Mapping Urban Food and Nutrition Insecurity: A Spatial Index for Barcelona

Page 1 of 24 Mapping Urban Food and Nutrition Insecurity: A Spatial Index for Barcelona Working Paper, July 2025 Authors: Garcia-Sierra, Marta¹; Cruz, Irene²; Andreu, Marta²; LaRota-Aguilera, Maria José;3 Ana Moragues-Faus;3 Domene, Elena² ¹ Epidemiology and Public Health Area, FISABIO – Public Health, Valencia, Spain. ² Institut Metròpoli, Universitat Autònoma de Barcelona, Barcelona, Spain. 3 Faculty of Economics and Business, University of Barcelona, Spain. Project Name: FOOD4GOOD—Addressing Food Insecurity and Sustainability Challenges to Build Health, Prosperous Societies and Regenerate the Environment (PID2021-125013NA-I00) Working Paper Series: FISABIO – Epidemiology and Public Health Area WP Number and Date: WP1 – November 2025 Acknowledgements: This research was funded by the project FOOD4GOOD – Addressing Food Insecurity and Sustainability Challenges to Build Health, Prosperous Societies and Regenerate the Environment (PID2021-125013NA-I00), financed by MCIN/AEI/10.13039/501100011033 and by the FEDER, European Union. The authors would also like to acknowledge the support of the FOOD4GOOD consortium. Abstract Despite growing attention to sustainable food systems, food insecurity is still under-addressed in European cities, often sidelined in urban policy agendas and constrained by limited conceptual and spatial frameworks. The aim of this study is to develop a Food Insecurity Vulnerability Index (FIVI) with a spatial foundation, estimated at the census tract level, using a model that balances simplicity with strong explanatory power. The selected indicators span the six dimensions of food security and nutrition as defined by HLPE (2020) — availability, access, utilisation, stability, sustainability, and agency — and are spatially operationalised and adapted to the context of a Mediterranean European urban area, specifically Barcelona. A generalised linear model (GLM) with a quasi-binomial distribution was used, accounting for zeroinflation in the dependent variable —the prevalence of childhood obesity among boys aged 0 to 12 years (%). This proxy for malnutrition was selected due to the high-quality data systematically collected during scheduled paediatric check-ups. The model includes socioeconomic and food environment predictors, fitted using data from Barcelona and projected to Madrid using an equivalent dataset and the estimated coefficients. Key predictors include a lower income, a younger population profile, and greater access to organic food outlets, all significantly associated with reduced vulnerability. In contrast, higher voter abstention—used as a proxy for agency—was positively associated with vulnerability. These findings underscore the relevance of socioeconomic status, agency, and food access in shaping food insecurity risk. The resulting hotspot and coldspot mapping based on the index enhances understanding of food insecurity dynamics and their links to territorial, socio-economic, ecological, and health factors, offering valuable insights for local-level planning and the monitoring of targeted interventions. Keywords Food insecurity; synthetic index; hotspot mapping; Barcelona; Madrid. Page 2 of 24 1. Introduction Despite increasing attention to sustainable and healthy food systems, food insecurity remains a largely overlooked issue in European cities (Garratt, 2020; Long et al., 2020; Penne & Goedemé, 2021). Food insecurity may appear to refer exclusively to developing countries, but moderate and severe food insecurity levels have also affected almost 9% of the population in Europe and North America, and could even reach values of 10–15% in specific regions (Carrillo-Álvarez et al., 2021; FAO 2020, 2021). This is particularly evident in Mediterranean contexts, where the persistence of malnutrition, diet-related diseases, and unequal access to healthy food environments is frequently underestimated (Moragues-Faus & Magaña-González, 2022; Marchetti & Secondi, 2022). Food insecurity—understood as inconsistent access to adequate and nutritious food—acts as a key driver of malnutrition, including undernutrition, overweight, obesity, and micronutrient deficiencies. Recent data from Mediterranean countries show that food insecurity is far from marginal. In Italy, 22.3% of the population is at risk of food poverty or food insecurity, with regional disparities ranging from 14.6% in Umbria to 29.6% in Abruzzo (Marchetti & Secondi, 2022). In Greece, the 2020 Income and Living Conditions Survey (SILC) reported that 13.2% of the population worried about not having enough food, 12.8% could not maintain a healthy and nutritious diet, and 6.2% were forced to skip meals (Hellenic Statistical Authority, 2021). In Spain, the first nationally representative survey using the Food Insecurity Experience Scale (FIES) found that 13.3% of households (2.5 million) experienced food insecurity in 2021 during the COVID-19 pandemic (Moragues-Faus & Magaña-González, 2022). Food insecurity was most prevalent among households with lower socioeconomic status, economic difficulties, and precarious employment, and was often associated with chronic illness or disability within the household. In Catalonia, the 2024 Health Survey reported that 2.6% of the population could not afford a meal with meat, fish, chicken, or a vegetarian equivalent every two days, with prevalence rising sharply among those with only primary education (8.0%) or in lower social classes (4.1%) (Schiaffino & Medina, 2025). The most recent data on malnutrition indicate that 33.8% of boys and 28.4% of girls aged 2–17 were affected by obesity, overweight, or underweight (IDESCAT, 2025). Simultaneously, food price anomalies have surged since the pandemic, with bread, cereals, and other food items exceeding €126 in 2023, compared to levels below €100 prior to 2020 (IDESCAT, 2025). These figures, while illustrative, point to deeper structural issues. Recent reviews have emphasized that food insecurity is shaped by the interplay of structural, spatial, and social factors, and that many existing indicators fail to capture its full scope (Righettini & Bordin, 2023). Global assessments already suggest that the world is off track to achieve Sustainable Development Goal 2 (“Zero Hunger”) by 2030, as the prevalence of food insecurity continues to rise even in high-income regions (FAO et al., 2024). In wealthy countries, food insecurity does not stem from food scarcity, but from deepening inequalities and increasing barriers to affording healthy diets, driven by poverty, job precarity, and the rising cost of living (Schiaffino & Medina, 2025). These conditions have fuelled new forms of deprivation, visible in the growing reliance on food assistance and charitable food distribution (Caraher & Furey, 2018). In Europe, around 21% of the population (94.6 million people) were at risk of poverty or social exclusion in 2023 (Eurostat, 2024). Such figures reflect a paradox of abundance: while food availability is not a structural problem, unequal access to healthy and affordable food has become a major determinant of diet-related inequalities. Urban areas are where these inequalities and barriers to food access are most evident. Over three-quarters of the world’s food-insecure population now live in cities (Battersby et al, 2024), yet urban food insecurity remains systematically underexplored. Food insecurity is now measured at multiple