Primary Care Professionals and Social Marketing of Health in Neighbourhoods: a Case Study Approach to Identify, Target and Communicate with ‘At Risk’ Populations

Primary Health Care Research and Development 2007; 8: 22–35 doi: 10.1017/S1463423607000047 Primary care professionals and social marketing of health in neighbourhoods: a case study approach to identify, target and communicate with ‘at risk’ populations

Jane Powell Faculty of Health and Social Care, University of the West of England, UK, Alan Tapp Business School, University of the West of England, Bristol, UK, Judy Orme Faculty of Health and Social Care, University of the West of England, UK and Marc Farr Dr Foster, London, UK and (formerly of Experian Ltd, Nottingham, UK); Lancaster University, UK

Aim: In this article the authors illustrate using a case study approach how primary care professionals can use the combination of geodemographic data with hospital episode statistics (HES) to predict the location of people ‘at risk’ of diabetes mellitus (Type 2 diabetes) in the population of England. This approach facilitates social marketing of those ‘at risk’. Method: Geodemographic segmentation data for all households was combined with HES for 2001–2002, to predict population groups ‘at risk’ of Type 2 diabetes. Using a case study approach and quantitative data analysis techniques, a profile of the undiagnosed and ‘at risk’ population of Slough Primary Care Trust was created at town, ward and street levels. Recent literature on social marketing was applied to predi- cate a discussion of the theory and practice of social marketing that was most likely to succeed in dealing with the prevention of Type 2 diabetes, via the reduction of obesity and overweight in the population. Discussion: The increase in lifestyle-related diseases, such as, Type 2 diabetes that are linked with the rise in overweight and obesity and create large disease management costs for the National Health Service (NHS) are of great concern to primary healthcare professionals and governments throughout the westernized world. Until recently, public and government responses have been very reactive in respect of population groups most in need of lifestyle change. Approaches to the identification of ‘sub-populations’ most at risk of Type 2 diabetes and targeting of these is of direct relevance to the preventive work of primary care professionals. Conclusion: Geodemographic data overlaid onto official NHS and other routinely col- lected data, can aid the identification and targeting of groups most vulnerable to overweight and obesity, through social marketing approaches including direct mail, telephone canvassing and door-to-door communication channels.

Key words: geodemographics; hospital episode statistics; MOSAIC; resources; social marketing; Type 2 diabetes Received: January 2006; accepted: August 2006

Introduction: Type 2 diabetes, lifestyle in the UK have been diagnosed with diabetes and and the ‘Obesity Time bomb’ 90–95% of these cases are Type 2 diabetics (Diabetes UK, 2005). In the UK and other westernized coun- Type 2 diabetes and obesity are inextricably linked tries the proportion of overweight and obese adults (Diabetes UK, 2005). Around 1.8 million people and children is increasing rapidly. While the trend has been upwards for decades, levels of obesity have Author for correspondence: Dr Jane Powell, Glenside Campus, increased at an alarming rate through the 1990s and University of the West of England, Bristol, Blackberry Hill, 2000s in particular. In England, successive National Stapleton, Bristol BS16 1DD, UK. Diet and Nutrition surveys in 1987 and 2001, have © 2007 Cambridge University Press

