Reliability of telemedicine for diagnosing and managing eye problems in accident and emergency departments.

OBJECTIVES

To assess the accuracy and efficiency of telemedicine in diagnosing and managing eye problems presenting to accident and emergency departments.

DESIGN

A controlled trial with a face-to-face and telemedicine phases, each involving 40 patients undergoing two consecutive consultations. In the face-to-face phase, both consultations were in person; in the telemedicine phase, observer 1 used videoconferencing technology at 384 kbit/s (separate nonslit lamp-torchlight and slit lamp examinations) and observer 2 saw the patient face to face. Setting The accident and emergency department at Moorfields Eye Hospital.Participants In total, 80 consenting new patients presenting to the department.

MAIN OUTCOME MEASURES

(1) Agreement levels between the two observers for each phase (judged by an independent masked investigator), (2) length of consultation, and (3) number of unnecessary recalls.

RESULTS

Agreement rates were as follows. Face-to-face phase: total agreement (30/40=75%), trivial disagreement (8/40=20%), clinically important disagreement (2/40=5%). Telemedicine phase (torchlight): complete agreement (16/40=40%), trivial disagreement (20/40=50%), clinically important disagreement (4/40=10%). Telemedicine phase (slit lamp): total agreement (23/40=58%), trivial disagreement (15/40=37%), clinically important disagreement (2/40=5%). Agreement levels in the telemedicine phase with torchlight examination were significantly lower (chi(2)=10.07, P=0.007) for any disagreement. Telemedicine consultations erred on the side of clinical caution and were no slower than face-to-face consultations (mean 6 min for observer 1 in both phases). Recalls were more likely (chi(2)=5.16, P=0.02) after telemedicine consultations with torchlight only (9/40) compared with face-to-face consultations (2/40). Although there were more significant disagreements using the telemedicine, in each case the telemedicine diagnosis and management erred on the side of safety; hence, no patient would have suffered by wrong management because of the consultation using telemedicine.

CONCLUSIONS

Telemedicine was found to be an accurate, safe, and efficient method of diagnosing and managing these patients, especially if slit lamp images were used. Advice using telemedicine erred on the side of caution, which resulted in more recalls.