VARIABILITY IN ESTIMATES OF THE BASIC REPRODUCTION NUMBER AND THE UTILITY OF THE POLYMOD DIARY STUDY

Abstract

The basic reproduction number, R¬0, represents the number of secondary infections caused by one typical individual in a susceptible population. R0 is considered one of the most important parameters for the study of infectious diseases because it quantifies a threshold below which disease spread is controlled. Estimates of R0 are highly variable, mostly because of differences in the mathematical models used for estimation. A major source of uncertainty in these models is how patterns of social contacts are represented. Mixing patterns in models generally depend on poorly substantiated, a priori assumptions and not on data. The Polymod diary study, conducted in 2005-6 in eight European countries, is the first large-scale, representative study aimed at quantifying social contacts relevant to the transmission of respiratory infections in an effort to substantiate mixing assumptions. However, there is some uncertainty as to whether the Polymod data properly captures social mixing patterns relevant to disease transmission.

This research reveals that Polymod-based models measles, mumps, and rubella seroprevalence data poorly in a systematic way. Across different settings and data collection methods, the Polymod data seems to underestimate transmission-relevant contacts in infants and overestimate them in schoolchildren. Potential reasons for the data's shortcomings and how they relate to R0 estimation are explored through the lens of demographic features and through examination of different subsets of the Polymod data. This work suggests that much more research needs to be done before the impact of social mixing patterns on disease transmission is properly understood.