Health Inequalities: Where do our deprived people live in Dumfries & Galloway?

Health Inequalities: Where do our deprived people live in Dumfries & Galloway? What is meant by deprivation? Deprivation is a concept that overlaps with, but is not synonymous with, poverty. A definition by Townsend is widely used: Deprivation takes many different forms in every known society. People can be said to be deprived if they lack the types of diet, clothing, housing, household facilities, fuel, and environmental, educational, working and social conditions activities and facilities that are customary in society. 1 This indicates the breadth of factors that contribute to deprivation or disadvantage. Deprivation in a rural area like Dumfries & Galloway shares many of these factors with urban areas but also has some distinct elements 2, such as a different pattern of housing, employment and a greater necessity for car ownership. In addition rural areas tend to be more heterogeneous than urban neighbourhoods. The importance of deprivation for the health service stems from the clear link between deprivation and health outcomes with deprived groups having poorer outcomes across a wide range of health measures such as deaths, disease rates or hospital admissions. 3,4 A related paper on inequalities (in production) shows some current deprivation gradients in outcomes for Dumfries & Galloway. This related paper also describes the factors underlying deprivation and its measurement in detail. How can deprivation be measured? Effectively there are two ways of assessing deprivation: by measures that relate to areas such as Carstairs deprivation categories or the Scottish Index of Multiple Deprivation, and measures pertaining to individuals such as social class or level of income. Area-based measures are widely used. They have substantial benefits including that they: 1. are easily measurable from routine data; 2. provide a simple means of categorising each area; 3. can enable direct comparisons between areas within Dumfries & Galloway or across Scotland; 4. are used for Arbuthnott allocations of NHS resources; 5. are frequently used for targeting resources or services towards areas assessed as deprived. However area-based measures of deprivation do have drawbacks that include: 1. Area-based measures are based on assumptions about which factors best represent material deprivation. One factor previously used to

measure deprivation is access to a car, yet that may not measure deprivation validly in a rural area where a car may be seen as essential. 2. Areas are not internally homogeneous, particularly in rural regions. Populations containing a mix of deprived and affluent households are likely to have middle ranking scores and this can hide the effects of deprivation. Even in neighbourhoods that are recognised as deprived or affluent, area-based measures can mislead. Some deprived people live in affluent areas and conversely not everybody living in an area recognised as deprived is individually deprived. In addition to the area-based measures it is important therefore to try to find a way of identifying those individuals within Dumfries & Galloway who are or may be deprived. This is partly achievable through use of the standard areabased measure, the Scottish Index of Multiple Deprivation. The Scottish Index of Multiple Deprivation The Scottish Executive first published the Scottish Index of Multiple Deprivation in 2003 and an updated index was issued in 2004 (SIMD2004). 5 The SIMD2004 uses data zones as geographical units, aggregated from census output areas. There are 193 data zones in Dumfries & Galloway containing an average of approximately 765 people in each data zone. Data zones are substantially smaller than postcode sectors (on which the Carstairs deprivation categories were based) so should have less heterogeneity than postcode sectors. The SIMD2004 uses 31 indicators of deprivation from six domains: income, employment, education, health, housing and access to services. It does not include access to a car. The related paper on inequalities gives full details of the SIMD2004. The SIMD2004 scores for the whole of Scotland are ranked in order and divided into either five (quintiles) or ten (deciles) equal sized groups for comparison. Though at a Scottish level there are approximately equal population numbers in the five or ten groups, it does not follow that this will be the case for an individual NHS Board like Dumfries & Galloway. We usually use five rather than ten groups in Dumfries & Galloway as this tends to be the standard and gives larger numbers in each group.

