Slums, social disparities and the COVID-19 pandemic: who were the people truly impacted during the pandemic in Brazil?

Vol.:(0123456789) Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 Discover Social Science and Health Research Slums, social disparities andtheCOVID‑19 pandemic: who were thepeople truly impacted duringthepandemic inBrazil? HeriedersonSávioDiasMoura1· LetíciaPerticarraraFerezin1· FernandaBruzadelliPaulinodaCosta1· RanderJuniorRosa1· MônicaChiodiToscanodeCampos1,2· NatachaMartinsRibeiro1· ThaísZamboniBerra1· TitiladeKehindeAyandeyiTeibo1· JulianaSoaresTenóriodeAraújo1· ArielaFehrTártaro1· RosaMariaPinheiroSouza3· FátimaMorales4· RicardoAlexandreArcêncio1 Received: 2 September 2024 / Accepted: 3 December 2024 © The Author(s) 2024 OPEN Abstract Background Brazil was heavily impacted by COVID-19 pandemic, which intensified social and health disparities, especially in vulnerable areas such as slums. Therefore, the aim of this study was to investigate which residents of slums in Brazil were most affected by COVID-19 pandemic, with a focus on food insecurity, loss of income, ethnic and racial inequalities, and other social and geographic disparities. Methods We conducted a cross-sectional study among the populations of slums in Brazilian capitals and the Federal District by applying a validated instrument. Inferential descriptive statistics were used, with the application of a logistic regression model. Results The study included 1,227 individuals from slums, most of whom were women, black or mixed-race, single, selfemployed, and users of the Brazilian public healthcare system. Slum residents in the North region experienced the highest income loss, while those in the South region had the lowest, though the South also had the highest food insecurity. People with incomes below one minimum wage were the most affected by income loss, while those with incomes above three minimum wages were the least affected. White individuals were less impacted by food insecurity. Government aid was associated with families experiencing both income loss and food insecurity. Conclusion This study highlights the impact of the COVID-19 pandemic among those with lower incomes living in slums, being most of them black or mixed-race women working in informal employment. It also shows that social and structural disparities have been exacerbated by the pandemic, particularly regarding race/ethnicity, income, and geographic region of residence. Keywords COVID-19· Slums· Pandemic· Social vulnerability 1 Background Brazil was heavily impacted by COVID-19 pandemic. However, the impact varied among social groups, being those at greater social disadvantage the most affected [1, 2]. * Heriederson Sávio Dias Moura, [email protected] | 1University ofSão Paulo atRibeirão Preto School ofNursing, SãoPaulo, Brazil. 2University ofBrasília, Brasília, Brazil. 3National School ofPublic Health Sergio Arouca, Oswaldo Cruz Foundation, RiodeJaneiro, Brazil. 4Department ofPreventive Medicine andPublic Health, University ofSeville, Seville, Spain. Vol:.(1234567890) Research Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 During the pandemic, the working class and slum residents were unjustly blamed for the spread of SARS-CoV-2, particularly due to their use of public transportation [3], ignoring their vulnerability and the lack of options for social isolation. While the middle and upper classes could protect themselves, workers remained exposed, reflecting the invisibility of structural inequalities and the precarization of labor [4] perpetuating the idea that the poor were responsible for their own condition. Although the first cases in Brazil emerged among high-income individuals [5], it was the less privileged groups, such as those living in slums, who were most vulnerable to the crisis, highlighting how the lack of access to resources and protection amplified social inequalities. Brazil is among the countries with the highest concentration of slums, being the leading country in Latin America with over 16 million people living in slums, according to estimates from the 2022 Demographic Census by the Brazilian Institute of Geography and Statistics (IBGE—Instituto Brasileiro de Geografia e Estatística) [6]. Slums are the result of a historical and social process, existing since the time of colonization in Brazil, and they took shape vigorously after the arrival of the Portuguese court in Brazil, during a process of urban reorganization, particularly in Rio de Janeiro. The growth of slums intensified with the rise of neoliberalism. They also emerged from the rural exodus of populations to urban areas, leading to economic stagnation or depression, high unemployment, poverty, informal economy, and lack of urban planning [7]. According to IBGE, slums are defined as “popular territories originating from various strategies employed by the population to meet, generally in an autonomous and collective manner, their housing needs and associated uses (commerce, services, leisure, culture, among others), in the face of the insufficiency and inadequacy of public policies and private investments aimed at ensuring the right to the city” [8]. It is noteworthy that the intense and rapid process of “peripherization” that accompanied COVID-19 in Brazil [9], especially in slum areas, resulted in disparities in access to healthcare services, reduction in employment, lack of social protection, and difficulties in accessing to new opportunities for social mobility [10]. The literature indicates that the impacts of the pandemic will be felt for many years, particularly among vulnerable populations, reversing any possibilitpáginy of social mobility by several years [11]. Many studies have been conducted to assess the detrimental impacts of COVID-19 pandemic and disparities in slums, with a focus on Latin American countries such as Colombia [12], as well as countries in Africa [13] and Asia [14, 15] A study conducted in Brazil revealed that there were more deaths in slum areas compared to non-slum areas [16]. Another study conducted in Argentina observed that demographic factors (age and gender), self-reported comorbidities (hypertension, diabetes), obesity, healthcare financing, and living in a slum were associated with a higher likelihood of contracting COVID-19 [17]. Additionally, there is evidence of the interactions of other factors with COVID-19 in these territories [18], which resulted in avoidable deaths. A study conducted in the United States highlighted that during times of crisis, there are significant disruptions in the social protection network, and in this process of exclusion, the historically most excluded and marginalized groups are the most adversely affected. Adding to this process is the issue of race/ethnicity, with black individuals being the most harmed within a socio-historical context of violence, segregation, exclusion, and denial of rights [19]. So far, studies have been evaluating the impact of COVID-19 pandemic in slums in general. However, even within slums, there are groups that are more disadvantaged than others, revealing an important knowledge gap, particularly regarding income loss and food insecurity. Adjacent issues such as food insecurity, income, race/ethnicity, age, gender, and other social and geographic disparities can be quite disparate among groups, underscoring the need for studies focusing on these aspects. Thus, the aim of this study was to investigate which individuals residing in slums in Brazil were most affected by COVID- 19 pandemic, with a focus on food insecurity, loss of income, ethnic and racial inequalities, and other social and geographic disparities. 2 Methods 2.1 Study design This is a cross-sectional [20, 21] study that is part of a larger research project titled “Social Thermometer - COVID-19 in Brazil”. Vol.:(0123456789) Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 Research 2.2 Scenario understudy The study was conducted in the capitals of the following Brazilian states: Acre, Alagoas, Amapá, Bahia, Ceará, Federal District, Espírito Santo, Goiás, Maranhão, Mato Grosso, Mato Grosso do Sul, Pará, Paraíba, Pernambuco, Piauí, Rio de Janeiro, Rio Grande do Norte, Rio Grande do Sul, Rondônia, São Paulo, Sergipe, and Tocantins. 2.3 Population andsamples The study population consisted of residents of slums, either native-born or naturalized Brazilian citizens, aged 18 or older, who had been living in urban areas characterized by inadequate housing and lack of proper infrastructure (such as slums, informal settlements, among others) for at least 6months during the COVID-19 pandemic. The pandemic period was considered to have started on March 11, 2020, when the WHO declared the COVID-19 pandemic [22], and continued until the time of the interview. In addition, participants were required to understand the language spoken in Brazil (Brazilian Portuguese). Given the specific characteristics of the population under analysis, living in a context of inequality and social exclusion, individuals were selected through sequential sampling [23], being included as soon as they were located in the slums and agreed to participate in the study. Additionally, this approach was used to manage samples of varying sizes without the need to maintain a fixed sample size. Nevertheless, to determine the sample size, calculations for finite populations were used, considering a random margin of error of 5%, a confidence level of 95%, a test power of 80%, and a variance of 50%. Additionally, a 10% increase was calculated to account for possible sample losses. The population size was based on data from the 2022 IBGE Demographic Census [6], which estimated that approximately 16.6 million people (8% of the Brazilian population) live in slums in the country. 2.4 Instruments anddata collection The study utilized a questionnaire developed by the National School of Public Health at NOVA University of Lisbon (ENSP-UNL – Escola Nacional de Saúde Pública da Universidade Nova de Lisboa), based on questions from the World Health Organization’s (WHO) [24] Risk Communication and Community Engagement (RCCE) strategy and the 2014 National Health Survey of Portugal [25]. This questionnaire was later validated and published in studies with various objectives and perspectives by ENSP-UNL researchers [26–28]. Notably, in Brazil, the questionnaire underwent cultural adaptation and validation by senior researchers using Delphi technique [29], and was titled “Social Thermometer COVID-19: Social Opinion” (https:// doi. org/https:// doi. or g/ 10. 6084/ m9. figsh are. 25917 571). The interviews were conducted from July 2022 to September 2023. The aforementioned instrument has been used in studies already published in journals with selective editorial policies, which had various objectives, including: analyzing the level of public trust in the Unified Health System [30]; identifying factors associated with changes in mental health status and self-perception during COVID-19 pandemic [31]; assessing public agreement with relaxation measures [32]; food insecurity among migrants and refugees [33]; investigating inequalities related to income loss and food insecurity among homeless individuals during COVID-19 pandemic in Brazil [34]; among others [35]. The instrument was hosted on the REDCap platform [36, 37]. To administer the instrument, a network of contacts was established for participant recruitment, consisting of professionals linked to research institutions, universities, and civil society actors. Additionally, the strategy of mobilization through Social Movements was adopted, which was essential for advancing the sample composition, facilitating partnerships with various civil society actors and scholars specializing in Social Movements and health surveillance in vulnerable territories. This involvement was particularly relevant, as Social Movements play a crucial role in connecting with slum areas and other marginalized populations, easing access to these communities through the strategies developed in these contexts. The instrument was administered by field interviewers associated with the research, using mobile devices (cell phones and/or tablets), with an average administration time of 20–30min. The interviewers were rigorously trained to ensure standardized administration and avoid measurement bias. Although interviewers could be recruited from Social Movements, this was not a requirement of the study. It is important to note that interviewers did not need an internet Vol:.