Locality deprivation and Type 2 diabetes incidence: A local test of relative inequalities

Abstract

There is increasing evidence that the socio-spatial context of the local area in which one lives can have an effect on health, but teasing out contextual influences is not a simple task. We examine whether the incidence of Type 2 diabetes in small areas in Tayside, Scotland is associated with deprivation in neighbouring areas, controlling for the deprivation of the area itself. As such, this is a genuinely ‘contextual’ variable situating each small area in the context of surrounding places. We test two opposing hypotheses. First, a ‘psycho-social’ hypothesis might suggest that negative social comparisons made by individuals in relation to those who surround them could lead to chronic low-level stress via psycho-social pathways, the physiological effects of which could promote diabetes. Thus, we would expect people living in deprived areas surrounded by less deprived areas to have an increased risk of diabetes, compared to those living in similarly deprived areas that are surrounded by equally or more deprived areas. Alternatively, a neo-materialist approach might suggest that the social, cultural and environmental resources in the surrounding environment will influence circumstances in a particular area of interest. Poorer areas surrounded by less deprived areas would benefit from the better resources in the wider locality, while less deprived areas surrounded by poorer areas may be hampered by the poorer resources available nearby. We refer to this as the ‘pull-up/pull-down’ hypothesis. Our results show that, as expected, area deprivation is positively related to diabetes incidence (p<0.001), whilst deprivation inequality between areas and their neighbours is negatively related (p=0.006). Type 2 diabetes is more common in deprived areas, but lower in deprived areas that are surrounded by relatively less deprived areas. On the other hand, less deprived areas that are surrounded by relatively more deprived areas have higher diabetes incidence than would be expected from the deprivation of the area alone. Our model results are consistent with a pull-up/pull-down model and lend no support to a ‘psycho-social’ interpretation at this local scale of analysis.

Introduction

The prevalence of Type 2 diabetes is rising in the UK (Gatling et al., 1998; Harvey, Craney, & Kelly, 2002) and throughout the developed world (Amos, McCarthy, & Zimmet, 1997; Passa, 2002; Wild, Roglic, Green, Sicree, & King, 2004). It occurs when the cells of the body become less sensitive to the action of the hormone insulin, when the body fails to produce enough insulin or, as is most often the case, as a combination of these two factors. Insulin is the key hormone involved with the storage and controlled release of the chemical energy available from food. Without insulin, glucose circulating in the blood is not taken up by the cells, which results in a chronic state of hyperglycaemia (increased blood glucose). In the short term, hyperglycaemia will manifest in extreme tiredness, weight loss, thirst and copious urination. However in the long term, hyperglycaemia results in micro- and macro-vascular damage and leads to a host of serious health problems including blindness, kidney failure, cardiovascular disease, and foot ulcers that, if infected, can lead to amputation. These serious consequences make Type 2 diabetes a key public health concern.

The underlying aetiology of Type 2 diabetes remains unclear, although the development of the condition has been shown to have a genetic component (Kaprio et al., 1993). Various individual risk factors have been identified, including poor diet and lack of exercise (UK Prospective Diabetes Study IV, 1988) and, as with many diseases, it is well established that the prevalence of Type 2 diabetes is related to material deprivation (Connolly, Unwin, Sherriff, Bilous, & Kelly, 2000; Evans, Newton, Ruta, MacDonald, & Morris, 2000; Meadows, 1995; Whitford, Griffin, & Prevost, 2003). More recently, and of particular relevance to this study, national-level analysis suggests that calorie consumption, obesity and diabetes mortality are positive related to income inequality in 21 developed nations (Pickett, Kelly, Brunner, Lobstein, & Wilkinson, 2005), and this suggests that there may be a relationship between the incidence of diabetes and the presence of socio-economic inequalities.

Our analysis in Tayside seeks, first, to examine whether the incidence (rather than prevalence) of Type 2 diabetes is related to deprivation and, second, whether variations in deprivation within the local context also influence the incidence of Type 2 diabetes. In so doing, we present an analysis which examines a potentially important contextual influence—the role of socio-economic conditions in areas neighbouring a particular place of interest. The premise is that the local ‘socio-spatial’ context that one lives in has a significant impact on health via a complex interplay of social, cultural and environmental interactions which take place within and between local areas. Indeed, ‘deprivation inequalities’ within wards in England & Wales have previously been shown to be associated with limiting long-term illness (Boyle, Gatrell, & Duke-Williams, 1999). This raises the intriguing question of whether living in a poor area surrounded by richer areas is better or worse for one's health than living in a poor area surrounded by similarly poor or worse off areas (Boyle, Gatrell, & Duke-Williams, 2004). Therefore, this paper explores whether the incidence of Type 2 diabetes in small areas is raised or lowered if surrounding areas are more or less deprived.

Two opposing models could be anticipated. A psycho-social interpretation might suggest that social comparisons would lead to worse health in areas surrounded by better off places, other things being equal. A neo-materialist model might argue that the proximity of less deprived areas confers material advantages, which lower the incidence of diabetes. This analysis therefore contributes to the debate concerning the relative importance of these opposing theoretical positions (Adler, 2006; Lynch, 2000; Lynch & Davey Smith, 2002; Lynch, Davey Smith, Kaplan, & House, 2000; Macleod, Davey Smith, Metcalfe, & Hart, 2005; Macleod, Davey Smith, Metcalfe, & Hart, 2006; Wilkinson (1994), Wilkinson (2000a), Wilkinson (2000b)), and stresses the importance of local contextual effects.

