Abstract

Various methods, mostly contact surveys and sensor-based approaches, have been used to measure social mixing for elucidating infectious disease transmission dynamics of respiratory infections. Large and very detailed contact datasets have been obtained using these methods, e.g., within the POLYMOD and SocioPatterns projects, and have been used to gain insights into infection transmission. Despite their wide use, systematic comparisons of these methods as well as research on measurement errors and biases are rare.

In my seminar, I will discuss how ‘potentially contagious contact’ has been defined for various diseases and methods. I will contrast different data collection methods, in particular survey and sensor methods, and elucidate their strengths and weaknesses. Next, I will report how contact data obtained from surveys differ from data measured by sensors. Finally, I will shed light on the problem of survey underreporting and discuss which kind of contact is more likely to remain unreported and by whom.