levels—individual, household, national, regional, and global—but few instruments exist for analysing it at the urban scale, where disparities are widening, often reflected in growing inequalities in health due to malnutrition. Such inequalities are partly mediated by the characteristics of local food environments, which shape people’s everyday opportunities to access, afford, and consume healthy foods, thus amplifying the effects of income and social disadvantage. Evidence consistently shows that low-income and racial or ethnic minority communities tend to live in poorer-quality food environments, where healthy options are scarce and cheap, unhealthy foods are more accessible, exacerbating both food insecurity and nutrition-related health inequalities (Larson et al., 2009; Walker et al., 2010; Odoms-Young et al., 2024). Page 3 of 24 This complexity—rooted in the multidimensional and context-specific nature of food insecurity—demands not only a solid conceptual foundation but also robust and adaptable measurement tools. This reinforces the need for spatially sensitive and locally adapted approaches that reflect the lived realities of urban populations. In response to this need, our work focuses on the development of a Food Insecurity Vulnerability Index (FIVI) that is both parsimonious and spatially grounded. We depart from the HLPE’s (2020) extended definition of food security, which encompasses six dimensions—availability, access, utilisation, stability, agency, and sustainability—and operationalise these into spatially-based indicators. Drawing on the insights of La Rota-Aguilera & Moragues-Faus’s (2025) adaptation of the HLPE framework to urban contexts, the aim is to construct a composite index that is simple and allows for the localisation and monitoring of food insecurity in urban settings. The remainder of this working paper is structured as follows. Section 2 reviews the theoretical framework and state of the art. Section 3 details the methodological approach for constructing a new food insecurity index using data from the urban area of Barcelona, including indicator selection, model development, and spatial mapping of hotspots and coldspots. Section 4 presents the results and the exercise of projecting the index onto a second urban area. Finally, Section 5 discusses the findings, implications, and limitations of the study. 2. Background and Theoretical Framework 2.1 Conceptualising and Measuring Food Insecurity Due to the multidimensional and context-specific nature of food insecurity—especially in urban settings— understanding and addressing it requires not only a solid conceptual foundation but also robust and adaptable measurement tools. The concept of food insecurity has evolved since it was first understood as the mere absence of hunger, and so have the approaches used to measure it. The most extended definition of food security refers to “a situation that exists when all people, at all times, have physical, social and economic access to sufficient, safe and nutritious food that meets their dietary needs and food preferences for an active and healthy life” (FAO, 2000:26). As defined by the Food and Agriculture Organization (FAO) and the High Level Panel of Experts on Food Security and Nutrition (HLPE), it encompasses six dimensions: availability, access, utilization, stability, agency, and sustainability HLPE (2020). Availability refers to the sufficient supply of safe, culturally appropriate food through domestic production or imports primarily measured at national or regional levels. At the EU level, for instance, the European Food Security Crisis Preparedness and Response Mechanism (EFSCM) monitors food supply and security in times of crisis. It tracks trends in key indicators, external alerts, and provides a qualitative assessment of the EU agri-food sector. All EU Member States benefit from this mechanism within the framework of the common market. As such, these dimensions are of limited relevance when measuring vulnerability to food insecurity at the local or urban scale. Food access concerns individuals’ or households’ physical and financial ability to obtain adequate food without compromising other basic needs, ensuring inclusion of vulnerable groups. Utilization involves adequate diets, clean water, sanitation, and healthcare to achieve nutritional well-being. Stability reflects the ability to maintain consistent access to food despite shocks or cyclical events. Sustainability relates to food systems that protect environmental, social, and economic resources for future generations. Finally, agency highlights people’s capacity to participate in decisions affecting their food security, supported by appropriate policies and institutions. Together, these dimensions offer a comprehensive framework for addressing the complex nature of food insecurity (FAO, 2021; WFP, 2020). The stability dimension poses challenges when applied to city-scale analysis within a European context. However, in line with the extended definition of food security, it should also reflect the ability to maintain access to adequate food at the household level, particularly during periods of economic or inflationary crisis. Events such as rising unemployment or the increasing cost of living can compromise food security, even when food is physically available (Balistreri, 2016). In such situations, ensuring a stable food supply for vulnerable populations often depends on interventions that buffer the impact of reduced purchasing power, such as food assistance services (Loopstra & Tarasuk, 2015). We therefore propose that this Page 4 of 24 dimension of stability should also include the use of food aid programmes —such as food vouchers, wallet cards, school meal schemes, soup kitchens, food banks, or pantries. These mechanisms help secure continued household access to adequate nutrition when financial shocks undermine people’s ability to afford food. As Gracia-Arnaiz (2022) suggests, the uptake of food assistance can serve as a proxy indicator of vulnerability to food insecurity, signalling unmet needs and disruptions in food access during economic stress. While food security is often addressed at national or global levels, its six dimensions—availability, utilization, stability, sustainability, agency, and access—can also be meaningfully interpreted at the local scale. Availability and utilization are shaped by local infrastructure, services, and supply chains, while stability can be reflected in the presence of food assistance programs that buffer economic shocks. Sustainability and agency are increasingly embedded in municipal food strategies and participatory governance. Access is arguably the dimension most relevant at the local scale, as it is directly shaped by the spatial distribution, affordability, and availability of food outlets in a given territory. Local food environments—understood as the physical and economic conditions that determine where, how, and what food is available—play a central role in shaping dietary opportunities and constraints. The food environment includes the foods available for people in their daily life, and factors such as nutritional quality, safety, price, convenience, labelling and promotion of these foods (Downs et al. 2020). Food environments can have a significant influence on both access to and consumption of nutritious, adequate, and affordable foods by individuals. Consequently, they may affect the quality of people’s