Testing the reliability of an eye-dedicated triaging system: the RESCUE

To calculate the reliability of an eye-dedicated triaging system named Rome Eye Scoring System for Urgency and Emergency (RESCUE).

There were four coding parameters: pain, redness, loss of vision, and risk of open globe. Each parameter is assigned a score, the sum of which allows color coding. There were 1000 consecutive patients divided into urgent (U) or non-urgent (NU) based upon diagnosis, need for treatment, hospitalization, and/or follow-up visit. Correlation between RESCUE triage scoring as assigned by the nurse on presentation and urgency as estimated retrospectively was calculated. Accuracy, sensitivity, and specificity have been calculated. False positives (FP) have been defined as patients assigned a RESCUE green or yellow code while retrospectively judged NU and false negatives (FN) have been defined as patients assigned a white code despite being considered U.

Of 1000 patients, 332 (33.2%) were classified as U and 668 (66.8%) NU. The difference in RESCUE scoring between U and NU patients was significant (p<0.001), as well as the correlation between RESCUE scoring and urgency status. Accuracy was 95% with 9.3% FP and 2.7% FN. Sensitivity was 90.7% and specificity 97.2%. Positive predictive value was 94.6%, and negative predictive value was 95.2%. All 32 hospitalized patients and 147/198 (74.2%) patients given a return appointment properly received a yellow or green code.

RESCUE accuracy, sensitivity, and specificity yield encouraging results, confirming the system''s ability to properly spot the most urgent cases. The concept of urgency in ophthalmology can be difficult to establish; nonetheless, an eye-dedicated triage can help in properly prioritizing urgent patients.