Car travel time and accessibility by bus to general practitioner services: a study using patient registers and GIS
Introduction
One of the guiding principles of the National Health Service in the UK, as in health care services in many other countries, is the ideal of equal access to health services for those in equal need. This is an impossible objective from a geographical point of view. Health services are inevitably concentrated in particular places, and are therefore more accessible to people nearby than to people who live further away. In this paper, we assess the differences in physical accessibility within a large rural population. Inequality in accessibility due to distance is, of course, just one aspect of the problem of equal access to health services. Access to services can be represented as a continuum, in which availability, accessibility, acceptability, contact and effectiveness all play a part (Hongvivatana, 1984; Ricketts & Savitz, 1994). In the real world, the proportion of people who benefit from health services is determined by variations in the availability or effectiveness of services and by difficulties in using them. Difficulties in use are caused by values and beliefs, social and cultural factors and, in some systems, the inability to pay. They are also caused by the difficulty of overcoming physical distance, and this tends to be a dominant factor in rural regions (Joseph & Bantock, 1982). Poor physical accessibility is known to reduce the use of services, and perhaps it leads to poorer health outcomes (Joseph & Phillips, 1984; Haynes, 1986; Watt, Franks, & Sheldon, 1993; Jones & Bentham, 1997). Low utilization of primary care services is of particular concern because primary care often acts as a gateway to secondary and tertiary services, and reduced use of general practitioner services might lead to lower use of hospitals at a later stage. More than 20% of the population of Britain lives in rural areas, but the extent and consequences of variations in physical accessibility to rural primary care services are poorly understood (Cox, 1995).
There is no accepted measure of physical accessibility. It is commonly measured in terms of distance to services or the costs of travel, but neither of these can fully capture the differences between ease of access in travelling by private car or public transport, on a motorway or on a minor road (Moseley, 1977). Arguably, the measure closest to people's experience is that of travel time. Recent developments in computing technology and digital map databases have made it possible to use Geographical Information Systems (GIS) to calculate travel times between locations in a more automated and sophisticated manner than was previously practical. An early study in North America used a raster-based GIS system to make travel time estimates for a grid of cells from road network overlays, adding individual cell values to produce estimates of car travel times to a particular hospital location (Furbee & Spencer, 1993). More recently, vector-based GIS have been widely employed to estimate travel times from the speed and other attributes of segments in road networks (e.g., Walsh, Page, & Gesler, 1997; Higgs & White, 1997; Naude, deJong, & van Teeffelen, 1999; Bateman, Lovett, & Brainard, 1999). The research described here also employs a vector-based GIS to calculate car travel times, but adds a further dimension by considering accessibility via scheduled bus services as well. Another innovation was to use GP patient registers to represent the spatial distribution of the population utilising primary health care services. These records provide a more current and spatially disaggregated (important in a rural region) source of demographic information compared to the 1991 Census. There have been suggestions that GP patient registers in some part of Britain (particularly large conurbations) are not sufficiently complete and reliable to be used for research purposes, but recent studies indicate that this is not now a problem in East Anglia. (Haynes, Lovett, Bentham, Brainard, & Gale, 1995a; Lovett et al., 1998).
The research therefore used the combination of GIS techniques and patient registers to evaluate the accessibility of the whole population of East Anglia (Cambridgeshire, Norfolk and Suffolk) to primary health care services. The study area is primarily rural, with four main urban centres (populations in the range 100,000–200,000) and several smaller towns (Fig. 1). The full study included GP surgeries, dental services and pharmacies, but only the results relating to GP services are reported here. The objectives were to develop new methods for measuring access to primary health care services, to compare patterns of access to services by public and private transport and to investigate the socio-economic characteristics of populations with the poorest access to services.
