In the growing field of digital sociology, a distinction is made between the methods built on the digital infrastructure's functionalities and the approaches which consist of digitizing traditional social science methods. In the view of Richard Rogers (2013), digital methods consist of repurposing the traces produced by digital infrastructure (i.e. links, likes, shares) as well as their native features (i.e. querying, web crawling, filtering) to understand social phenomena. In addition to the new digital methods, a wide range of researchers are embracing the digitization model by implementing traditional methods in a digital context. Between these two extremes, there are many hybrid approaches. To move beyond this opposition between digital methods and the digitization of methods, Marres and Gerlitz (2015) suggest the term 'interface methods' to refer to inventive methodologies that are driven neither by the possibilities offered by digital infrastructure nor by the implementation of traditional methods in the digital context. In other words, interface methods require researchers to constantly adapt measurements, tools, and databases to the object being studied. This adaptation translates into a 'tailor-made' methodology capable of considering the digital platform functionalities in the conceptualization of the research design, but without forgetting the methodological assets of the social sciences in which digital methods often have their roots (Marres, 2017).

In this thesis, I am interested in applying these hybrid and innovative methods to the study of health and social media.

This thesis in articles is made up of two theoretical chapters, three empirical chapters, and a concluding chapter. Chapter 1 presents the theoretical background and methodology of the thesis. Chapter 2 focuses on the network analysis and its use for understanding and planning health interventions, both online and offline. This chapter complements the methodological part of Chapter 1 by showing the specific role of networks and their analyses in the health field. Chapter 3 provides a mapping of the scientometric landscape of research on health and social media. This initial research gives, among other things, an account of the different fields and epistemologies involved, the topics covered, and their evolutions over time in the study of health and social media. Chapters 4 and 5 deal with case studies of health topics on social media. More specifically, Chapter 4 is an analysis of media coverage of cancer types on Facebook and Instagram. Chapter 5 looks at the controversies surrounding vaccination and anti-vaccination on Twitter. Finally, Chapter 6 concludes with the main contributions of the thesis and outlines the concept of the 'digital sociology of health.'