Food insecurity among adults and adolescents in Australia, Canada, Chile, Mexico, the United Kingdom, and the United States before and during the Coronavirus Disease 2019 Pandemic

Abstract

Background: Household food insecurity is a major public health concern with negative consequences for health and well-being among adults and adolescents. During the Coronavirus disease 2019 (COVID-19) pandemic, shocks and stabilizers may have affected the prevalence and severity of household food insecurity overall and among subgroups of the population. For example, restrictions to reduce the spread of COVID-19 led to loss of employment, potentially increasing risk of food insecurity, whereas policies such as income supplementation for those who lost employment may have had a stabilizing effect. Comparing changes in household food insecurity across multiple countries over time can shed light into how packages of economic and social policy responses implemented or revised by national level governments during the pandemic may have influenced household food insecurity, providing insights into strategies to ameliorate household food insecurity, including in times of crisis.

Objectives: The objectives of this thesis were to 1) characterize changes, or lack thereof, in the prevalence and severity of food insecurity among adults and adolescents in Australia, Canada, Chile, Mexico, the United Kingdom, and the United States from before (2019) to during (2020) the pandemic (Chapter 5); 2) examine national level government economic and social policy responses implemented or revised in these countries to, in part, mitigate the impacts of the COVID-19 pandemic and related safety measures (Chapter 6); and 3) determine whether economic and social policy responses appeared to mitigate anticipated increases in food insecurity during the COVID-19 (Chapter 7). The countries of interest were identified based on the similarity in the languages spoken, food environments, and culture among other characteristics.

Methods and results: Three inter-related studies addressed the thesis objectives, with two of the studies integrating consultation with a collaborative working group to facilitate the relevance and applicability of study findings by considering country-specific contexts. Two studies drawing upon the International Food Policy Study considered food insecurity data reported by adults aged 18 to 100 years and directly by adolescents aged 10 to 17 years, enabling comparison in trends among the two population subgroups over time. Among adults, food insecurity was measured using the 18-item Household Food Security Survey Module, enabling categorization of the households in which participants lived as food secure or food insecure over the past 12 months. Among adolescents, the 10-item Child Food Insecurity Experiences Scale was used, enabling consideration of the number of food insecurity experiences (i.e., no, few, many, or several) they had over the past 12 months.

The first study (Chapter 5) drew upon repeat annual cross-sectional data from adults (n = 63,278) and adolescents (n = 23,107) to examine changes in food insecurity prevalence and severity from before (2019) to during (2020) the COVID-19 pandemic in the countries of interest. Based on adjusted binary and multinomial logistic regression models, adults in Australia (adjusted odds ratio (AOR): 0.81; 95% CI: 0.72-0.92) and Canada (AOR: 0.87; 95% CI: 0.77-0.99) were less likely to live in food-insecure households during (2020) relative to before (2019) the pandemic. There was little indication of changes in the odds of adults living in food-insecure households in the United Kingdom (AOR: 0.90; 95% CI: 0.79-1.02) and the United States (AOR: 1.05; 95% CI: 0.93-1.19) in 2020 relative to 2019, while in Mexico, adults were more likely (AOR: 1.15; 95% CI: 1.02-1.31) to live in food-insecure households in 2020. Findings related to changes in the severity of food insecurity were similar. Adolescents in Mexico were more likely (AOR: 1.43; 95% CI: 1.19-1.71) to report experiencing food insecurity in 2020 compared to 2019. In Australia (adjusted relative risk ratio (ARRR): 2.24; 95% CI: 1.65-3.02) and the United States (ARRR: 1.39; 95% CI: 1.04-1.86), adolescents had a higher relative risk ratio of many compared to no experiences of food insecurity in 2020 compared to 2019. No differences in the prevalence or number of experiences of food insecurity among adolescents were observed in other countries.