Section snippets
Primary care facility locations
Details of GP surgeries were obtained from the four health authorities covering the study area. Information was obtained on provision existing in Autumn 1997 and all the surgery addresses were converted to 100 m resolution grid references via the 1997 release of the Postcode to Enumeration District Directory (Gatrell, Dunn, & Boyle, 1991; Raper, Rhind, & Shepherd, 1992; Martin, 1992). In total, the database contained details of 481 main or branch surgeries. No distinction was made between main
Matching surgery locations to the road network
The Arc/Info GIS software used in the research stored a road network as a series of digitised lines (often termed arcs or segments) which connected to each other at nodes. The nodes therefore typically represented junctions or points where the road changed from one type to another (e.g., from single to dual carriageway). In order to calculate travel times along roads it was necessary to assign each GP surgery to its nearest node on the network. All of the longer arcs were split so that the
Travel times to the nearest surgery
Fig. 5 shows the distribution of GP surgeries in the study area and around its margin. Concentrations of surgeries in the urban centres are evident. Many of the smaller towns had several main or branch surgeries, and most of the larger villages had at least one surgery. Fig. 5 also shows the estimated car travel times to the nearest GP surgery. All the main centres of population were within 5 min of a surgery, and a journey time of 10 min included almost all settlements on the network of A and B
Conclusions
For the majority of the population, access to general practitioner services was relatively good in this rural part of the UK. Fully 90% of the population in Autumn 1997 lived within 10 min car drive of primary medical care services. Most residents could reach a GP surgery by bus, although for 13% of the population this was not possible. For 5% of the population, the car journey to the nearest GP surgery was longer than 10 min and there was no suitable bus service each weekday. A total of 114,175
Acknowledgements
This research was funded by the NHS Executive Eastern Region. We thank the following for their assistance with data or guidance: Bob Beaumont, Suffolk County Council; Chris Beer, Cambridgeshire County Council; Julian Bester, East Norfolk HA, Chris Edmunds, Suffolk HA; Amanda Hayes, Cambridge & Huntingdon HA; David Lea and Caroline Murphy, North West Anglia HA; and Liz Joyce, Norfolk County Council. For 1991 Census data, we acknowledge the source: The 1991 Census, Crown Copyright, ESRC Purchase.
References (40)
- et al.
Using isochrone surfaces in travel-cost models
Journal of Transport Geography
(1997) - et al.
Defining localities for health planningA GIS approach
Social Science and Medicine
(1996) - et al.
Changes in service provision in rural areas. Part 1The use of GIS in analysing accessibility to services in rural deprivation research
Journal of Rural Studies
(1997) - et al.
Measuring potential physical accessibility to general practitioners in rural areasA method and case study
Social Science and Medicine
(1982) - Bateman, I. J., Brainard, J. S., & Lovett, A. A. (1995). Modelling woodland recreation demand using geographical...
- et al.
Developing a methodology for benefit transfers using geographical information systemsModelling demand for woodland recreation
Regional Studies
(1999) - et al.
Limiting long term illness and its associations with mortality and indicators of social deprivation
Journal of Epidemiology and Community Health
(1995) - et al.
ScreeningThe inadequacy of population registers
British Medical Journal
(1989) - et al.
Linkage of the 1981 and 1991 UK Censuses using surface modelling concepts
Environment and Planning A
(1995) - Cambridgeshire County Council. (1997a). Cambridgeshire passenger transport map 1997/8. Cambridgeshire County Council,...
Accessing the data via SASPAC91
Using GIS to determine travel times to hospitals
Geo Info Systems
The relative utility of the central Postcode Directory and Pinpoint Address Code in applications of geographical information systems
Environment and Planning A
The inverse care law
Lancet
The geography of health services in Britain
Cited by (222)
Mapping and measuring urban-rural inequalities in accessibility to social infrastructures
2024, Geography and SustainabilityPassive transit accessibility: Modelling and application for transit gap analysis and station area assessment
2024, Journal of Transport GeographyOn determining the weight of edges in map-representing graphs-applications of heuristic methods in planning escape routes
2022, Journal of Traffic and Transportation Engineering (English Edition)