An evaluation of the accuracy of community-based perimetry.

We assessed the accuracy of five office-based perimetric technicians in the examination of 14 patients with prechiasmal and chiasmal defects unknown to them. A pre-test followed by two days of instruction and supervised practice; the technicians were then retested on the same patients (post-test). An independent evaluator scored their results. The identification of any field defect rose significantly (P = .039) from 69% in the pre-test to 95% in the post-test. The identification of the hemianopic configuration of a defect rose from 45% in the pre-test to 84% in the post-test (P = .11); initial identification of nasal steps was 77%, falling to 67% in the post-test. Adequate definition of extent, depth, and slope of defects was rare (5%) on the pre-test, but rose significantly (P = .032) to 57% on the post-test. The monitored patient examinations were essential for correcting recurrent flaws in technique. This study shows that such teaching efforts, including emphasis on identifying hemianopic defects, are necessary to raise visual field examination to the level of a consistently reliable diagnostic determinant.