diets, nutrition status, and health outcomes (Wang et al., 2022; Eskandari et al. 2022; Swinburn et al., 2013; Caspi et al. 2012; Engler-Stringer et al. 2014). Food environments shape patterns of food acquisition and consumption, directly affecting access to healthy and affordable foods, and potentially altering the quality of their use. Some recent research reinforces the connection between food security and food environment. Evidence from peri-urban Flanders (Inaç et al., 2024) showed that perceived availability and affordability of fruit and vegetables, neighbourhood social cohesion, and housing conditions significantly influence the odds of experiencing food insecurity. Findings from the US highlight how under-resourced food environments disproportionately affect low-income and racial/ethnic minority populations, contributing to higher rates of food insecurity and diet-related chronic diseases such as obesity, diabetes, and cardiovascular conditions (Odoms-Young et al., 2024). Systematic reviews further reinforce this association, underscoring the role of the local food environment in shaping food insecurity outcomes. Households experiencing food insecurity tend to have greater exposure to unhealthy food options, rely more frequently on convenience stores and small retailers for food purchases, and report limited access to fresh and nutritious foods such as fruits and vegetables (Bezerra et al., 2024). Despite a growing body of evidence, there remains a significant gap in the availability of spatially grounded indices—based on secondary data—that can localize and monitor food insecurity in urban areas. Existing tools often fail to capture the complex, multi-dimensional nature of food insecurity as it manifests across different urban contexts. Yet such indices are essential for identifying vulnerable neighbourhoods, informing place-based interventions, and guiding resource allocation in municipal/urban food policies. To address this gap, we propose the development of a Food Insecurity Vulnerability Index (FIVI), designed to capture the spatial dimensions of food insecurity at the local scale through a parsimonious set of indicators. This study builds on a previous analytical framework grounded on the HLPE’s six‐dimensional approach applied to the case of Barcelona (LaRota‐Aguilera et al. 2025), and outlined in Table 1. Page 5 of 24 Table 1. Conceptual framework for measuring food insecurity in urban areas, including proposed dimensions, attributes, and indicators. Dimension Attribute Indicator Availability: Having a quantity and quality of food sufficient to satisfy the dietary needs of individuals, free from adverse substances and acceptable within a given culture, supplied through domestic production or imports. Sufficiency Healthy food availability Access: Having personal or household financial means to acquire food for an adequate diet at a level to ensure that satisfaction of other basic needs are not threatened or compromised; and that adequate food is accessible to everyone, including vulnerable individuals and groups. Economic access Income Social access Foreign Youth Physical access Healthy food access Stability: Having the ability to ensure food security in the event of sudden shocks (e.g. an economic, health, conflict or climatic crisis) or cyclical events (e.g. seasonal food insecurity). Food aid School meal program Food aid program Utilization: Having an adequate diet, clean water, sanitation and health care to reach a state of nutritional well-being where all physiological needs are met Adequate diet Prevalence of stunting among children under 5 y Prevalence of malnutrition among children under 5 y (wasting and overweight) Prevalence of anaemia in women 15-49 y by pregnancy status Agency: Individuals or groups having the capacity to act independently to make choices about what they eat, the foods they produce, how that food is produced, processed, and distributed, and to engage in policy processes that shape food systems. The protection of agency requires socio-political systems that uphold governance structures that enable the achievement of food security and nutrition for all. Capacity to act Voter abstention (proxy of political participation) Sustainability: Food system practices that contribute to long-term regeneration of natural, social and economic systems, ensuring the food needs of the present generations are met without compromising the food needs of future generations. Sustainability Organic food access Source: Adapted from HLPE (2020) definition of food security encompassing six dimensions, and extended with insights from LaRota‐Aguilera et al. (2025). Page 6 of 24 2.2 Rationale for a Spatial, Localised Food Insecurity Index Although food insecurity has been widely studied, spatially grounded indices that allow for fine-grained analysis at the subnational or urban scale remain relatively rare. The majority of existing tools rely on socalled experience-based food insecurity scales, such as the FAO’s Food Insecurity Experience Scale (FIES) or the USDA food insecurity scales, which are based on householdor individual-level surveys (CarrilloAlvarez et al., 2021; Ballard et al., 2014; Balistreri, 2016). At the global level, the FIES represents an important advance in psychometrically validated measurement (Ballard et al., 2014), but its reliance on surveys and national-level resolution limits its utility for urban and neighbourhood-level analysis. Moreover, as Ballard et al. (2014) note, the FIES provides a valuable tool for advancing knowledge on the relationship between the lived experience of food insecurity and indicators of malnutrition. However, it is not itself a measure of malnutrition and cannot detect nutritional deficiencies or obesity, making it unsuitable for assessing nutrition-specific outcomes of food security policies or programmes. These limitations highlight the need for approaches capable of capturing the spatial heterogeneity of food insecurity within urban areas. At a more conceptual level, recent reviews of food security indices (Manikas et al., 2023; Al-Ansari et al., 2025) highlight the importance of integrating complementary indicators that reflect multiple dimensions of food security (availability, access, utilisation, stability), while arguing that experience-based indicators are especially suited for rapid assessments but insufficient for structural or spatial analyses. A further challenge of these indices stems from potential biases in ranking, which show profound divergences across tools (AlAnsari et al., 2025). Poudel and Gopinath (2021) compared outputs from several global food security indicators—from FAO, UNDP, IFPRI, and USDA—between 1991 and 2018, and found major variations in prevalence estimates. They provide further evidence that macro-structural drivers—such as GDP growth, literacy, urbanisation, and internet access—play differentiated roles depending on income levels, reinforcing the need for context-sensitive tools that can be applied at multiple scales, including the urban. Similarly, in a recent scoping review of individualand household-level measures of food insecurity in high-income countries, Carrillo-Alvarez et al. (2021) identified 23 different instruments and highlighted recurring correlates of food insecurity—sociodemographics, health, social stressors, and environmental factors—along with persistent inequalities by gender, race, and geography. While valuable, these