Predictive Modelling Of Visual Acuity Upon Diabetic Retinopathy In Type 2 Diabetes Mellitus

ABSTRACT

@#The worldwide upsurge of Type 2 Diabetes Mellitus (T2DM) warrants the attention of public health, especially in complications. Diabetic Retinopathy (DR), the commonest ocular complication, contributes to the growing incidence of blindness. Recommendations that the visual acuity (VA) assessment guidelines should be used in DR screening was not routinely practiced. This study aims to model the predictive effect of VA - a simple economic clinical assessment - upon DR in patients with T2DM. A case-control study was conducted in Hospital Universiti Sains Malaysia with subjects recruited from the ophthalmology clinic. The primary variable during analysis was VA, and the outcome variable was DR. Models with eight control variables which included age, gender, and duration of DM were developed. The predictive effect measured by logistic regression showed that when unadjusted; four variables had a significant association with DR, at p-value<0.25; they were the duration of DM, systolic blood pressure, glycosylated haemoglobin and VA. From the eight different predictive models, the estimated adjusted odds ratio produced ranges from 6.09 to 11.64. Our study shows that VA has a predictive effect upon DR in T2DM patients. We suggest VA assessment, to be on par with the monitoring of blood pressure and blood glucose.