indicated steep rises in levels of overweight and health problems that are considered ‘preventative’ obese people in just a few short years. In 1987, and have an unequal spread in the population.The obesity prevalence rates of 8% in men and 12% link between health and social marketing is com- in women were recorded. By 2001, prevalence plex and multi-levelled, but there has been work increased to 17% of men and 20% of women establishing a theoretical dimension to marketing (Office of National Statistics, 2003). The Health as a positive force for better health, via the ‘social Survey for England in 2002 indicated that 41% of marketing paradigm’ (Hastings and Saren, 2003; men and 33% of women were overweight and 25% National Social Marketing Centre for Excellence, of men and 20% of women were obese (Department 2005). Lefebvre and Flora (1988) (cited in Naidoo of Health, 2003). Diabetes UK estimate that a fur- and Wills, 2005) propose the key components of ther one million people in the UK have Type 2 dia- social marketing as consumer orientation, identifi- betes, but are unaware of this and are undiagnosed. cation of key audience through segmentation and Obesity is defined through calculation of Body analysis, voluntary and mutually beneficial exchange, Mass Index (BMI). This is a person’s weight in kilo- formative research, clear objective setting, channel grams divided by the square of the person’s height in analysis, a marketing mix of product, place, pro- metres. In the UK, people with a BMI between 25 motion and monitoring evaluation. In particular, and 30 are categorized as overweight, and those with as these authors point out, marketers are keen to an index above 30 are categorized as obese (Royal influence consumer behaviour – in this case the College of Physicians et al., 2004) The rising tide of need is to influence people to adopt healthier eat- obesity has recently been described as a health time ing and exercise habits to avoid obesity (Lefebvre bomb that needs defusing (Chief Medical Officer, and Flora, 1988; Naidoo and Wills, 2005). 2002). Increased prevalence of overweight and obese This article is primarily concerned with the idea of adults is of great concern to governments because of market segmentation or grouping people together. the link with ill-health through diseases, such as Type Segmentation based on a range of attributes is 2 diabetes, some cancers, coronary heart disease, central to effective social marketing (Lefebvre and chronic ill-health, renal failure, osteoarthritis, foot Flora, 1988; Naidoo and Wills, 2005).While the lan- problems and eye problems. Type 2 diabetes has guage of public health discourse may not express increased in the younger population creating poten- it in these terms, the idea of ‘market segmentation’, tial resource allocation problems for the public sec- is now increasingly commonplace in public health tor. Future disease management is likely to be a very policy. In particular, there is a realization that costly process for the National Health Service achievement of the Derek Wanless fully engaged (NHS) in England if measures are not taken to scenario depends upon the engagement of ‘deprived reduce it soon (Wanless, 2004). As a result, tackling and marginalized’ communities; if it is to be suc- obesity is now regarded as of paramount importance cessful. Some sections of the population experience by governments in developed countries, not just in barriers that thwart and prevent attempts to live the UK, but worldwide (International Obesity healthy lifestyles because the circumstances and Taskforce, 2002; World Health Organisation, 2004). context of their lives are not conducive to health While the debates about causes and treatments go improvement. The Department of Health White on, government and public bodies in the UK have Paper ‘Choosing health – Making healthy choices united around practical measures to tackle obesity as easier’ not only recognizes that people need sup- an urgent public priority and to invite the population portive environments within which to change to become ‘fully engaged’ with their own good behaviour, but also the need for individually tailored health. The opportunity cost of not achieving full- health improvement plans (Department of Health, engagement was estimated in the first Wanless 2004).‘Choosing Health’ recognizes the implemen- Report at £20 billion by 2022 (Wanless, 2002). tation of personalized ‘health guides’ that is practical plans for improving individual health that fit into the context of peoples lives (Department of Social marketing Health, 2004). In addition, it argues for the importance of context in an individual healthy lifestyle Primary healthcare professionals are increasingly acknowledging that this includes a whole host of interested in ‘social marketing’ as a way of tackling socio-economic and socio-cultural factors, as well Primary Health Care Research and Development 2007; 8: 22–35

as peoples’ own attitudes and beliefs surrounding diagnosed and undiagnosed population in a targeted good health. and appropriate way.From a health marketing per- spective, this gives public policy makers and health professionals a platform for ‘understanding well- Tackling obesity – bringing together being’ and a practical targeting tool from which a public health prevention policy and social marketing strategy can be designed. Hence social marketing in this article we combine geodemographic data with hospital episode statistics (HES) data recorded in The emphasis on prevention and health promo- 2001–2002 for Type 2 diabetes (strongly linked to tion and the close relationship between health and overweight and obesity), to generate detailed social marketing are central themes of the recent geodemographic profiles of population groups at White Paper (Department of Health, 2004). The national, city,town, ward and street levels. Our aim White Paper may,in a sense, be viewed as a market- is to illustrate a useful approach in an appropriate ing strategy which stimulates demand for health, way for primary care professionals and not to provides accessible and credible information, guar- re-explore the previously reported Slough case study antees a supply of local services or healthier products (Farr and Evans, 2005). and ensures a local environment that makes healthy Geodemographic data can be applied to under- choice easier (House of Commons Health Com- stand the context in which tailored behavioural mittee, 2004). However, there is also recognition that interventions could be designed and how and where making healthy choices is easier for some people resources might be targeted to deliver behaviour than others and in the case of obesity,we can begin change with social marketing initiatives. At the by examining socio-environmental factors and their present time, geodemographic profiling combined impact on the obesogenic environment.A complex with routine NHS data has not been widely used in multi-layered environment is revealed, with four a predictive manner to identify ‘Type 2 hotspots’ groups of factors contributing to increased obesity at various levels of population aggregation. levels identified.These are: firstly the presence of ‘at The authors see this work as a first step in risk’ groups in terms of medical and socio-economic demonstrating how existing social marketing and factors; secondly, the presence of conditions likely public health policy theory can be deployed in the to contribute to obesity levels, for example, poverty, ‘obesity field’. A pre-understanding of the socio- social marginalization. Thirdly, the prevalence of environmental factors underpinning obesity allows risk behaviours, for example, low levels of physical us to consider more in-depth models that could activity,high-energy diets, or poor diets. Finally the underpin the creation of health intervention pro- presence of psychological or environmental risk grammes that may have a chance of success. Figure 1 factors, for example, lack of social support, exposure is tentatively proposed as an explanation of how to food advertising, poor access to healthy food demographics and ‘psychographics’ (an individual’s (Mulvihill and Quigley, 2003). psychological and attitudinal make-up) may under- If, as current health policy identifies, social mar- pin behaviours linked to Type 2 diabetes and obesity. keting is an important approach in future behavioural change strategies, then it is vital that market Research design and methods segmentation and targeting are effected as precisely as possible. This may mean segmentation that is A predictive, secondary data analysis was run achieved using a large number of relevant popula- using HES of Type 2 diabetes for 2001–2002 with tion characteristics. 61 appended geodemographic MOSAIC groups The central purpose of this article is to illustrate and type codes for all households in England. to an audience of primary care professionals, the worth of ‘an approach’ to identification and targeting of ‘at risk sub-populations’ for Type 2 diabetes GB HES for Type 2 diabetes 2001–2002 in a preventative sense. The approach combines a This dataset contains entries for 12.8 million segmentation database of the UK population with overnight hospital admissions to all hospitals in health services data, to give direction to public Great Britain (GB) in 2001–2002. Each data record health policy on reduction of Type 2 diabetes in the for admissions contains a diagnostic code indicating Primary Health Care Research and Development 2007; 8: 22–35

