Targeting home fire safety checks Andy Mobbs - London Fire Brigade Risk Information Team February 2008.

Slides:



Advertisements
Similar presentations
Community Newspaper Readership
Advertisements

1 UNIVERSITIES of AUSTRALASIA BENCHMARKING RISK MANAGEMENT BILL DUNNE DIRECTOR RISK MANAGEMENT UNSW. PROUDLY SPONSORED BY UNIMUTUAL.
1 Impact of Changes in the Telephone Environment On RDD Telephone Surveys Mary Cay Murray Abt Associates Inc Erin Foster Abt Associates Inc Jessica Cardoni.
1 of 20 Information Dissemination Audiences and Markets IMARK Investing in Information for Development Information Dissemination Audiences and Markets.
Overcoming Indigenous Disadvantage in Australia Gary Banks Chairman, Productivity Commission OECD WORLD FORUM Statistics, Knowledge and Policy Measuring.
THE 2004 LIVING CONDITIONS MONITORING SURVEY : ZAMBIA EXTENT TO WHICH GENDER WAS INCORPORATED presented at the Global Forum on Gender Statistics, Accra.
Disability status in Ethiopia in 1984, 1994 & 2007 population and housing sensus Ehete Bekele Seyoum ESA/STAT/AC.219/25.
1 Introduction to Safety Management April Objective The objective of this presentation is to highlight some of the basic elements of Safety Management.
THE COMMONWEALTH FUND National Scorecard on U.S. Health System Performance: Complete Chartpack Cathy Schoen, Senior Vice President Sabrina K. H. How, Research.
CLOSING THE DIVIDE: HOW MEDICAL HOMES PROMOTE EQUITY IN HEALTH CARE Results from the Commonwealth Fund 2006 Health Care Quality Survey THE COMMONWEALTH.
Local Employment Dynamics: OnTheMap Version 4 New Jersey State Data Center Annual Network Meeting Stephen Tibbets LEHD Program June 17, 2010.
Department of State Health Services (DSHS) House Human Services Committee August 8, 2006.
Training and Development Agency for Schools (TDA) Alison Kitson, Programme Leader (CPD)
NOT PROTECTIVELY MARKED SCD5 C.A.I.C – Risk Factors Metropolitan Police Service SCD5 - Child Abuse Investigation Command Common Risk Factors in the Physical.
Leeds Institute of Health Sciences Trajectories of alcohol use from Year 9 to Year 12 Longitudinal Study of Young People in England (LSYPE): one-day introductory.
Improving the commercial sectors access to government data Visiting Professor Birkbeck College University of London SUF Users Forum 18 th February 2007.
Comparing Results from the England and Wales, Scotland and Northern Ireland Longitudinal Studies: Health and Mortality as a case study Census Microdata.
Looking forward to the 2006/07 HBAI publication: New analyses and improvements Peter Matejic (DWP) Households Below Average Income ESDS Government FRS.
Estimating Prevalence of Diabetes and Other Chronic Diseases for Small Geographic Areas Peter Congdon, Geography, QMUL.
Households Below Average Income 2008/09
The Census Area Statistics Myles Gould Understanding area-level inequality & change.
Domestic violence, sexual assault and stalking:
Options appraisal, the business case & procurement
1 European benchmarking with the CAF ROME 17-18th of November 2003.
Understanding Multiyear Estimates from the American Community Survey Updated February 2013.
1 Understanding Multiyear Estimates from the American Community Survey.
© Tarek Hegazy – 1 Basics of Asset Management Prof. Tarek Hegazy.
Marketing Research at the Turn of the Decade: Four Key Trends that Affect Research Presentation to the Ottawa Chapter of the Marketing Research and Intelligence.
Fire research is about people... and so is economics inFIRE Conference Presentation Dr Ganesh Nana Chief Economist BERL 01 March 2012.
The basics for simulations
Management Matters in Healthcare. 1 Agenda Measuring management practices in healthcare 2 Describing management across hospitals 3 Drivers of management.
Effectively applying ISO9001:2000 clauses 6 and 7.
Part 3 Marketplace Dynamics
Customer-Driven Marketing Strategy Creating Value for Target Customers
Real Estate Market Analysis
Poverty and Health BCHLA Webinar Dr. Brian O’Connor, MD, MHSc April 17, 2013.
Evaluation of an intervention to increase online filing of individuals’ tax returns Peter Lumb September 2009.
MARKETING MANAGEMENT 13th edition
Saint Paul Fire Department
Employment Ontario Program Updates EO Leadership Summit – May 13, 2013 Barb Simmons, MTCU.
2011 WINNISQUAM COMMUNITY SURVEY YOUTH RISK BEHAVIOR GRADES 9-12 STUDENTS=1021.
2011 FRANKLIN COMMUNITY SURVEY YOUTH RISK BEHAVIOR GRADES 9-12 STUDENTS=332.
Partnership Approaches to Preventing and Tackling Child Poverty 17 October 2011 James Hardy Children Schools and Families Locality Manager.
All-Age Integrated Commissioning Strategy (Health and Wellbeing) CAS Voluntary Sector Forum workshop 17 July 2014.
Joint Strategic Needs Assessment 2012/13 Doncaster.
Cross-jurisdictional Policing Perspectives Jon White, ANZPAA CEO 13 November 2012.
Historical Changes in Stay-at-Home Mothers: 1969 to 2009 American Sociological Association Annual Meeting Atlanta, GA August 14-17, 2010 Rose M. Kreider,
LFB - Responding to New Challenges in an Uncertain World Ron Dobson CBE QFSM FIFireE – London Fire Commissioner.
THINK! Child road safety campaign: Tales of the Road Using MAST (Road Safety Analysis tool) to inform development of campaign strategy.
LOCALISING CHILD POVERTY TARGETS: A TOOL KIT FOR LOCAL PARTNERS.
What the quarterly Labour Force Survey can tell us about the economic circumstances of people with sight loss Sue Keil RNIB.
Predicting the numbers of older people “helped to live at home” A needs-based model and its implications Project undertaken for ADSS Eastern Region – Project.
1 Case Study 2: Choosing between ACS 1-, 3-, and 5-Year Estimates 2009 Population Association of America ACS Workshop April 29, 2009.
Chapter 14 Fraud Risk Assessment.
Making London a safer city The Difficulties In Reaching And Effectively Communicating With Vulnerable Groups Within The Community A London Perspective.
Page 1 Boscombe Strategic Assessment 21 st July 2011.
2011 Census results for Edinburgh summary results for Edinburgh City Centre CEC Planning Information, Services for Communities, February 2014.
Public Service Promise September Partners The Vision Partners share a commitment to developing services organised for the convenience of local.
Modelling Crime: A Spatial Microsimulation Approach Charatdao Kongmuang School of Geography University of Leeds Supervisors Dr. Graham Clarke, Dr. Andrew.
Views of Commercial Users Barry Leventhal MRS Census & Geodemographics Group ONS Summer Workshop 24 th July 2013.
Blackpool Council Health OSC December NHS Blackpool Approach to Social Marketing Insight Choose Well Richard Emmess.
Bromley by Bow ward is the most deprived ward in Tower Hamlets and is within the 5% most deprived in the country Large ethnic minority community: 40%
Urbanization as a Social Determinant of Health Marilyn Rice, MA, MPH, CHES Senior Advisor in Health Promotion Coordinator, Urban Health & Health Determinants.
What’s new in the Child Poverty Unit – Research and Measurement Team Research and Measurement Team Child Poverty Unit.
Improving Services to Businesses Rita Wilson Strategic Director Lichfield District Council West Midlands Business Matters Project Lead Brian Higgs Corporate.
Using Statistical Analysis to Target home fire safety visits. EFA 7 th Round Table, Brussels – September 2008.
The Leeds Joint Health and Wellbeing Strategy Explaining the approach to creating the refreshed Joint Health and Wellbeing Strategy for Leeds and.
Too high a price: Injuries and accidents in London Caroline Lowdell, Justine Fitzpatrick, Ruth Wallis, Jenny Mindell and Bobbie Jacobson.
supported by a local government initiative Improving together esd-toolkit Scottish TLC 26 th October Nicki Gill Sheila Apicella.
Making London a safer city 26 th September 2007 Introduction Nick Collins, Assistant Commissioner Risk London Fire Brigade.
Presentation transcript:

