# Time to Blindness and Associated Factors Among Glaucoma Patients Under Treatment at Governmental Hospitals in Amhara Region, Ethiopia; Application of Gamma Shared Frailty Accelerate Failure Time Model

## TL;DR

This study examines how long it takes for glaucoma patients in Ethiopia to go blind and identifies factors that influence this time.

## Contribution

The study applies a gamma shared frailty accelerated failure time model to analyze time to blindness in glaucoma patients in Ethiopia.

## Key findings

- Age, education, and urban residence are associated with longer time to blindness.
- Poor IOP control and comorbidities like hypertension and diabetes are linked to shorter time to blindness.
- Hospital differences significantly impact patient outcomes, suggesting a need for standardized treatment protocols.

## Abstract

In Ethiopia, glaucoma and related cases account for 5.2% of the irreversible blindness and is the fifth main cause of blindness. The aim of this study was to investigate the time to blindness and associated factors among glaucoma patients under treatment at government hospitals in Amhara region, Ethiopia.

A retrospective cohort study using a gamma shared frailty accelerated failure time model was conducted at 5 randomly selected public hospitals in Amhara region. About 1102 glaucoma patients, whose follow ups, from January 2018 to December 2023 were randomly selected using stratified random sampling technique.

The results of the multivariable Gamma Shared Frailty accelerate failure time model indicates that age [ϕ = 0.694; 95% CI: (0.408, 0.773)], literate glaucoma patients [ϕ = 1.163, 95% CI: (1.062, 1.206)], urban glaucoma patients [ϕ = 1.157; 95% CI: (1.049, 1.274)], non‐hypertensive glaucoma [ϕ = 0.812; 95%: (0.673, 0.979)], non‐diabetic glaucoma patients [ϕ = 0.949; 95% CI: (0.533, 0.687)], housewives [AOR = 0.772; 95% CI: (0.428, 0.975)], farmers Glaucoma patients (ϕ = 0.969; 95% CI: (0.735, 0.999)], moderate glaucoma stage [ϕ = 0.791; 95% CI: (0.419, 0.890)], advanced stage glaucoma patients [ϕ = 0.959; 95% CI: (0.703, 0.998)], glaucoma patients with cup‐disk ratio (CDR) > 0.7 [ϕ = 0.931; 95% CI: (0.739, 0. 998)] and glaucoma patients with normal IOP (ϕ = 1.058; 95% CI: (1.019, 1.285)] significantly affected the time to blindness for glaucoma patients.

Age, lack of education, and residing in a rural area were all associated with a shorter time to blindness. Conversely, patients with better IOP control, no concurrent hypertension or diabetes, and those with a lower cup‐to‐disc ratio (CDR) showed a longer time to blindness. The significant frailty effect confirmed that there is substantial variation in patient outcomes across different hospitals, likely due to differences in healthcare quality and provision. To minimize this disparity and ensure consistent, high‐quality care, it is critically important to standardize treatment protocols across all facilities.