Elsevier

Health & Place

Volume 8, Issue 1, March 2002, Pages 3-13
Health & Place

Increasing the sophistication of access measurement in a rural healthcare study

https://doi.org/10.1016/S1353-8292(01)00031-4Get rights and content

Abstract

This paper considers the problem of deriving realistic access measures between population demand and health service locations, in the context of a rural region of England. The paper reviews approaches used in earlier work by the authors and others, and considers new public transport information systems that are now becoming available. An application is presented which incorporates the modelling of both private and public transportation travel times for access to district general hospitals in Cornwall. This information has been assembled from published timetables in order to evaluate the use of more sophisticated access measures that might be used when such data becomes more generally available. The work is set within the context of an ongoing substantive research programme concerned with health outcomes in the rural South West of England.

Introduction

This paper examines the use of a series of increasingly sophisticated measures of access to health care locations, based on the authors’ previous work on rural inaccessibility in the South West of England. Beginning with simple and widely used indicators, we consider the use of a geographical information system (GIS) for the measurement of potential accessibility in various ways, and identify the particular limitations and weaknesses of such approaches. The empirical sections of the paper demonstrate our population surface-based approach to the integration of private and public transport access measures, and include a preliminary analysis of the implications for the health outcome modelling using the example of limiting long term illness. We anticipate an increasing ability to incorporate public transport information with the emergence of online timetable systems.

The rest of this paper comprises six sections: first, we briefly review the types of measures of physical access that have seen some usage in health studies. In section three, the very real GIS implementation challenges of constructing adequate representational models using these measures are considered in some detail. The need for the incorporation of public transport information is considered in section four and some current developments in timetable information systems described. We then illustrate these issues with reference to a study of rural accessibility and deprivation in the South West of England, from which the dimensions of rurality and deprivation are discussed more fully in Martin et al. (2000), and Barnett et al. (2001). The modelling includes the use of published public transport information for the County of Cornwall, in order to evaluate how imminent digital data sources may be incorporated into access modelling. In section six, the potential impact of the different approaches on studies of health outcomes is considered in relation to the distribution of limiting long term illness (LLTI). The paper concludes with some observations on the potential to take advantage of advances in timetabling systems and the remaining obstacles which need to be overcome in increasing the sophistication of access measurement in health care studies. In particular, we consider the extent to which simple and more sophisticated measures of accessibility may display the same relationships with ill health.

Section snippets

Accessibility measurement

Even the briefest review of the health care literature reveals that the term ‘accessibility’ as applied to health care services is used in many different senses. It encompasses concepts of the quality and variety of services provided, referral mechanisms, waiting lists and physical access to services across the transportation network, and has been characterized by Rogers et al. (1999) as ‘providing the right service at the right time in the right place’. In this section we review some of the

Accessibility modelling

In this section we consider the construction of a GIS data model for the measurement of accessibility to health care service locations which may provide the access variables required in more substantive health studies. Lovett et al. (2000) attempt to approach a very similar general problem to that explored here. Their work provides a useful contrast to some aspects of the approach adopted here, and provides a similar application to a different geographical region, also in the context of a rural

The use of public transport information

Although representing a significant increase in sophistication relative to crow-fly or indirect indicators of accessibility, an important failing of many approaches to the modelling of road networks and drive-times is that they relate only to travel by private car. For many of the most vulnerable groups, who are often those of greatest interest in health care studies, the availability of public transportation access measures is more important, although the issues involved may be complex.

Modelling rural accessibility in the South West of England

When the number of demand points is very large (or indeed demand is spatially continuous) there are good arguments in favour of producing an accessibility surface rather than computing individual network-based journey times, and this approach has been followed here: this modelling scenario is illustrated at various stages in Fig. 1. In order to assemble travel cost surfaces for the South West Region, as required by the study of which this paper describes one part, 1991 Census populations were

Impacts on health modelling

In this section, we briefly consider the relationships between the alternative access measures and a health outcome measure, taking as an example LLTI as recorded by the 1991 Census. An objective of our ongoing substantive work is to examine the effects of each of these alternative access measures within the multi-level modelling framework described by Barnett et al. (2001), and to examine a range of health outcomes including LLTI, mortality and key conditions identified from health episode

Conclusions

Despite the apparent lack of association between the more sophisticated distance measures and the LLTI ratio, perhaps the most important point is to recognise that the same relationships observed for population density and crowfly distance do not hold when the more complex measures are used. Undoubtedly, the car and bus travel times modelled here are much more strongly related to the actual transportation barriers to be overcome when travelling to the nearest hospital, and population density or

Acknowledgements

The work reported in this paper was supported by South and West Regional Health Authority research and development grant C/MV/20/04.97/Roderick. S Barnett was in receipt of Medical Research Council studentship G610/47. The authors acknowledge the helpful comments of two anonymous referees.

References (23)

  • G. Bentham et al.

    Evaluation of a mobile branch surgery in a rural area

    Social Science and Medicine

    (1992)
  • E.B. Parker et al.

    Measuring access to primary medical caresome examples of the use of geographical information systems

    Health and Place

    (1998)
  • J. Ball et al.

    Visualizing stochastic catchments in geographical networks

    The Cartographic Journal

    (1994)
  • S. Barnett et al.

    A multilevel analysis of the effects of rurality and social deprivation on premature limiting long term illness

    Journal of Epidemiology and Community Health

    (2001)
  • P.J. Boyle et al.

    Geographical variation in the referral of patients with chronic end stage renal failure for renal replacement therapy

    Quarterly Journal of Medicine

    (1996)
  • I. Bracken et al.

    Linkage of the 1981 and 1991 Censuses using surface modelling concepts

    Environment and Planning A

    (1995)
  • Cornwall Community Health Council, 2000. Patients on wheels: transport and access to health services in Cornwall Truro:...
  • Cornwall County Council, 1999. Cornwall public transport timetable 1999–2000 Truro, Cornwall County...
  • A.C. Gatrell et al.

    Health and health care applications

  • A.P. Jones et al.

    Accessibility and health service utilization for asthma in Norfolk

    England Journal of Public Health Medicine

    (1998)
  • A.E. Joseph et al.

    Accessibility and utilizationgeographical perspectives on health care delivery

    (1984)
  • Cited by (0)

    View full text