García-Fiñana M, et al. - In the present study, researchers assessed the proposed multivariate approach to identify patients who develop sight-threatening diabetic retinopathy (STDR) within a 1-year screen interval and examined the effect of simple stratification rules on prediction. Using a novel multivariate longitudinal discriminant approach, a 7-year dataset (2009-2016) from people with diabetes (PWD) was analysed. Data were analyzed from 13,103 PWD. Findings suggested that a multivariate risk model can be used to accurately predict progression to STDR in PWD while maintaining desirable specificity. Although simple stratification rules can achieve good sensitivity levels, the present study shows that their lower specificity (high false-positive rate) would require a higher frequency of eye examinations.