Composite healthy lifestyle, socioeconomic deprivation, and mental well-being during the COVID-19 pandemic: a prospective analysis

The study investigates whether a composite of modifiable pre-pandemic lifestyle factors is associated with reduced risk of depression, anxiety, and low personal well-being during the COVID-19 pandemic, and whether socioeconomic deprivation modifies these associations. The context is the documented psychological and social harms of the pandemic and related restrictions, which disproportionately affected older adults and socioeconomically deprived groups. Prior work has focused on single lifestyle behaviors or population-level recommendations; evidence on combined lifestyle factors and their interaction with deprivation in the pandemic setting was lacking. The purpose is to inform targeted interventions at both individual and population levels to promote mental well-being, particularly among vulnerable groups, during current and future pandemics.

Pre-pandemic evidence links individual lifestyle factors (BMI, smoking, alcohol use, physical activity, sleep, diet) with depression and anxiety, as well as cardiometabolic outcomes and mortality. Some studies during COVID-19 suggested single factors like physical activity or diet related to lower psychological distress. Lifestyle factors can interact synergistically, yet the role of combined lifestyle patterns in pandemic-related mental health had not been established. Socioeconomic deprivation worsens mental health and may exacerbate unhealthy lifestyles, but whether deprivation modifies the lifestyle–mental health association was unclear.

Design and data: Prospective analysis using the English Longitudinal Study of Ageing (ELSA), a nationally representative cohort of adults ≥50 years in England. Baseline for exposures and covariates was wave 9 (2018–2019, pre-pandemic). Outcomes were assessed in two ELSA COVID-19 substudy waves (June/July 2020 and November/December 2020; 94% longitudinal response). Ethical approval obtained from the National Research and Ethics Committee. Population: Of 5594 participants with COVID-19 substudy data, 5213 (93.2%) experienced self-isolation or shielding during lockdown; 5049 (96.8%) had complete baseline data on lifestyle, deprivation measures, and covariates and were included. Participants who reported neither self-isolation nor staying at home at any of the three queried time points (April, June/July, November/December 2020) were excluded (n=381, 6.8%). Exposures—Composite lifestyle score (7 factors): Based on UK guidelines and prior evidence, unhealthy category received 1 point each: unhealthy BMI (<18.5 or ≥25 kg/m²), current smoking, high alcohol intake (daily or almost daily), low physical activity (moderate or vigorous activity <1 time/week), sedentary time ≥7 h/day, short or long sleep (<7 or >9 h/day), and insufficient fruit/vegetable intake (<5 portions/day). Sum created an unweighted score (0–7). Lifestyle categories: favorable (0–1), intermediate (2–3), unfavorable (4–7). Socioeconomic deprivation: Group-level Index of Multiple Deprivation (IMD) and individual-level factors (education categorized low/middle/high; income and wealth quintiles). Primary analyses used IMD categorized into low (quintile 1, most deprived), intermediate (quintiles 2–4), and high (quintile 5, least deprived). Outcomes: During the pandemic, depressive symptoms by CES-D 8 (score ≥4), anxiety by GAD-7 (score ≥10), and personal well-being by ONS-4 (score ≤4). CES-D and GAD measure symptoms and are not clinical diagnoses but are validated for population studies. Covariates: Pre-pandemic measures (wave 9): age, sex, ethnicity, marital status, employment, disability, comorbidities (chronic lung disease, asthma, arthritis, osteoporosis, cancer, Parkinson’s disease, dementia, hypertension, diabetes), education, income, wealth, and pre-pandemic mental health (history of psychiatric disorders, depression, anxiety, and personal well-being). Sensitivity analyses additionally adjusted for pre-pandemic loneliness (UCLA-3) and social isolation (composite score). Statistical analysis: Cox proportional hazards models using study wave as timescale; proportional hazards assumptions checked with Schoenfeld residuals. Associations assessed for lifestyle score and category with each mental health outcome, adjusting for covariates including socioeconomic characteristics. Interaction between lifestyle category and IMD tested (multiplicative). Stratified analyses by deprivation groups and combined nine-category lifestyle×deprivation analysis; HR for trend per one increment change in lifestyle category and in lifestyle×deprivation category estimated. Sensitivity analyses: (1) deprivation characterized by individual-level education, income, wealth; (2) associations for each lifestyle factor mutually adjusted; (3) excluding participants with history of mental disorders; (4) additional adjustment for loneliness and social isolation. Two-sided tests with p<0.05; software SAS 9.4 and R 4.2.2.

