Interobserver agreement using Goldmann applanation tonometry and dynamic contour tonometry: comparing ophthalmologists, nurses and technicians.

Abstract

To compare agreement between ophthalmologists and non-ophthalmologists (nurses and ophthalmic technicians) when measuring the intraocular pressure (IOP) using Goldmann applanation tonometry (GAT) and Pascal dynamic contour tonometry (DCT). Patients attending for their routine glaucoma outpatient appointment were invited to participate. IOP was measured in one eye by either two ophthalmologists (observer group 1), two non-ophthalmologists (observer group 2) or one ophthalmologist and one non-ophthalmologist (observer group 3). All were experienced in using GAT and some experienced in using the DCT. The order of tonometer and staff was randomised. Agreement was calculated by Bland-Altman analysis, with the mean difference and 95% limits of agreement (LoA) of measurements calculated for each observer group. One hundred eyes were measured within each observer group. The mean difference (95% LoA) in IOP measurements were GAT: group 1=-0.20 (4.9) mm Hg, group 2=0.6 (5.4) mm Hg and group 3=0.0 (3.7) mm Hg; DCT: group 1=0.8 (7.7) mm Hg, group 2=0.3 (4.2) and group 3=0.0 (5.2) mm Hg. The DCT consistently over-read the GAT for all observer groups. Ophthalmologists show good levels of agreement with each other when using GAT, while technicians/nursing staff show better agreement when using the DCT. The DCT may be a better tonometer to use if permanently delegating IOP measurements to non-ophthalmologists, but measurements cannot be interchanged with the GAT.