The psycho-social interpretation is informed, to some extent, by the ‘income inequality’ debate. Income inequality refers to the difference in earnings between people at the top and bottom of society and this type of national-level measure has been associated with both mortality and morbidity outcomes (e.g. Kaplan, Pamuk, Lynch, Cohen, & Balfour, 1996; Kennedy, Kawachi, Lochner, & Prothrow-Smith, 1996; Ross, Wolfson, Dunn, Berthelot, Kaplan, & Lynch, 2000; Sanmartin et al., 2003; Waldman, 1992; Wilkinson, 1992). Although this broad hypothesis has attracted criticism (e.g. Deaton, 2003; Gravelle, Wildman, & Sutton, 2002), proponents argue that income inequality may be detrimental to health via key psycho-social pathways associated with a person's standing in society, and the hypothesis has been used to explain why substantial health inequalities exist in developed nations, despite the decline in absolute poverty that most have experienced. Wilkinson (1992), Wilkinson (1996), Wilkinson (1999), a key proponent of the income inequality thesis, believes that in the presence of such social inequalities, poorer people in developed societies compare themselves unfavourably with the rest of society and that this comparison is damaging to health. The argument centres on social class identity within the national class structure and he argues that through the social tension and weak social affiliation resulting from such comparisons, individuals who perceive themselves as poor may experience chronic low-level stress as a result of the psycho-social impact of their perceived relative social position. Such low-level stress may result in an unintentional physiological response, akin to flight-or-fight, which can affect a host of neuroendocrinic, physiological and immunological variables (Brunner, 1997). This thesis is particularly relevant to our study as over stimulation of the neuroendocrine pathways through chronic stress could well influence the development of Type 2 diabetes (Brunner & Marmot, 1999) as they play an important role in homeostasis and the mobilisation of energy within the body. It has also been suggested that, in addition to the physiological effects of stress, relative poverty may have detrimental effects on health via ‘comfort’ behaviours (e.g. smoking or high alcohol consumption) and the inability to affect positive lifestyle change as a result of the lack of control operated over one's own life. Other such behaviours associated with relative poverty include sedentarism, increased calorie intake, and poor food choice which are also thought to diabetogenic (Pickett et al., 2005). It therefore seems relevant to examine the role of relative inequalities in relation to diabetes incidence.

Most previous studies that have identified relationships between health outcomes and income inequality have been conducted for large geographical areas, such as nations (Wilkinson, 1996), American states (Kennedy, Kawachi, & Prothrow-Smith (1996a), Kennedy, Kawachi, & Prothrow-Smith (1996b)), or metropolitan areas (Lynch et al., 1998). Much of the income inequality argument rests on the comparisons that individuals are expected to make with the rest of society: Wilkinson believes that social comparisons centre on social class identity within the national class structure, which are best measured across whole societies. However, social comparisons may occur at different scales and relate to different processes (Atkinson & Kintrea, 2004) and it seems intuitively plausible that people's perception of their social status and position will also be formed in relation to those they live among in local neighbourhoods. Indeed, there is mounting evidence from multi-level analyses of the significant role of the neighbourhood context in health outcomes (Pickett & Pearl, 2001). Therefore it would seem entirely possible that the psycho-social factors implicated in the development of diabetes at the societal level are also relevant in the local context—if people compare themselves to those in society in general, it would seem logical that they also compare themselves to those who live around them, particularly if the circumstances of their neighbours are noticeably better or worse than their own.

In the absence of reliable small area income data,¹ we therefore analyse ‘relative deprivation inequality.’ If psycho-social factors are important at this scale of analysis, we might expect those living in deprived areas that are surrounded by less deprived areas to have a higher incidence of diabetes than would otherwise be expected, as a result of negative social comparisons. On the other hand, those who live in less deprived areas that are surrounded by poorer areas may make positive comparisons and have lower levels of diabetes than would otherwise be expected. This is not intended to be a direct test of the income inequality thesis, but it is clearly informed by similar processes.

The contrasting neo-materialistic interpretation of the effects of socio-economic inequalities on health would emphasise the impact of poverty throughout the life-course of an individual and the associated under-investment within the areas in which they live. In this interpretation, a person's material well-being and life opportunities, usually indicated by variables such as car ownership, home ownership and educational attainment, are held to be the critical factors underpinning health outcomes and they are expected to be far more influential than the possible psychological effects of relative social position within a social hierarchy (Shaw, Dorling, Gordon, & Davey-Smith, 2004). From a geographical perspective, the areas in which poor people live have, in general, poorer physical, social and health infrastructures when compared to less deprived areas and this may be expected to have an effect on health (Macintyre, Ellaway, & Cummins, 2002). Building on such work, the ‘pull-up/pull-down’ hypothesis (Boyle, Gatrell, & Duke-Williams (2004), Boyle, Norman, & Rees (2004); Gatrell, 1997) suggests that the positive or negative social and environmental resources in the surrounding area will influence circumstances in a particular area of interest. For example, those living in a deprived area that is surrounded by relatively less deprived areas may benefit from the better local services, recreation facilities and fresh food availability, allied with greater social regard for health-promoting behaviours concerning diet, exercise and smoking in these neighbouring areas. This may lead to lower incidences of diabetes than would otherwise be expected (‘pull-up’). On the other hand, the health of those in a less deprived area, which is surrounded by more deprived areas may be negatively influenced by the surrounding cultural, social and environmental circumstances (‘pull-down’). In summary, the ‘pull-up/pull-down’ hypothesis, and the ‘psycho-social’ hypothesis as engendered by negative social comparisons, would seem to predict opposing consequences for the geographical distribution of diabetes incidence resulting from the presence of relative socio-economic inequalities between nearby small areas. By examining relative deprivation inequality (the difference in deprivation between an area and surrounding nearby areas), we can assess whether our results for diabetes incidence are consistent with either a pull-up/pull-down or a psycho-social effect and, as such, our paper is a genuine test of one aspect of socio-spatial context.