Overarching environment of sedentary Socio-cultural everyday life environment of UK

Poor education about nutrition, healthy diet and the importance of exercise Low income – purchase of high fat, high salt foods Geodemographics Time poor – lack of exercise, processed foods target by locality Socio-culture – peer weight – lack of pressure to change Geography – rural urban, use of car, walking Psychology access to fresh fruit and vegetables target through individual characteristics Poor self-image – overeating and lethargy

Figure 1 Factors leading to obesity

the health problem that necessitated an overnight from a consumer segmentation database, which stay in hospital. HES also record patient date of provides coverage of all of the UK’s 46 million adult birth, ethnicity, Primary Care Trust (PCT) code, residents and 23 million households using the general practitioner (GP) code, local authority Electoral Roll, lifestyle survey information, con- code and postal code. sumer credit activity, post office address file, Shareholder’s Register, house price and council tax information and Office of National Statistics local MOSAIC geodemographic classifications for all area statistics. MOSAIC classifies consumers by GB households household or by postcode which allows optimization of use of the segmentation depending on application. Geodemographic dataset – MOSAIC UK MOSAIC UK is a commercial dataset owned by Experian Ltd.The household version of MOSAIC Procedure groups households together into clusters described according to their geographic and demographic The MOSAIC data is linked to the HES informa- characteristics. Household MOSAIC is based on tion using a simple postcode link.To maximize the census data, housing and financial data. A total of match-rate the postcodes are forced across eight 400 variables are used to build MOSAIC profiles digits in each file. For example W1 1AA becomes and some of these are updated annually. These W__1_1AA.Where postcodes have changed a ret- variables have been selected as inputs to the classifi- rospective file is also held that makes it possible to cation on the basis of their volume, quality,consist- update old postcodes which may have been ency and sustainability. In order to be included into entered on the HES file. the classification data must meet one or more of four criteria. First, that it allows identification and descrip- tion of consumer segments that are not necessarily Results distinguished solely by the use of Census data. Second, the data ensures accuracy of the MOSAIC HES for Type 2 diabetes were matched against code by either household address or postcode.Third, geodemographic codes for England. In this article data is current and lastly it improves discrimination we report on these figures aggregated for the UK. of segments and allows for the identification of a Table 1 and Table 2 demonstrate the geodemo- wide range of consumer behaviours. Fifty-four per graphic profile of Slough versus the UK average. cent of the data used to build MOSAIC is sourced Table 3 demonstrates incidence of diabetes by from the 2001 Census.The remaining 46% is derived ‘MOSAIC Group’ and Table 4 incidence of diabetes

Primary Health Care Research and Development 2007; 8: 22–35

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.234, on 25 Sep 2021 at 02:41:55, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1463423607000047 26 Jane Powell et al. UK average) % ofpopulationof Sloughin each populationMOSAIC of UK in % ofGroup average each MOSAIC index for Group Mean versus UK Slough (100 ts. les fi fi nances are easier. fi Slough PCT and UK pro – Ranked MOSAIC Groups Ranked MOSAIC uenced by urban consumption patterns. fl mature homes. Children are growing up and people are cosmopolitan in tastes and liberal attitudes. enterprising and may well have exercised their right to buy. mostly younger age groups now raising children. communities, responsible workers with unsophisticated tastes. accommodation. mostly reliant on the council for accommodation and bene after locations, affording luxuries and premium quality products. locations, affording after owner-occupation. council estates where there is little relatively active in their lifestyles. in C who are successfully established in comfortable, Families 28.66 16.26 176 E single and mostly well educated, these Young, H who though not well-educated are practical and People B with focus on career and home, Families D 17.28 living in close-knit inner city and manufacturing town People 9.51I 18.54 Elderly people subsisting on meagre incomes in council 11.81F and are rewards who are struggling to achieve People 6.45 16.14 146 1.67 15.41 147 2.30 115 11.71 2.85 5.30 132 59 43 A careers with rewarding who live in sought People G live in large on lower incomes who often Families J Independent pensioners living in their own homes who areK living in rural areas where country life has not been People 3.31 1.21 2.11 0.00 10.41 6.70 6.89 5.47 32 18 31 0 Table 1 Table Total 100 100 100

