Divergence bias in Hess compared to Harms screen strabismus testing.

The Hess and the Harms screen test each have different testing distances. While the Harms screen test is usually performed at 2.5 m, the Hess screen test is performed at 0.5 m. The geometry of the closer testing distance of the Hess screen test requires an increase of the convergence angle by 6°. This study investigates the quantitative differences between the two frequently employed screen tests. Ocular deviation of 18 normal subjects and 36 patients with congenital or acquired paralytic or concomitant strabismus were assessed with a complete orthoptic examination including alternate prism cover testing at near (nPCT) and far (fPCT), as well as Hess and Harms screen testing. One-way ANOVA was used for statistical analysis. The Hess test recorded more overall exodeviation compared to the Harms test for patients (mean difference -3.50°, 95% limits of agreement (CI) = [-4.79, -2.21], p < .001), and controls (mean difference -1.78°, CI = [-2.99, -0.56], p = .004). For vertical deviations, there was no statistically significant difference between the two tests for patients (mean difference +0.75°, CI = [-0.41, +1.91], p = .251), and controls (mean difference -0.28°, CI = [-0.68, -0.11], p = 0.231). This study emphasizes the importance to consider the divergence bias when comparing the Hess to the Harms screen test, which is likely explained by the greater vergence demand dependent on the closer testing distance. The exodeviation shift tended to be more pronounced in patients than controls, which may imply that patients with strabismus have an impaired convergence drive