Emerging infectious diseases disrupt human and wildlife communities, often with unpredictable outbreaks in time and space. By accounting for contact patterns, considering interaction networks can improve our ability to predict disease spread in multi-host communities. At the same time, all interaction networks are embedded within landscapes that shape their structure, and are composed of species with different functional traits. In this talk, I will highlight our efforts to understand how plant-pollinator networks are shaped by landscape simplification, how this impacts pathogen prevalence in bee communities, and how transmission of pathogens in networks of bees and flowers with different functional traits may play a fundamental role in structuring epidemiological patterns.