Fig 1.
Location of HIMAQs air monitoring sites, Perth, Western Australia, 2012.
Fig 2.
Spatial distribution of modelled PM2.5 for HIMS participants (wave II) across metropolitan Perth.
Table 1.
Descriptive statistics of CVD risk factors (2001–2004).
Table 2.
Description of the study population.
Table 3.
Descriptive statistics of residential air pollutant exposure as long term annual concentrations (2001–2004).
Fig 3.
Effect sizes of the association between long-term exposure to air pollution and risk factors for cardiovascular disease, per IQRw increase in air pollutant exposure under the main model.
Fig 4.
Smoothed association between fine particulate matter (PM25) and HDL cholesterol (top panel) and triglycerides (bottom panel) adjusted for the main confounders. Grey shade indicate the 95% confidence interval. Smoothed association estimated using a generalized additive model which included the same covariates as for the main model. Association is illustrated as the linear predictor of outcome for a typical cohort member and presented on the original scale of the outcome. That is, expected mean response for an individual with mean age, BMI, SEIFA, and lifestyle scores; never smoker, attended high school, and with no history of cvd or taking treatments to lower blood pressure or cholesterol; Note that the estimated association has the shape (only scaled up/down) for individuals with differing covariate values as the main model does not include interaction terms.
Table 4.
Effect sizes and 95% CI of the association between long-term exposure to air pollution and risk factors for cardiovascular disease, per IQRw increase in air pollutant exposure.
Fig 5.
Effect modification of the association between long-term exposure to air pollution and risk factors for cardiovascular disease, per IQRw increase in PM2.5 exposure under the main model.