Premium
Information technology in ophthalmology. www.retinarisk.com
Author(s) -
Stefansson E.
Publication year - 2015
Publication title -
acta ophthalmologica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.534
H-Index - 87
eISSN - 1755-3768
pISSN - 1755-375X
DOI - 10.1111/j.1755-3768.2015.0254
Subject(s) - medicine , diabetic retinopathy , diabetes mellitus , risk assessment , optometry , epidemiology , ophthalmology , computer science , computer security , endocrinology
Summary A huge amount of epidemiological data has been accumulated in ophthalmology as in other fields of medicine. The use of these data in health care tends to be simplistic and makes too little use of information technology and mathematics. We have used published reports on risk factors for progression of diabetic retinopathy, DR, to create an algorithm that predicts an individual's risk of sight threatening DR. The prediction is based on the type and duration of diabetes mellitus, blood glucose and pressure, DR stage and gender; see www.retinarisk.com . It predicts 80% of the risk as seen as the area‐under‐curve on ROC plots. The software provides individual predictions based on these risk factors. It educates the patient about DR risk and allows estimation of the effect of improving risk factors such as blood glucose. It has motivational and educational value. Current norm in DR screening is annual examination (every other year for eyes without DR). Individualized risk assessment allows individualized determination of the frequency of eye screening. High risk eyes are screened more frequently and low risk eyes less frequently. This improves safety for the high risk eyes and saves on resources for the low risk eyes. When compared with annual screening this provides cost savings of about 50%.