Deprivation in Dumfries & Galloway as measured by SIMD2004 Figure 1 shows the distribution of data zones and population in Dumfries & Galloway. Out of 193 data zones, Dumfries & Galloway has 15 in the 20% most deprived in Scotland. This is equivalent to a 1.2% share of all the worst areas in Scotland. The number of people who live in these most deprived areas is approximately 12,000, which is 8.1% of the Dumfries & Galloway population. The figure for Scotland is 20% of the population in the 20% most deprived areas, so we have significantly fewer people living in our worst areas. Figure 1: Distribution of SIMD2004 Data Zones and population in Dumfries & Galloway Number of Data zones Percent of population Number of Data Zones 100 80 60 40 20 0 23% 40% 22% 78 7% 44 43 8% 13 15 1 2 3 4 5 SIMD04 Quintile (1 = Advantaged, 5 =Deprived) 50% 40% 30% 20% 10% 0% Percent of Population Dumfries & Galloway also has 13 data zones that are ranked in the top 20% most advantaged in Scotland, which is equivalent to a 1.0% share of the best areas in Scotland. The number of people who live in these most advantaged areas is approximately 10,400, which is 7.0% of the Dumfries & Galloway population.

The neighbourhood areas that have been identified as having the most and least area-based deprivation using the SIMD2004 are: Deprived = 15 Data Zones Stranraer x 3 around Dick s Hill Wigtownshire x 1 Whithorn Dumfries x 5 North West Dumfries past A75 Dumfries x 2 Loreburn Area, Buccleuch St Upper Nithsdale x 3 Kirkconnel and Kelloholm Annandale & Eskdale x 1 Annan Advantaged = 13 Data Zones Stranraer x 1 North West of harbour Stewartry x 1 farmland North of Dalbeattie, Haugh of Urr Dumfries x 2 Maxwelltown Dumfries x 5 Georgetown Dumfries x 1 Marchmount Annandale & Eskdale x 1 Langholm Annandale & Eskdale x 1 Gretna Annandale & Eskdale x 1 Annan The deprived data zones include the four recognised areas in North West Dumfries, Upper Nithsdale, Stranraer and the Machars, which are all areas served by the Building Healthy Communities initiative. For the first time however it includes areas of central Dumfries and Annan as deprived. Figure 2 shows a map of Dumfries & Galloway with the distribution of SIMD2004 quintiles. Figure 2: Map of SIMD 2004, with Zoom-ins for Dumfries and Stranraer

Deprivation in localities in Dumfries & Galloway as measured by SIMD2004 Tables 1 and 2 below show the number of data zones and population in each locality as well as for the whole of Dumfries & Galloway and for Scotland. Table 1: Number of data zones in each locality by SIMD2004 quintile Advantaged >>>>>>>>>>>>>>>>>> Deprived Area Band 1 Band 2 Band 3 Band 4 Band 5 Total Scotland 1301 1301 1301 1301 1301 6505 Dumfries & Galloway 13 44 78 43 15 193 Annandale & Eskdale 3 19 18 8 1 49 Dumfries & Nithsdale 8 16 23 18 10 75 Stewartry 1 9 19 2-31 Wigtownshire 1-18 15 4 38 Table 2: Population in each locality by SIMD2004 quintile Advantaged >>>>>>>>>>>>>>>>>> Deprived Area Band 1 Band 2 Band 3 Band 4 Band 5 Total Scotland 1030654 994871 992946 1011756 1031784 5062011 Dumfries & Galloway 10,353 33,824 59,135 32,480 11,973 147,765 Annandale &Eskdale 2,656 14,334 13,142 6,077 665 36,874 Dumfries & Nithsdale 6,239 12,187 17,863 13,177 8,072 57,538 Stewartry 613 7,303 14,265 1,651-23,832 Wigtownshire 845-13,865 11,575 3,236 29,521 Table 2 indicates that the largest numbers of people living in both the most deprived and most advantaged areas are both found in Dumfries & Nithsdale. The distribution of the population in Dumfries & Galloway as a whole is symmetrical between deprived and advantaged (see Figure 1). Similarly the population in Dumfries and Nithsdale is fairly balanced between deprived and advantaged, while the populations of Annandale & Eskdale and Stewartry are skewed towards the advantaged quintiles and that of Wigtownshire is skewed towards the deprived. Deprivation of Individuals in Dumfries & Galloway In tackling inequalities, most of the focus is on those living in deprived rather than advantaged circumstances. It is important to know as accurately as possible where deprived people live so that services can work to address the inequalities they are experiencing. As discussed earlier, area-based measures can give an indication of where some deprived individuals live and this indication will be better if the geographical unit is smaller. For example the four areas of relative deprivation where Building Healthy Communities works have been locally recognised as such for many years and can be

