Population Subgroups at Risk of Unhealthy Changes in Food and Beverage Consumption During COVID-19 Lockdowns

The global COVID-19 pandemic has affected many facets of individuals’ lives, including their food and alcohol consumption [1]. Lockdown periods and social distancing measures limit the ways in which alcohol and food products can be accessed and consumed [1], potentially resulting in changes in the patterns of consumption of these products. Greater intake of unhealthy products may impact individuals’ physical and mental health, as well as their COVID-related risks through impaired immune and physiological functioning [2‐7]. Greater consumption of unhealthy products during lockdown has been linked to people using alcohol and ‘comfort’ foods as a way to deal with the increased stress experienced [1, 8‐10].

In Australia, the context of the present study, lockdowns, and social distancing requirements were periodically enacted during the pandemic as key strategies for reducing the spread of the virus [11]. Research to date suggests this resulted in changes in both food and alcohol consumption patterns among some segments of the Australian population [12]. For example, Australian women were found to be more likely to report increased alcohol consumption during COVID-19 lockdowns [13‐15], while young men were found to be more likely to report reduced consumption [16, 17]. These findings have been attributed to females facing additional stressors during lockdown, such as taking on greater caring duties [16].

International studies have linked factors such as such as gender, age, education, living circumstances (being alone or a parent), and COVID-related beliefs to changes in diet and alcohol consumption during lockdowns [1, 8, 10, 18, 19]. While the findings for age, gender, and education have varied across countries [1, 10, 19, 20], it has been proposed the differences may be due to variations in working from home, employment certainty, and childcare responsibilities during lockdowns, which can differ according to one or more of these demographic characteristics [18, 19]. In terms of COVID-related beliefs, believing that a healthy diet will reduce the risk of being infected with COVID-19 has been linked to positive dietary changes during lockdown [20, 21]. This is aligned with the Health Belief Model that emphasises the importance of perceived susceptibility to negative outcomes when making health-related decisions [22].

Research outside of lockdowns suggests that if a person engages in one unhealthy behaviour, they are more likely to engage in multiple unhealthy dietary behaviours, such as the combination of alcohol consumption and eating unhealthy foods [23, 24]. Most work to date on dietary and alcohol consumption changes during COVID-19 has focused on either alcohol or food separately, with little attention given to simultaneous changes across these behaviours and the resulting implications for public health interventions [1]. To address this deficit, the aim of this exploratory study was to examine whether people changed their frequency of consuming unhealthy foods (sweet snacks, salty snacks, sugary beverages, alcoholic beverages) during lockdown and, if so, whether demographic characteristics and COVID-related health beliefs were associated with these changes. Identifying population subgroups most at risk of unhealthy dietary changes can inform the development of interventions designed to promote and facilitate healthy consumption behaviours during lockdowns.

The frequency analyses demonstrated that most respondents did not change their frequency of consuming alcohol, sweet snacks, and sugary beverages from before to during lockdown (a detailed breakdown is available in Supplementary Table 1). A comparable proportion of respondents increased or decreased their consumption of these products, ranging from 13% each way for alcohol to 21% each way for sweet snacks (see Table 2). Slightly more respondents increased (22%) compared to decreased (18%) their frequency of consuming salty snacks.

Consistent with the descriptive analyses, the generalised linear models only revealed a main effect of time for salty snacks, with an increase in frequency of consumption over time (see Table 3). Across the product categories, the models identified several demographic and COVID-19 belief variables associated with increased frequency of consumption. Those with children at home were more likely to increase their frequency of consuming all four product categories (alcohol B = .51, sweet snacks B = .37, salty snacks B = .41, and sugared beverages B = .50), whereas those who believed an unhealthy diet would increase COVID-19 severity consumed these products less frequently (alcohol B =  − .46, sweet snacks B =  − .31, salty snacks B =  − .34, and sugared beverages B =  − .44). Older people increased their frequency of alcohol consumption (B = .01), but reduced their frequency of consuming sweet snacks (B =  − .01), salty snacks (B =  − .02), and sugared beverages (B =  − .02). Male gender was associated with increased frequency of consumption of alcohol (B = .65), salty snacks (B = .13), and sugared beverages (B = .49). A belief that alcohol use would increase COVID-19 severity was associated with decreased frequency of alcohol consumption (B =  − .14) but increased frequency for the remaining products (sweet snacks B = .16, salty snacks B = .20, and sugared beverages B = .12). Higher education was linked to more frequent use of alcohol (B = .09) and less frequent consumption of sugared beverages (B =  − .06). Last, living with more people was linked to more frequent consumption of salty snacks (B = .04).

The results of the present study are largely consistent with previous Australian findings that overall frequency of consuming alcohol and unhealthy food products did not substantially increase during COVID-19 lockdowns [11, 13, 14, 16, 17, 28]. Like previous research, the findings indicate specific population subgroups exhibited unhealthy changes in dietary patterns [6, 11, 13‐16, 29]. These previous studies examined food and alcohol separately, and the present work builds on their findings by identifying males and those with children as being more likely to demonstrate unhealthy changes across several unhealthy product categories during lockdown periods. Consumption patterns according to the remaining demographic variables were less consistent.

In the present study, respondents with children at home were consistently more likely to consume all the examined unhealthy products more frequently, which is aligned with previous findings [28, 30, 31]. Australian parents have self-reported consuming more alcohol to try and manage stress associated with concern for their children and changes to parenting responsibilities during lockdown [30]. Similarly, parents have described snacking more during lockdown due to boredom and increased caregiving-related stress [31]. Some parents in Australia also reported using the extra spare time associated with lockdowns to bond with their children using cooking activities, including preparing and eating unhealthy ‘treats’ such as cakes and biscuits [32].

The finding that males were more likely than females to report increased frequency of consuming alcohol during COVID-19 contradicts previous Australian alcohol research finding the opposite [13‐17]. This discrepancy could reflect the different measures used: some of these studies asked respondents whether they consumed more, less, or the same amount of alcohol as they did before lockdown [13‐15] and others estimated the quantity of alcohol consumed [16, 17]. Potentially, males are more likely to increase the frequency but not the quantity of alcohol consumed in lockdown compared to females. In the present study, males were also more likely than females to eat more salty snacks and sugary beverages, which conflicts with a study involving a younger student sample that observed the reverse [29].

Given the protracted lockdowns experienced in some parts of Australia and evidence to suggest that unhealthy behaviours developed in response to a traumatic event can last for years [33], this more frequent engagement in multiple unhealthy dietary behaviours by some population segments could increase susceptibility to a range of adverse health conditions. Such conditions include infectious diseases such as COVID-19 through reduced immune function, respiratory impairment, chronic inflammation, greater likelihood of comorbidities, and metabolic and endocrine dysregulation [2‐4].

Consistent with international research [21, 34] and the Health Belief Model [22], respondents who recognised the potential for unhealthy foods/beverages or alcohol to increase the risk of severe COVID-19 were less likely to increase their frequency of consuming these products respectively. This suggests a role for greater education to inform the public of how health-related risks can be reduced during lockdowns by managing lifestyle factors, such as by consuming healthy diets and reducing alcohol use. Bhoyroo et al. [12] found that when Australians reflected on their lockdown experiences, they retrospectively desired more public health messaging to encourage healthy behaviours during this time, indicating that such an approach would be considered acceptable. The results of the present study suggest that effective messaging may need to (i) include concrete recommendations and support to minimise any increases in intake of unhealthy foods and beverages, (ii) address certain health beliefs, and (iii) focus on males and those living with children.