Purchase this article with an account.
Susan Kelly, Yi Pang, Chandra Engs, Lauren Foley, Nellie Salami, Audra Sexton; Test-retest Repeatability For Contrast Sensitivity In Children and Young Adults. Invest. Ophthalmol. Vis. Sci. 2011;52(14):1895.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Contrast sensitivity (CS) testing has become an important tool used to assess patient vision. Currently two letter CS tests,the Pelli-Robson and the Mars chart are often used to assess CS. Both these charts measure patient sensitivity to peak spatial frequencies. These tests are not only quickly administered but have validity and good test-retest repeatability. There are chart-based CS tests that measure sensitivity to contrast across the full range of spatial frequencies that are also rapid but their accuracy and reliability have been less well examined. The present study measured the test-retest repeatability for the Vector Vision CSV-1000E system in visually normal children and young adults.
Twenty-one visually normal adults (mean age=27.96 yrs, range = 22-38) and 16 children (mean age= 7.7 yrs, range =5-12) were tested on the the Vector Vision 1000E chart on two separate visits (average inter-visit duration = 15 wks for adults and 2 wks for children). All subjects were free of ocular pathology, had best-corrected visual acuity of 20/20 or better and wore their habitual correction if needed. The testing procedure followed that recommended by the manufacturer. The chart self-calibrates to a mean luminance of 85 cd/m2 and testing was conducted at 8 feet.
Data were converted to log CS scores for all data analysis. Test-retest repeatability was determined with a Bland-Altman analysis. Both the 95% limits of agreement (and their 95% confidence intervals) and the correlation of repeatability (COR) were calculated. The average test-retest difference was -0.07 log CS for adults and -0.014 for children. The negative values indicate performance improved on the second administration of the test. The 95% limits of agreement (LoA) ranged from + 0.425 to+0.757 for adults and +0.482 to +0.91 for children.
Both children and adults completed the test in under 3 minutes. Although the average test-retest difference in logCS between visits is small, it is 3 times that which has been reported for the Pelli-Robson or the Mars chart in adults. The LoA reported for the Pelli-Robson test in visual normals and clinical populations ranges from + 0.14 to +0.2. This is in contrast with the LoA measured in the present study which are 2 to 4 times greater. These values are also larger than those reported for a healthy, older population using the Vector Vision test (Pomerance and Evans, '94). It is possible that test-retest variability varies with the subject's understanding of the instructions; variability in CS estimates will occur if they guess the location of the pattern after they no longer see it.
This PDF is available to Subscribers Only