Data that can be used to support surveillance of public health are becoming available to the analysts at increasing volumes and varieties. Widening of the availability is accompanied by a rapid rise of the volume, specicity, and complexity of questions being asked about data. Those processes jointly exacerbate the challenges related to statistical reliability of the models used in the analyses, and to statistical signicance of the ndings. Efcient and effective implementations of data aggregation strategies can tackle several of such challenges, including those related to dealing with large volumes of high-dimensional data, or dealing with low density of useful evidence in data due to sparseness of patterns of interest and due to high specicity of analytic questions.