Analysis of reliability indices from Humphrey visual field tests in an urban glaucoma population

Purpose: Visual field assessment is extremely important in glaucoma management, but interpretation is affected by the quality of the patient's performance. The authors have investigated the reliability of visual field performance by a randomly selected sample of the chronic glaucoma population at an urban tertiary care practice. Methods: Patient reliability in Humphrey automated visual field testing was studied in 106 randomly selected chronic open-angle glaucoma patient charts, which provided 768 tests (mean, 7.2 ± 4.8 fields; range, 2–18 fields). Reliability criteria were established as less than 20% fixation losses, less than 33% false-negative error, and less than 33% false-positive error, as recommended by Humphrey Instruments, Inc (San Leandro, CA). Results: Patients performed reliably in 61% of right eye fields, 58% of left eye fields, and 59.5% overall. Of the 106 patients, only 35 (33%) were always reliable in both eyes, whereas 8 (7.5%) were always unreliable in both eyes. The most common cause of unreliability was fixation loss (39%), whereas false-positive error (5%) and false-negative error (9%) were less frequent. A more severely depressed mean deviation correlated significantly with poorer performance on the three reliability indices, with false-negative error having the greatest correlation, followed by fixation loss and false-positive error. Corrected pattern standard deviation correlated closely only with false-negative error. Prolonged test time also correlated with all three reliability indices. Age was a significant factor for fixation loss but not for false-negative or false-positive error. Conclusions: The authors conclude that fewer than two thirds of the Humphrey visual fields were reliable with the authors' urban tertiary care population of patients with glaucoma. Relaxing the fixation loss criterion to less than 33% improved the rate of reliability to approximately 75%. The severity of glaucomatous visual field defects, test time, and age were identified as factors influencing the reliability of the Humphrey visual fields.