Primary Health Care Research and Development 2007; 8: 22–35

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.234, on 25 Sep 2021 at 02:41:55, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1463423607000047 Primary care professionals and social marketing of health 27 (Continued) UK average) Type Type (100 Populationof Slough % ofin each populationMOSAICType of UK in of Slough in each Population Population % of each MOSAIC MOSAIC Type each of UK in Penalty % Mean MOSAIC average index for Slough versus UK Geodemographic breakdown for Slough PCT uent Blue Collar 172 0.14 2 000 876 3.39 0.01 4 ﬂ A of Success Symbols 01 Global Connections02 Cultural Leadership03 Corporate Chieftains04 Golden Empty Nesters05 Privilege Provincial 006 12 543 High Technologists 28207 Seclusion Semi-Rural B 430 Families Happy 0.0008 2350 412 In Moving Just 0.01 0.45 0.2309 Fledgling Nurseries10 Upscale New Owners11 0.35 Making Good Families 1.9312 0.34 340 4593 Families Middle Rung 206 56913 774 707 610 087 2133 813 753 Burdened Optimists 69414 In Military Quarters 4138 0.58 1041 0.96 1.31 885 1.38 1C 345 396 1 265 3.77 Suburban Comfort 6610 42015 0.50 1.75 Close to Retirement 1.76 0.5716 2.28 Conservative Values 3.40 2.1417 0 Small Time Business18 0.00 5.43 1368 Sprawling Subtopia 0.00 0.07 0.03 70719 1018 526 040 1 511942 Original Suburbs 49120 2 321 Asian Enterprise 063 678 182 265 0.04 0.00 1.20 5677 1.72 2.56 0.17D 0.03 1.12 Ties of Community 1158 699 0.54 3.4921 Rows Respectable 0 34 0.40 1895 17 22 1 25 Af 303 0.15 1.9623 4.66 Industrial Grit 20 13524 823 1 875 85 Coronation Street 500 0.65 16 0.06 6069 0.1425 Centre Refuge Town 1 1.56 639 20.77 657 0.2326 0.22 0.20 Asian South 3.18Industry 1 781 39627 1981 Minorities Settled 458 2.78 0.57 10 827 113E 198 4.98 3.02 315 Urban Intelligence 921 1 3.36 29 547 68 28 875 Counter Cultural Mix 105 776 56929 97 City Adventurers 4286 0.00 2.6230 0.07 277 8.89 New Urban Colonists 0.09 0.16 1.31 131 449 283 0.03 0.76 Caring Professionals 4432 0.01 Dinky Developments33 0.29 2.45 221Transition 3.52 Gown Town 34 629 735 University Challenge 1 0 35 500 575 9962 725 503 936 0.12 681 2 359 262 0.04 3.26 15 0 2.54 0.18 1.23 139 0.85 3.99 5 991 0.52 695 0.60 0 0.42 8.18 1580 59 1.68 678 0.00 424 726 135 0.00 0.01 1.15 561 1.49 0.02 848 203 606 0.04 229 488 323 1.23 0.95 456 842 1.03 0.43 0.83 723 294 045 0.77 11 6 19 0.01 0.50 0.03 210 0.11 0.12 2.04 0.00 3 15 0.00 54 989 59 0 0 Table 2 Table Primary Health Care Research and Development 2007; 8: 22–35