approaches fall short of providing standardised, replicable, and validated indices that capture the multidimensional and structural drivers of food insecurity across diverse urban contexts. Together, these findings underscore the absence of spatially explicit, comparable, and scalable tools for assessing food insecurity in cities. At the urban level, Odoms-Young et al. (2024) argue that under-resourced neighbourhood food environments disproportionately affect low-income and minority populations, and call for moving beyond individual and household drivers to address broader structural determinants—such as neighbourhood conditions, systemic poverty, and policy-level influences. They emphasise the need for context-sensitive tools that capture these multilevel drivers of urban food insecurity and support equity-oriented interventions. Urban food environment indices, such as the Retail Food Environment Index and its modifications, are widely used in health studies (Odoms-Young et al., 2024), but they capture only the distribution of food outlets rather than food insecurity itself, and they lack standardised, validated links to food insecurity outcomes. Several efforts illustrate both progress and limitations. Local governments and research institutions in the U.S., UK, and Canada have created neighbourhoodor municipality-level indices (e.g., Toronto Food Strategy; Feeding America’s county-level estimates). However, these instruments often rely on proprietary or survey-based data, lack methodological transparency, and are rarely replicable beyond their original context (Loopstra & Tarasuk, 2015). Similarly, the USDA Food Access Research Atlas employs secondary data to map low-income, low-access census tracts, but it provides only indirect measures of potential access rather than actual food insecurity experiences. These limitations highlight the opportunity to develop a transparent, replicable, and spatially sensitive tool that can operate at fine urban scales. Given the growing awareness of the spatial dimensions of food insecurity, this section makes the case for a new, context-sensitive, and exportable index. It highlights the need for a parsimonious model tailored to urban settings and underscores the added value of spatial analysis—particularly hotspot and coldspot mapping—for understanding territorial disparities and informing place-based policy responses. To address Page 7 of 24 these limitations, this study develops the Food Insecurity Vulnerability Index (FIVI), a spatially grounded parsimonious tool designed to locate populations experiencing chronic food insecurity. The FIVI assesses five of the six dimensions of food security (access, stability, utilisation, agency, and sustainability) and enables the identification of areas of high incidence (hotspots) within urban settings. The notion of parsimonious here refers to a model that captures the complexity of food insecurity with a minimal yet sufficient number of variables, facilitating its application across diverse urban contexts. By integrating indicators from multiple domains—socioeconomic, health, and food environment—the FIVI provides a contextually sensitive and exportable framework that captures the spatial patterns of food insecurity at the census tract level. It therefore serves as a valuable tool for urban policy and planning, allowing local governments to identify vulnerable areas, monitor changes over time, and design targeted interventions to ensure equitable access to adequate and nutritious food. 3. Methodology 3.1 Variable selection Selected indicators spanned demographic, socioeconomic, health, and food environment domains, consistent with the six dimensions —availability, access, utilization, stability, sustainability, and agency— included in the definition of food security and nutrition (HLPE, 2020). Initially, some 20 indicators were extracted from on a thorough literature review conducted during the last trimester of 2023 (Table 2). From this long list, around 12 indicators were collected in cartographic format and assessed for statistical suitability (Table 1). The final FIVI, however, is a curated index composed of six predictors—household income, foreign population, young population (16–29 years), voter abstention, availability of healthy food outlets, and availability of organic food outlets—with the dependent variable defined as the prevalence of obesity in boys aged 0–12, expressed as a proportion. We prioritised long-term, preferably annually updated indicators available for Spain at the census tract level, or those that could be readily constructed. Variable selection involved correlation analysis and multicollinearity testing using the Variance Inflation Factor (VIF), accepting predictors with VIF values below 5 to ensure robustness. Page 8 of 24 Table 2. Main indicators identified from literature review for measuring food insecurity in urban areas. Risk Factor Indicators selected from literature review References Data availability Food environment Presence of community gardens & food coops Drisdelle et al. (2020); Morgan (2020); Taylor & Ard (2015); Westbury et al. ( 2021 ) Yes Access to organic food (density or proximity of organic food stores) Downs et al. (2020); Wang et al. (2019); Eskandari et al. (2022); Swinburn et al. (2013); Caspi et al. (2012); Engler-Stringer et al. (2014); Turner et al. ( 2020 ) Yes Total access to food (density or proximity of food stores) Yes Poverty Household income Moragues-Faus & Magaña-González (2022)-Spain; Leitz (2018)-US; Garratt (2020)-Europe Yes Poverty rate Yes Population receiving social benefits No High housing costs (residential vulnerability ) Housing affordability Leitz (2018)-USA; Garratt (2020)-Europe Yes Unemployment or underemployment Unemployment rate Moragues-Faus & Magaña-González (2022)-Spain; Leitz (2018)-US; Garratt (2020) - Europe Yes Limited access to affordable food Affordability of healthy food options No Inadequate social support Access to school meal programs Leitz (2018) - US; Garratt (2020) - Europe Yes Availability or use of food assistance programs Yes Education level Education level or school drop -out Moragues-Faus and Magaña-González (2022)-Spain; Leitz (2018)-US; Garratt (2020) - Europe No Social and cultural inequalities (racial and ethnic minorities, immigrants, individuals with disabilities) Elderly population Leitz (2018)-US; Garratt (2020)-Europe García et al. (2020) - Barcelona; Lee & Frongillo (2001) Yes Foreign population (immigration) Moragues - Faus and Magaña - González (2022) - Spain Yes Gender (head of household) Moragues - Faus and Magaña - González (2022) - Spain No Homeless population Yes Political participation Olabiyi, O. M. (2020) Yes Health conditions Prevalence of mental disorders in the general population Coleman-Jensen et al. (2013); Moragues-Faus and Magaña-González (2022)-Spain; Eskandari et al. (2022); Swinburn et al. (2013); Caspi et al. ( 2012 ) Yes Prevalence of stunting among children under 5 y NLIS (WHO, 2017; 2019) Yes Prevalence of malnutrition among children under 5 y (wasting and overweight) NLIS (WHO, 2017; 2019) Yes Prevalence of anaemia in women 15 - 49 y by pregnancy status NLIS (WHO, 2017; 2019) Yes Page 9 of 24 Data were retrieved at scale of the census tract for selected indicators. The city of Barcelona has some 1.6 million inhabitants and was administratively divided into 1,067 census tracts in 2021/22, each comprising approximately 1,500 individuals. The