**Background**
The COVID-19 pandemic and associated restrictions caused sudden lifestyle changes, increased psychological stress, and widespread weight gain. Prior to the pandemic, chronic stress was already linked to weight gain, and food cravings — intense desires for specific foods, often high in sugar and fat — were known mediators of stress-related eating. This study aimed to investigate whether higher perceived stress during the pandemic was associated with greater weight gain, whether poor prior mental health amplified this effect, and whether food cravings and comfort eating mediated the stress–weight gain relationship.
**Methods**
Between 6 January and 2 February 2021, UK adults (n=179 after exclusions; 75% female; 85% aged 18–55; mean BMI 26.10±5.46 kg/m²) completed an anonymous online survey via Qualtrics. Participants reported demographic information, weight change since March 2020 (categorical: lost >3kg, lost up to 3kg, stable, gained up to 3kg, gained 3–6kg, gained >6kg), changes in eating behaviours (hunger, food cravings, meals, snacking), dietary consumption (16 food/drink categories), physical activity, and reasons for increased eating. Perceived stress was measured using the Perceived Stress Scale (PSS-10; Cronbach's α=0.87). Participants also reported their mental health prior to COVID-19 (coped well; some stress/anxiety/depression but not diagnosed; previously diagnosed; prefer not to say) and the impact of 12 specific COVID-19-related lifestyle changes. Participants were categorised into higher stress (PSS ≥19.4) or lower stress (PSS ≤19.3) groups based on the mean score (19.3±6.9). Ordinal regression, binary regression, and chi-square analyses were used; odds ratios (OR) and 95% confidence intervals (CI) were reported.
**Key Results**
- Weight gain: 50% of females and 27% of males reported weight gain; 10% of females vs. 5% of males gained >6kg. The higher stress group was significantly more likely to report greater weight gain (χ²=14.639, p=0.012, phi=0.286). Gender and weight gain were significantly associated (χ²=11.986, p=0.035, phi=0.259).
- Eating behaviours: 52% reported increased food cravings, 53% increased daytime snacking, and 53% increased evening snacking. Increased food cravings (OR=3.225, p=0.002), increased number of meals (OR=2.978, p=0.012), increased daytime snacking (OR=2.568, p=0.012), and eating due to stress/anxiety/low mood (OR=2.192, p=0.026) were significant predictors of weight gain in the ordinal model (full model χ²=70.877, p<0.001).
- Stress and eating: Higher stress participants were twice as likely to report increased food cravings (OR=2.3, p=0.011) and daytime snacking (OR=1.9, p=0.041).
- Dietary consumption: Increased high-sugar foods (OR=3.148, p=0.005) and processed foods (OR=2.501, p=0.032) were linked to weight gain; increased high-fibre foods (OR=0.351, p=0.034) and water (OR=0.455, p=0.033) were linked to weight loss (full model χ²=75.291, p<0.001). Higher stress participants were twice as likely to increase high-sugar foods (OR=1.9, p=0.045) and processed foods (OR=2.5, p=0.012).
- Food cravings and comfort eating: Participants with increased cravings were 11.2 times more likely to increase high-sugar food consumption (p<0.001), 6.3 times more likely to increase processed food consumption (p<0.001), and 6.0 times more likely to increase daytime snacking (p<0.001).
- Physical activity: 47% reduced moderate/strenuous activity; reduced activity was linked to weight gain (χ²=25.354, p=0.005, phi=0.271) but not to stress levels (χ²=1.40, p=0.50).
- Predictors of high stress: Prior diagnosis of stress/anxiety/depression (OR=5.767, p=0.004), some prior stress/anxiety/depression without diagnosis (OR=3.663, p=0.004), and female sex (OR=3.228, p=0.014) were significant predictors. No individual COVID-19 lifestyle change was significantly linked to higher stress in the binary regression model (full model χ²=58.95, p<0.001).
- Gender differences: Females reported more COVID-19 lifestyle changes (53% affected by ≥2 changes vs. 39% of males) and were more likely to report less money for food (16% vs. 9%), less access to healthy food (12% vs. 7%), and less time to cook (11% vs. 5%).
**Clinical Implications**
This study confirms that higher perceived stress during the pandemic was significantly associated with weight gain, mediated primarily through increased food cravings and consumption of high-sugar and processed foods. The findings underscore that prior mental health status is a major risk factor: individuals with a pre-pandemic psychiatric diagnosis were nearly six times more likely to experience high stress during the pandemic. The disproportionate impact on women — who reported higher stress, more lifestyle disruptions, and greater weight gain — suggests that gender-sensitive approaches to weight management are needed. The strong link between food cravings and both snacking and comfort food consumption (OR 6–11) indicates that weight loss interventions should shift focus from simply restricting these downstream behaviours to addressing craving control, potentially through stress reduction techniques. Limitations include the cross-sectional design (precluding causal inference), reliance on self-report, a relatively small convenience sample (n=179, predominantly female and young-to-middle-aged), and categorical rather than continuous measures which may obscure subtle effects.