Improving Visual Field Testing Reliability by Using an Educational Patient Leaflet.

Introduction Reliability indices such as fixation losses (FL), false positives (FP), and false negatives (FN) determine the clinical utility of standard automated perimetry (SAP). Patient education can improve test performance, but low-cost, scalable approaches are needed. Our study aimed to evaluate whether a one-page patient leaflet delivered immediately before testing improves Humphrey Swedish Interactive Thresholding Algorithm (SITA) Fast 24-2 reliability. Methods We conducted a single-center, non-randomized quality-improvement study in glaucoma follow-up clinics. A total of 39 patients were included in this study. Each eye's most recent pre-leaflet SAP test was paired with its first post-leaflet SAP test under identical parameters (Humphrey Field Analyzer, SITA Fast 24-2, size III, 31.5 asb). The primary endpoint was composite "full reliability" (FL ≤10%, FP ≤10%, FN ≤10%) per eye, analyzed with McNemar's test and effects reported as the difference in paired proportions with 95% CI. Family-wise error across eyes was controlled by Holm-Bonferroni. The secondary endpoints were continuous changes in FL%, FP%, and FN% per eye, analyzed using the Wilcoxon signed-rank test with Hodges-Lehmann (HL) median change. Analyses were performed per eye without across-eye pooling, and only pairwise complete cases were used. This project was registered locally as a quality-improvement study and adhered to the principles outlined in the Declaration of Helsinki. Results  We analyzed 39 right and 37 left eyes (patient mean age: 66±15.5 years; mean inter-test interval 6.3 months). For our primary outcome of full reliability, the proportion of reliable tests increased from 31% to 54% in the right eye and from 41% to 65% in the left eye, with both remaining significant after Holm-Bonferroni. For our secondary outcome, FN% decreased significantly (right -4.5 pp and left -4.0 pp), while FL% declined in the right eye (-16.4 pp) with no significant change in the left. FP% showed small, non-significant increases. Conclusions A brief, one-page leaflet delivered immediately before SITA Fast testing increased the proportion of fully reliable fields and reduced false-negative responses in routine practice. This low-cost intervention is straightforward to implement and may reduce repeat testing. Confirmation under SITA Faster with a randomized design using visit-level endpoints is warranted.