Text this: Comparison and bias analysis of medically attended acute gastroenteritis incidence estimates derived from electronic health record surveillance versus cross-sectional surveys