formally identified using area-based measures of deprivation. The SIMD2004 picks out two additional areas of relative deprivation in central Dumfries and Annan. An individual measure that captured all the complexities of deprivation would give a much better picture of where deprived people live in the region. There is however no such individual measure. Indications of two aspects of individual deprivation may be obtained from the SIMD2004. The SIMD2004 gives the score and rank for each of the six domains that comprise the index at data zone level. It also gives for each data zone the number of current income-deprived and employment-deprived individuals. The definition of income-deprived individuals is based on different types of benefit claims while employment-deprived individuals are derived from unemployment or incapacity indicators. Some people who are eligible do not register for benefits or unemployment claims and the definitions used are likely to undercount the true numbers of people who may be income or employment deprived in Dumfries & Galloway. Numbers of income or employment-deprived individuals cannot be related to any other indices, e.g. of deaths or health interventions since they are only given as counts. However they do give the best available information on where income or employment-deprived individuals live in Dumfries & Galloway. A total of 18,543 individuals are classed as current income deprived, representing 12.5% of the population of Dumfries & Galloway. A total of 10,284 persons are regarded as employment deprived, which makes up 11.8% of the working age population. Obviously there will be a considerable overlap between those classified as current income deprived or employment deprived; however there is no way to define the magnitude of this overlap. It is important to examine both the proportions and numbers of people who are classed as current income deprived or employment deprived in the quintiles of the distribution of SIMD2004. Figures 3 and 4 show these relations for current income deprivation and Figures 5 and 6 for employment deprivation.

Figure 3: Proportions of people in Dumfries & Galloway classed as current income deprived by SIMD2004 quintile 35% 31.6% 30% 25% Percent of Population 20% 15% 10% 11.3% 17.4% 6.2% 5% 3.0% 0% 1 2 3 4 5 SIMD2004 Quintile (1=Advantaged, 5=Deprived) Figure 4: Numbers of people in Dumfries & Galloway classed as current income deprived by SIMD2004 quintile 8000 7000 6700 6000 5654 Number of People 5000 4000 3000 3782 2000 2098 1000 0 309 1 2 3 4 5 SIMD2004 Quintile (1=Advantaged, 5=Deprived)

Figure 5: Proportions of working age population in Dumfries & Galloway classed as employment deprived by SIMD2004 quintile 30% Percent of Working Age Population 25% 20% 15% 10% 5% 5.5% 7.3% 10.8% 15.3% 25.6% 0% 1 2 3 4 5 SIMD Quintile (1=Advantaged, 5=Deprived) Figure 6: Numbers of working age population in Dumfries & Galloway classed as employment deprived by SIMD2004 quintile 4000 3710 3500 Number people of Working Age 3000 2500 2000 1500 1000 1454 2941 1835 500 344 0 1 2 3 4 5 SIMD Quintile (1=Advantaged, 5=Deprived)

