Purpose To evaluate the repeatability of clinical decision-making by one glaucoma specialist and determine the influence of intraocular pressure (IOP) variation on those decisions. Patients and methods 40 patients were selected in whom consultant decisions were appropriate concerning management. These notes were reviewed on three separate occasions, each 3 months apart. The final examination was changed to include clinical findings with the IOP, either the same, ±2 or ±4 mm Hg different from the recorded IOP. A forced choice clinical decision was made on each occasion: continue present treatment, change medical treatment, or recommend surgery. The clinical decisions were then compared. Results Our results showed that when the presented IOP was the same there was an 80% agreement in the management decision (j 0.7). When the presented IOP was lowered by 2 or 4 mm Hg the agreement was 70 and 85%, respectively. When the presented IOP was increased by 2 or 4 mm Hg the agreement was 65 and 70%, respectively. None of these changes were significantly different. Similarly there was no evidence of a trend towards disagreement with IOP change (v 2 trend ¼ 1.3 P ¼ 0.25). Conclusions Algorithms for decision-making in glaucoma are complex. Large differences between specialists are recognised. This is a first report of within specialist agreement. The impact of within measurement error differences in single IOP measurements was negligible. Review of comments suggests that the main reason for disagreement was patient preference, which was absent with note review.