Abstract
In a study using US nationwide claims data from 2008-2017, the median time between visual field tests for patients with glaucoma was once every 1.59 years (interquartile range: 1.14-3.03 years). This is less frequent than the AAO-recommended minimum annual screening. We conducted simulation modeling using real-world clinical data to determine how long it would take to identify trend-based glaucoma visual field progression at this testing frequency and compare it to AAO-recommended annual screening.
Simulation modeling using retrospective clinical data SUBJECTS: Patients with glaucoma in the Duke Glaucoma Registry (DGR) METHODS: To determine how long it would take to identify glaucoma progression at testing frequencies in this insured US nationwide population, we conducted simulation modeling using real-world visual field data from the DGR. Models were created to evaluate the time it would take to identify progression for patients with testing frequency at the 25
, 50
, and 75
percentiles in the previously published US nationwide claims data and at the AAO-recommended minimum annual screening. The simulations assumed baseline mean deviation (MD) of -5, -10, or -15 dB and a true rate of progression of -0.5 or -1.0 dB/year. Progression was defined as having a negative MD slope estimate with a p-value <0.05. We simulated 1000 sequences of visual fields for each test group.
Time to detect glaucoma progression RESULTS: Across the testing scenarios, the time to detect progression in 80% of eyes ranged from 15.2 to 24.3 years for patients receiving visual field testing at the 25
percentile (3.03 years between tests), 9.6 to 17.5 years for patients receiving visual field testing at the 50
percentile (1.59 years between tests), 8.0 to 14.9 years for patients receiving visual field testing at the 75
percentile (1.14 years between tests), and 7.1 to 14.1 years for patients receiving annual testing.
These results suggest that either visual field testing needs to be done more frequently or other methods of identification of progressive glaucoma vision loss need to be developed. Clinical decision support systems could help tailor the frequency of visual field testing to the needs of individual patients.