A computerized version of the Lancaster red-green test.

PURPOSE: To compare results from a computerized version of the Lancaster red-green test with those of the conventional test.

METHODS: Consecutive adult patients with noncomitant strabismus were tested with the conventional Lancaster red-green test and with a computerized version of the same. The computerized test was administered by means of a 40-inch monitor at a working distance of 50 cm or a projector and screen at a working distance of 1 meter. Agreement between the measured horizontal, vertical, and torsional deviations in the conventional test and both computerized versions was evaluated with the mountain plot, Bland-Altman plot, and Deming regression analysis models.

RESULTS: A total of 82 patients were tested. Agreement of measured horizontal deviation in the conventional test was better with the projector version of the test (limits of agreement: right eye, -4.6(Δ) to 3.4(Δ); left eye, -4.9(Δ) to 3.5(Δ)) than the monitor version (limits of agreement: right eye, -10(Δ) to 4.2(Δ); left eye, -8.9(Δ) to 4.1(Δ)). The measured vertical and torsional deviation in the conventional test showed good agreement with both versions of the computerized test (limits of agreement <5(Δ) for vertical measurements and <3° for torsional measurements). Agreement was similar for right and left eyes.

CONCLUSIONS: The vertical and torsional deviations measured with both computerized versions of the test were in good agreement with those obtained with the conventional test. For measured horizontal deviations, the projector version had better agreement than the monitor version.