To effectively inform infectious disease control strategies, accurate knowledge of the pathogen’s transmission dynamics is required. The infection incidence, which describes the number of new infections in a given time interval (e.g., per day or per week), is fundamental to understanding transmission dynamics, and can be used to estimate the time-varying reproduction number and the severity (e.g., the infection fatality ratio) of a disease. The timings of infections are rarely known and so estimates of the infection incidence often rely on measurements of other quantities amenable to surveillance. Case-based surveillance, in which infected individuals are identified by a positive test, is the pre-dominant form of surveillance for many pathogens, and was used extensively during the COVID-19 pandemic. However, there can be many biases present in case-based surveillance indicators due to, for example, test sensitivity and specificity, changing testing behaviours, and the co-circulation of pathogens with similar symptom profiles. Without a full understanding of the process by which surveillance systems generate data, robust estimates of the infection incidence, time-varying reproduction number, and severity based on these data cannot be made. Here we develop a mathematical description of case-based surveillance of infectious diseases. By considering realistic epidemiological parameters and situations, we demonstrate potential biases in common surveillance indicators based on case-based surveillance data. The description is highly general and could be applied to a diverse set of pathogens and situations. The mathematical description could be used to inform inference of infection incidence using existing data, with a full understanding of where bias and uncertainty will be present in any such analysis. Future surveillance strategies could be designed to minimise these sources of bias and uncertainty, providing more accurate estimates of a pathogen’s transmission dynamics and, ultimately, more targeted application of public health measures.
