Hong Kong Med J 2008;14(Suppl 3):S9-12
A pilot study to elucidate and validate public health–related environmental effect estimates from statistical modelling of daily health outcome counts
CM Wong, TQ Thach, S Ma
Department of Community Medicine, 5/F, William MW Mong Block, Faculty of Medicine Building, The University of Hong Kong, 21 Sassoon Road, Hong Kong SAR, China
1. Results from statistical modelling of short-term health effects of air pollution can be used to derive the number of health outcomes avoidable for a given reduction in air pollutant concentrations. Effect estimates obtained from daily time-series studies can be used to produce cost curves for reduction in pollutants and benefit in terms of the avoidance of bad health outcomes. This forms the basis for setting air quality guidelines and assessment of air quality interventions.
2. In studies involving time-series data and complex statistical modelling, simulation can be used to obtain data, which mimic the complex structure of the data and the power functions for various effect sizes. This supports the estimation of sample sizes required for the studies, which is often neglected at present.