Research Repository

A joint modelling approach for clustered recurrent events and death events

Bao, Y and Dai, H and Wang, T and Chuang, SK (2013) 'A joint modelling approach for clustered recurrent events and death events.' Journal of Applied Statistics, 40 (1). 123 - 140. ISSN 0266-4763

Full text not available from this repository.

Abstract

In dental implant research studies, events such as implant complications including pain or infection may be observed recurrently before failure events, i.e. the death of implants. It is natural to assume that recurrent events and failure events are correlated to each other, since they happen on the same implant (subject) and complication times have strong effects on the implant survival time. On the other hand, each patient may have more than one implant. Therefore these recurrent events or failure events are clustered since implant complication times or failure times within the same patient (cluster) are likely to be correlated. The overall implant survival times and recurrent complication times are both interesting to us. In this paper, a joint modelling approach is proposed for modelling complication events and dental implant survival times simultaneously. The proposed method uses a frailty process to model the correlation within cluster and the correlation within subjects. We use Bayesian methods to obtain estimates of the parameters. Performance of the joint models are shown via simulation studies and data analysis. © 2013 Copyright Taylor and Francis Group, LLC.

Item Type: Article
Subjects: Q Science > QA Mathematics
Divisions: Faculty of Science and Health > Mathematical Sciences, Department of
Depositing User: Jim Jamieson
Date Deposited: 12 Feb 2013 08:04
Last Modified: 02 Sep 2019 21:15
URI: http://repository.essex.ac.uk/id/eprint/5499

Actions (login required)

View Item View Item