The COVID-19 pandemic caused significant disruptions to daily routines, leading to increased social isolation, job insecurity, financial strain, and altered food environments. These changes have created a large-scale, global experiment on human health behaviors. While small European studies suggested negative impacts (increased sedentary behavior, snacking, weight gain), the evidence remains inconclusive. Major life events strongly influence health behaviors like alcohol consumption, sleep, diet, and physical activity. These behaviors interact to affect body weight, a major public health concern. Understanding pandemic-related health behavior changes is crucial for predicting long-term consequences. This study uses data from the ZOE COVID Study to examine self-reported impacts of the pandemic on diet and health behaviors using a composite disruption index (DI), analyzing its association with weight change and exploring the relationship between pre-pandemic behaviors and pandemic-related changes.

The introduction adequately references existing literature showing the impact of major life events on health behaviors such as alcohol intake, sleep, diet, and physical activity. It also cites studies linking these behaviors to weight gain and obesity-related morbidities. The review highlights the lack of comprehensive data on the pandemic's effect on health behaviors, justifying the need for this research. Specific studies cited include those examining the impact of COVID-19 lockdowns on diet and lifestyle in various European countries (France, Italy, Poland), as well as research on the association between life events and changes in alcohol consumption, sleep patterns, and physical activity.

This retrospective longitudinal observational cohort study utilized data from the ZOE COVID Study app. Participants (896,286) completed a retrospective questionnaire assessing diet quality, habits, and lifestyle behaviors at two time points: peri-pandemic (previous month) and pre-pandemic (February 2020). Data included demographics, comorbidities, and (for UK participants) the Index of Multiple Deprivation (IMD). After exclusions (replicates, <18 years old, pregnant women, non-UK/US participants, implausible anthropometric data), 896,286 peri-pandemic and 291,871 pre- and peri-pandemic participants remained. The UK cohort was split into discovery (n=380,847) and validation (n=448,321) sets. A disruption index (DI) was calculated using five metrics: diet quality score (DQS), alcohol frequency, physical activity, snacking frequency, and weekday sleep duration. The DI ranged from 0 (no disruption) to 5 (change in all five domains). Body weight change was analyzed, and structural equation modeling examined associations between lifestyle behavior changes and weight change. Factor analysis identified 'healthier' and 'less healthy' pre-pandemic lifestyle patterns to investigate differential responses to the pandemic.

Large inter-individual changes were observed in 46 health and diet behaviors, but no mean population change. Younger, female, and socioeconomically deprived participants showed greater disruption. Greater disruption was associated with more variable weight change, including both gain and loss. Individuals with less healthy pre-pandemic habits improved diet quality and lost more weight than those with initially healthier habits. Structural equation modeling revealed that changes in physical activity and diet quality were strongly associated with weight change. While the average change in most variables was minimal, a large proportion of participants exhibited individual changes (e.g., 37% increased DQS, 31% increased fruit and vegetable portions). Those most adherent to less healthy pre-pandemic habits showed more improvements in diet and weight loss, regardless of deprivation level.

The study's findings challenge the notion that the pandemic uniformly negatively impacted health behaviors. While disruption was greater in specific demographics, the overall population showed minimal average change. The significant inter-individual variation highlights the complexity of the pandemic's influence. The observation that individuals with less healthy pre-pandemic lifestyles exhibited greater improvement during the pandemic may reflect a larger potential for positive change. The significant role of physical activity and diet quality in weight change confirms the importance of these factors in public health interventions. Future research should investigate higher-level factors (e.g., working from home, food availability, societal influences) that mediate individual health behavior changes. The study’s limitations include self-reported data, potential recall bias, and the potential influence of the study’s sample on the observed behaviours.

This large study reveals the diverse and often positive impacts of the COVID-19 pandemic on lifestyle behaviors. While some demographics experienced greater disruption and increased susceptibility to weight changes, many individuals improved their diet and lifestyle. The study highlights the importance of targeted public health initiatives focusing on physical activity and diet quality, while acknowledging the need to address existing health inequalities. Future research should explore the sustainability of these changes and their long-term effects on metabolic health. Investigating the role of higher-level disruptions in shaping individual behaviors is also crucial.

The study relied on self-reported data, which is susceptible to recall bias. The data collection methods were slightly modified during the study, which could have introduced some bias. The disruption index was relatively crude, incorporating only five behaviors. The study sample may not fully represent the general population. Regression to the mean might have influenced the observed changes in some subgroups.