Global Polymod

Global Polymod: multi site study to comprehensively profile social mixing patterns in low and middle-income countries 

Funding body: National Institute of Health - Emory University
Grant Holder: Alessia Melegaro
Project Duration: 2019-2024

Project Description: The aim of the project is to collect social contract data that serves to modernize the study of infectious diseases in poor countries. 


A1. Importance of modeling infectious diseases

Mathematical models of infectious disease transmissione are increasingly important in informing global health policy and investment strategies. In addition, social mixing patterns are a fundamental parameter in the calculation of the force of infection (i.e. the rate of susceptible individuals becoming infected) in disease transmission models (parameterization). 

Ten years ago, direct data on social contacts were systematically collected for eight European populations as part of the highly-cited POLYMOD study. The POLYMOD study has informed many models and policies since its findings and data were published in 2008. In Thailand, age-assortative data from the POLYMOD study were used to assess whether a decrease in pertussis transmission was due to vaccine policy or factors associated with transmission. In Australia, Campbell et all used disease transmission models, parameterized by the POLYMOD data, to assess the impact of several potential vaccine strategies, specifically assessing the indirect protection of infants too young to be vaccinated. There is, however, a lack of information (collected by standard methods) in low-income countries. A multisite social mixing study conducted in low income countries will inform both epidemiologic and economic impact models assessing control strategies for disease transmission.


A2. Importance of sensors in social-mixing studies

Sensors are becoming increasingly important in social mixing studies as they can objectively record contact between individuals. It has been noted in previous studies that different data collection methods (online vs. paper) yield different results. Contrary to social contact diaries, sensor based studies are not affected by recall bias and can gather data on human proximity and interactions without being subject to differences in memory between individuals. Very few studies have been conducted using sensor-based methods, specifically in low income settings and among infants less than six months of age. Using sensord in low and middle-income countries to collect data from individuals will corroborate the data collected from diaries and provide a more accurate representation of social contact patterns.