Triaging Ophthalmic Emergencies: A 4-Parameter Tool embedded in an Emergency Department’s Electronic Health Record System

Abstract

Purpose : To assess the reliability of a 4-parameter triaging system embedded in the electronic health record (EHR) system in predicting ophthalmic urgency when conducted by emergency department (ED) nurses.

Methods : A 4-parameter triaging system previously found to be highly reliable in detecting urgent ophthalmic pathology at an ophthalmic ED was adapted for use by non-ophthalmic providers: (1) Descriptive and visual guidelines for use were developed, and (2) The triaging system and developed guidelines were embedded into the ED's EHR. All patients presenting to the ED with an ophthalmic complaint were assigned a score between 0 and 12 by an ED triaging nurse that assessed the patient’s presentation in four categories: eye redness, eye pain, risk of open globe, and vision loss. Color coding was assigned based on the cumulative triaging score: 0–3 was classified as green, 4–7 as yellow, and 8–12 as red. The triaging score was compared against the ophthalmic consultant’s assessment in the same categories and assigned diagnosis.

Results : During the study period, 250 patients were assessed. Of these, 72 (28.8%) patient presentations were classified as urgent (a code of yellow or red), with the remainder classified as non-urgent. Patients were triaged using the tool within 9.1 ± 5.3 minutes of arriving at the ED. The sensitivity of the system was 0.958(95% CI 0.88-0.99) and specificity 0.781(95% CI 0.71-0.84). The correlation between urgency level as coded by triage nurses and eye care providers was significant with a Spearman’s rank coefficient 0.67 (P<0.0001): 82.0% patients were coded at the identical level of urgency by both providers; 6.0% were undercoded by one category level, 11.6% were overcoded by one category level, and 0.4% were overcoded by two category levels.

Conclusions : The triaging tool as administered by ED nurses using the guidelines developed is a simple, rapid and highly sensitive method of detecting urgent ocular pathology. Such a system holds the potential to decrease the cost of eye care to patients and the health care system more broadly. By reliably referring patients with non-urgent ocular conditions to outpatient appointments in collaboration with ophthalmic providers, wait times and length of stay may be considerably reduced for all-comers, while valuable and costly ED resources would be conserved for patients with truly urgent conditions.

This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.