Purchase this article with an account.
A.R. O'Connor, H. Morris, E.E. Birch, Functional Significance of Stereopsis InvestigatorGroup; The Relationship Between Distance Stereotests . Invest. Ophthalmol. Vis. Sci. 2006;47(13):2447.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Measuring stereoacuity is an important part of the visual examination and several tests are in use in clinical practice. The introduction of tests to measure stereoacuity at distance have provided additional clinical information. However little is known about how these tests relate to each other which is important in determining whether the tests can be used interchangeably.
Stereoacuity was measured in 98 subjects, 66 had no ocular motor deviation and 32 had a manifest deviation. All measurements were carried out at 2 sites in Liverpool and Dallas. The stereoacuity of the whole group was measured with 2 different tests, the FD2 (free space test of real depth range 200–5"), and the Randot (random dot test with polarised glasses, range 800–60"). In addition the 45 subjects from Liverpool (40 with no ocular motor deviation) also had their stereoacuity assessed using the Test Chart 2000 Pro (computer based random dot with red/green glasses, range 254–13"). A Bland Altman analysis was used to compare the results between the stereotests.
The median values obtained on each test were FD2 20", Randot 60" and Test Chart 2000 25". The Bland Altman bias shown in the table demonstrates the variability between two tests. The Bland Altman plot showing the percentage difference against the average between the FD2 and Randot shows considerable variation. However, the Randot is consistently higher than the FD2 and demonstrates the best agreement at the higher values of stereoacuity. The other plots show less agreement but with few subjects at the lower end of the stereoacuity range.
The results show similar magnitude of bias between the tests but the plots highlight that this is inconsistent. The scale differences between the tests contribute to the variation in the mean bias but the lack of agreement across the range of stereoacuities shows that they are measuring different aspects of binocular vision. This demonstrates that these routinely performed tests are not interchangeable.
This PDF is available to Subscribers Only