(1234567890) Research Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 connection during the interviews; however, at the end of the process, they were required to go to an internet access point to upload the collected data to the REDCap platform [36, 37], where all research data were stored. Before beginning the questionnaire, the Informed Consent Form (ICF) was presented and read to the participants, with the interview starting only after their agreement and signature on the form. The ICF was structured in two copies, which were fully read and initialed on each page by both the participant and the responsible interviewer. The instrument was applied only once per participant, meaning each person answered the questionnaire just one time. This ensured data consistency, preventing repetition and guaranteeing the validity and reliability of the collected information. In the case of illiterate participants, a fingerprint was collected, and both the participant and the interviewer received a copy of the ICF. The interviewer also advised the participant on the importance of keeping the document. It is worth noting that participation in the research was entirely voluntary. 2.5 Study variables 2.5.1 Dependent variables Table1 presents the dependent variables of this study, as derived from the questionnaire, along with their corresponding response patterns. Additionally, it includes the versions used in the binary logistic regression analyses, which were dichotomized into values of 0 and 1. Individuals who did not respond to the dependent variables were excluded from the binary logistic regression models. 2.5.2 Independent variables Table2 presents the independent variables considered in the study, according to the original response options for the questionnaire question or the listed response category, along with the corresponding dichotomization adopted. 2.6 Statistical analysis Initially, the consistency and standardization analysis of the database was conducted using Microsoft Office Excel 2010. To characterize the study participants, absolute frequency (n) and relative frequency (%) analyses were performed. The data were tabulated in spreadsheets using Microsoft Office Excel 2010 and subsequently imported for analysis into RStudio, version 4.1.1 [38]. The normality of the distribution of quantitative variables was assessed using the Shapiro-Wilk test, considering that variables with a p > 0.05 were assumed to have a normal distribution. Next, five variables of interest were selected for the construction of the maps: employment/occupation status (formal or informal) (yes or no), use of the SUS (yes or no), financial difficulty in acquiring food due to COVID-19 pandemic (yes or no), receipt of government aid (yes or no), and temporary loss of income due to COVID-19 pandemic (yes or no). The dichotomized responses were grouped according to the macroregion (North, Northeast, South, Southeast, or Central-West) of the participants’ residence, in order to visualize in which Brazilian macroregions populations residing in slums and urban communities experienced greater social inequality. The variables were organized in spreadsheets using Microsoft Office Excel 2010, and then the maps were created with the information using ArcGIS 10.5, employing shapefiles provided by IBGE [39]. Two binary logistic regressions were conducted, with the independent variables of the final model being selected in two stages. In the first stage, statistically associated variables with the outcomes were identified through univariate analyses using the chi-square test, considering a significance level of p ≤ 0.20 [40, 41] for inclusion in multivariate models, as well as those theoretically important. The crude Odds Ratios (OR) were calculated along with their respective 95% confidence intervals (95% CI). In the second stage, multicollinearity was assessed to avoid including correlated variables in the model. The presence of multicollinearity was tested using the variance inflation factor (VIF), one of the most commonly used measures, defined by the following expression: VIF j= 1 1−R 2 j Vol.:(0123456789) Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 Research Table 1 Operationalization of the dependent variable under study Dependent Variables Original response options from questionnaire Dichotomization Financial difficulty in acquiring food during to the COVID-19 pandemic I. Had financial difficulty before the pandemic and no longer do II. Did not have financial difficulty before the pandemic and now do III. Always had financial difficulty IV. Never had financial difficulty V. Prefer not to answer 0 = had financial difficulty before the pandemic and no longer do or never had financial difficulty in acquiring food 1 = did not have financial difficulty before the pandemic and now do or always had financial difficulty in acquiring food Temporary income loss due to the COVID-19 pandemic I. Did not lose II. Partially lost III. Totally lost IV. Had no income V. Prefer not to answer 0 = did not lose or had no income 1 = partially lost or totally lost income Vol:.(1234567890) Research Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 Table 2 Operationalization of the study’s independent variables Independent variables Original answer options for the question or answer category Dichotomization Age 18–29years old 0 = No 1 = Yes 30–59years old 0 = No 1 = Yes 60years or older 0 = No 1 = Yes Sex/gender Male Female Other 0 = Female 1 = Male Race/color/ethnicity White 0 = No 1 = Yes Black/mixed race 0 = No 1 = Yes Asian 0 = No 1 = Yes Indigenous 0 = No 1 = Yes No race/color/ethnicity declared 0 = No 1 = Yes Marital status Married or in a stable union 0 = No 1 = Yes Divorced or legally separated 0 = No 1 = Yes Single 0 = No 1 = Yes Widowed 0 = No 1 = Yes Employment/occupation Student Public employee Private sector employee Self-employed Microentrepreneur Business owner Family farmer Domestic worker Retired Unemployed Informal work/casual work Other 0 = Does not have an employment/occupation (unemployed) 1 = Has some type of employment/occupation Vol.:(0123456789) Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 Research Table 2 (continued) Independent variables Original answer options for the question or answer category Dichotomization Main form of housing Own Provided Rented Leased Street Shelter Other 0 = Does not have housing (Street) 1 = Has some type of housing Education No schooling 0 = No 1 = Yes Incomplete elementary education to complete elementary education 0 = Não 1 = Yes Incomplete secondary education to complete secondary education 0 = No 1 = Yes Higher education incomplete or more 0 = No 1 = Yes Monthly family income No income 0 = No 1 = Yes Less than one minimum wage 0 = No 1 = Yes One to three minimum wages 0 = No 1 = Yes Above three minimum wages 0 = No 1 = Yes Government aid Pandemic emergency aid Family Grant Continuous Benefit Do not receive government aid Other 0 = Did not receive any type of government aid 1 = Received some type of government aid Use of the Unified Health System (SUS— Sistema Único de Saúde)Uses the SUS Does not use SUS 0 = No 1 = Yes Visit from Community Health Agent (ACS— Agente Comunitário de Saúde)Receives visit from ACS Does not receive visit from ACS 0 = No 1 = Yes Vol:.