The current research observed dietary changes among individuals LWBC during the COVID-19 pandemic. Specifically, fruit, vegetable, and alcohol consumption decreased whilst snacking increased. Changes in dietary habits were influenced by sociodemographic, health-related, and COVID-19-related factors. Women and individuals with limited social contact were more likely to reduce fruit and vegetable intake during the pandemic. Women and individuals with poorer sleep quality, limited social contact, shielding requirements, and without higher education were more likely to increase snacking during the pandemic. Individuals with poorer sleep quality, greater anxiety and depression levels, and regular social contact were more likely to decrease alcohol consumption during the pandemic.
The first research question aimed to identify changes in the consumption of fruit and vegetables, snacking, and alcohol among individuals LWBC from before to during the pandemic. Within this sample, fruit and vegetable consumption appeared to decrease from before to during the pandemic which is supported by existing research(45). Food availability throughout the pandemic has been impacted by disruption to the food supply chain(46). In the UK, these food availability concerns led to hoarding and stock piling behaviour(47, 48) which seemed to result in relative preference towards non-perishable foods over fresh foods, such as fresh fruit and vegetables(46). Furthermore, the pandemic has impacted the UK economy, including higher unemployment rates(49). These economic consequences have been detrimental for many UK households(50) and have impacted consumer behaviour, including reduced purchases of fresh fruit and vegetables(46) perhaps as they tend to be perceived as more expensive(51). The pandemic has also impacted the mental health of individuals living with and beyond cancer, including increases in stress and anxiety levels(52). These adverse mental health outcomes are associated with unhealthy behaviours including increased snacking and a reduction in fruit and vegetable consumption(45, 53, 54). Previous research has observed an association between ‘natural disasters’ and reduced fresh produce consumption(55), explained via a combination of higher anxiety levels, increased food prices, and decreased availability, which could be likened to the pandemic.
Existing literature also supports the finding that snacking increased during the pandemic for people LWBC. The pandemic may have resulted in behavioural changes due to social restrictions imposed, changes to working patterns, or from COVID-19-related stress and the coping strategy implemented(56). During the pandemic, increased snacking has been observed alongside elevated stress, boredom, and emotional eating(56–58). Given the high palatability of snacks, individuals may have demonstrated emotional eating behaviours and increased snacking as a comfort mechanism to deal with COVID-19 related stress, depression or anxiety(57, 59). Foods high in saturated fat, salt, and sugar (HFSS) also tend to have long shelf lives, be cheaper and more readily accessible. Individuals may have ‘stocked-up’ on HFSS foods through less frequent shopping trips(57). Home confinement may have increased sedentary and screen time, further contributing to increased snacking(60). Initial shifts in snacking behaviours may then have been maintained through habit formation(61).
The observed finding that alcohol consumption decreased among individuals living with and beyond cancer during the pandemic is less expected. A systematic review of general and clinical populations observed an overall trend towards increased alcohol consumption(62), which contradicts the current findings. However, alcohol consumption may have decreased due to reduced socialising following the social restrictions imposed by the government, and the current sample may have been comprised of primarily social drinkers. Furthermore, research suggests that following a cancer diagnosis individuals living with and beyond cancer reduce overall consumption of alcohol to improve survival chances and reduce likelihood of re-diagnosis(63). This protective behaviour may therefore have been established before the pandemic which could have limited any negative changes during the pandemic.
The second research question aimed to explore the sociodemographic, health-related, and COVID-19-related factors associated with dietary changes during the pandemic. The current findings observed that females, compared to males, consumed more snacks and less fruit and vegetables during the pandemic compared to before. Past research has demonstrated that women have a higher tendency to increase their snacking behaviour in response to negative affect(64, 65) whereas men tend to decrease their snacking behaviour(64–66). Negative affect generally refers to affective states such as anxiety, stress, and boredom which are all suggested to have increased during the pandemic(67). Therefore, females may be more behaviourally responsive to COVID-19-related stressors, resulting in unhealthy dietary changes. Research during the pandemic observed a significant interaction by gender between change in adherence to the Mediterranean diet, including intakes of fruits and vegetables. Specifically, men showed a stronger association between positive change in adherence compared to women. This supports the observed finding that females consumed fewer fruits and vegetables during the pandemic compared to males.