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.234, on 25 Sep 2021 at 02:41:55, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1463423607000047 28 Jane Powell et al. UK average) Type Type (100 Populationof Slough % ofin each populationMOSAICType of UK in of Slough in each Population Population % of each MOSAIC MOSAIC Type each of UK in Penalty % Mean MOSAIC average index for Slough versus UK ts 2246 1.84 838 640 1.42 0.27 130 fi ciaries 0 0.00 274 729 0.47 0.00 0 fi (Continued) ed Dependency 153 0.13 550 905 0.93 0.03 13 fi Table 2 Table F Borderline Welfare 35 Bedsit Bene 36 Metro Multiculture37 Upper Floor Families38 Living Block Tower 39 Digni 40 1960 Sharing a Staircase 150G Municipal Dependency 53841 on Bene Families 42 1.61 Horizons Low 0.1243 Legacy Ex-industrial 0H 0.44 Blue Collar Enterprise 44 Resilience Rustbelt 855 13545 990 476 320 Older Right to Buy 0.0046 White Van Culture47 184 953 New Town Materialism 1.45 0 1.68I Subsistence 0.26 Twilight 48 0 2497 271 681 770 17 0.31 in Flats Old People 87249 Income Elderly Low 50 0.00 0.23 Cared for Pensioners 0.02 1 534 025 0.46J 0.00 2.05 14.67 0.56 Perspectives Grey 0.2951 150 Memories Sepia 863 1026 152 698 498 2.60 Childfree Serenity 111 53 1 1950 High Spending Elders 1988 536 003 354 1 145 0.00 501061 7 54 Retirement Bungalow 141 0.1255 2.88 0.71 0.84 Small Town Seniors56 Attendants Tourist 0.02 3.30 3.37 2.60 2.54 995 66 0K Rural Isolation 0 0 57 292 729 Summer Playgrounds 0.00 822 569 408 995 03258 Greenbelt Guardians59 0.82 0.00 0.90 0.16 0.05 10 0.05 Villagers Parochial 0.0060 0 0.50 Symphony Pastoral 0.00 1.39 0.9661 0 Upland Hill Farmers 0.33Total 0 0 436 605 636 283 090 79 826 085 22 0.00 0 0.05 618 551 1 0 0.10 0.18 486 739 0.00 0 1.03 0 0.48 0.00 1.40 1.05 138 570 2.52 0.00 25 0.00 138 51 216 87 0.00 1058 857 0.16 0.02 0.00 0.23 0.00 0.23 0.03 121825 995 000 794 1.79 801 241036 80 11 100 0.00 0 1.68 1.35 0.00 13 0 0.41 0.00 59059067 0.00 0 0.00 0.00 0 100 0 0 0 0 0.21 100

Primary Health Care Research and Development 2007; 8: 22–35

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.234, on 25 Sep 2021 at 02:41:55, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1463423607000047 30 Jane Powell et al. 100) Incidenceof diabetes2 across eachMOSAIC Incidence 2 across each of diabetesType MOSAIC MOSAIC of UK in each PopulationType MOSAIC of UK in each 2 across each Population of diabetes for diabetes in penetration deviation score Mean average Type MOSAIC UK Types MOSAIC Type in UKType in UK Type (average UK incidence of diabetes by ranked types MOSAIC ed Dependency 13 950 1.76 550 905 0.93 2.53 189 ﬁ Ex-industrial LegacyEx-industrial Income ElderlyLow Upper Floor Families 35 468 15 480 15 584 HorizonsLow Small Town SeniorsOlder Right to Buy 4.48 1.95Coronation Street 26 171Childfree Serenity 1.97 30127 26High Spending Elders 114 25 955 13 1 612 534 025 10Metro Multiculture 088 3.30 MinoritiesSettled 822 995 New Town 3.29 3.80Materialism 855 135 14 947 2.60 20 946 3.27 1.72 13 621 1.27 1.39 1 486 739 1.45 1 1 698 501061 498 1.89 2.64 2.31 1.72 1 500 575 2.52 826 085 1.88 605 636 2.54 2.88 1.82 2.54 1 536 003 990 1.40 476 1.03 172 991 695 1.76 140 1.77 1.74 136 2.60 1.68 1.73 1.68 1.65 1.67 131 130 132 1.36 1.51 129 1.37 123 124 102 112 102 Cared for Pensioners in FlatsOld People MemoriesSepia 32 984 Asian South Industry RetirementBungalow 15 013Digni 21 322 15 910 11198 4.16 1.89 2.69 2.01 1.41 569 032 292 729 725 936 618 551 283 090 0.96 0.50 1.23 1.05 0.48 5.80 5.13 2.94 2.57 3.96 432 382 219 192 295 Industrial Grit 31 854 4.02 2 359 262 3.99 1.35 101 Conservative Values 26 504 3.34 1 639 657 2.78 1.62 120 Asian Enterprise 14 570 1.84 776 569 1.31 1.88 140 White Van White Van Culture 30 551 3.85 1988 145 3.37 1.54 115 Town Centre RefugeTown Living Block Tower 9100 AttendantsTourist 3309 1.15 2410 0.42 0.30 503 681 184 953 0.85 138 0.31 570 0.23 1.81 1.79 1.74 135 133 130 43 49 20 37 25 38 42 55 45 56 24 52 53 16 46 36 27 47 50 48 51 26 54 39 Table 4 Table Types Ranked MOSAIC Target % Base % % Penalty 23 Index

