Risk Prediction Model for Progression of Age-Related Macular Degeneration

The aim of this paper was to develop a risk prediction model for the progression of age-related macular degeneration (AMD) in Koreans using systemic and environmental factors.

The study sample included 10,890 individuals 50 years of age or older; 318 (2.92%) presented with early AMD findings in baseline examinations. Re-examinations were performed in 157 (49.37%) who were followed up for 4.4 years. The multivariate analysis of covariates included demographic and environmental factors. After using these data to develop a risk prediction model, the individual algorithm was made, and receiver operating characteristic curves were calculated to assess the predictive ability of the risk model for AMD progression.

The individual algorithm to predict the AMD progression risk based on systemic and ocular factors was as follows: Y = -9.565 + 1.709 (drusen locationcenter) + 0.795 (drusen locationparacentral) + 1.074 (both eyes) + 0.094 (drusen sizeintermediate) + 0.034 (drusen sizelarge) + 0.614 (drusen number10-20) + 2.278 (drusen number>20) + 0.577 (hyperpigmentation) + 0.725 (hypopigmentation) + 0.079 (male) - 0.025 (age) - 0.921 (SMKex) + 1.574 (SMKcurrent) + 0.363 (total protein) + 1.626 (globulin), where SMK means smoking status. The C statistics for the model was 0.84 (0.75-0.92) indicating a good predictive power.

A comprehensive risk prediction model for AMD progression was made to calculate the individual AMD progression risk using personal systemic and environmental factors.