- Sample: 5049 participants (mean age 68.1±10.9 years; 57.2% female; 96.1% White). Lifestyle categories: favorable 41.6%, intermediate 48.9%, unfavorable 9.5%. Pandemic-period prevalence: depression 26.6%, anxiety 9.2%, low well-being 8.1%. - Lifestyle category and outcomes (adjusted): HR for trend per one-step shift toward unfavorable category: depression 1.17 (95% CI 1.09–1.26); anxiety 1.23 (1.07–1.42); low well-being 1.39 (1.20–1.61). Compared with favorable: intermediate HRs—depression 1.14 (1.04–1.26), anxiety 1.23 (1.01–1.50), low well-being 1.25 (1.00–1.56); unfavorable HRs—depression 1.41 (1.21–1.65), anxiety 1.51 (1.12–2.05), low well-being 2.04 (1.51–2.75). - Number of healthy lifestyle factors: Clear dose–response; vs 0–2 factors, having all 7 was associated with lower risks: depression HR 0.52 (0.39–0.69), anxiety 0.58 (0.35–0.98), low well-being 0.42 (0.23–0.75). Per 1-point increase in lifestyle score: depression 1.11 (1.07–1.16), anxiety 1.11 (1.03–1.19), low well-being 1.20 (1.10–1.30). - Individual lifestyle factors (mutually adjusted): Higher risks observed for unhealthy BMI (obese vs normal—depression 1.31 [1.14–1.51], anxiety 1.33 [1.00–1.78], low well-being 1.40 [1.03–1.91]), underweight and depression 1.87 (1.25–2.80); current smoking (depression 1.38 [1.18–1.61], anxiety 1.67 [1.18–2.35], low well-being 2.08 [1.46–2.96]); low physical activity (<1×/wk) associated with higher risk (regular activity protective: depression 0.64 [0.67–0.72], anxiety 0.69 [0.54–0.88], low well-being 0.67 [0.61–0.86]); adequate sleep (7–9 h/d) protective for depression 0.80 (0.71–0.90) and anxiety 0.69 (0.54–0.87). Fruit/vegetable intake showed modest inverse association with depression 0.88 (0.79–0.98). Sedentary time showed no significant independent associations after adjustment. - Socioeconomic deprivation (IMD): Per increment in deprivation category, higher risk of depression 1.25 (1.15–1.35), anxiety 1.57 (1.34–1.84), low well-being 1.22 (1.03–1.44). - Joint lifestyle×deprivation: No significant interactions; similar lifestyle–risk gradients across deprivation groups. Compared with least deprived + favorable lifestyle, most deprived + unfavorable lifestyle had highest risks: depression 2.80 (2.05–3.82), anxiety 3.66 (2.02–6.64), low well-being 3.43 (1.83–6.42). Per increment across combined categories: depression 1.07 (1.05–1.10), anxiety 1.11 (1.07–1.16), low well-being 1.13 (1.07–1.18).

The study demonstrates that adherence to multiple healthy lifestyle factors prior to the pandemic is associated with lower risks of depression, anxiety, and low personal well-being during COVID-19. The associations are dose-responsive and persist after adjusting for sociodemographics, comorbidities, and pre-pandemic mental health, indicating that combined behaviors—not just single factors—relate to mental health resilience in this context. Socioeconomic deprivation independently elevates risk, but the lack of significant interaction suggests that beneficial effects of healthy lifestyles are similar across deprivation strata. Together, unfavorable lifestyle and deprivation exert additive adverse associations on mental health. The findings align with mechanistic and genetic evidence supporting roles for BMI, smoking, physical activity, and sleep in mental health and underscore the potential of comprehensive lifestyle promotion to mitigate pandemic-associated mental health burdens, alongside policies addressing deprivation.

A composite of healthy lifestyle behaviors (healthy BMI, non-smoking, moderate alcohol use, regular physical activity, limited sedentary time, adequate sleep, sufficient fruit/vegetable intake) was associated with substantially lower risks of depression, anxiety, and low well-being during the COVID-19 pandemic in older adults in England. Socioeconomic deprivation independently increased risks, with no evidence of interaction with lifestyle. These results support individual-level interventions that target multiple lifestyle factors and population-level strategies to reduce deprivation. Future research should investigate causality (e.g., via randomized or quasi-experimental designs), mechanisms linking lifestyle to mental health in crisis contexts, and the effectiveness of multicomponent lifestyle interventions across socioeconomic groups.

- Observational design limits causal inference; the authors note that further research is needed to assess causality. - Mental health outcomes were measured with validated symptom scales (CES-D 8, GAD-7, ONS-4) rather than clinical diagnoses, which may affect clinical generalizability. - Lifestyle behaviors and some covariates were self-reported, potentially introducing measurement error. - The cohort comprises adults aged ≥50 years in England (and excluded those not self-isolating/staying at home during queried periods), which may limit generalizability to younger populations or other settings.