S34.2: Missing value ‐ different kinds of imputations and their effects on outcome analyses of a randomised interventional study in 2807 patients

Abstract

Age -related macular degeneration (AMD) is the most common cause for the loss of central vision associated with aging. It is a complex disorder caused by many factors. Beside choroidal neovascularization and pigment epithelial detachment, geographic atrophy (GA) is a frequent cause of severe destruction of sharp, central vision in patients with AMD. Pathophysiologic mechanisms underlying the atrophic processes and causes which influence the dynamics of GA are poorly understood at present. Time course data on the development of GA from 114 patients recruited in the FAM -Study were used for the analysis. The data provide evidence for differences of speed regarding GA growth for individual patients. There are also differences in GA growth for individual eyes within patients. Two linear mixed models for describing the dynamics of GA were investigated and discussed. Both models help to detect mechanisms which determine the dynamics of GA and allow to identify variance components and assessments of hypotheses on the data observed. Main purpose is to chose the best model for describing reality. The first model predicts a linear increase in GA growth. The other one involves an exponential growth, which means a more rapid development. These models are essential for the planning of future clinical trials to determine things which reduce the speed of GA growth. Information about sources of variance (e.g. various methods of measurement) are important for the study design. Time course data provide evidence for strong individual effects on the progression of GA. This may imply the role of prognostic factors which are not assessed so far. Future research activities could focus on the implementation of models for investigating that a proposed classification based on geometric features of the GA lesion can explain differences in GA dynamics between patients.