Food insecurity, a global issue extending beyond mere hunger to encompass nutritional deficiencies and access uncertainties, has been exacerbated by the COVID-19 pandemic. The Food and Agriculture Organization of the United Nations (FAO) reported a significant increase in global food insecurity in 2020. Brazil, experiencing rising food insecurity since 2013, witnessed a sharp increase during the pandemic, particularly impacting vulnerable populations. The economic crisis triggered by the pandemic has raised concerns about its long-term effects on social and health issues, including mental health problems like stress, anxiety, and depression. Studies have increasingly demonstrated a link between food insecurity and mental health disorders, highlighting the importance of investigating these factors' interplay during and after crises. This study aimed to assess the prevalence of household food insecurity before and during the COVID-19 pandemic and its association with perceived stress among adults in Criciúma, Southern Brazil, a city with sociodemographic characteristics better than the national average but still exhibiting disparities.

Existing literature extensively documents the global prevalence of food insecurity and its persistent nature. The FAO's estimates highlight the dramatic increase in individuals lacking adequate food access in 2020. Similarly, Brazil has shown a growing trend in household food insecurity since 2013, despite prior progress. The COVID-19 pandemic significantly worsened this situation, with national surveys revealing a substantial increase in food insecurity across the country. Regional disparities exist, with Southern Brazil showing lower rates than other regions but still experiencing increases. The economic ramifications of the pandemic triggered widespread discussion about its repercussions on various social and health issues, including a rise in food insecurity rates globally and particularly in developing countries like Brazil. Concurrent with the increase in food insecurity, several studies reported a rise in mental health issues, such as stress, anxiety, and depression, often associated with the insecurity and fear generated during the pandemic. A growing body of research demonstrates a clear association between food insecurity, depression, and stress, with one study linking even mild food insecurity to increased mental health disorders and reduced well-being across diverse income levels and countries. This evidence underscores the importance of understanding the combined influence of food insecurity and mental health problems.

This study utilized data from two population-based studies conducted in Criciúma, Santa Catarina, Southern Brazil: "Health of the Criciúma Population" (2019, pre-pandemic) and "Mental COVID" (October 2020-January 2021, during the pandemic). Both studies employed a two-stage sampling process, randomly selecting census tracts and households within those tracts. The sample size for each study was calculated to ensure sufficient power and accuracy. The Health of the Criciúma Population study used the complete Brazilian Food Insecurity Scale (EBIA), a 14-question scale measuring food insecurity over a three-month period. This allowed for classification into four levels of food insecurity: food security, mild, moderate, and severe. The Mental COVID study employed the short-form EBIA due to time constraints imposed by the pandemic situation. This shorter version identifies food insecurity but doesn't distinguish between the degrees of severity. The Perceived Stress Scale (PSS), validated for the Brazilian population, measured perceived stress levels over the previous month. The PSS uses a Likert-type scale and categorizes individuals based on quintiles, with the highest quintile classified as having perceived stress. Sociodemographic and behavioral characteristics, including sex, age, skin color, schooling, income, employment, marital status, household crowding, overweight (BMI ≥25kg/m² for adults and ≥27kg/m² for older adults), and diet quality (based on a food-frequency questionnaire), were included as potential confounders. Statistical analysis involved descriptive statistics, chi-square tests to compare groups, and Poisson regression with robust variance to assess the association between food insecurity and perceived stress, controlling for confounders using a three-level hierarchical model. All analyses accounted for the complex sampling design using the svy prefix in Stata version 16.1. Ethical approvals were obtained before data collection.

A total of 1683 adults participated (820 in 2019 and 863 in 2020-2021). In 2019, household food insecurity prevalence was 25.8% (95%CI: 22.3; 29.6), decreasing slightly to 21.6% (95%CI: 18.9; 24.4) in 2020. Perceived stress prevalence was approximately 38% in both years. Before the pandemic, food insecurity was significantly associated with perceived stress (crude PR = 1.41, 95%CI: 1.12; 1.76; adjusted PR = 1.29, 95%CI: 1.02; 1.63). This indicates that individuals in food-insecure households were 29% more likely to experience perceived stress. However, during the pandemic, no significant association was found between food insecurity and perceived stress in either crude or adjusted models. Sociodemographic factors showed differing associations with food insecurity and stress, with some disparities in prevalence decreasing between 2019 and 2020. For example, there was a higher prevalence of food insecurity among less educated, unmarried, and non-white individuals in both years, but the gap decreased between the two periods. Similarly, there was a decrease in the prevalence of stress among individuals living in overcrowded houses, those who were unmarried and those with black skin color between the two study periods. The prevalence of stress was higher among females and those with poorer diet quality in both years. Younger individuals exhibited higher levels of stress before the pandemic, while there was no significant difference by age in 2020.

The findings reveal a concerningly high prevalence of food insecurity in Criciúma before and during the pandemic, affecting approximately one-quarter of the population. The prevalence of perceived stress also remained high (around 40%) in both periods. The significant association between food insecurity and perceived stress observed before the pandemic, which was no longer present during the pandemic, suggests that the COVID-19 pandemic may not have negatively impacted food insecurity and stress prevalence in this specific population. This is likely due to government interventions implemented in 2020, such as emergency aid and food distribution programs, which could have mitigated the impact of the pandemic on food security and reduced stress related to food access. The strong economy of Criciúma may have also played a role in buffering the negative economic consequences of the pandemic. The study's findings align with previous research showing a pre-pandemic association between food insecurity and stress. The lack of association during the pandemic could reflect the protective effect of government initiatives. The pathways linking food insecurity to stress are likely multi-faceted and complex, involving individual, collective, and national levels. Poor diet quality, a consequence of food insecurity, can exacerbate stress through nutrient deficiencies impacting the body's ability to cope with stress.

This population-based study demonstrates that food insecurity significantly increased the likelihood of perceived stress before the COVID-19 pandemic, but this association was not observed during the pandemic. This suggests that government interventions may have played a protective role. However, considerable social inequalities persist, with vulnerable populations remaining disproportionately affected by both food insecurity and stress. Maintaining and expanding social and health policies, such as Brazil's National Food and Nutrition Policy, is crucial to ensure basic life rights and address these issues.

This study has some limitations. Different versions of the EBIA scale were used in the two studies. The 2019 study used the full version, enabling a more nuanced assessment of food insecurity levels, while the 2020-2021 study used the short form, limiting the detail of assessment. The sample may overrepresent women and older adults due to the timing of interviews, potentially underestimating the prevalence of food insecurity and stress among working-age men. The cross-sectional design limits the ability to establish causal relationships.