Health & Place xxx (xxxx) xxx Please cite this article as: Dalia Mattioni, Health & Place, https://doi.org/10.1016/j.healthplace.2019.102244 1353-8292/© 2019 Elsevier Ltd. All rights reserved. Healthy diets and the retail food environment: A sociological approach Dalia Mattioni a, * , Allison Marie Loconto b , Gianluca Brunori a a Department of Agriculture, Food and Environment, University of Pisa, Via del Borghetto 80, 56124, Pisa, Italy b Institut National de la Recherche Agronomique (INRA), Laboratoire Interdisciplinaire Sciences Innovations Societes (UMR LISIS 1326 - CNRS, ESIEE, INRA, UPEM), Universite Paris-Est Marne-la-Vallee, 5 Blvd Descartes, 77454, Marne-la-Vallee Cedex 02, France 1. Introduction The current global surge in non-communicable diseases, and its link to dietary factors, has brought to the fore the importance of encouraging healthy diets (Development Initiatives, 2018; Willett et al., 2019). In addition to education and communication, food environments have been singled out as important contributors to eating habits (GloPan, 2016; HLPE, 2017). According to the ANGELO framework (Analysis Grid for Environments Linked to Obesity), food environments are made up of various “components” that fall into four main types: physical, economic, political and socio-cultural (Swinburn et al., 1999). The physical environment refers to the type of food available in a variety of outlets and settings: from schools and workplaces to neighbourhoods. The economic environment refers to the relative price of food, while the political food environment includes the rules related to how food is sold (including labels and standards). The socio-cultural environment refers to societal beliefs and values around food, including the media envi- ronment that affects food advertisements. Ultimately, the overall food environment that emerges from these four types of environments de- termines which foods are available, affordable and desirable to in- dividuals in their surroundings thus contributing strongly to shaping dietary patterns (Swinburn et al., 2013). This article focuses on one specific aspect of the physical food environment: the retail food envi- ronment (RFE). The RFE consists of those food outlets where consumers eat out or that sell the food they eat at home; these can range from hypermarkets to street vendors. Research on the link between the RFE and dietary patterns has a long history in high-income countries (HICs) due to earlier incidences of high levels of obesity rates compared to low-and middle-income countries (LMICs) and an earlier modernization of the food system that increased the availability, affordability and desirability of unhealthy foods. De- cades of research on the posited link between the residential RFE (measured in terms of residential proximity and density of retail outlets) and individual dietary patterns has yielded mixed results, particularly when using proximity measures. Density measures have shown more consistent results, such as the (mild) positive correlation between den- sity of fast food outlets and poor dietary patterns (Caspi et al, 2012; Ni Mhurchu et al., 2013; Black et al., 2014). These mixed results have justified the need to consider and analyze other spatial and social vari- ables that mediate the impact of the RFE on dietary patterns and health outcomes (Lytle, 2009; Cummins et al., 2014). In this article we respond to the outstanding question: is there a role for food practices in mediating the influence of the RFE and does the RFE influence eating practices? Our study contributes to research on RFEs in two ways: firstly, by exploring the nature of the social factors that mediate the link between the RFE and dietary patterns by using social practice theory. The paper contends that it is the food practices that individuals are engaged in that influence where they shop, i.e. the way they interact with the food environment, and ultimately, their dietary patterns. At the same time, individual food practices are influenced by “notions” of food transmitted by food outlets. This does not mean that the way outlets are distributed in space – and particularly the relative density of healthy/unhealthy outlets in the RFE - does not have an influence on where they decide to shop. Instead, it is the nature of their influence on social practices that this paper analyzes. We suggest that food outlets act as powerful cultural intermediaries that influence the meaning and competences that in- dividuals have around food. This forms the second area of contribution of the paper. We begin our analysis by identifying key gaps in current research on the link between the RFE and healthy dietary patterns. We then introduce social practice theory to advance our understanding of this link, which is demonstrated through a case study of Farmers’ Markets in Costa Rica. We conclude with some methodological and policy implications for further research and practice. 1.1. Current RFE research and existing gaps In the face of mixed results on the direct link between RFEs and di- etary patterns, a number of urban sociologists and geographers have pointed out the importance of examining consumer daily routines. Various studies have shown that individuals do not necessarily purchase * Corresponding author. Via Pomponia Grecina 10, 00145, Rome, Italy. E-mail addresses: daliamattioni@hotmail.com, d.mattioni2@mailbox.unipi.it (D. Mattioni). Contents lists available at ScienceDirect Health and Place journal homepage: http://www.elsevier.com/locate/healthplace https://doi.org/10.1016/j.healthplace.2019.102244 Received 13 June 2019; Received in revised form 7 November 2019; Accepted 8 November 2019