The geography of inequalities in access to healthcare across England: the role of bus travel time variability

TL;DR

The paper tackles the problem of accurately measuring access to healthcare by incorporating travel time variability (TTV) in public transport, using England-wide bus timetable data and a routing engine to compute hourly travel times from every LSOA to the nearest hospitals and GPs. It introduces a TTV metric defined as the standard deviation of travel times across hourly departures, and analyzes spatial patterns and inequality at both the local (LSOA) and regional (LAD) scales, employing Moran’s I, LISA, and Gini coefficients. Key findings show strong spatial clustering of TTV, an urban–rural divide with rural areas experiencing higher TTV and longer mean times, and nuanced relationships between TTV and deprivation that vary by region and destination type; a four-way categorization of TTV/IMD distributions helps identify targeted policy needs. The study highlights the value of incorporating dynamic, timetable-based TTV into accessibility assessments while acknowledging the gap to real-time operational data, and it suggests that such dynamic measures can inform transport policy aiming to level up access to essential healthcare services.

Abstract

Fair access to healthcare facilities is fundamental to achieving social equity. Traditional travel time-based accessibility measures often overlook the dynamic nature of travel times resulting from different departure times, which compromises the accuracy of these measures in reflecting the true accessibility experienced by individuals. This study examines public transport-based accessibility to healthcare facilities across England from the perspective of travel time variability (TTV). Using comprehensive bus timetable data from the Bus Open Data Service (BODS), we calculated hourly travel times from each Lower Layer Super Output Area (LSOA) to the nearest hospitals and general practices and developed a TTV metric for each LSOA and analysed its geographical inequalities across various spatial scales. Our analysis reveals notable spatial-temporal patterns of TTV and average travel times, including an urban-rural divide, clustering of high and low TTV regions, and distinct outliers. Furthermore, we explored the relationship between TTV and deprivation, categorising LSOAs into four groups based on their unique characteristics, which provides valuable insights for designing targeted interventions. Our study also highlights the limitations of using theoretical TTV derived from timetable data and emphasises the potential of using real-time operational data to capture more realistic accessibility measures. By offering a more dynamic perspective on accessibility, our findings complement existing travel time-based metrics and pave way for future research on TTV-based accessibility using real-time data. This evidence-based approach can inform efforts to "level up" public transport services, addressing geographical inequalities and promoting equitable access to essential healthcare services.

Figures (13)

• Figure 1: Diagram showing how R5 routing engine works. A typical calculated journey begins with a walk from the population weighted centroid of an LSOA (dark blue) to a bus stop, where a bus journey (light blue) starts , and concludes with a walk from the destination bus stop to the final destination.
• Figure 2: Maps of LSOAs in England showing (a) average travel time and (b) travel time variability (TTV) to hospitals and GPs on 30th May.
• Figure 3: Local Indicators of Spatial Association (LISA) clustering map for travel time variability (TTV) to hospitals and GPs from LSOAs in England. High-high clusters (red) are mainly observed in rural areas across the country, while low-low clusters (dark blue) are concentrated in major urban centres (see zoomed in maps of selected metropolitan areas). Additionally, high-low outliers (yellow) and low-high outliers (light blue) are present, though they are less visually prominent on a map of this scale. These spatial patterns highlight the geographical inequalities of TTV, with urban areas generally exhibiting more consistent travel times and rural areas experiencing greater variability.
• Figure 4: Local Indicators of Spatial Association (LISA) clustering map for travel time variability (TTV) to hospitals from LSOAs in a large metropolitan area in North West England consisiting of Liverpool and Manchester. High-low outliers (yellow) can be clearly observed, suggesting these particular LSOAs experience significantly higher TTV than the surrounding LSOAs.
• Figure 5: Scatterplot of travel time variability (TTV) versus average travel time for trips to hospitals and GPs from each LSOA in England, with points color-coded by settlement type (urban/rural). The plot reveals a general positive correlation between TTV and average travel time, with urban areas exhibiting lower values for both metrics, while rural areas tend to show higher variability and longer travel times. This pattern is less distinct for hospital access with some urban areas exhibiting high average travel times while certain rural areas displaying low TTV, suggesting access to hospitals is influenced by additional factors beyond the rural-urban classification, including the limited number of hospitals and their uneven geographical distribution.Overall, the distribution highlights the urban-rural divide in accessibility from both dimensions.