Epidemiological context of signalling

Abstract

Epidemiology has become the conscience and soul of signal detection and interpretation from assuring insights into the strengths and limitations of the data source to articulating the disease natural history and its relationship to other diseases, outcomes and treatments. The epidemiologist has struggled to bring several issues into focus including the analysis and interpretation of surveillance data, the integration of signalling and causality criteria, the statistical issues of multiple comparisons and multiple looks at the data, whether one significant disproportion is a signal or whether it should persist over time before it is 'declared'. However, epidemiology, like signalling, has its proper place with its own strengths and limitations. This paper will attempt to bring everyone up to speed on the role of epidemiology in the context of signalling and how signalling fits into the broader context of epidemiology, how perceptions and biases may affect our application of these methods and why context and communication have become so important in this field.