Primary Health Care Research and Development 2007; 8: 22–35

by ‘MOSAIC Type’.Table 3 and Table 4 reveal what ‘Ties of community’ and ‘Blue Collar Enterprise’ are perhaps the most important results of this analy- are also working class-based cultural groups with sis.Those ‘groups’ and ‘types’ that have the highest distinct sub-cultures, but once more likely to be Type 2 counts are the ones with the highest indices. characterized by strong socially driven attitudes to An index of 100 indicates the UK average penetra- diet and exercise that will need hard work to break tion of the disease. The Groups with the highest down. However these groups have indices that are penetrations of Type 2 diabetes (over 100) are, in only just above the national average. order, ‘Twilight Subsistence’, ‘Grey Perspectives’, At the bottom of the chart ‘Rural Isolation’, ‘Municipal Dependency’,‘Welfare Borderline’,‘Ties ‘Urban Intelligence’, ‘Symbols of Success’ and of Community’, ‘Blue Collar Enterprise’. People ﬁnally ‘Happy Families’ all have incidences of who live in areas characterized as ‘Twilight Subsist- Type 2 diabetes that are much lower than the ence’ have a high-average age, and a lower than national average, perhaps suggesting that less average income.They may not be well educated, and resources will need to be deployed in these popu- will typically have lower social class backgrounds. lation clusters. These people may not be particularly mobile and Table 4, highlights the same data as Table 3, this are likely to have low aspirations with the remainder time split into ‘MOSAIC Types’,which are smaller of their lives.‘Grey Perspectives’, though of similar clusters with more ‘extreme’ underlying demo- advanced years, will be better off and better edu- graphics. As can be seen, the indices can be much cated than the ‘Twilight Subsistence’ group, and it higher or lower than for Groups, with an index of is interesting to note the dramatic drop in diabetes 432 indicating an incidence of diabetes of 4.32 that occurs. While age is the key underlying factor times the national average. This has important of the top two groups, the next two, ‘Municipal implications for the extra return on investment Dependency’ and ‘Welfare Borderline’ groups com- that may be expected for tightly targeted activities. prise people who have very low average incomes Finally, it is interesting to note the high indices and low horizons in life.They exhibit predominantly attached to predominantly Asian geographic types, working class cultures that will strongly inﬂuence suggesting the adverse impact Asian-based diets attitudes to diet and exercise, and will be less well have had on prevalence of Type 2 diabetes. educated than UK averages. The concern with Table 5 illustrates how the combination of datasets these groups lies particularly with their modest MOSAIC and HES, can allow one to drill down to average age, indicating poor weight control in early postcode sectors to highlight ‘hotspots’ for Type 2 stages in life. diabetes prevalence. In this case study for Slough

Table 5 Demonstration of geodemographics as a targeting system: Slough PCT – Zone Ranking

Postal sector D43 Settled but poor older % Slough % Penalty Index people in low-rise social housing, often found in declining industrial areas

SL 2 1 Long Readings Lane, Slough 143 57.43 12,484 11.1 1.145 515 SL 3 8 Sutton, Slough 106 42.57 14 624 13.1 0.725 326 SL 1 1 Upton, Slough 0 0.00 4340 3.9 0.000 0 SL 1 2 Chalvey, Slough 0 0.00 12 002 10.7 0.000 0 SL 1 3 Salt Hill, Slough 0 0.00 12 319 11.0 0.000 0 SL 1 4 Manor Park, Slough 0 0.00 890 0.8 0.000 0 SL 1 5 Dorney, Slough 0 0.00 10 343 9.2 0.000 0 SL 1 6 Blumﬁeld Cr, Slough 0 0.00 6142 5.5 0.000 0 SL 1 9 Wood Lane, Slough 0 0.00 2630 2.3 0.000 0 SL 2 2 Lynch Hill, Slough 0 0.00 7429 6.6 0.000 0 SL 2 5 Petersﬁeld Road, Slough 0 0.00 13 677 12.2 0.000 0 SL 3 0 Colnbrook, Slough 0 0.00 3533 3.2 0.000 0 SL 3 7 Slough, Slough 0 0.00 11573 10.3 0.000 0 Total 249 100.00 111986 100.0 0.222 100

Primary Health Care Research and Development 2007; 8: 22–35

PCT area down to a street level which is powerful resources at Type 2 diabetes prevention vis-a-vis information on which to base resource allocation other competing uses of scarce resources. and targeting decisions at a policy level. The combination of datasets method is very useful to all PCTs in terms of identiﬁcation of at risk Discussion groups and targeting of resources and effort.Table 6 illustrates ‘hotspots’ and low-spots’ at PCT area Limitations of the approach level. This can aid primary care professionals to Our analysis has highlighted age, weight, ethnicity recognize the need to target or not target scarce and poverty, as the key variables impacting on

