Whole-population trends in obesity across dimensions of inequality in England, 2019-25: a retrospective, longitudinal cohort study of 54 million adults.

Obesity prevalence in England has risen to 30.3%, with significant disparities across age, sex, and socioeconomic status, particularly affecting disadvantaged groups.

Summary and findings

This study measured trends in the incidence and prevalence of obesity among adults in England from 2019 to 2025. A total of 4,131,555 individuals presented with obesity, with an overall age- and sex-standardised incidence of 22 per 1000 person-years. By 2025, the obesity prevalence reached 30.3%, up from 26.3% at the study's start.

Abstract

<h4>Background</h4>Obesity is one of the 21st century's greatest public health challenges. Evidence on how inequalities intersect to shape the obesity burden is scarce, particularly since the COVID-19 pandemic. We aimed to investigate trends in the incidence and prevalence of obesity among adults in England, and to examine variation by age, sex, socioeconomic status, ethnicity, and geographical region.<h4>Methods</h4>In this retrospective, longitudinal cohort study, we analysed whole-population, individual-level, anonymised, electronic health records with life-course data on the entire population of adults aged 18-99 years in England, accessed via the National Health Service England Secure Data Environment. We estimated age- and sex-standardised incidence and prevalence rates of obesity (BMI ≥30·0 kg/m² or clinician-assigned diagnosis) from 2019 to 2025 and used negative binomial regression to examine trends by age, sex, socioeconomic status, ethnicity, and geographical region (defined using middle layer super output areas: neighbourhood-level units comprised of 5000 to 15 000 individuals).<h4>Findings</h4>Between Nov 1, 2019, and April 30, 2025, we identified 54 892 390 adults with records in the National Health Service England Secure Data Environment. 4 131 555 people had a first presentation of obesity during the study period, of whom 2 278 485 (55·1%) were women and 1 853 070 (44·9%) were men. 3 106 740 (75·2%) of individuals were White, 482 690 (11·7%) were Asian or Asian British, and 291 920 (7·1%) were Black, Black British, Caribbean, or African. The median age at first presentation of obesity was 43 years (IQR 31-58), and mean BMI was 33·4 kg/m² (SD 10·7). The overall age- and sex-standardised incidence of first-recorded obesity was 22 per 1000 person-years (95% CI 22-22) and increased by 4% over the study period (incidence rate ratio [IRR] 2024-2025 vs 2019-2020 1·04, 95% CI 1·01-1·07), with marked variation across subgroups. The steepest increases over time occurred in those aged 20-29 years (IRR 1·16, 95% CI 1·08-1·25) and in those aged 30-39 years (1·19, 1·13-1·25). Obesity incidence was 35% greater in the most socioeconomically deprived quintile compared with the least deprived quintile (IRR 1·35, 95% CI 1·28-1·42), with greater disparities in women (1·54, 1·45-1·64), particularly Asian women (1·94, 1·86-2·02). By 2025, overall obesity prevalence reached 30·3%, rising from 26·3% at the start of the study. There was considerable variation in obesity prevalence by age, sex, socioeconomic status, and ethnicity, ranging from 4·3% in the least socioeconomically deprived White men aged 18-19 years to 66·1% in the most deprived Black women aged 60-69 years, which was nearly double that in the least-deprived White women in the same age group (60-69 years; 34·5%). Geographical disparities were striking, with prevalence varying between 8·5% and 48·1% (a nearly 6-fold difference), with the largest increases over time observed in most deprived regions.<h4>Interpretation</h4>Obesity affects nearly one in three adults in England, with disproportionate burden among disadvantaged groups and widening disparities since the COVID-19 pandemic. Increasing rates among individuals of childbearing age risk perpetuating intergenerational cycles of health inequality. The intersection of sociodemographic determinants underscores the preventable nature of obesity and the need to address dimensions of inequality in prevention efforts.<h4>Funding</h4>Wellcome Trust, Health Data Research UK, British Heart Foundation Data Science Centre.

Limitations

Not reported in abstract.