Research Questions
Does long-term, time-varying exposure to ambient air pollution and green-space deprivation increase the incidence of coronary heart disease (CHD) in the UK Biobank cohort? Are the associations independent of traditional cardiovascular risk factors, and do they differ by age or sex?
Objectives
1.To construct annual, 1 km-resolution, time-dependent estimates of PM!.!, NO!, O! and NDVI for all participants by linking residential address histories (baseline + every recorded move) to open UK-AIR and EMEP4UK databases, replacing the static 2010 UK Biobank built-in pollution variables.
2.To estimate hazard ratios for incident CHD (ICD-10 I20-I25) associated with each pollutant and with green-space using multivariable Cox models with time-varying covariates, adjusting for smoking, BMI, Townsend deprivation, hypertension, diabetes and calendar year.
3.To quantify potential effect modification by baseline age (< 60 vs ! 60 years) and sex, and to compare risk predictions from time-varying exposure models with those from conventional lifelong mean exposure models.
Scientific Rationale
Most large prospective studies assign a single long-term average exposure to each individual, ignoring residential mobility and temporal changes in air quality. The UK Biobank's own pollution variables are limited to 2010 annual means and coarse spatial resolution, leading to exposure misclassification. By integrating high-resolution, open-government atmospheric data updated yearly, this project will provide more accurate effect estimates for CHD, inform stricter air-quality policies and demonstrate a reproducible framework for dynamic environmental exposure assessment in population cohorts.