Background
How rurality is defined matters, both from a policy and service delivery perspective, and for the populations and communities who live in rural places.1 Definitions of rurality have an important role in health contexts. A fit-for-purpose definition allows the health status of rural populations to be accurately monitored. This may identify rural-urban health inequities, providing the impetus for targeted strategy, policy and interventions for the equitable allocation of resources.2-6 There is, however, no internationally agreed definition of ‘rural’. Definitions are dependent on context, can change over time, and have become increasingly blurred.1 To date, Aotearoa New Zealand (NZ hereafter) has lacked an urban-rural classification designed for use in health research and policy.
Defining rural in geography
Geographers have long contested the definition of rurality.7 The two main ways of conceptualizing and defining rurality are: (1) a socio-cultural approach; and (2) a descriptive, data driven approach.8 Socio-cultural definitions consider particular cultural characteristics of communities in order to define places as rural or urban.9 Descriptive approaches are technical and quantitative ways to empirically describe socio-spatial characteristics, and classify places as rural if they meet pre-defined criteria.8 Both approaches have limitations, particularly when used alone, and can result in anomalies. Socio-cultural approaches assume that population density affects behaviour and attitudes and that values and behaviours differ between rural and urban residents, despite contradictory evidence such as supposedly ‘rural’ communities in the East End of London.8 Conversely, descriptive definitions of rurality have been strongly critiqued as providing an inadequate view of the social construction that is rurality.1, 10, 11 It is also argued that they assume a clear geographic distinction between rural and urban areas, when in fact borders are often blurred, contested, and subjective,11 and complex ‘black-box’ methods reify methodologies and lend results a misleading aura of authenticity.1 The core concepts and measures of rurality have not changed markedly since the 1970s.1 Population size and proximity to metropolitan areas are still the most commonly used input variables in rurality measures. However, there is growing recognition that ‘rurality’ is a fluid concept that is context-dependent.1 Rurality is increasingly viewed as a social construct that is defined by discourse.8 Social representation is used to consider how people construct themselves as being rural, recognizing that rurality is ‘a state of mind’ and in the eye of the beholder.8 The development of a meaningful classification of rurality needs to effectively balance ‘technical’ and ‘discourse’ approaches so that classifications are the result of a clear, transparent, and replicable process, and also make sense on the ground. Heuristic, on the ground understandings of rurality are as important as purely geographic data-driven approaches, and the plurality of rural must be embraced.9
The statistician George Box’s argues that “all models are wrong, but some are useful”.12 Descriptive, technical approaches are important, especially in research and policy contexts where quantitative measures are needed to consistently define populations or designate policies.1 However, it is somewhat naïve to think that a purely technical approach will produce the most fit-for-purpose classification of rurality – a concept which is multifaceted, nuanced, and dependent on context.1, 7-9
Defining rural for health
Internationally, a wide range of approaches to describing areas as urban or rural in a health context have been taken by both researchers and government departments. In the United States of America, there are five key measures of rurality for epidemiological studies, all based on a combination of population size, density, and distance or commuting patterns. 13 Canada has at least four different rurality classifications used in health research - all based on a combination of population size, density, and distance.14 While exact thresholds cannot be universally applied, factors of population size, density, and distance are key considerations in international geographic classifications of rurality. For example, the Modified Monash Model (MMM) was developed in Australia as a framework for distributing rural health workforce recruitment and retention funding. Although six indicators of rural and remote workforce retention were examined, population size and distance to metropolitan centre were found to be more sensitive indicators of the need for recruitment and retention incentives, leading to a more parsimonious classification.6
The United States Rural Policy Research Institute15 offers guidance on addressing the plurality of rurality when developing a classification for health. It acknowledges that a transparent data-driven geographic approach is preferred to intuition or personal experience. However, it also argues that geographic approaches must be combined with qualitative evaluation and ‘ground truthing’ to ensure the final classification has face validity. As NZ’s Minister of Health, Andrew Little, noted in his keynote address at the 2021 NZ National Rural Health Conference,16 the definition of rural is “not just semantic” and has real implications in terms of policy decisions and resource allocation. Poorly defined rural-urban divisions lead to poorly defined and implemented policies.1 While descriptive approaches remain an important aspect of classifying rurality to allow meaningful analysis of health data, and thereby inform health policy,1 they must involve a qualitative aspect to ensure face validity.9
Defining rural in the NZ health context
The definition of rurality is also an essential component of research that explores rural-urban health inequities. Such inequities, exacerbated by deprivation and ethnicity, have been clearly described internationally.17-19 However, the same inequities have not been as clearly demonstrated in NZ. Many health practitioners, academics, and other informed stakeholders argue that this is due to the definitions of 'rural' used. Internationally, it has been demonstrated that different rurality classifications result in inconsistent categorization of areas and populations, impacting the results of epidemiological studies and health service research and thereby potentially masking inequities.20-25 This is an example of the Modifiable Area Unit Problem (MAUP), which has shown that the results of analysis can vary according to the size, number and configuration of spatial units that are used.26 Furthermore, aggregation methods can greatly influence results, even when the size of the total population is similar between methods.27 Similarly, the choice of rurality classification also influences results, as different classifications aggregate together different populations into rural or urban categories. This is problematic in NZ, where a wide range of usually generic definitions of rurality, have been used in the health research literature.28
A review of the last 20 years of NZ health research28 revealed the use of over 30 classifications of rurality. This complicates discussions of rural health, and highlights the need for a clear, consistent, and fit-for-purpose definition of rurality in NZ that can be routinely used to accurately monitor rural-urban differences in health outcomes. Statistics New Zealand’s (Stats NZ) Urban Rural Experimental Profile (UREP)29 is a commonly used rurality classification in the NZ health literature. Although, the National Health Committee in 2010 found little difference in health outcomes between rural and urban residents,30 this conclusion is likely an artefact of the way the UREP was used in their analysis.31 When the UREP was modified to better represent ‘rural’ as understood by NZ’s rural health community, the relative incidence of heart disease in the ‘rural’ population increased from 62% to 166% of the urban incidence.30-32 One issue is that, within an urban-rural binary, the UREP classifies as ‘independent urban communities’ numerous places that are, in a health discourse, invariably considered rural. Independent urban communities are often rural towns with a considerably smaller population than metropolitan centres. They do not have a significant functional relationship with main urban areas, and tend to have higher levels of socioeconomic deprivation. The UREP category ‘Rural areas with high urban influence’ is also problematic. These places are, generally, NZ’s most affluent, with ‘a significant proportion’ of residents working in the adjacent ‘large urban area’.29 In 2018, Stats NZ, updated its Statistical Standard for Geographic Areas (SSGA18)33 creating Statistical Area 1s (SA1s) as the smallest geographic unit for census population data. In 2020 Stats NZ’s Urban Accessibility (UA) classification34 replaced the UREP. The UA was designed to recognize the impact that proximity to urban centres has when determining gradations of rurality. However, the UA remains a ‘generic’ classification that was not specifically designed for health outcome analyses, and complexities around rural and urban fringes11 have not been considered from a health perspective. The UA is therefore likely to continue masking rural-urban health inequities.
There has been a pressing need for a rural-urban classification which supports the consistent analysis of national health data and is likely to have significant uptake and use among health researchers and policymakers. The objective of this paper is to describe the development and validation of a Geographic Classification for Health (GCH) that is not only descriptive and technically robust for use within policy and research contexts, but also aligns with a heuristic sense of what is understood to be rural in the NZ health context.