A retrospective study demonstrated slightly elevated baseline fasting plasma glucose levels to be a robust marker of incident diabetes risk, even when those levels fell within the normal range.