Table 6 Predicted prevalence of Type 2 diabetes by PCT

Ranking PCT Estimated number of Total population people with diabetes

231 Staffordshire Moorlands 5HR 4421 116 076 232 Mansfield District 5AM 4403 96 162 233 South Liverpool 5HC 4386 99 091 234 Chiltern and South Bucks 5G4 4381 166 544 235 Exeter 5FR 4377 130 402 236 Hartlepool 5D9 4372 88 004 237 Sedgefield 5KE 4360 89 734 238 Doncaster East 5EK 4351 109 222 239 Broadland 5JL 4348 121016 240 Ashfield 5FA 4338 90 723 241 Slough 5DM 4325 121 337 242 North Peterborough 5AF 4292 99 651 243 Witham, Braintree and Halstead TAG 4268 125 892 244 Basildon 5GR 4251 101 853 245 South Wiltshire 5DJ 4236 119 457 246 West Wiltshire 5DH 4233 121969 247 Langbaurgh 5KN 4228 97 144 248 Hambleton and Richmondshire 5KH 4226 117 767 249 Derwentside 5KA 4193 89 331 250 Cherwell Vale 5DV 4188 129 011 251 Oldbury and Smethwick 5MG 4174 89 147 Top 10 PCTs 319 Newbury and Community 5DK 3415 89 151 320 Fylde 5HE 3166 71100 321 Maldon and South Chelmsford 5GL 3145 83 887 322 North East Oxfordshire 5DT 2930 67 189 323 Eden Valley 5D5 2559 68 703 324 Royston, Buntingford and Bishop’s Stortford 5GK 2541 64 047 325 Uttlesford 5GN 2454 68 980 326 Western Isles 15 1138 25 940 327 Shetland 14 911 21910 328 Orkney 13 729 19 111 Bottom 10 PCTs 1 Bro Taf QW2 34 916 709 662 2 Lothian 05 33 647 783 081 3 Greater Glasgow 09 32 237 866 755 4 Eastern (NI) 9501 31 782 662 066 5 North Wales QW4 31177 668 067 6 Gwent QW1 29 574 559 650 7 Lanarkshire 10 28 929 552 164 8 Morgannwg QW5 24 887 488 626 9 Grampian 02 23 706 521 377 10 Dyfed Powys QW3 21113 496 391

Primary Health Care Research and Development 2007; 8: 22–35

Type 2 diabetes. In this analysis it is important not a direct approach that can use the segmentation to to overclaim diabetes as a surrogate for obesity.This best advantage. New media such as Email or mobile is true amongst younger age groups, but for over have yet to be well linked with geodemographics, so 65s diabetes may occur more frequently regardless the best traditional channels include telephone can- of the obesity of individuals. Of more interest there- vassing, door-to-door and direct mail. Door-to-door fore is the rather stark segmentation that emerges is particularly well suited to geodemographics, as according to class, with ‘lower social class’ groups door-to-door media is itself geographically based much more prone to obesity than ‘middle classes’. (Tapp, 2005). Perhaps the most important finding, is the exhibit One of the Wanless Report’s main themes is to that segmentation systems based on demographics encourage deprived or marginalized communities and geography seem to offer precise targeting to improve their own health. MOSAIC analysis opportunities and hence a powerful way for focus- clearly shows how important deprived communities ing scarce health resources. Another qualification are at the front line of the obesity issue in the UK. that might limit the power of combining geodemo- Primary care professionals therefore could have a graphic and HES data, is that individual attitudes role in using these communication channels to sup- and beliefs concerning health status are captured port and empower deprived and marginalized com- indirectly rather than directly using this approach. munities to improve their own health. For those ‘at Interestingly, while there are instances of psycho- risk’ of Type 2 diabetes our behavioural intervention logical unhappiness being linked to overeating, the model for reducing obesity suggests these commu- World Health Organisation (WHO) suggest that nication channels are used to mount a two-pronged by and large overeating is not the key issue leading initiative that ‘targets’ both diet (a little) and phys- to obesity, as we eat less on average in terms of ical activity (mostly). As far as primary care pro- calorific value, today, than we did 50 years ago.The fessionals are concerned door-to-door canvassing central issue is lack of physical exercise, within the in the form of ‘brief intervention’ might be both context of westernized lives being made more and supportive and cautionary in advising overweight more sedentary with respect to how we live our or obese people in terms of sudden increased phys- everyday lives. A typical westernized life now ical activity and improved diet. If more informa- involves a lack of in-built exercise to such a pro- tion is required, direct mail, with its extra physical found extent that most physical activity involves space, may be better suited. ‘artificial’ intervention in daily routines. If this line of argument is accepted, and it is by the WHO, then Future work the current obsession with dieting is tackling the problem from the wrong end (WHO, 2004).The role Figure 2 is offered as a way of structuring the causal of social marketing is to work out how to design linkages, and may be a start point for future research. and implement behavioural intervention pro- It is vital that the public are involved in the imple- grammes that have the best chance of success in mentation of a social marketing strategy to reduce increasing physical activity. However, these inter- obesity. Social marketers emphasize that ‘listening ventions would also need to be tailored appropri- to the consumer’ or ‘consumer involvement’ are all ately to their audience if they are to be successful important in terms of initiating sustained behaviour in motivating and sustaining people in increased changes in consumers (Lefebvre and Flora, 1988; physical activity and again household MOSAIC Naidoo and Wills, 2005). This stance is in keeping would be a useful tool to facilitate appropriate with the current view of public health profession- approaches tailored to the specific sub-populations. als; that they are engaged in partnership with users of services and that they should enable ‘user involvement’ in the delivery of public healthcare. Communications channels The approach identified in this article is not complete without some discussion of the commu- Conclusion nications channels which use of MOSAIC profiles open up, as these are as crucial to success of target- This article has reported on a case study approach ing. The best media will be those which facilitate in which HES for Type 2 diabetes were combined Primary Health Care Research and Development 2007; 8: 22–35

