Can we empirically derive a geographic definition of 'coastal' for use in cancer data reporting? An ecological modelling study using England's national cancer registry.

Background: Reducing avoidable systematic differences in population health requires first understanding which populations are currently disadvantaged. Although the health of coastal communities in England has been of concern for some years, an operationalised definition of 'coastal' is lacking. This study aims to use national cancer statistics to define and validate a small area-level definition of 'coastal' that could be used to better report cancer-related health inequalities in England. Methods: Information on the geography and demography of English populations at the Lower Super Output Area (LSOA) level were used to define a suite of candidate coastal variables that considered foreshore proximity, resident population location, rurality and deprivation. Adjusted linear models of LSOA-level statistics of cancer incidence, prevalence and mortality in England (2016 to 2020) were used to identify candidate coastal variable(s) that explained the greatest proportion of variation in cancer outcomes after adjustment. Results: The candidate 'G_25_5' (LSOA's designated as 'coastal' if 25% or more of postcodes were within 5 km of the coastline) was selected as the candidate that explained the most residual variation in cancer incidence and prevalence after adjustment. This variable would assign 7377 2011 LSOAs as coastal, whose populations summed to 12.3 million people (22% of England's population, in 2016). This candidate variable was not significantly associated with cancer mortality. Conclusions: The coastal variable that we identify can explain some of the 'coastal excess' in poor cancer outcomes. We propose that this variable is now embedded into health inequalities reporting and adopted as the working definition of 'coastal' implicated in NHS England's 'Core20PLUS5' approach for use in cancer data reporting.