Individuals with no higher education, compared to those with higher education, were found to increase snack consumption during the pandemic compared to before. Human capital theory indicates that education may increase income(68), meaning individuals with lower education levels likely have lower income compared to those with higher education. Considering the financial impact of the pandemic, including increased unemployment(49) and food shortages(47, 48), individuals that are less socioeconomically advantaged may be more likely to change consumer behaviour and prioritise options with longer shelf lives including snacks(46). Likewise, these disadvantaged individuals may be more likely to experience COVID-19-related stress due to financial insecurity(69) which may encourage emotional eating and a preference for HFSS foods(70).
Several health-related and COVID-19-related variables were also associated with increased snacking among individuals living with and beyond cancer during the pandemic. Shielding and limited social contact were associated with increased snacking during the pandemic, compared to before. These individuals are likely experiencing higher levels of social isolation compared to others(71). One article concluded that periods of isolation, even less than 10 days, can have long-term effects on health, including psychiatric symptoms(72). As mentioned previously, psychiatric symptoms have been associated with emotional eating which can lead to increased snacking(73). Individuals with worse sleep also consumed more snacks during the pandemic. This is supported by a plethora of research highlighting that insufficient sleep increases snacking(23). Interestingly, research suggests that the hedonic aspects of regulating energy intake override the homeostatic factors(74).
Reduced alcohol consumption was predicted by various health-related and COVID-19-related factors. Individuals with greater levels of anxiety and depression consumed less alcohol during the pandemic compared to before. Research primarily observes the opposite relationship, with poorer mental health predicting increases in alcohol consumption(75). Perhaps individuals with poorer mental health may be less likely to be drinking socially compared to those with better mental health during the pandemic. However, more research is necessary to understand this association. Individuals with poorer sleep quality also consumed less alcohol during the pandemic. Research commonly finds an association between worse sleep quality and alcohol consumption(76), which does not support the current finding. However, these individuals may have tried to rectify their poor sleep during the pandemic by reducing their alcohol intake. This potential explanation warrants further longitudinal exploration. Individuals with regular social contact consumed less alcohol during the pandemic compared to before. Some research does not support this finding as, for example, decreases in adolescent alcohol consumption are primarily associated with reductions in social engagement(77). However, more frequent social contact during the pandemic was associated with lower anxiety symptoms(78). Therefore, stress-related mental health may act as a mediator between social contact and reduced alcohol consumption. Furthermore, this increase could be viewed as alcohol acting as a coping mechanism(79, 80) for increased feelings of loneliness for those with less social contact.
To our knowledge, this is the first paper to prospectively investigate the dietary changes among individuals living with and beyond cancer during the COVID-19 pandemic. A strength of the current study is the large sample size compared to previous research investigating dietary patterns among this population which provides greater power to detect differences. The longitudinal design also allows the tracking of behaviour change over time, allowing behavioural patterns and determinants to be identified. Nonetheless, the findings should be viewed in the context of several limitations. The sample may not be representative of the target population as some subgroups may have been unequally represented. The subgroup analyses comparing participants that were excluded and included in the sample demonstrated differences in age, ethnicity, and education. These differences increase the likelihood of selection bias and question the generalisability of the findings. Considering the outcome measures were self-reported and the sensitive nature of several measurements collected, including mental health and health behaviours, the research may be subject to self-report bias or social desirability bias whereby participants present themselves in a more positive light. Finally, as we included only one pre-pandemic timepoint we cannot determine whether the effects observed were a response to the pandemic or due to a pre-existing behavioural trend. These are important considerations when interpreting the current findings.
To conclude, this study suggests that individuals living with and beyond cancer made unhealthy dietary changes for fruit, vegetable, and snack consumption but made healthy changes to their alcohol consumption in response to COVID-19. These changes appear to be different based on sociodemographic, health-related, and COVID-19-related factors which suggests that the pandemic has not equally impacted everyone. These results have important implications for both individuals LWBC and the general population by providing a more detailed understanding of the risk factors for dietary change during the pandemic. Specifically, the findings suggest a need for more targeted interventions post-COVID based on the risk factors identified. This could allow resources to be prioritised amongst these at-risk groups to prevent or reduce any further negative impact of the pandemic. Future research should focus on understanding the potential mediators explaining the observed associations to better understand the pathways to unhealthy dietary change.