Research to date has established that there is a relationship between high
mortality rates and area deprivation in Britain. However, the majority of t
his research has looked at the regional level and the few studies that have
looked at a smaller area level have tended to focus on London or the North
. At a national level a relationship between housing tenure and mortality h
as also been found. This paper considers the relationship between mortality
and place and in particular housing, at ward level in a city in the South
East of England. It is found that, in Oxford, there is no straightforward r
elationship between housing tenure and mortality rates. Rather, it is pocke
ts of poverty within Oxford, expressed in different types of housing, that
are associated with high mortality rates. Whilst the very poorest live in t
he worst quality and least preferred housing, the extent of this relationsh
ip is likely to differ in different areas, according to the historical and
current patterns of housing provision. A method of identifying such areas n
ationally is needed, as a reliance on national studies, particular those on
tenure, is likely to obscure the true picture and extent of geographical i
nequalities in health. (C) 1999 Elsevier Science Ltd. All rights reserved.