Details of talk
|Title||Statistical considerations in neighbourhoods and health research|
|Presenter||Karen Lamb (Deakin University)|
|Session||Biostatistics and Bioinformatics|
The places in which we live are thought to shape our health and behaviours. From the availability of parks and recreational facilities to access to supermarkets and fast food outlets, the resources available to us within our neighbourhoods may affect behaviours such as physical activity and diet, both of which are associated with a number of chronic diseases. The study of place effects on health and behaviour is challenging due to a number of important statistical and epidemiological considerations. These include challenges such as how to define exposure to neighbourhood attributes; methodological issues associated with dealing with spatial data such as spatial autocorrelation and the modifiable areal unit problem; and, finally, issues around attempts to establish a causal relationship between neighbourhood exposures and health or behaviour. In this presentation, I will highlight these issues with examples from my work in this area of research.