In the second study (Chapter 6), packages of economic and social policy responses to, in part, mitigate the impacts of the pandemic and safety measures to limit spread of the virus in the countries of interest were characterized. Potentially relevant policy responses implemented or revised by national governments between January 2020 and December 2022 were identified using eight policy response trackers. A theory of change articulating the hypothesized influence of different types of policy responses on household food insecurity guided the inclusion criteria. Two rounds of screening were conducted to identify the final set of 142 relevant economic and social policy responses. Hypotheses about how policy packages influenced household food insecurity in 2021 and 2022 were developed based on observed changes in household food insecurity from 2019 to 2020 and broader contextual events. Consideration of the influence of policies on food insecurity in 2021 and 2022 integrated evidence from other sources. Financial-based policy responses (e.g., unemployment protection) were emphasized in Australia, Canada, and the United Kingdom. In Chile, Mexico, and the United States, there was a more equal distribution of financial-based and resource-based (e.g., food assistance) policy responses. Across most countries and years, policy packages provided long periods of coverage (six months or more). Mexico was an exception in 2021 with a mix of policy responses that had short- and long-term coverage.

The third study (Chapter 7) drew upon repeat cross-sectional data from 104,418 adults and 44,759 adolescents from 2018 (adults only), 2019, 2020, 2021, and 2022 to determine whether economic and social policy responses mitigated anticipated increases in food insecurity among adults and adolescents during the COVID-19 pandemic. Three lines of inquiry were used to realize this objective. First, year-to-year changes in food insecurity among adults and adolescents and in the known determinants (e.g., difficulty making ends meet) and consequences (e.g., poor quality diets) of food insecurity were compared. Alignment in changes would instill confidence that the nature of food insecurity did not change during the pandemic and set the foundation for the subsequent two lines of inquiry. Second, among adults living in households, the predicted prevalence of food insecurity in 2020, 2021, and 2022 had the pandemic not occurred was estimated (this analysis was not conducted for adolescents given the lack of data for 2018). Hypothesized changes in food insecurity among adults living in households for each country by year given that the pandemic occurred and considering broader contextual events (e.g., global inflation, Brexit in the United Kingdom) were developed to guide assessment of whether economic and social policy responses mitigated anticipated increases in food insecurity. Observing that the observed proportion of adults in food-insecure households was lower or did not differ from the predicted proportion would suggest that economic and social policy responses mitigated anticipated increases in food insecurity. Third, adjusted multinomial or binary logistic regression models with interactions for the survey year and selected characteristics that indicate high risk of experiencing food insecurity (e.g., households with children) assessed changes in the odds or relative risk ratio of food insecurity among adults living in households and adolescents during (2020 to 2022) relative to before (2019) the COVID-19 pandemic among the subgroups selected. Subgroups with selected characteristics were anticipated to be more likely to experience food insecurity during the pandemic. If this finding was not observed, it would suggest that economic and social policy responses mitigated anticipated increases in food insecurity among those at high risk during the pandemic.

Alignment between changes in food insecurity among adults living in households and adolescents and its known determinants and consequences ranged across countries and over time. The variation in alignment was likely due to the range of factors that could influence food insecurity and the unlikelihood that a one percent change in the proportion of food insecurity would equate to a one percent change in the singular category examined for each determinant and consequence. The observed proportions of food insecurity among adults living in households were lower than predicted had the pandemic not occurred in Australia, Canada, and the United States at various time points, suggesting that policy responses to address pandemic-related disruptions mitigated increases in food insecurity that were otherwise anticipated to due to shocks related to COVID-19 safety measures. Subgroups that were anticipated to be more likely to experience food insecurity during (2020 to 2022) relative to before (2019) the COVID-19 pandemic largely did not, with indication that some subgroups were better protected by policies introduced during the pandemic compared to beforehand.

Conclusions: Although some increases in food insecurity among adults were observed through the pandemic in the countries examined, it appears that economic and social policies introduced by national governments mitigated otherwise large, anticipated increases due to loss of income and other shocks to households. The lack of increases in the proportions of adolescents experiencing food insecurity in most countries reinforces the notion that policy packages influenced food insecurity during the pandemic. Nonetheless, household food insecurity remains prevalent and increased in later years, likely due to the expiration of policy responses, inflation, and other events globally.

Though long-term economic and social policies to address household food insecurity are lacking in the countries examined, this work indicates it is possible for governments to ameliorate this serious public health problem using policy levers. Packages of policy responses that focused on financial-based support and/or scaled up existing policy responses mitigated anticipated increases in food insecurity among adults in 2020. Policy packages also appear to have contributed to the observed prevalence of food insecurity among adults living in households being the same or lower than the predicted prevalence through the pandemic. These findings are salient as the world works to achieve the 17 Sustainable Development Goals by 2030 against the backdrop of ongoing crises anticipated to sharply increase risk of household food insecurity, including the climate emergency and conflicts.