Trends in childhood obesity for upper tier local authorities in England between 2007/08 and 2023/24: a latent trajectory analysis
No Thumbnail Available
Authors
John Rahilly
Oliver Mytton
Mario Cortina-Borja
Contact
Check for full-text access
Issue Date
06-May-26
Type
Conference Abstract
Language
Keywords
Neighbourhood health & place-based working
Alternative Title
Abstract
Background
There is marked local variation in the prevalence of children living with obesity across England. Identifying areas which diverge from the majority over time may offer insights to inform intervention.
Method
We used routinely published National Child Measurement Programme data on the prevalence of children in Reception and Year 6 living with obesity (BMI ≥ 95th centile) for 150 Upper Tier Local Authorities (LAs) in England for each year (2007/2008 to 2023/2024), excluding 2019/20 and 2020/21. We fitted linear Latent Growth Mixture Models to identify distinct classes with different trajectories. We then used logistic regression to test whether socio-demographic factors would predict class membership, using routinely collected data: Index of Multiple Deprivation [IMD] and Income Deprivation Affecting Children Index [IDACI] (2010, 2015, 2019); and primary school ethnicity (2010 - 2024).
Results
For both Reception and Year 6, obesity trajectories were separated into two classes. Reception Class I (majority group): “Moderate & Stable” (132/150, 88%); and Class II: “High & Declining” prevalence (18/150, 12%). Year 6 Class I (majority group): “Moderate & Rapidly Increasing” (135/150, 90%) and Class II: “High and Gradually Increasing” (15/150, 10%). Eleven LAs were Class II for both Reception and Year 6, with the majority assigned to Class II trajectories located within London or the South-East (17/18 for Reception, 14/15 for Year 6). Universally, Class II membership was associated with higher deprivation (ORs 1.09 [1.02 – 1.17] to 1.22 [1.13 – 1.34]) and a higher proportion of the child population being from ethnic minority groups at multiple time points (ORs 1.04 [1.00 – 1.08] to 1.41 [1.24 – 1.67]). It was also associated with a relative reduction in deprivation over time (1.05 [1.03 – 1.08] to 1.12 [1.08 – 1.18]) and changes in the proportions of the ethnic composition of the primary school population.
Conclusion
A small number of LAs, predominantly located in the South-East of England, were identified as following trajectories that diverged from the majority for both Year 6 and Reception children. These authorities should be subject to further analysis to identify and understand any shared modifiable factors, which could have contributed to this trend.