Building a health and environment geographical information system :an evaluation, looking at childhood cancer in Northern England

Abstract

The aim of this research was to evaluate a relatively young technology, Geographical Information Systems (GIS), in a specific applications environment. The application adopted was that of searching for environmental causes of childhood cancer, in particular that of Acute Lymphoblastic Leukaemia (ALL), in Northern England. It is also relevant in terms of the WHO's intention to develop a Health and Environment GIS, and therefore the research aims to satisfy their recommendations for pilot studies. The subject matter of this thesis therefore covers two very high profile topics, which it is believed will mutually benefit from the research carried out. Firstly, very little is known about the aetiology of ALL, and thus any new methodology which is introduced to help analyse sensitive issues of causation is welcomed not only by those in the medical field but also the public. The application was made possible with the provision of detailed cancer data for Northern England and a weak but interesting hypothesis that environmental factors may be an attributable mechanism for causation. Key questions which are asked include; Where are incidences of ALL located? Why are they there? Is there a cluster? and What could be the cause? Secondly a Geographical Information System, in this case the proprietary software package ARC/INFO, was considered an excellent medium for tackling this spatial epidemiological problem. Especially with its capability to store large volumes of diverse data, and its inherent flexibility to deal with spatial information pertaining to health and environmental factors. More importantly the application itself offered a means of evaluating the implementation of a GIS. Establishing the advantages and pitfalls which accompany all stages of 'The GIS Process' and an invaluable documentation of the experiences acquired as an initiator, developer and implementor of this new technology. In addition, this research offers fresh ideas and techniques for improving those areas of the technology which appear to be lacking in these early phases of its development. The problems of spatial analysis in GIS and the provision of useful tools such as 'pattern spotters', 'relationship seekers' and 'error handlers' are discussed as alternative techniques. To ensure an exciting future for GIS technology in application environments the latter and other key areas of research which should be persued are highlighted in this thesis.