reference period for the analysis is 2021–2023; specifically, health data from 2023 were used to minimize the effects of the COVID-19 pandemic (e.g., disruptions to regular check-ups). All variables were harmonised to 2021/22 census tracts. Indicators were selected to ensure data generality and homogeneity across Barcelona while maximizing territorial detail to capture internal variations within the city. Socioeconomic and demographic data—household income, proportion of foreign population, and proportion of young residents (aged 16–29)—were retrieved from the Spanish National Institute of Statistics (INE). The income indicator corresponds to the household income median, based on the 2022 edition of the Atlas de Distribución de Renta de los Hogares (INE, 2022). The foreign population indicator is defined as the percentage of residents originating from lowor middle-income countries, according to World Bank classifications, using data from the Ministerio del Interior, 2021 and Estadística del Padrón Continuo a 1 de enero de 2020, INE. Voter abstention data were obtained from municipal statistical offices, in turn based on the Ministerio del Interior (2020), and referring to the 2019 municipal elections for both Barcelona and Madrid City Councils. In Spain, foreign residents can vote in municipal elections if they are citizens of countries that grant reciprocal voting rights to Spaniards through a treaty.1 Additionally, all EU citizens residing in Spain who meet the same requirements as Spanish voters and formally declare their intention to vote are also eligible to participate in local elections. The indicators on food access include two variables: HealthyFA and OrganicFA. The HealthyFA (Healthy Food Access) indicator captures the presence of healthy food outlets per census tract, including grocery shops (butchers, fishmongers, fruit and vegetable shops, and organic stores), organic supermarkets, and general supermarkets which might or might not offer organic produce. The OrganicFA (Organic Food Access) indicator refers specifically to the availability of specialised organic shops, organic supermarkets, food cooperatives and groups. Data on food outlets for Barcelona were obtained from the Census of groundfloor commercial premises in the city of Barcelona (Barcelona City Council, 2022). For Madrid, data were sourced from the Census of Premises and Activities (Madrid City Council, as of September 2023), except for organic food outlets, which were retrieved from the Mapping of Alternative Food Networks (MAE – Madrid Agroecológico, 2024). The classification of food retail outlets was based on an adaptation of prior work by Garcia et al. (2020) and Garcia-Sierra et al. (2021) in Barcelona, and Bilal et al. (2018) and Díez et al. (2019) in Madrid. Using the opportunities provided by each city’s retail census, a harmonised classification system was created to enable comparability across both urban foodscapes. This is further described in the Supplementary Information SI.2. Finally, health indicators, used as dependent variable, were based on those from the WHO Nutrition Landscape Information System (NLIS) (WHO, 2017; 2019), and which further coincide with the Sustainable Development Goal (SDG) Indicators for SDG2: End hunger, achieve food security and improved nutrition, and specifically those from 2.2. These indicators were retrieved from the Information System for Research in Primary Care (SIDIAP; www.sidiap.org) from Catalonia, Spain (Bolíbar et al., 2012). The SIDIAP database is a large pseudo-anonymised database comprising the electronic health records from primary care visits collected since 2005 in centres managed by the Catalan Health Institute. It covers approximately 5.5 million individuals, representing around 74% of the population of Catalonia. SIDIAP includes individual-level demographic information, clinical variables, immunisations, referrals to specialists, prescriptions and dispensation records, sick leave history, and any acute or chronic health conditions recorded during primary care visits. This study was approved by the ethics committee of the Jordi Gol i Gurina Institute for Research in Primary Care (IDIAPJGol: 23/238-P). Health indicators included:  Prevalence of iron-deficiency anaemia in women of reproductive age (15-49 years), by age and pregnancy status (n, % crude and standardized)  Prevalence of iron-deficiency anaemia and severe anaemia in children aged 6-59 months (under 5 years), by age and gender (n, % crude and standardized) 1 These currently include Bolivia, Cape Verde, Chile, Colombia, South Korea, Ecuador, Iceland, Norway, New Zealand, Paraguay, Peru, the United Kingdom, and Trinidad and Tobago. Page 16 of 24 References Ajuntament de Barcelona (2022). Estratègia d’alimentació sostenible i saludable de Barcelona 2030. Al-Ansari, M. A., Nabeel, H., Abdella, G. M., & El Mekkawy, T. (2025). Moving Beyond Indices: A Systematic Approach Integrating Food System Performance and Characteristics for Comprehensive Food Security Assessment. Foods, 14(10), 1834. Ayuntamiento de Madrid (2023). Estrategia Alimentaria de la Ciudad de Madrid. Balistreri, K. S. (2016). A decade of change: measuring the extent, depth and severity of food insecurity. Journal of family and economic issues, 37(3), 373-382. Ballard, T. J., Kepple, A. W., Cafiero, C., & Schmidhuber, J. (2014). Better measurement of food insecurity in the context of enhancing nutrition. Ernahrungs Umschau, 61(2), 38-41. Battersby, J., Hatab, A. A., Ambikapathi, R., Chicoma, J. L., Shulang, F., Kimani‑Murage, E., ... & Resnick, D. (2024). Strengthening urban and peri‑urban food systems to achieve food security and nutrition, in the context of urbanization and rural transformation, HLPE Report No. 19 (Vol. 19). Food and Agricultural Organization of the United Nations (FAO)-Giulio Napolitano. Bezerra, M. S., Lima, S. C. V. C., De Souza, C. V. S., Seabra, L. M. J., & Lyra, C. D. O. (2024). Food environments and association with household food insecurity: a systematic review. Public Health, 235, 4248. Bilal, U., Jones-Smith, J., Diez, J., Lawrence, R. S., Celentano, D. D., & Franco, M. (2018). Neighborhood social and economic change and retail food environment change in Madrid (Spain): The heart healthy hoods study. Health & Place, 51, 107-117. Bolíbar, B., Aviles, F. F., Morros, R., del Mar Garcia-Gil, M., Hermosilla, E., Ramos, R., ... & PrietoAlhambra, D. (2012). Base de datos SIDIAP: La historia clínica informatizada de Atención Primaria como fuente de información para la investigación epidemiológica. Medicina Clínica, 138(14), 617-621. Candel, J. J., & Pereira, L. (2017). Towards integrated food policy: Main challenges and steps ahead. Environmental Science & Policy, 73, 89-92. Caraher, M., & Furey, S. (2018). Growth of food banks in the UK (and Europe): Leftover food for leftover people. In The economics of emergency food aid provision (pp. 25–48). Palgrave Macmillan. Carrillo-Alvarez, E., Salinas-Roca, B., Costa-Tutusaus, L., Milà-Villarroel, R., & Shankar Krishnan, N. (2021). The measurement of food insecurity in high-income countries: a scoping review. International Journal of Environmental Research and Public Health, 18(18), 9829. Caspi, C. E., Sorensen, G., Subramanian, S. V., & Kawachi, I. (2012). The local food environment and diet: A systematic review. Health & Place, 18(5), 1172–1187. Darmon, N., & Drewnowski, A. (2015). Contribution of food prices and diet cost