An improved method to estimate frequency of false positive answers in computerized perimetry

Reliability of patient performance in static computerized perimetry is important for evaluation of results. False positive answers tend to falsely increase measured threshold sensitivity. The frequency of false positive responses is traditionally measured by adding extra questions, catch trials, to the test. Catch trials are few and limited because of time constraints, leading to inexact estimates. We developed an improved method for estimation of false positive answers by using information already available in current ordinary computerized visual field testing, without increasing test time. We here describe the method and evaluate it in a prospectively collected material of 49 glaucoma eyes of 49 patients. The results show that the new method reduces measurement errors considerably and significantly as compared with the traditional catch trials method. Test-retest change was only half with the new method as compared to the traditional method of catch trials. Furthermore, it can reduce test time by eliminating the need to use catch trials to estimate the frequency of false positive responses.