Joint Modeling of Longitudinal Visual Field Changes and Time to Detect Progression in Glaucoma Patients: A Secondary Data Analysis.

Background: Glaucoma causes irreversible damage to the optic nerve and can lead to blindness if it is not treated appropriately. Evaluation of longitudinal changes in the visual field (VF) and detecting progression in a timely manner are critical for effective disease management. This study aimed to identify factors associated with VF impairment and disease progression using a Bayesian joint model.

Methods: A total of 129 glaucoma patients (228 eyes) were recruited from an ongoing cohort study initiated in 1998 at the Rotterdam Eye Hospital in the Netherlands. Standard Automated Perimetry (SAP) was performed for each patient at regular 6-month follow-up intervals. Covariates included sex, age at baseline, mean intraocular pressure (IOP), and disease severity. A Bayesian joint model was employed, integrating a linear mixed effects model (LMM) for longitudinal mean deviation (MD) values and a Cox proportional hazards model for progression time. The statistical analyses were conducted using R software and the 'JMbayes2' package.

Results: Progression was observed in 33.8% of eyes. A significant association was found between MD changes and progression risk (α=-0.39, P<0.001). Older age (P=0.01), early-stage disease (P<0.001), and higher mean IOP (P<0.001) were associated with an increased risk of progression.

Conclusion: Considering longitudinal MD changes, age at baseline, mean IOP, and disease severity were significantly associated with the time to progression detection. Sex was not found to be a significant factor in glaucoma progression.