to socioeconomic disparities in diet quality and health: a systematic review and analysis. Nutrition reviews, 73(10), 643-660. Díez, J., Cebrecos, A., Rapela, A., Borrell, L. N., Bilal, U., & Franco, M. (2019). Socioeconomic inequalities in the retail food environment around schools in a Southern European context. Nutrients, 11(7), 1511. Doustmohammadian, A., Mohammadi-Nasrabadi, F., Keshavarz-Mohammadi, N., Hajjar, M., Alibeyk, S., & Hajigholam-Saryazdi, M. (2022). Community-based participatory interventions to improve food security: A systematic review. Frontiers in nutrition, 9, 1028394. Downs, S. M., Ahmed, S., Fanzo, J., & Herforth, A. (2020). Food environment typology: Advancing an expanded definition, framework, and methodological approach for improved characterization of food environments toward sustainable diets. Foods, 9(4), 532. Page 17 of 24 Drisdelle, C., Kestens, Y., Hamelin, A. M., & Mercille, G. (2020). Disparities in access to healthy diets: how food security and food shopping behaviors relate to fruit and vegetable intake. Journal of the Academy of Nutrition and Dietetics, 120(11), 1847-1858. Engler-Stringer, R., Le, H., Gerrard, A., & Muhajarine, N. (2014). The community and consumer food environment and children's diet: A systematic review. BMC Public Health, 14(1), 522. Eskandari, F., Lake, A. A., Rose, K., Butler, M., & O'Malley, C. (2022). A mixed‐method systematic review and meta‐analysis of the influences of food environments and food insecurity on obesity in high‐income countries. Food science & nutrition, 10(11), 3689-3723. Eurostat. (2024). Persons at risk of poverty or social exclusion by age and sex [Data set]. https://ec.europa.eu/eurostat/databrowser/view/ilc_peps01n__custom_11655562/bookmark/table?lang=en &bookmarkId=1fe5fbf6‐51a6‐4b4b‐a0d5‐b9e715618a34 FAO (2000). The state of food insecurity in the world 2000. FAO. Retrieved from: https://www.fao.org/agrifood-economics/publications/detail/en/c/122102/ FAO, IFAD, UNICEF, WFP, & WHO. (2024). The state of food security and nutrition in the world 2024: Financing to end hunger, food insecurity and malnutrition in all its forms. https://doi.org/10.4060/cd1254en FAO., IFAD., UNICEF., WFP., & WHO. (2021). The state of food security and nutrition in the world 2021: Transforming food systems for food security, improved nutrition and affordable healthy diets for all. FAO. https://doi.org/10.4060/cb4474en Garcia, X., Garcia-Sierra, M., & Domene, E. (2020). Spatial inequality and its relationship with local food environments: The case of Barcelona. Applied Geography, 115, 102140. García-Sierra, M., Domene, E., Borràs, M., Ruíz, N., Sánchez, A., & Andreu, M. (2021, December). Food environments, childhood and health: Exposure in the school environment in Barcelona [Research report]. Institut d’Estudis Regionals i Metropolitans de Barcelona (IERMB). https://www.institutmetropoli.cat/ca/estudi/entorns-alimentaris-infancia-salut-exposicio-lambit-escolarbarcelona/ Garratt, E. (2020). Food insecurity in Europe: Who is at risk, and how successful are social benefits in protecting against food insecurity?. Journal of Social Policy, 49(4), 785-809. Gracia-Arnaiz, M. (2022). The precarisation of daily life in Spain: Austerity, social policy and food insecurity. Appetite, 171, 105906. Hellenic Statistical Authority. Press Release. Food Security. 2022 Survey on Income and Living Conditions. Piraeus. 2023. Available online: https://www.statistics.gr/documents/20181/52ab7204-642ecf81-f1ed-87f62f059174 (accessed on 20 December 2023). HLPE. 2020. Food security and nutrition: building a global narrative towards 2030. A report by the High Level Panel of Experts on Food Security and Nutrition of the Committee on World Food Security, Rome. Retrieved from: http://www.fao.org/3/ca5162en/ca5162en.pdf IDESCAT. (2025). Indicadors dels objectius de desenvolupament sostenible de les NU per a Catalunya. Institut d’Estadística de Catalunya. Retrieved from: https://www.idescat.cat/dades/ods/nu/ Inaç, Y., D’Hooghe, S., De Ridder, K., Dury, S., Van de Weghe, N., De Clercq, E. M., ... & Vandevijvere, S. (2024). Determinants of food insecurity among adults residing in peri-urban municipal settings in Flanders, Belgium. BMC Public Health, 24(1), 2034. LaRota‐Aguilera, M. J., & Moragues-Faus, A. (2025). Reclaiming Food Insecurity in European Urban Policies: Lessons From Public and Community‐Based Initiatives. Urban Planning, 10. Larson, N. I., Story, M. T., & Nelson, M. C. (2009). Neighborhood environments: disparities in access to healthy foods in the US. American journal of preventive medicine, 36(1), 74-81. Page 18 of 24 Lee, J. S., & Frongillo, E. A. (2001). Factors associated with food insecurity among U.S. elderly persons: Importance of functional impairments. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 56(2), S94-S99. Long, M. A., Gonçalves, L., Stretesky, P. B., & Defeyter, M. A. (2020). Food insecurity in advanced capitalist nations: A review. Sustainability, 12(9), 3654. Loopstra, R., & Tarasuk, V. (2015). Food bank usage is a poor indicator of food insecurity: Insights from Canada. Social Policy and Society, 14(3), 443-455. Manikas, I., Ali, B. M., & Sundarakani, B. (2023). A systematic literature review of indicators measuring food security. Agriculture & food security, 12(1), 10. Marchetti, S., & Secondi, L. (2022). The economic perspective of food poverty and (in) security: An analytical approach to measuring and estimation in Italy. Social indicators research, 162(3), 995-1020. Moragues-Faus, A. (2024). Urban food governance for just transitions: Reflections from European city networks. Local Environment, 29(1), 1–18. Moragues‐Faus, A., & Magaña‐González, C. R. (2022). Alimentando un futuro sostenible: Estudio sobre la situación alimentaria en Barcelona 2022. Universidad de Barcelona; Fundación Daniel y Nina Carasso. https://www.ub.edu/alimentandounfuturosostenible/documents/informe‐alimentacion_una‐pag.pdf Odoms-Young, A., Brown, A. G., Agurs-Collins, T., & Glanz, K. (2024). Food insecurity, neighborhood food environment, and health disparities: state of the science, research gaps and opportunities. The American journal of clinical nutrition, 119(3), 850-861. Penne, T., & Goedemé, T. (2021). Can low-income households afford a healthy diet? Insufficient income as a driver of food insecurity in Europe. Food Policy, 99, 101978. Poudel, D., & Gopinath, M. (2021). Exploring the disparity in global food security indicators. Global Food Security, 29, 100549. Prats, M., Smid, S., & Ferrín, M. (2024). Lack of trust in institutions and political engagement. Documents de travail de l'OCDE sur la gouvernance publique. Rehm, C. D., Peñalvo, J. L., Afshin, A., & Mozaffarian, D. (2016). Dietary intake among US adults, 19992012. Jama, 315(23), 2542-2553. Righettini, M. S., & Bordin, E. (2023). Exploring food security as a multidimensional topic: twenty years of scientific publications and recent developments. Quality & Quantity, 57(3), 2739-2758. Schiaffino A, Medina A. Enquesta de salut de Catalunya, 2024. Barcelona: Direcció General de Planificació en Salut; 2025. Sonnino, R., & Coulson, H. (2021). Unpacking the new urban food agenda: The changing dynamics of global governance in the urban age. Urban Studies, 58(5), 1032-1049. Sreehari, E., & Babu, L. D. (2023). Global Hunger Index: A multistage coefficient estimation analysis using machine learning techniques for a hunger free society. Journal of