and extend it to the problem of surrogate endpoints. A Markov chain Monte Carlo algorithm is implemented for estimating parameters of interest. Several quantities of clinical relevance are proposed to determine the efficacy of a treatment. I illustrate the joint model for time to event and repeated measures with analysis of data from a randomized trial comparing risperidone to placebo for treatment of persons with schizophrenia. I extend the work to handle the possibility of multiple surrogate endpoints. The use of this methodology in validating surrogate endpoints from multi-trial study is discussed and sensitivity analyses are carried out.
590 $a School code: 0098.
650 4 $a Biology, Biostatistics. $3 1018416
650 4 $a Statistics. $3 517247
690 $a 0308
690 $a 0463
710 2 0 $a The Johns Hopkins University. $3 1017431
773 0 $t Dissertation Abstracts International $g 60-11B.
790 $a 0098
790 1 0 $a Zeger, Scott, $e advisor
791 $a Ph.D.
792 $a 2000
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9950618