Figures 3 and 5 show the expected relations with SIMD2004 quintiles: the proportions of current income-deprived and employment-deprived people are greater in more deprived quintiles. However Figures 4 and 6 reveal a surprising and crucially important result: the overall numbers of current income-deprived and employment-deprived people are much higher in quintiles 3 and 4 than in quintile 5. The reason underlying this is that many more people live in quintiles 3 and 4 than in quintile 5 in Dumfries & Galloway (see Figure 1). The implication for Dumfries & Galloway is that the large majority of income or employment-deprived people in our region live in areas that are not recognised as deprived. There is little difference whether current income or employment is used as a measure of individual deprivation. The numbers affected differ but the distribution patterns remain similar. This finding is of huge practical importance. We might have expected that the majority of income or employment-deprived people live in the most deprived areas, as identified by quintile 5 of SIMD2004. That in itself would extend the areas of recognised deprivation into new territory in central Dumfries and Annan, as we have seen. At the very least, we might expect most of the deprived individuals who do not live in quintile 5 data zones to live in quintile 4, the other set of data zones that are below the mean for deprivation. Yet this turns out not to be so. Table 3 shows the numbers of current income or employment-deprived people who live in quintiles 5, 4 and any of the other more affluent quintiles 1-3. It can be seen that only just over half of current income-deprived and 46% of employment-deprived people live in data zones that could in any way be classed as deprived. Table 3: Numbers of current income or employment-deprived people living in data zones in quintiles 4 or 5 of SIMD2004 Current income deprived Employment deprived Quintile 5 (20% most deprived data zones) Number Percent Number Percent 3,782 20.4 1,835 17.8 Quintile 4 5,654 30.5 2,941 28.6 Quintiles 4 + 5 9,436 50.9 4,776 46.4 Non-deprived quintiles (1 + 2 + 3) 9,107 49.1 5,508 53.6 Total 18,543 100.0 10,284 100.0 Deprivation of Individuals in localities in Dumfries & Galloway Figures 7 and 8 show the proportions and numbers of people who are classed as current income deprived and Figures 9 and 10 for those identified as employment deprived in our localities.

Figure 7: Proportion of people who are current income deprived by locality Proportion of population in Dumfries and Galloway who are income deprived by LHP 18.0% 16.0% 16.4% 14.0% 12.9% Percent of Population 12.0% 10.0% 8.0% 6.0% 4.0% 10.5% 10.2% 2.0% 0.0% Annandale & Eskdale Dumfries & Upper Nithsdale Stewartry Wigtownshire Figure 7 shows that the highest proportion of people who are income deprived live in Wigtownshire. Figure 8 however indicates that the actual numbers of people affected are substantially greater in Dumfries & Nithsdale. In each locality the large majority of income-deprived people live outwith the most deprived data zones in quintile 5, which is where most of the resources to address inequalities are currently focussed. Figure 8: Number of people who are current income deprived by locality separated into those in most deprived quintile 5 and others in quintiles 1-4 Quintile 1=Advantaged, Quintile 5=Deprived 8000 7000 SIMD04 Quintile 5 SIMD04 Quintiles 1-4 4952 6000 Number of people 5000 4000 3000 2000 3683 2420 3706 1000 0 187 2467 1128 Annandale & Eskdale Dumfries & Upper Nithsdale Stewartry Wigtownshire

Figure 9: Proportion of working age population in Dumfries & Galloway who are employment deprived by locality Proportion of working age population in Dumfries and Galloway who are employment deprived by LHP 16.0% 14.5% Percent of Working Age Population 14.0% 12.0% 10.0% 8.0% 6.0% 4.0% 2.0% 9.8% 12.6% 10.0% 0.0% Annandale & Eskdale Dumfries & Upper Nithsdale Stewartry Wigtownshire Figures 9 and 10 show that employment deprivation has a similar distribution to current income deprivation, with the highest proportion in Wigtownshire but the largest numbers in Dumfries & Nithsdale. The proportions of the Annandale & Eskdale and Stewartry populations who are current income or employment deprived are very similar, however the numbers of individuals involved are considerably higher in Annandale & Eskdale. Figure 10: Number of working age population in Dumfries & Galloway who are employment deprived by locality separated into those in most deprived quintile 5 and others in quintiles 1-4 Quintile 1=Advantaged, Quintile 5=Deprived 7000 6000 SIMD04 Quintile 5 SIMD04 Quintiles 1-4 4952 Number of people 5000 4000 3000 2000 3683 2420 3706 1000 0 84 1255 496 Annandale & Eskdale Dumfries & Upper Nithsdale Stewartry Wigtownshire

