Public health surveillance of chronic diseases: suitability of spatio-temporal methods

Health surveillance is better established for infectious than non-communicable diseases (NCDs). The recent recognition of clusters of congenital anomalies (microcephaly) associated with Zika virus highlights the need and importance of a knowledge and understanding of disease occurrence in order to identify data signals in space and time and how less dramatic increases in incidence could be detected. The NHS has one of the most detailed database of health records worldwide, offering a unique opportunity to test and develop methods to detect spatio-temporal signals such as disease clusters or unusual trends. Early detection of such data signals is essential from a public health perspective to warrant further investigation of potential risk factors, including environmental pollutants and extreme climatic events, and to implement relevant prevention or protection measures.

Building on the unique expertise of the UK Small Area Health Statistics Unit, we will test the strengths and limitations, including accuracy and interpretability of results, of existing enhanced surveillance tools using data on known past signals; to assess their capacity to detect unidentified clusters and to define the optimal approach for prospective surveillance of NCDs.

The research findings could help improve monitoring and prevention of NCDs.