Cleaner Production, 429, 139515. Swinburn, B., Vandevijvere, S., Kraak, V., Sacks, G., Snowdon, W., Hawkes, C., ... & INFORMAS. (2013). Monitoring and benchmarking government policies and actions to improve the healthiness of food environments: a proposed G overnment H ealthy F ood E nvironment P olicy I ndex. Obesity reviews, 14, 24-37. Walker, R. E., Keane, C. R., & Burke, J. G. (2010). Disparities and access to healthy food in the United States: A review of food deserts literature. Health & place, 16(5), 876-884. Wang, L., Huang, W., Zhao, C., Hu, Y., & Cui, S. (2022). Exploring the environment-nutrition-obesity effects associated with food consumption in different groups in China. Journal of environmental management, 317, 115287. Page 19 of 24 WHO (2017) Global Nutrition Monitoring Framework: operational guidance for tracking progress in meeting targets for 2025. Geneva: World Health Organization; 2017. Retrieved from: https://iris.who.int/server/api/core/bitstreams/9855b752-f8e2-443b-af04-f051c8ecedb2/content WHO (2019) Nutrition Landscape Information System (NLiS) country profile indicators: interpretation guide, second edition. Geneva: World Health Organization; 2019. Retrieved from: https://www.who.int/publications/i/item/9789241516952 World Food Programme. (2020). The State of Food Security and Nutrition in the World 2020: Transforming food systems for affordable healthy diets. Rome: FAO, IFAD, UNICEF, WFP and WHO. https://doi.org/10.4060/ca9692en Page 20 of 24 Supplementary Information SI.1. Health indicators in a nutshell. Prevalence of iron-deficiency anaemia:  Indicator of prevalence of iron-deficiency anaemia in women of reproductive age (15-49 years), by age and pregnancy status (n, % crude and standardized)  Indicator of prevalence of iron-deficiency anaemia and severe anaemia in children aged 6-59 months (under 5 years), by age and gender (n, % crude and standardized) Prevalence of malnutrition:  Indicators of prevalence of malnutrition in children under 5 years – Population aged 0-5 years with stunting, wasting, overweight, or underweight, by age and gender (n, % crude and standardized)  Indicators of prevalence of malnutrition in in school-age children (6-12 years) – Population aged 6-12 years with overweight and obesity, by age and gender (n, % crude and standardized)  Indicators of prevalence of malnutrition in adults aged 18-74 years – Population aged 18-74 years with underweight, overweight, and obesity, by age and gender (n, % crude and standardized) Indicator Calculation Notes: Indicator of prevalence of iron-deficiency anaemia in women of reproductive age (15-49 years), by age and pregnancy status (n, % crude and standardized) Definition: Women of reproductive age (15–49 years) with iron-deficiency anaemia by age and pregnancy status (n, % crude and standardized) Calculation of anaemia in women of reproductive age (15-49 years), by pregnancy status: (Number of women aged 15 to 49 years with a diagnostic code for iron-deficiency anaemia DD50 OR with a haemoglobin concentration <120 g/L for non-pregnant women and lactating women at sea level, AND <110 g/L for pregnant women at sea level) / (Women aged 15−49 years assigned to the primary care team and who had haemoglobin concentration assessed during a specified period) * 100. Calculation at the scale of census tract from SIDIAP data (n, % crude and standardized): SIDIAP data: age (years), gender, pregnancy status (non-pregnant women, lactating women, pregnant women), haemoglobin concentration in clinical analytics, and diagnostic code (CIM-10) for irondeficiency anaemia DD50 (latest year with available data). Territorial division: Barcelona’s census tracts. Indicator of prevalence of iron-deficiency anaemia and severe anaemia in children aged 6-59 months (under 5 years), by age and gender (n, % crude and standardized) Definition: Children aged 6−59 months (under 5 years) with iron-deficiency anaemia and severe anaemia by gender (n, % crude and standardized) Calculation of anaemia in children aged 6-59 months: (Number of children aged 6−59 months with a haemoglobin concentration <110 g/L at sea level) / (Total number of children aged 6−59 months assigned to the primary care team and who had haemoglobin concentration assessed) * 100. Page 21 of 24 Calculation at the scale of census tract from SIDIAP data (n, % crude and standardized): SIDIAP data: age (months), gender, haemoglobin concentration in clinical analytics, and diagnostic code (CIM-10) for iron-deficiency anaemia DD50 (latest year with available data). Disaggregation by gender. Territorial division: Barcelona’s census tracts. Indicators of prevalence of malnutrition in children under 5 years – Population aged 0-5 years with stunting, wasting, overweight, or underweight, by age and gender (n, % crude and standardized) Definition: Children aged 0-5 years with stunting, wasting, overweight, or underweight by gender (n, % crude and standardized). Calculation of stunting (WHO indicator): (Population aged 0-5 years with stunting according to WHO tables) / (Population aged 0-5 years assigned to primary care) * 100. Calculation of wasting/underweight (WHO indicator): (Population aged 0-5 years with wasting according to WHO tables) / (Population aged 0-5 years assigned to primary care) * 100. Calculation of overweight (WHO indicator): (Population aged 0-5 years with overweight according to WHO tables) / (Population aged 0-5 years assigned to primary care) * 100. In children aged under 5 years these indicators are defined as follows:  stunting – sex-specific height-for-age <-2 SD of the WHO Child growth standards median;  wasting – sex-specific weight-for-height <-2 SD of the WHO Child growth standards median;  overweight – sex-specific weight-for-height >+2 SD of the WHO Child growth standards median; and  underweight – sex-specific weight-for-age <-2 SD of the WHO Child growth standards median. The WHO Child growth standards can be found for age groups here: WHO tables for the age groups of 0-2 and 2-5 years. Note: (*) For the purposes of this study, an intermediate calculation of overweight (excluding obesity) is also included. Calculation at the scale of census tract from SIDIAP data (n, % crude and standardized): SIDIAP data: age (months), gender, weight (in kilograms), height (in meters). Disaggregation by gender. Territorial division: Barcelona’s census tracts. Indicators of prevalence of malnutrition in in school-age children (6-12 years) – Population aged 6-12 years with overweight and obesity, by age and gender (n, % crude and standardized) Definition: Children aged 6-12 years with overweight and obesity (n, % crude and standardized). Calculation of overweight (WHO indicator): (Population aged 6-12 years with overweight or obesity according to WHO tables) / (Population aged 6-12 years assigned to primary care) * 100. Calculation of obesity (WHO indicator): (Population aged 6-12 years with obesity according to WHO tables) / (Population aged 6-12 years assigned to primary care) * 100. In school-age children and adolescents aged 5–19 years these indicators are defined as follows:  overweight – sex-specific BMI-for-age (equivalent to BMI 25 kg/m2 at 19 years) >+1 SD of the WHO Child growth standards median; and Page 22 of 24  obesity – sex-specific BMI-for-age (equivalent to BMI 30 kg/m2 at 19 years) >+2 SD of the WHO Child growth standards median. In children and adolescents (5 to 19 years), BMI is age and sex-specific. The