(1234567890) Research Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 where R2 j is the coefficient of multiple correlation resulting from the regression of Xj on the other p – 1 regressors. The higher the degree of dependence of Xj on the remaining regressors, the stronger the dependence and the higher the value of R2 j . A cutoff value of VIF > 10 was adopted [42]. The selection of independent variables for the final model was performed using the Backward stepwise selection method, starting with a full model (including all independent variables with VIF < 10) and sequentially removing variables one by one while monitoring the model’s performance. The best model was selected based on the lowest Akaike Information Criterion (AIC) value [43]. For the final model, adjusted Odds Ratios (aOR) were calculated along with their respective 95% CI [44]. After selecting the final model based on the lowest AIC value, the Hosmer–Lemeshow test was applied to validate the model and assess the reliability and accuracy of the predictions generated by the binary logistic regression analysis. Additionally, the likelihood ratio test was used to evaluate the model’s adequacy, identify significant variables, validate the statistical significance of the results, compare different models, and provide a more robust interpretation of the statistical analysis results. This test compared the null model (reference) with the full model and also with the final model. Furthermore, the predictive ability and accuracy of the models were evaluated by the area under the ROC (Receiver Operating Characteristic) curve, along with their respective 95% CI [44]. All analysis and validation tests were performed using RStudio, version 4.1.1 [38]. 3 Results At the end of the study, 1,309 individuals were recruited, but after applying the inclusion and exclusion criteria, only 1,227 participants met the requirements and continued in the study (Table3). The response rate was 98.09% for the variable “Financial difficulty in acquiring food during the COVID-19 pandemic” and 94.27% for “Temporary income loss due to the COVID-19 pandemic”. Table3 presents the socioeconomic characteristics of the slum populations included in the study. The median age of the participants was 39years (minimum = 18 and maximum = 85), with the majority being women (n = 730; 59.5%), black or mixed-race (n = 897; 73.1%), single (n = 560; 45.6%), self-employed (n = 273; 22.2%), with their own housing (n = 769; 62.7%), having completed secondary education (n = 388; 31.6%), with a family monthly income of one to two minimum wages (n = 481; 39.2%), not receiving any type of government aid (n = 781; 63.7%), using the SUS (Unified Health System) (n = 1,169; 95.3%), and receiving visits from community health agents (n = 644; 52.5%). Most individuals reported never having financial difficulty in acquiring food (n = 491; 40%) and did not lose income due to the COVID-19 pandemic (n = 467; 38.1%). Figure1 illustrates the main disparities observed among the people living in slums in the five Brazilian macro-regions. There are differences between the regions regarding employment, whether formal or informal. The Southeast macroregion has the highest number of employed people (96%), while the Midwest region has the lowest proportion (75%). Regarding the use of the SUS, most residents, regardless of the macro-region, depend on the SUS. One notable difference between the macro-regions was the difficulty in obtaining food during the pandemic. Residents in the South reported the greatest difficulties (59.52%), whereas in the Southeast, this percentage was the lowest among the macro-regions (19.8%). Regarding social benefits, the North (42.69%) and Midwest (50%) macro-regions reported receiving government aid the most; residents of the Southeast reported the least aid (14.57%). Residents of the Northeast were the most impacted by income loss (57.72%), while those in the South were the least affected (32.65%). The individuals that experienced greater financial difficulty in acquiring food during COVID-19 pandemic were (Table4): (i) aged 30–59 (aOR = 1.41; 95% CI= 1.06–1.89), (ii) single (aOR = 1.54; 95% CI= 1.16–2.05), and (iii) those who received some form of government aid (aOR = 1.45; 95% CI= 1.09–1.94), while the ones that appeared to be more protected were (Table4): (i) white individuals (aOR = 0.28; 95% CI= 0.11–0.71), (ii) with some form of employment/occupation (aOR = 0.65; 95% CI= 0.45–0.91), (iii) with incomplete higher education or more (aOR = 0.30; 95% CI= 0.19–0.47), (iv) with a monthly family income of one to three minimum wages (aOR = 0.34; 95% CI= 0.25–0.45), and v) with a monthly family income above three minimum wages (aOR = 0.06; 95% CI= 0.01–0.17). For the model validation presented in Table4, the accuracy of the model through the area under the ROC curve was found to be 0.76. The Hosmer–Lemeshow test (p = 0.30) and likelihood ratio test (p < 0.01) were also evaluated. Vol.:(0123456789) Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 Research The individuals that appeared to be more associated with income loss due to COVID-19 pandemic (Table5) were: the ones with a monthly family income of less than one minimum wage (aOR = 1.51; 95% CI= 1.16–1.97) and those who received some form of government aid (aOR = 1.38; 95% CI= 1.06–1.79). Conversely, the individuals that appeared to be less associated (Table5) were: (i) aged 60 or older (aOR = 0.54; 95% CI= 0.37–0.79), (ii) widowed individuals (aOR = 0.53; 95% CI= 0.29–0.94), (iii) those with incomplete primary education or complete primary education (aOR = 0.62; 95% CI= 0.47–0.82), and (iv) those with a monthly family income above three minimum wages (aOR = 0.46; 95% CI= 0.27–0.77). For the model validation presented in Table5, the accuracy of the model through the area under the ROC curve was found to be 0.62. The Hosmer–Lemeshow test (p = 0.90) and likelihood ratio test (p < 0.01) were also evaluated. 