New Zealand Statistical Association

NZSA 2009

Victoria University of Wellington

Brigid Betz-Stablein

Massey University

Modelling glaucoma with censored and error prone predictors

Glaucoma is the second leading cause of blindness in the world, accounting for 12.3% of blindness worldwide. Glaucoma is caused by a build-up of fluid in the eye, which can lead to permanent, irreparable vision loss. While glaucoma subjects’ eyes are more likely than normal eyes to lack spontaneous venous pulsation (SVP), this can be induced by applying pressure to the eye using an ophthalmodynamometer. Recently, research undertaken by The Lions Eye Institute (Perth, WA) has suggested that the pressure measured, ophthalmodynamometric force (ODF), is related to the severity of glaucoma. The amount of data available is limited; therefore it is important to employ efficient statistical models to investigate this relationship. Recorded ODF values are subject to measurement error, which is complicated by heteroscedastic variation and the fact that the data is also censored at zero. We employed hierarchical models within a Bayesian framework to investigate the effect of change in errorless ODF on increased excavation of the optic disk, a measure of glaucoma progression. Our results showed that size of the effect of ODF on glaucoma progression increased compared to models that do not account for measurement error and censoring. Therefore change in ODF may provide more information than originally thought to assist in glaucoma prognostication, and may play a larger role in assessing the likelihood of glaucoma progression.
Contact Us | Section Map | Disclaimer | RSS feed RSS FeedBack to top ^

Valid XHTML and CSS | Built on Foswiki

Page Updated: 31 Jul 2009 by haywoodj. © Victoria University of Wellington, New Zealand, unless otherwise stated. Header image used and relicensed under Creative Commons. Original author: Djof.