WHO Child growth standards for BMI-for-age (5-19 years) can be found here: WHO table for the age group of 5 to 19 years. Calculation of Body Mass Index (BMI) = weight (in kilograms)/height2 (in meters). Note: (*) For the purposes of this study, an intermediate calculation of overweight (excluding obesity) is also included. Calculation at the scale of census tract from SIDIAP data (n, % crude and standardized): SIDIAP data: age (months), gender, weight (in kilograms), height (in meters). Disaggregation by gender. Territorial division: Barcelona’s census tracts. Indicators of prevalence of malnutrition in adults aged 18-74 years – Population aged 18-74 years with underweight, overweight, and obesity, by age and gender (n, % crude and standardized) Definition: Population aged 18-74 years with moderate and severe thinness, underweight, overweight, and obesity by gender (n, % crude and standardized). Calculation of underweight (WHO indicator): (Population aged 18-74 years with a BMI < 18.5) / (Population aged 18-74 years assigned to primary care) * 100. *Intermediate calculation of overweight: (Population aged 18-74 years with a BMI > 25 AND ≤ 30 ORor a diagnostic code for abnormal weight gain) / (Population aged 18-74 years assigned to primary care) * 100. For overweight in adults, it will be considered that the patient has a last value of BMI (Body Mass Index) >25 or the diagnostic code (CIM-10) R63.5 abnormal weight gain. Calculation of obesity: (Population aged 18-74 years with a BMI > 30 OR a diagnostic code for obesity) / (Population aged 18-74 years assigned to primary care) * 100. For obesity in adults, it will be considered that the patient has a last value of BMI (Body Mass Index) >30 OR one of the following diagnostic codes (CIM-10): E66 obesity, E66.0 obesity secondary to excess calories, E66.1 drug-induced obesity, E66.2 extreme obesity with alveolar hypoventilation, E66.8 morbid obesity (BMI > 40), E66.9 unspecified obesity. Calculation of overweight (WHO indicator): (Population aged 18-74 years with overweight or obesity) / (Population aged 18-74 years assigned to primary care) * 100. Calculation of Body Mass Index (BMI) = weight (in kilograms)/height2 (in meters). The population aged 75 years and older does not have a clear reference value, which is why they are excluded from this analysis. Note: (*) For the purposes of this study, an intermediate calculation of overweight (excluding obesity) is also included. Calculation at the scale of census tract from SIDIAP data (n, % crude and standardized): SIDIAP data: age (years), gender, weight (in kilograms), height (in meters), diagnostic code (CIM-10) for abnormal weight gain R63.5 (latest year with available data) and diagnostic codes (CIM-10) for obesity E66 (latest year with available data). Disaggregation by gender. Territorial division: Barcelona’s census tracts. Page 23 of 24 SI.2. Socioeconomic and food environment indicators. Data sources The population data by single year of age at the census tract level, corresponding to the annual population census as of 1 January 2022 and 1 January 2023, were obtained from the National Statistics Institute (INE). The socioeconomic data on household income (median) were sourced from the Household Income Distribution Atlas (Atlas de distribución de renta de los hogares) of the National Statistics Institute (INE, 2021). The data sources for the analysis of food environments in Madrid and Barcelona are detailed below in Tables 1.SI and 2.SI, respectively. Table 2.SI. Data sources for the analysis of food environments in Barcelona. Food environment, Barcelona Source Year Scale Food retail outlets Cens d'activitats econòmiques en planta baixa de la ciutat de Barcelona (Ajuntament de Barcelona, 2023) 2022 Coord. UTM Organic food outlets Cens d'activitats econòmiques en planta baixa de la ciutat de Barcelona (Ajuntament de Barcelona, 2023) 2022 Coord. UTM Cooperatives and consumer groups Institut Metropoli (producción propia) 2019 Coord. UTM Community urban gardens Base de dades d’horts urbans de l’Ajuntament de Barcelona (Direcció d'Espais Verds i Biodiversitat Medi Ambient i Serveis Urbans ‒ Ecologia Urbana, Ajuntament de Barcelona, 2021) 2021 Coord. UTM Municipal urban gardens Base de dades d’horts urbans de l’Ajuntament de Barcelona (Direcció d'Espais Verds i Biodiversitat Medi Ambient i Serveis Urbans - Ecologia Urbana, Ajuntament de Barcelona, 2021) 2021 Coord. UTM Tabla 1.SI. Data sources for the analysis of food environments in Madrid. Food environment, Madrid Source Year Scale Food retail outlets Censo de Locales y Actividades (Ayuntamiento de Madrid ‒ Datos 09.2023) 2023 Coord. UTM Organic food outlets Censo de Locales y Actividades (Ayuntamiento de Madrid) ‒ Datos 09.2024; Mapeo Alternative Food Networks (MAE ‒ Madrid Agroecológico) 2023 Coord. UTM Cooperatives and consumer groups Mapeo Alternative Food Networks (MAE ‒ Madrid Agroecológico) 2023 Coord. UTM Community urban gardens Ayuntamiento de Madrid 2023 Coord. UTM School gardens Ayuntamiento de Madrid 2023 Coord. UTM Municipal urban gardens Ayuntamiento de Madrid 2023 Coord. UTM Classification algorithms for food retail outlets The classification of food retail outlets is based on an adaptation of previous work by Bilal et al. (2018) and Díez et al. (2019) in Madrid, and by Garcia et al. (2020) and Garcia-Sierra et al. (2021) in Barcelona. A comparable classification has been developed for the two urban foodscapes using the opportunities provided by each city’s census data. The classification algorithms for food retail outlets in both cities are shown in Figure 1A (Madrid) and Figure 2A (Barcelona). The identification of outlets specialising in organic food is based on business codes [In Madrid: ID 472901 = Herbalist. In Barcelona: ID 2003000 = Herbalist], a list of specialist supermarkets and major chains (Garcia-Sierra et al., 2021), and keywords — such as eco, bio, vegan, vegeta, etc. Page 24 of 24 Figure 1.SI. Classification algorithms for food retail outlets in Madrid. Source: Adapted from Bilal et al. (2018) and Díez et al. (2019). Notes: Codes ID 561002 and ID 561007 are categorised as “Takeaway (e.g., small fast-food businesses for takeaway)” or “Fast Food (large chain)” in the first case, and as “Café (e.g., sale of packaged food, hot chocolate, ice cream, etc.)” in the second case. Codes ID 472902 to 472904 (ice cream parlours) are also included in this latter category. Figure 2.SI. Classification algorithms for food retail outlets in Barcelona. Source: Adapted from GarciaSierra et al. (2021). Yes │No 47.11.03? Yes │No Supermarket name? Yes │No 47.21? FRUITS & VEG Yes │No 47.22? BUTCHERIES 47.23? FISHMONGERS No 47.24? BAKERIES Any 47.1 or 47.2 CNAE code Any 47.1 ? UNSPECIALIZED FOOD STORES More than one 47.2? SUPERMARKETS CONVENIENCE STORES SMALL GROCERS SPECIALIZED STORES OTHER SPECIALIZED Codi_Principal_Activitat Codi = 1 'Actiu'? Codi = 1 'Comerç al detall'? Codi_Grup_Activitat Codi = 1 'Quotidià alimentari'? Codi_Activitat_2022 1000020? 1003000? 1002000? 1005000? 1007000? 1006000? Yes 1000030? 1400003? Yes 1000010Takeaway Fast Food FOOD SERVICES Peix i marisc Codi = 2 'Serveis'? Codi_Sector_Activitat Codi_Activitat_2022 Carn i Porc Altres Plats preparats SPECIALIZED STORES UNSPECIALIZED FOOD STORES Autoservei / Supermercat Fruites i verdures Pa, pastisseria i làctics Ous i aus Codi_Grup_Activitat Codi = 14 'Restaurants, bars i hotels (Inclòs hostals, pensions i fondes)'