4 Discussion The study aimed to investigate which residents of slums in Brazil were most affected by the COVID-19 pandemic, with a focus on food insecurity, loss of income, ethnic and racial inequalities, and other social and geographic disparities. The effects of intersectionality (gender, race/ethnicity and class) on the vulnerability of black and mixed-race women living in slums in Brazil are revealed with the socioeconomic data from this study. The majority of participants in this study were women, black or mixed-race, single, self-employed, with completed secondary education, a monthly family income of one to two minimum wages, and did not receive any form of government aid. According to the Brazilian Institute of Social and Economic Analysis (IBASE – Instituto Brasileiro de Análises Sociais e Econômicas) [45], approximately 67% of individuals living in slums are black, and among the 6.3 million Brazilian women living in these areas, 69% are black. Thus, in demographic terms, it is legitimate to assert that slums in Brazil are predominantly black, with a notable presence of black women, who suffer from the precariousness of their living conditions due to the historically entrenched structural vulnerabilities in the country [46–48]. We should highlight the concerning scenario, based on the data from this study, regarding the high absence of social protection, food insecurity, and income loss due to COVID-19 pandemic among slum populations in Brazil, particularly in the North and Northeast regions. These areas have experienced an increase in social vulnerability. As evidenced by the suspension of the Bolsa Família program benefits for 158,452 people in March 2020, the effects were disproportionately felt, especially in the Northeast region, which already suffers from a high regional unemployment rate (17.9%) and a significant proportion of families living in extreme poverty (51%). Consequently, this situation exacerbates the rise in hunger related to poverty [49–51]. Furthermore, the finding that individuals in the South region of Brazil faced greater financial difficulties in acquiring food while reporting lower income loss during COVID-19 pandemic reveals the complexity of the economic consequences of the pandemic in slums of the Global South [12]. Specifically, in Brazil, when looking at the favelas of the Southern macro-region, this disparity can be attributed to the fact that metropolitan areas, such as Porto Alegre [52], the capital of Rio Grande do Sul, where the research took place, have peripheral slum sectors with high socioeconomic inequality and various social and environmental problems. This significantly impacts the ability of families to secure income and food security. In our study, individuals living in slums, aged between 30 and 59years, experienced greater financial difficulties in acquiring food during the pandemic. We should take into consideration individuals of this age range living in slums in Brazil who, while often considered economically active, faced distinct challenges in terms of access to essential services, as well as social, economic, and health impacts. Macchia [17] discovered that younger individuals (35.6 ± 20.8years) living in slums were more hospitalized than those living outside slums (50.0 ± 23.9years). Such evidence supports our findings and reveals an age group that, despite being composed of adults in the productive phase of their careers, was impacted in various aspects, including income loss and food insecurity. Financial difficulty in acquiring food among single individuals can be attributed to various complex factors, with one of the main ones being the lack of a financial support network during the pandemic. Having the support of a partner or family [53] can represent a significant advantage in purchasing food and managing household expenses. For people living alone, the proportional costs of housing, transportation, and other basic expenses can consume a considerable portion of available income, leaving limited resources for adequate nutrition. Additionally, job instability and the often informal nature of work [12, 15] among single individuals living in slums can further exacerbate food insecurity, making them more vulnerable to economic crises, such as those caused by COVID-19 pandemic. Vol:.(1234567890) Research Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 Author contributions H.S.D.M., R.J.R., T.Z.B., J.S.T.A., A.F.T., R.M.P.S. and R.A.A. participated in data collection. H.S.D.M., L.P.F., F.B.P., T.Z.B. and R.A.A. participated in data analysis and interpretation. H.S.D.M., L.P.F., F.B.P.C., R.J.R., M.C.T.C., N.M.R., T.Z.B., J.S.T.A., T.K.A.T., A.F.T., R.M.P.S., F.M. and R.A.A. contributed to the critical revision of the manuscript and the final approval of the version to be submitted. Funding This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001; CAPES:process: 88887.657730/2021-00; the São Paulo Research Foundation (FAPESP): process: 2021/08263-7; and the National Council for Scientific and Technological Development (CNPq): process: 405902/2021-2. Data availability The data used for this research is available upon reasonable request to the authors. Requests for access to the data can be sent to [email protected]. Declarations Ethics approval and consent to participate The research was approved by the Research Ethics Committee (CEP – Comitê de Ética em Pesquisa) of the School of Nursing of Ribeirão Preto at the University of São Paulo (EERP-USP – Escola de Enfermagem de Ribeirão Preto da Universidade de São Paulo), with the CAAE number: 57933622.4.1001.5393 (Opinion: 5.512.199). Before starting the administration of the questionnaire, the Informed Consent Form (ICF) was read to the study participants, and only after their agreement and signature was the interview initiated. Informed consent was obtained from all individual participants included in the study. Competing interests The authors declare no competing interests. Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ bync- nd/4. 0/. References 1. Estrada CD, Nóbrega L. Covid-19: Boletim indica um Brasil desigual frente à pandemia. Boletim do Observatório Covid-19 Fiocruz. Agência Fiocruz de Notícias. Fiocruz. 2022. https:// portal. fiocr uz. br/ notic ia/ co vid- 19- bolet imindica- umbr asildesig ualfren tepande mia. Accessed 18 Jul 2024. 2. World Bank Group. Pobreza e desigualdade no Brasil: pandemia complica velhos problemas e gera novos desafios para população vulnerável. World Bank. 2022. https:// www. world bank. org/ pt/ news/ pressrelea se/ 2022/ 07/ 14/ pobre za- e- desig ualda deno- brasilpande miacompl icavelhos- probl emas-e- geranovos- desafi ospara- popul acaovulne ravel. Accessed 18 Jul 2024. 3. de Moreira MF, Meirelles LC, Cunha LAM. Covid-19 no ambiente de trabalho e suas consequências à saúde dos trabalhadores. Saúde Debate. 2022;45:107–22. 4. Lacerda Neto MS, Guimaraes Junior SD, Monteagudo PM, Ferreira JB. Racismo, trabalho e psicologia: provocações ético-políticas à luz da pandemia pela COVID-19. Psicol Am Lat. 2021;35:105–12. 5. de Oliveira WK, Duarte E, de França GVA, Garcia LP. Como o Brasil pode deter a COVID-19. Epidemiol Serv Saúde. 2020;29:e2020044. 6. Confederação Nacional de Municípios. IBGE anuncia retorno da utilização do termo favela no censo demográfico. Agência CNM de Notícias. 2024https:// cnm. org. br/ comun icacao/ notic ias/ ibgeanunc iaretor noda- utili zacaodo- termofavela- nocenso- demog rafico. Accessed 18 Jul 2024. 7. Magalhães JCR. Histórico das favelas na cidade do Rio de Janeiro. Revista Desafios do Desenvolvimento. 2010 Nov 19. https:// www. ipea. gov. br/ desafi os/ index. php? option= com_ conte nt& view= artic le& id= 1111: catid= 28& Itemid= 23. Accessed 18 Jul 2024. 8. Instituto Brasileiro de Geografia e Estatística. Favelas e Comunidades Urbanas. Notas metodológicas n. 01. Sobre a mudança de Aglomerados Subnormais para Favelas e Comunidades Urbanas. Ministério do Planejamento e Orçamento. Instituto Brasileiro de Geografia e Estatística - IBGE. Diretoria de Geociências. Coordenação de Geografia. Instituto Brasileiro de Geografia e Estatística. 2024. https:// bibli oteca. ibge. gov. br/ index. php/ bibli otecacatal ogo? view= detal hes& id= 21020 62. Accessed 18 Jul 2024. 9. Chioro A, Calife K, dos Barros CR, S, Martins LC, Calvo M, Stanislau E, etal. Covid-19 em uma Região Metropolitana: vulnerabilidade social e políticas públicas em contextos de desigualdades. Saúde Debate. 2021;44:219–31. 10. Hill RV, Narayan A. COVID-19 and inequality: a review of the evidence on likely impact and policy options. Working paper, Centre for Disaster Protection, London. 2020. https:// www . disas terpr otect ion. org/ publi catio nscentre/ c ovid- 19- andinequ ality-a- review- of- theevide nceon- likelyimpact- andpolicy- optio ns. Accessed 18 Jul 2024. 11. Antunes JLF. Condições socioeconômicas em saúde: discussão de dois paradigmas. Rev Saúde Pública. 2008;42:562–7. 12. Molina-Betancur JC, Martínez-Herrera E, Pericàs JM, Benach J. Coronavirus disease 2019 and slums in the Global South: lessons from Medellín (Colombia). Glob Health Promot. 2021;28:65–9. 13. Kibe PM, Kisia L, Bakibinga P. COVID-19 and community healthcare: perspectives from Nairobi’s informal settlements. Pan Afr Med J. 2020;35:106. Vol.:(0123456789) Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 Research 14. Auerbach AM, Thachil T. How does Covid-19 affect urban slums? Evidence from settlement leaders in India. World Dev. 2021;140:105304. 15. Wirastri MV, Morrison N, Paine G. The connection between slums and COVID-19 cases in Jakarta, Indonesia: A case study of Kapuk Urban Village. Habitat Int. 2023;134:102765. 16. Brotherhood L, Cavalcanti T, Da Mata D, Santos C. Slums and pandemics. J Dev Econ. 2022;157:102882. 17. Macchia A, Ferrante D, Battistella G, Mariani J, de Quirós FGB. COVID-19 among the inhabitants of the slums in the city of Buenos Aires: a population-based study. BMJ Open. 2021;11:e044592. 18. Carbajo AE, Cardo MV, Pesce M, Iummato LE, Bárcena Barbeira P, Santini MS, etal. Age and socio-economic status affect dengue and COVID-19 incidence: spatio-temporal analysis of the 2020 syndemic in Buenos Aires City. PeerJ. 2023;11:e14735. 19. Gauthier GR, Smith JA, García C, Garcia MA, Thomas PA. Exacerbating inequalities: social networks, racial/ethnic disparities, and the COVID-19 pandemic in the United States. J Gerontol B Psychol Sci Soc Sci. 2021;76:e88-92. 20. Rothman KJ, Greenland S, Lash TL. Modern epidemiology. 3rd ed., thoroughly rev. and updated. Philadelphia: Wolters Kluwer Health/Lippincott Williams & Wilkins. 2008. https:// www. acade mia. edu/ 37995 138/ Epide miolo gia_ Moder na_ 3a_ Edi% C3% A7% C3% A3o. Accessed 23 Mar 2023. 21. Romanowski FN de A, Castro MB de, Neris NW. Manual de Tipos de Estudo. Centro Universitário de Anápolis. Pró-Reitoria de Pós-Graduação, Pesquisa, Extensão e Ação Comunitária. Programa de Pós-Graduação em Odontologia. 2019. http:// repos itorio. aee. edu. br/ bitst ream/ aee/ 15586/1/ MANUAL% 20DE% 20TIP OS% 20DE% 20EST UDO. pdf. Accessed 1 Mar 2024. 22. Organização Pan-Americana da Saúde. OMS declara emergência de saúde pública de importância internacional por surto de novo coronavírus. 2020. https:// www. paho. org/ pt/ news/ 30-1- 2020- whodecla respublic- healthemerg encynovel- coron avirus. Accessed 18 July 2024. 23. Etikan I. Comparison of convenience sampling and purposive sampling. Am J Theor Appl Stat. 2016;5:1. 24. World Health Organization. Risk Communication and Community Engagement (RCCE) Action Plan Guidance COVID-19 Preparedness and Response. 2020. https:// www. who. int/ publi catio ns/i/ item/ riskcommu nicat ion- and- commu nityengag ement- (rcce)- actionplan- guida nce. Accessed 18 Jul 2024. 25. Instituto Nacional de Estatística. Inquérito Nacional de Saúde - 2014. Lisboa, Portugal. 2016 Jun. https:// www . ine. pt/ xport al/ xmain? xpid= INE& xpgid= ine_ publi cacoe s& PUBLI CACOE Spub_ boui= 26371 4091& PUBLI CACOE Smodo= 2& xlang= pt. Accessed 18 July 2024. 