1. Using geo-demographic segmentation tools to help inform health insight planning Simon Orange, Public Health Information Analyst Yorkshire and Humber Public Health Observatory, spo500@york.ac.uk 01904 328225

2. “Promoting and delivering public health intelligence” Segmentation A process of looking at the audience or ‘market’ and seeking to identify distinct sub- groups (segments) that may have similar needs, attitudes or behaviours. The aim of any segmentation should be to define a small number of groups so that: –all members of a particular group are as similar to each other as possible –they are as different from the other groups as possible

3. “Promoting and delivering public health intelligence” Types of Segmentation Demographic - Divides people into groups based on variables such as age, gender, race. Geographic - Divides people into groups based on different geographical units such as regions. Geo-demographic - Combines demographics and geography into a segmentation. Identify groups of small geographic areas that have similar demographic profiles. Behavioural - Divides people into groups based on the way they respond to, use, or know of a product or service.

4. “Promoting and delivering public health intelligence” Geodemographics Geodemographics has been defined as the ‘analysis of people by where they live’. Use of geography to help us draw general conclusions about the characteristics and behaviours of the people who live in them. Similar people live in similar places, do similar things and have similar lifestyles ‘birds of a feather flock together’.

5. “Promoting and delivering public health intelligence” Geodemographic Segmentation http://booth.lse.ac.uk/ Charles Booth’s 1898-99 Poverty Map of London Webber and Craig (1976 & 78) Office of Population Censuses and Surveys (OPCS) commissioned three classifications based on the 1971 census. Key variables used included unemployment, students, two car households, industry sector, social class, age, migration, tenure, overseas immigration, overcrowding and household amenities Commercial sector saw potential benefits of area classification. The British Market Research Bureau (BMRB) restructured Webbers classification and renamed it ‘A Classification of Residential Neighbourhoods’ (ACORN). http://www.geog.leeds.ac.uk/people/old/d.vickers/d.vick ers_thesis_c2.pdf

6. “Promoting and delivering public health intelligence” Tool Creation Data is standardised Clusters are created Number of clusters are chosen Variables are chosen and refined Clusters are named and profiled Clusters are named and profiled V1 V2 0% 20% 40% 60% 80% 100% 5101520253035 Cluster numbers % of variance explained

7. “Promoting and delivering public health intelligence” Tools Available

8. “Promoting and delivering public health intelligence” Segmentation Profiles

9. “Promoting and delivering public health intelligence” Example of Profiling cervical screening for Yorkshire and the Humber:

10. “Promoting and delivering public health intelligence” Example of Profiling cervical screening for Bradford:

11. “Promoting and delivering public health intelligence” Segmentation Profiles Baseline ProfileTarget Profile Secure Families Struggling families Burdened Singles All Leeds Deliveries Under 18 Leeds Deliveries Source: HES data

12. “Promoting and delivering public health intelligence” Mapping

13. “Promoting and delivering public health intelligence” Health Intelligence Practice Profiles

14. “Promoting and delivering public health intelligence” Identifying areas with excess expected prevalence/incidence

15. “Promoting and delivering public health intelligence” Comparison

16. “Promoting and delivering public health intelligence” How do they add value? Population health profiling: –Understanding the characteristics of small geographical areas. –Mapping and visualisation. Targeting health interventions by identifying areas with excess expected prevalence/incidence etc. Measurement of health inequalities by: –Explaining variation in health determinants, outcomes or services. –Providing a more granular measure of health inequality. Social marketing and communications Support data linkage, and non-disclosable data sharing. As a base for generating additional insight - on top of the off-the-shelf descriptions.

17. “Promoting and delivering public health intelligence” What geodemographics can not tell us. Birds of a feather may not always flock together Census 2001 data dependent Analysis of the profiles is usually based on probabilities, rather than hard data. Not really intended to be used as a trend monitoring tool. Cannot quickly detect change Geodemographics in themselves do not provide the answer. They are simply one of a range of tools and approaches that can be used to generate insight and health intelligence to support social marketing decisions.

18. Simon Orange, Public Health Information Analyst Yorkshire and Humber Public Health Observatory, spo500@york.ac.uk 01904 328225