Abstract
This study assessed the impact of race and ethnicity on longitudinal test variability and time to detect glaucoma progression using standard automated perimetry (SAP) and optical coherence tomography (OCT). The sample consisted of 47,003 SAP tests from 5402 eyes and 25,480 OCT tests from 4125 eyes, with 20% of participants self-identifying as Black or African American and 80% as White; 29% as Hispanic or Latino and 71% as Not Hispanic or Latino. Variability was measured using standard deviations of residuals from linear regression models for SAP mean deviation (MD) and OCT retinal nerve fiber layer (RNFL) thickness over time. Results showed significantly greater SAP variability in Black or African American (1.80 ± 1.30 dB) compared to White participants (1.56 ± 1.21 dB; P < 0.001) and in Hispanic or Latino (1.81 ± 1.46 dB) compared to Not Hispanic or Latino individuals (1.52 ± 1.10 dB; P < 0.001). OCT variability was higher in Black or African American (2.3 ± 1.5 μm) compared to White (2.1 ± 1.3 μm; P < 0.001) and in Not Hispanic or Latino (2.2 ± 1.3 μm) compared to Hispanic or Latino (2.1 ± 1.2 μm; P = 0.029). Increased SAP variability delayed progression detection, while OCT showed minimal differences. These findings suggest that higher perimetric variability in Black or African American and Hispanic or Latino may affect glaucoma progression detection using SAP.