Personalized risk-based screening for diabetic retinopathy: A multivariate approach versus the use of stratification rules

Abstract

Aims: To evaluate our proposed multivariate approach to identify patients who will develop sight-threatening diabetic retinopathy (STDR) within a 1-year screen interval, and explore the impact of simple stratification rules on prediction.

Materials and methods: A 7-year dataset (2009-2016) from people with diabetes (PWD) was analysed using a novel multivariate longitudinal discriminant approach. Level of diabetic retinopathy, assessed from routine digital screening photographs of both eyes, was jointly modelled using clinical data collected over time. Simple stratification rules based on retinopathy level were also applied and compared with the multivariate discriminant approach.

Results: Data from 13 103 PWD (49 520 screening episodes) were analysed. The multivariate approach accurately predicted whether patients developed STDR or not within 1 year from the time of prediction in 84.0% of patients (95% confidence interval [CI] 80.4-89.7), compared with 56.7% (95% CI 55.5-58.0) and 79.7% (95% CI 78.8-80.6) achieved by the two stratification rules. While the stratification rules detected up to 95.2% (95% CI 92.2-97.6) of the STDR cases (sensitivity) only 55.6% (95% CI 54.5-56.7) of patients who did not develop STDR were correctly identified (specificity), compared with 85.4% (95% CI 80.4-89.7%) and 84.0% (95% CI 80.7-87.6%), respectively, achieved by the multivariate risk model.

Conclusions: Accurate prediction of progression to STDR in PWD can be achieved using a multivariate risk model whilst also maintaining desirable specificity. While simple stratification rules can achieve good levels of sensitivity, the present study indicates that their lower specificity (high false-positive rate) would therefore necessitate a greater frequency of eye examinations.

Keywords: cohort study; diabetic retinopathy; observational study; primary care.

Figure 1

Boxplots for the predicted probability of developing sight‐threatening diabetic retinopathy (STDR) within 1 year when applying our multivariate discriminant model (left panel). The boxplots display the distribution of the predicted probabilities (1.5 × interquartile range below the first quartile, first quartile, median, third quartile and 1.5 × interquartile range above the third quartile). Values greater than 1.5 × interquartile range above the third quartile (or lower than 1.5 × interquartile range below the first quartile) are shown separately as plotted points (outliers). Receiver‐operating characteristic (ROC) curve (right panel) with sensitivity (85.4%) and specificity (84.0%) illustrated by the green dot

Figure 2

Boxplots for the predicted probability of developing sight‐threatening diabetic retinopathy (STDR) within 1 year for patients with no diabetic retinopathy (DR) at the time of prediction (left), mild non‐proliferative DR (NPDR)/background DR (BDR) in one eye (middle), and mild NPDR/BDR in both eyes (right)

Figure 3

Prediction accuracy of the two‐episode stratification rule versus our personalized risk model