26. Laires PA, Dias S, Gama A, Moniz M, Pedro AR, Soares P, etal. The association between chronic disease and serious COVID-19 outcomes and its influence on risk perception: survey study and database analysis. JMIR Public Health Surveill. 2021;7:e22794. 27. Pedro AR, Gama A, Soares P, Moniz M, Laires PA, Dias S. COVID-19 barometer: social opinion – what do the Portuguese think in this time of COVID-19? PJP. 2020;38:42–50. 28. Soares P, Rocha JV, Moniz M, Gama A, Laires PA, Pedro AR, etal. Factors associated with COVID-19 vaccine hesitancy. Vaccines (Basel). 2021;9:300. 29. Zarili TFT, Castanheira ERL, Nunes LO, Sanine PR, Carrapato JFL, Machado DF, etal. Técnica Delphi no processo de validação do Questionário de Avaliação da Atenção Básica (QualiAB) para aplicação nacional. Saude soc. 2021;30:e190505. 30. de Araújo JST, Delpino FM, Berra TZ, Moura HSD, Ramos ACV, do Nascimento MC, etal. Level of trust of the population in the unified health system in the midst of the COVID-19 crisis in Brazil. Int J Environ Res Public Health. 2022;19:14999. 31. Rosa RJ, de Araújo JST, Berra TZ, Ramos ACV, Moura HSD, do Nascimento MC, etal. Factors associated with status and self-perceived mental health changes in the face of the COVID-19 pandemic in Brazil. PLOS Glob Public Health. 2023;3:e0001636. 32. Silva RV dos S, Moura HSD, Delpino FM, Nascimento MC do, Ramos ACV, Berra TZ, etal. Concordância às medidas de flexibilização durante a pandemia de covid-19 no Brasil. Ciência, Cuidado e Saúde. 2023. https:// perio dicos. uem. br/ ojs/ index. php/ Cienc CuidS aude/ artic le/ view/ 66049. Accessed 18 Jul 2024. 33. Moura HSD, Berra TZ, Rosa RJ, dos Silva RVS, de Soares DA, de Araújo JST, etal. Health condition, income loss, food insecurity and other social inequities among migrants and refugees during the COVID-19 pandemic in Brazil. BMC Public Health. 2023;23:1728. 34. Moura HSD, Canatto MCD, Berra TZ, dos Silva RVS, Rosa RJ, de Campos MCT, etal. Food insecurity, income loss, healthcare access, and other exacerbated social inequalities among people experiencing homelessness during the COVID-19 pandemic in Brazil (2021–2023). Discov Public Health. 2024;21:17. 35. Gil JDC, Aguiar PMV, Azeredo-Lopes S, Soares P, Moura HSD, Arcêncio RA, etal. Physical Distancing and Mental Well-Being in Youth Population of Portugal and Brazil during the COVID-19 Pandemic. Portuguese J Public Health. 2022;40:91–100. 36. Harris PA, Taylor R, Minor BL, Elliott V, Fernandez M, O’Neal L, etal. The REDCap consortium: Building an international community of software platform partners. J Biomed Inform. 2019;95:103208. 37. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–81. 38. R Core Team. R: A language and environment for statistical computing. 2021. https:// www.rproje ct. org/ about. html. Accessed 16 Jan 2022. 39. Instituto Brasileiro de Geografia e Estatística. Malha Municipal. IBGE. 2022. https:// www. ibge. gov. br/ geoci encias/ organ izacaodo- terri torio/ malhasterri toria is/ 15774- malhas. html. Accessed 15 Mar 2023. 40. Chowdhury MZI, Turin TC. Variable selection strategies and its importance in clinical prediction modelling. Fam Med Community Health. 2020;8: e000262. 41. Hosmer Jr DW, Lemeshow S, Sturdivant RX. Applied logistic regression. John Wiley & Sons; 2013. 42. Zuur AF, Ieno EN, Elphick CS. A protocol for data exploration to avoid common statistical problems. Methods Ecol Evol. 2010;1:3–14. 43. Wagh YS, Kamalja KK. Zero-inflated models and estimation in zero-inflated Poisson distribution. Commun Stat Simul Comput. 2018;47:2248–65. 44. Šimundić A-M. Measures of diagnostic accuracy: basic definitions. EJIFCC. 2009;19:203–11. 45. Motta A, Brandão RC. Favelas – uma condição urbana de caráter nacional – Ibase. 2022. https:// ibase. br/ favel asuma- condi caourbana- decarat ernacio nal/. Accessed 18 July 2024. 46. Madeira Z, de Gomes DDO. Persistentes desigualdades raciais e resistências negras no Brasil contemporâneo. Serv Soc Soc. 2018. https:// doi. org/ 10. 1590/ 0101- 6628. 154. Vol:.(1234567890) Research Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 47. Passos L, Souza L. Vulnerabilidades cruzadas: as mulheres e suas experiências diversificadas. Rev katálysis. 2021;24:198–209. 48. Santos GC, Brisola EBV, Moreira D, Tostes GW, Cury VE. Impacto do Racismo nas Vivências de Mulheres Negras Brasileiras: Um Estudo Fenomenológico. Psicol cienc prof. 2023;43:e249674. 49. de Oliveira JTC, de Camargo AM, Machado BOB, de Oliveira AR, Fiates GMR, de Vasconcelos FAG. ”Hunger and rage (and the virus) are human things”: reflections on solidarity in times of Covid-19. Rev Nutr. 2021;34:e200183. 50. de Ribeiro-Silva RC, Pereira M, Campello T, Aragão É, de Guimarães JMM, Ferreira AJ, etal. Implicações da pandemia COVID-19 para a segurança alimentar e nutricional no Brasil. Ciênc Saúde Coletiva. 2020;25:3421–30. 51. de Sousa LRM, Segall-Corrêa AM, Ville AS, Melgar-Quiñonez H. Food security status in times of financial and political crisis in Brazil. Cad Saúde Pública. 2019;35:e00084118. 52. Augustin AC, Soares PRR. Desigualdades intraurbanas e a Covid-19: uma análise do isolamento social no município de Porto Alegre. Cad Metrop. 2021;23:971–92. 53. Joshi A, Arora A, Amadi-Mgbenka C, Mittal N, Sharma S, Malhotra B, etal. Burden of household food insecurity in urban slum settings. PLoS ONE. 2019;14:e0214461. 54. Ministério da Previdência Social. Pensão por morte. Instituto Nacional do Seguro Social - INSS. 2023. https:// www . gov. br/ inss/ ptbr/ direi tos-e- dever es/ penso es/ pensaopor- morte. Accessed 18 Jul 2024. 55. Biggs MA, Becker A, Schroeder R, Kaller S, Scott K, Grossman D, etal. Support for criminalization of self-managed abortion (SMA): A national representative survey. Soc Sci Med. 2024;340:116433. 56. do Gurgel AM, dos Santos CCS, de Alves KPS, de Araujo JM, Leal VS. Estratégias governamentais para a garantia do direito humano à alimentação adequada e saudável no enfrentamento à pandemia de Covid-19 no Brasil. Ciênc Saúde Coletiva. 2020;25:4945–56. 57. Natalino M, Pinheiro MB. Nota Técnica No 67. Proteção social aos mais vulneráveis em contexto de pandemia: algumas limitações práticas do auxílio emergencial e a adequação dos benefícios eventuais como instrumento complementar de política socioassistencial. Instituto de Pesquisa Econômica Aplicada. Diretoria de Estudos e Políticas Sociais. 