Duncan Lee and Rich Mitchell have been awarded a grant from ESRC to try and solve a problem with the methods used to investigate spatial variation in health.
There is great interest in how and why the risks of good or poor health seem to vary from place to place. There are lots of studies which have tried to work out how the characteristics of places and the people who live there may explain why health is worse in some areas than in others. When scientists do this kind of work, they need to take into account the fact that places which are physically closer together tend to be similar to each other; this similarity can have an influence on the statistical tests used and if this problem is ignored, it can lead to the wrong conclusions about how the characteristics of places and the people who live there are related to risks of poor health.
There is a problem however, in that the existing techniques are not sophisticated enough to tell the difference between places that are right next to each other and which actually are very similar, and those which are right next to each other but which are actually quite different. This is a particular problem in cities, where you can get two neighbouring areas that have very different characteristics.
In this project, we intend to develop a method which is able to spot when areas are close together but have different characteristics and when areas which are close together are indeed quite similar. The method will then apply the right kind of statistical approach in each situation. Our project has three parts. First, we will develop the new method. Then we will test it in a way that allows us to see how much better our new approach is than the existing technique. Finally, we will use the method to look at how and why the risks of three different health problems vary in the central belt of Scotland. The health problems will be alcohol problems, lung problems and breast cancer. For each of these health problems we will use our new technique to explore what characteristics of places and the people who live in them, might raise or lower risk. Rather than discovering new factors that might be involved in the risks of these disease, we expect to be able to gain a better understanding of the relative importance of factors which have already been identified. This might help the health service to know what level of ill health to expect in an area, given its characteristics, and hence plan services better, or it might help to identify what aspects of places and their residents need to be changed or helped to reduce risk of poor health.
So, overall the project will make a contribution to methods in this kind of science and will also provides a useful study of some big health problems.
The project starts in October 2010 and will run for 2 years.