Test-retest repeatability and agreement of the quantitative contrast sensitivity function test: towards the validation of a new clinical endpoint

Abstract

Purpose: The purpose of this study is to investigate test-retest reliability and agreement of the quantitative contrast sensitivity function test (qCSF) in the retina clinic.

Methods: A total of 121 right eyes of 121 patients were tested and consecutively re-tested with qCSF in the retina clinic. Outcomes included area under the logarithm of contrast sensitivity function curve (AULCSF), contrast acuity, and contrast sensitivity thresholds at 1-18 cycles per degree (cpd). Test-retest means were compared with paired t-test, variability was compared with the Brown-Forsythe test, and intraclass correlation coefficient (ICC) and Bland Altman plots evaluated reliability and agreement.

Results: Mean test-retest differences for all qCSF metrics ranged from 0.02 to 0.05 log units without statistically significant differences in variability. Standard deviations ranged from 0.08 to 0.14. Coefficients of repeatability ranged from 0.16 to 0.27 log units. ICC > 0.9 for all metrics except 1cpd (ICC = 0.84, all p < 0.001); AULCSF ICC = 0.971.

Conclusion: qCSF-measured contrast sensitivity shows great test-retest repeatability and agreement in the retina clinic.

Keywords: Agreement; Clinical endpoints; Contrast sensitivity; Functional vision; Quantitative contrast sensitivity function; Repeatability; Retina; Validation; Visual function.