2021;16. 58. Fleury S, Menezes P. Pandemia nas favelas: entre carências e potências. Saúde debate. 2021;44:267–80. 59. Bambra C, Riordan R, Ford J, Matthews F. The COVID-19 pandemic and health inequalities. J Epidemiol Community Health. 2020;74:964–8. 60. Crepalde MM. Desigualdades socioespaciais e a dinâmica da COVID-19 no município do Rio de Janeiro, no período de 2020 - 2022. Socio-spatial inequalities and the dynamics of COVID-19 in the municipality of Rio de Janeiro, during the Period from 2020 to 2022. 2024. https:// www. arca. fiocr uz. br/ handle/ icict/ 63583. Accessed 18 Jul 2024. 61. Guarnieri FV, Melo-Silva LL. Cotas Universitárias no Brasil: Análise de uma década de produção científica. Psicologia Escolar e Educacional [online]. 2017;21:183–93. 62. Batista WM, Santos JCS, dos Santos LCN, da Silva ALR. Sistema de justiça criminal brasileiro e o racismo institucional: racialização e criminalização da população negra. Revista Brasileira de Sociologia do Direito. 2022;9:93–119. 63. Maas NM, Mendoza-Sassi RA, Meucci RD, Cesar JA. Insegurança Alimentar em famílias de área rural do extremo sul do Brasil. Ciênc saúde coletiva. 2020;25:2605–14. 64. de Morais DC, Dutra LV, do Franceschini SCC, Priore SE. Insegurança alimentar e indicadores antropométricos, dietéticos e sociais em estudos brasileiros: uma revisão sistemática. Ciênc Saúde Coletiva. 2014;19:1475–88. 65. Trivellato PT, de Morais DC, Lopes SO, da Miguel ES, do Franceschini SCC, Priore SE. Insegurança alimentar e nutricional em famílias do meio rural brasileiro: revisão sistemática. Ciênc Saúde Coletiva. 2019;24:865–74. 66. dos Santos TG, da Silveira JAC, Longo-Silva G, Ramires EKNM, de Menezes RCE. Tendência e fatores associados à insegurança alimentar no Brasil: Pesquisa Nacional por Amostra de Domicílios 2004, 2009 e 2013. Cad Saúde Pública. 2018;34:e00066917. 67. dos Santos LP, Schäfer AA, de Meller FO, Harter J, Nunes BP, da Silva ICM, etal. Tendências e desigualdades na insegurança alimentar durante a pandemia de COVID-19: resultados de quatro inquéritos epidemiológicos seriados. Cad Saúde Pública. 2021;37:e00268520. 68. Panigassi G, Segall-Corrêa AM, Marin-León L, Pérez-Escamilla R, de Sampaio MFA, Maranha LK. Insegurança alimentar como indicador de iniqüidade: análise de inquérito populacional. Cad Saúde Pública. 2008;24:2376–84. 69. de Souza NN, de Dias MM, Sperandio N, do Franceschini SCC, Priore SE. Perfil socioeconômico e insegurança alimentar e nutricional de famílias beneficiárias do Programa Bolsa Família no município de Viçosa, Estado de Minas Gerais, Brasil, em 2011: um estudo epidemiológico transversal. Epidemiologia e Serviços de Saúde. 2012;21:655–62. 70. da Guerra LDS, Espinosa MM, Bezerra ACD, Guimarães LV, Lima-Lopes MA. Insegurança alimentar em domicílios com adolescentes da Amazônia Legal Brasileira: prevalência e fatores associados. Cad Saúde Pública. 2013;29:335–48. 71. Álvares L, Amaral TF. Food insecurity and associated factors in the Portuguese population. Food Nutr Bull. 2014;35:395–402. 72. Maia I, Monjardino T, Frias B, Canhão H, Cunha Branco J, Lucas R, etal. Food insecurity in Portugal among middle- and older-aged adults at a time of economic crisis recovery: prevalence and determinants. Food Nutr Bull. 2019;40:504–13. 73. Maia I, Monjardino T, Lucas R, Ramos E, Santos AC. Household food insecurity and socio-demographic determinants in young adults: findings from a Portuguese population-based sample. Int J Public Health. 2019;64:887–95. 74. Mainardes F, Raiher AP. (In) Segurança Alimentar no Brasil: Prevalência e Fatores Associados. Cadernos de Ciências Sociais Aplicadas. 2018. https:// doi. org/ 10. 22481/ ccsa. v15i25. 3978. 75. Qian Y, Fan W. Who loses income during the COVID-19 outbreak? Evidence from China. Research in Social Stratification and Mobility. 2020;68: 100522. 76. Syafiq A, Fikawati S, Gemily SC. Household food security during the COVID-19 pandemic in urban and semi-urban areas in Indonesia. J Health Popul Nutr. 2022;41:4. 77. Ukegbu P, Nwofia B, Ndudiri U, Uwakwe N, Uwaegbute A. Food insecurity and associated factors among university students. Food Nutr Bull. 2019;40:271–81. 78. Alie S, Sulaiman N, Mohd Nor F, Mesbah SF. Demographic factors, food security, health-related quality of life and body weight status of adolescents in rural area in Mentakab, Pahang, Malaysia. Mal J Nutr. 2019;25:297–307. 79. Almeida JA, Santos AS, Nascimento MADO, Oliveira JVC, Silva DGD, Mendes-Netto RS. Factors associated with food insecurity risk and nutrition in rural settlements of families. Ciênc Saúde Coletiva. 2017;22:479–88. Vol.:(0123456789) Discover Social Science and Health (2024) 4:75 | https://doi.org/10.1007/s44155-024-00137-6 Research 80. Andreyeva T, Marple K, Moore TE, Powell LM. Evaluation of economic and health outcomes associated with food taxes and subsidies. JAMA Netw Open. 2022;5:e2214371. 81. Martins APB, Canella DS, Baraldi LG, Monteiro CA. Cash transfer in Brazil and nutritional outcomes: a systematic review. Rev Saude Publica. 2013;47:1159–71. 82. Sukartini NM, Purwono R, Surjaningrum ER, Win T, Wangge G, Chrisnahutama A, etal. Do tax and subsidy on unhealthy food induce consumer consumption for healthy food? Evidence from experiment in Surabaya, Indonesia. J Public Health Res. 2023;12:22799036221147360. 83. Sott MK, Bender MS, da Silva Baum K. Covid-19 outbreak in Brazil: health, social, political, and economic implications. Int J Health Serv. 2022;52:442–54. 84. Aquino EML, Silveira IH, Pescarini JM, Aquino R, de Souza-Filho JA, Rocha AS, etal. Medidas de distanciamento social no controle da pandemia de COVID-19: potenciais impactos e desafios no Brasil. Ciênc Saúde Coletiva. 2020;25:2423–46. 85. Scherf EL, da Silva MVV, Fachini JS. The management (or lack thereof) of COVID-19 in Brazil: implications for human rights and public health. Int J Human Rights Healthc. 2021;14:158–74. 86. Marins MT, Rodrigues MN, da Silva JML, da Silva KCM, Carvalho PL. Auxílio Emergencial em tempos de pandemia. Soc estado. 2021;36:669–92. 87. Sott MK, Bender MS, Baum KS. Covid-19 outbreak in Brazil: health, social, political, and economic implications. Int J Health Serv. 2022;52:442–54. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.