Conclusions It is important to be aware that the measures studied - current income and employment deprivation - are elements of deprivation from the SIMD2004. Deprivation however is a complex concept and these measures may not capture all individuals who experience some aspect of deprivation as defined earlier. This analysis of deprivation in Dumfries & Galloway leads to a number of important findings: 1. It identifies for the first time the numbers of current income and employment-deprived people in our region. Using the definitions of deprivation arising from the way the SIMD2004 is modelled, there are approximately 18,500 current income-deprived and 10,300 employment-deprived individuals, clearly with major overlap between these groups. 2. The analysis shows where people classed as current income or employment-deprived live within Dumfries & Galloway at a data zone level. It is possible to identify the exact numbers of current incomedeprived or employment-deprived by data zone. This may enable targeting of initiatives or services much more accurately than before. 3. Two additional areas of relative deprivation in Dumfries & Galloway have been identified in addition to the four recognised areas. These are in central Dumfries and a data zone in Annan. 4. We now know that the large majority of income or employmentdeprived individuals in Dumfries & Galloway live outwith recognised areas of deprivation. Table 3 showed that the most deprived quintile (which itself includes places other than the recognised areas of deprivation) contains a mere 20% of current income deprived and 18% of the employment deprived. Even when this is extended to include the bottom two quintiles, there are still only around half the income and employment-deprived people accounted for. An implication of this is that the Building Healthy Communities Initiative, which serves the four currently recognised area of deprivation (all in quintile 5) can only be reaching a small number of the total who are income or employment deprived within Dumfries & Galloway. It can be estimated that approximately 5,700 current income-deprived people live in the four areas served by BHC. This leaves approximately 12,850 income-deprived people who live outwith the BHC areas. 5. Finally, the distribution of income and employment-deprived people between our localities is now known. Dumfries and Nithsdale accounts for the largest number of income and employment-deprived people by a substantial margin from Wigtownshire, with Annandale & Eskdale

and then Stewartry having lower numbers. It is important however to remember that there are deprived people who live in Stewartry, despite Stewartry having no data zones falling into the most deprived quintile of SIMD2004. Future Implications These findings highlight the clamant need to do more to address deprivation and tackle inequalities. Some immediate implications are: Focussing on recognised areas of deprivation is still worthwhile. It gives a way of engaging with the largest concentrations of deprivation in the region. Central Dumfries and areas of Annan should now be recognised as additional areas of relative deprivation in the region. Future resource planning should take these newly recognised areas of deprivation into account. This gives six recognised areas of deprivation within Dumfries & Galloway. The findings also indicate the need for new approaches in reaching out to the majority of current income or employment-deprived people who do not live in any of the recognised areas of deprivation. These are complex issues and require thought and discussion. Two aspects seem to be involved. First, we need practical methods to identify individuals who are experiencing deprivation but do not live in deprived areas, and may indeed live in some of the most affluent areas. The SIMD2004 does not identify people as individuals, but merely gives counts in each data zone. It is not obvious how people experiencing deprivation might be identified. They may live in small neighbourhoods of relative deprivation, perhaps part of a village street, a few houses or as families or individuals. Second, once a means of identifying individuals experiencing deprivation is determined, we need to find interventions to use that would make a difference to health status. This could mean health and other services trying to better address the needs of individuals as well as the communities they live in. There may be a role for further work with communities but on a more local level and there may also be greater needs for working with individuals or families. It seems likely that the localities will have a major role in these aspects of the work. The benefits for the health service would be increasing the effectiveness of our actions to reduce health inequalities. How can we hope to bring about genuine reductions in health inequalities in Dumfries & Galloway when we currently focus on less than a third of people who experience income or employment deprivation?

References 1. Townsend P. Deprivation. Journal of Social Policy 1987; 16:125-46. 2. Riddet C. Profiling Dumfries and Galloway: information, issues, opportunities, action. Health Promotion Service: Dumfries & Galloway, 1997. 3. Acheson D (chair). Independent Inquiry into Inequalities in Health. The Stationery Office: London, 1998. 4. McLaren G, Bain M. Deprivation and Health in Scotland: insights from NHS data. http://www.show.scot.nhs.uk/publications/isd/deprivation_and_health/ (last accessed 15.08.2006). 5. Office of the Chief Statistician. Scottish Index of Multiple Deprivation 2004: technical report. http://www.scotland.gov.uk/library5/society/siomd-00.asp (last accessed 15.08.2006). Dr Andrew Carnon Ananda Allan Consultant in Public Health Senior Health Intelligence Analyst August 2006