Daily changes and short-term exposure patterns in time series studies of air pollution and acute health effects

Mohammed, Nuredin Ibrahim (2016). Daily changes and short-term exposure patterns in time series studies of air pollution and acute health effects. University of Birmingham. Ph.D.

Preview

Mohammed16PhD.pdf
PDF - Accepted Version
Download (4MB)

Abstract

This thesis investigated the effects of daily changes in exposure (delta) and short-term exposure patterns on the relationship between air pollution and health in time series studies. Using data from London and Hong Kong, delta was defined as the difference in particulate matter (PM10) concentration between successive days. Short-term exposure pattern series were defined based on number of peaks in PM10 within rolling weekly blocks. The mathematical equivalence of identifiable models for delta with conventional distributed lag model was derived and alternative model specifications were proposed. Measurement error and missing data exhibited more impact on delta than the absolute metrics in simulation studies. Evidence of association for delta PM10 with mortality was found only in Hong Kong which attenuated towards the null with more rigorous adjustment for weather. The pattern analysis approach hypothesized, in addition to amount (dose) and duration of exposure, epidemiological studies ought to take patterns of exposure into account. However, convincing evidence was not found for the effect of short-term exposure patterns on mortality risk estimates both in London and Hong Kong. Refining the definition of exposure patterns and methodological improvements including analysing data from multiple cities are highly recommended in related studies in the future.

Type of Work:

Thesis (Doctorates > Ph.D.)

Award Type:

Doctorates > Ph.D.

Supervisor(s):