Social Networks and Infectious Disease Ecology in Northeast Madagascar

Abstract

Network methods have transformed infectious disease ecology and epidemiology by explicitly modeling heterogeneity in contact patterns, which allows for a deeper understanding of the complex social processes that drive infectious disease transmission. Networks can help identify new transmission pathways and characterize an individual’s infection risk. Studying the behavioral processes that generate disease-relevant networks can also inform public health efforts to alter a network’s structure during outbreaks. In this dissertation, I worked with an international team of scientists to investigate infectious disease transmission dynamics among smallholder farmers in northeast Madagascar. Farming households in this region of Madagascar engage in a wide range of economic activities, from households that primarily practice subsistence farming to households that primarily grow cash crops like vanilla. I used a rich dataset containing information on participants’ social networks, economic activities, mobility patterns, and infectious disease exposure to investigate the biosocial processes that shape infection risk at local (Chapters 2 and 3) and regional (Chapter 4) scales.In Chapter 2, I used transmission potential networks that reflected social, environmental, and zoonotic transmission pathways to identify the most important pathway for the spread of a prevalent and genetically diverse gastrointestinal protozoan, Blastocystis spp. DNA metabarcoding of fecal samples from humans and domesticated animals revealed six genetically distinct subtypes of Blastocystis. Infection patterns were best predicted by networks representing environmental transmission pathways, suggesting that Blastocystis is most likely spread through exposure to contaminated water and soil in this setting. In Chapter 3, I applied regular equivalence blockmodeling to identify social-epidemiological roles associated with viral exposure. Participants were clustered into three consistent role categories that captured equivalent patterns of interaction across networks reflecting shared free time and exchanges of food and farmwork. The role categories included individuals who had many reciprocated ties across networks, individuals who sent many ties with few reciprocated, and individuals who had few connections. Participants’ viral exposure was characterized by analyzing dried blood spot samples using Phage ImmunoPrecipitation Sequencing and VirScan, which identifies antibody signatures to over 300 virus species and subtypes. Role categories performed better at predicting patterns of viral exposure than single measures of network centrality, and the analysis provided important insights into the coupled dynamics of exposure and susceptibility in facilitating infectious disease transmission. In Chapter 4, I focused on regional-level transmission dynamics and tested the mobility tradeoff hypothesis, which proposes that market integration – the transition from subsistence to market-based livelihoods – increases regionally distributed resources and thus provides the means and incentives for wider travel. This chapter emphasizes the role of mobility networks in infectious disease outbreaks and considers the processes that generate these networks from the perspective of human behavioral ecology. Using survey-based mobility data and serological data on SARS-CoV-2 infection patterns from 34 communities, we found that greater market integration was associated with longer and more frequent regional travel as hypothesized. However, regional mobility patterns were not associated with SARS-CoV-2 infection patterns; instead, factors such as gender and occupation were the most important predictors of infection.Together, these findings demonstrate how network perspectives help unveil the behavioral processes that link rural livelihood strategies and infectious disease transmission dynamics across local and regional scales. By integrating ecological, behavioral, and epidemiological data, this dissertation advances a network-based approach to studying infectious disease transmission in contexts undergoing rapid social and economic change.