Effective public health interventions should narrow inequalities by having a positive impact on the health of populations most in need. However, this often does not occur, and interventions may sometimes widen inequalities. These intervention generated inequalities may be more pronounced for certain classes of intervention, such as those reliant on voluntary behaviour change, which may result in lower uptake among more disadvantaged groups. This has been shown in interventions such as those to promote breast feeding, seatbelt use and cancer screening. Furthermore, the effectiveness of interventions may be socially patterned, as found by studies of smoking restrictions and bicycle helmet legislation. These observations are of particular concern given the potential existence of ‘deprivation amplification’, whereby the most socioeconomically disadvantaged experience poorer health and poorer access to resources. A version of the ‘inverse care law’ may therefore operate for many interventions, whereby the provision of facilities that are spatial in nature may be poorest in areas with greatest need. One way of investigating such issues within evaluations is to explore the underlying geographical distribution of resources in relation to the context of the areas in which they are provided.
Achieving social equity in environmental planning and decision making processes is increasingly recognised as an important component of environmental policies in many countries. Building on the concept of environmental inequity in the USA – defined as the apparent unequal geographical distribution of benefits or burdens among those in poverty or for minority groups – early work focused on consideration of disamenities such as hazardous facilities[14, 15]. Subsequently, access to amenities has been considered, with studies examining the locations of parks, playgrounds[17, 18] and public health facilities, often suggesting that disadvantaged populations may have poorer access to given resources. A valuable component of environmental equity is therefore that of spatial equity.
An important current area of research is the exploration of the provision of resources that support physical activity. People who engage regularly in physical activity are less likely to experience a range of preventable chronic conditions including obesity. Consequently, interventions have been developed in order to increase activity levels in target groups or populations with mechanisms including the use of information, targeted behavioural and social programmes, and changing the physical environment and planning policy associated with it. For example, interventions to encourage walking and cycling have included publicity campaigns to increase awareness, financial incentives, and improvements to footpaths and cycle routes. Disadvantaged groups, such as those with low socio-economic status and ethnic minorities, often have higher levels of obesity and have sometimes been shown to live in areas with poorer access to facilities for physical activity or healthier eating. This emphasises the importance of understanding how need and provision vary spatially so that interventions can be located to serve these high-need yet often overlooked populations. Nevertheless, decisions about where to locate infrastructure can be difficult to reconcile with the location of target populations due to issues of land availability and other context-specific circumstances. There remains a paucity of evidence regarding the delivery of effective interventions due to limited evaluation of interventions.
Methods using geographic information system (GIS) technology may be employed to assess the spatial equity of amenities, resources and infrastructure. Studies have calculated the distances between residential locations and amenities and the number of facilities available per capita. However, the utility of these findings is limited, as not all have considered the underlying geography of need, so it is not always known if a facility favours or disadvantages a certain type of population. Some studies have addressed this; Nicholls[17, 29] evaluated the locations of disadvantaged populations (non-white populations, children, economically disadvantaged) with and without access to parks within 800m of home, finding that parks tended to be located in areas with greatest need. Such literature is generally concerned with the location of existing infrastructure rather than the evaluation of recent modifications to the built environment. Nevertheless, method and learning from these studies may be used to inform the evaluation of infrastructural interventions to target population health behaviours.
Process evaluation is central to determining whether interventions perform as intended and can therefore aid understanding of how the context within which they are developed may affect their eventual effectiveness. Spatial equity is the first step in a process towards reducing health inequality via structural or area-based interventions and should therefore be evaluated accordingly. If the intervention does not achieve a basic level of ‘availability’ through accurate targeting, then the next steps - uptake, efficacy, long-term compliance and health outcomes - are unlikely to be achieved. It is thus important to examine the potential reach of interventions, and the implications of their presence in order to give insight and possible explanation for the outcomes. UK guidance recommends that components of planning, implementation and operation should be reviewed as a necessary precursor to a full evaluation of health interventions to understand the local context, at whom a given intervention is aimed, and the components of an intervention. It has been suggested that evaluations might be further strengthened by integration of geographic data and we argue this is particularly so for spatial equity analysis. This may ultimately be used to give an early indication that an intervention might not eventually be effective because it is poorly targeted.
In this paper we develop and apply methods to evaluate how the spatial location of infrastructure relates to the underlying geography of population need, using the case study of a recent government-funded programme in England, ‘Healthy Towns’. This government-funded programme aimed to provide interventions to encourage dietary and physical activity behaviour change to combat obesity. We investigate whether infrastructure developed from this was best located in relation to areas of need, according to the socio-demographic characteristics of neighbourhoods.