To investigate whether the Index of Multiple Deprivation 2000 (IMD) is more strongly related to inequalities in health in rural areas than traditional deprivation indices. To explore the contribution of the IMD domain "geographical access to services" to understanding rural health variations.