British cultural Resistance factors bloody-mindedness: Identify and target ‘don’t tell me what most vulnerable to do with my life’ groups

Health education: diet and exercise Peer pressure not to bother. ‘Exercise isn’t cool’?

Social marketing Develop relationships One-to-one approach Listen and respond

Develop programmes Create environment in which exercise and good diet are accessible and socially acceptable

Figure 2 A behavioural intervention model for reducing obesity

with household MOSAIC to identify neighbour- interventions that are sent through appropriate hood, wards and streets for targeted interventions communication channels. in health prevention of Type 2 diabetes in the area covered by Slough PCT. The theoretical and practical benefits of the combination of NHS data and Acknowledgements geodemographic profiling seem to be considerable in terms of identification, targeting and use of com- Jane Powell, Alan Tapp and Judy Orme were sup- munication channels for ‘getting health messages ported by internal University of the West of across’ in an appropriate and tailored way. Primary England, Bristol funds. Marc Farr, Dr Foster, healthcare professionals could use the approach London, UK was employed at Experian UK when outlined in this article to devise tailored behavioural some of the analysis for this article was under- interventions including social marketing to various taken.Thanks to Experian UK, Business Strategies segments of the population at country, city, towns, Division for providing access to the data on which ward, neighbourhood and street level. Changing this article is based. one’s life to overcome overweight and obesity is very difficult. There are clearly big social, cultural and psychological barriers to reducing obesity.The first premise of social marketing is that these barriers References must be understood before attempting behaviour change. From a practical standpoint, tools such as Chief Medical Officer. 2002: Annual report 2002. London: MOSAIC can be combined with routine datasets Department of Health. to allow precise targeting of particularly vulnerable Department of Health. 2003: The health survey for England groups.This means scarce resources can be precisely 2002. London: HMSO. deployed to areas of most need. MOSAIC can in Department of Health. 2004: Choosing health: making healthy choices easier. London: HMSO. turn aid public health professionals and social Diabetes UK. 2005: Type 2 diabetes and obesity: a heavy burden. marketers to understand the geodemographic London: Diabetes UK. factors that influence lifestyles that lead to Farr, M. and Evans, A. 2005: Identifying unknown diabetics obesity; helping in efficient resource allocation using geodemographics and social marketing. Journal of and the design of appropriate programmes and Direct Data and Digital Marketing Practice 7, 47–58. Primary Health Care Research and Development 2007; 8: 22–35

Hastings, G. and Saren, M. 2003: Introduction to a special issue Ofﬁce of National Statistics. 2003: The national diet and nutrition on social marketing. Marketing Theory 3, 291–92. survey: adults aged 19–64 years. Newport: ONS. House of Commons Health Committee. 2004: Obesity. London: Royal College of Physicians Royal College of Paediatrics and The Stationary Ofﬁce. Child Health and Faculty of Public Health. 2004: Storing up International Obesity Taskforce. 2002: Retrieved 24 April 2004 problems: the medical case for a slimmer nation. Report of a from http://www.iotf.org joint working group. London: Royal College of Physicians. Lefebvre, R.C. and Flora, J.A. 1988: Social marketing and public Tapp, A. 2005: Principles of direct and database marketing.FT health intervention. Health Education Quarterly 15, 299–315. Prentice Hall: Harlow. Mulvihill, C. and Quigley, R. 2003: The management of obesity Wanless, D. 2002: Securing our future health: taking a long term and overweight: an analysis of reviews of diet, physical activity view. London: HM-Treasury. and behavioural approaches. London: Health Development Wanless, D. 2004: Securing good health for the whole population. Agency. London: HM-Treasury. Naidoo, J. and Wills, J. 2005: Public health and health promotion: World Health Organisation. 2004: Global strategy on diet, phys- developing practice. Edinburgh: Bailliere Tindall Elsevier. ical activity and health. Geneva:World Health Organisation. National Social Marketing Centre for Excellence. 2005: Social marketing pocket guide. London: Department of Health.

Primary Health Care Research and Development 2007; 8: 22–35