Targeting home fire safety checks Andy Mobbs - London Fire Brigade Risk Information Team February 2008

Reducing accidental fires in the home Modernisation of the fire service put clear focus on prevention as well as response Strategy for HFSVs started in 2003 Yearly target of approx 65,000 visits With 3.5 million homes in London and around 6,500 fires in the home, visits must be effectively targeted to reduce risks 3.25 percent of homes visited to date

Problems with targeting – where Incident locations vary year-to-year Accidental dwelling fires for Southwark 2005, 2006 & 2007

Problems with targeting – who CLG at risk groups defined from those fires that cause greatest harm Broad descriptions cannot easily be translated into real people Older people Children and young people Ethnic minorities or faith groups People with learning or physical disabilities People with alcohol or drug use problems People with mental illness People in poor housing

Solution Develop a risk model that combines our incident data with what we know about people, their society and their environment in such a way that we can predict the real level of risk (incident likelihood) Use consumer profiling to describe (and locate) at risk groups by lifestyle ….which we call iRAT

What is iRAT? An Incident Risk Analysis Toolkit A statistical approach to analysing the occurrence of incidents A way of combining data about incidents, people, society and geography Prioritising households by lifestyles A systematic way to deliver prevention activities in the areas of highest likelihood to those who experience incidents most often

How is iRAT used? Outputs from the statistical modelling are presented as risk rankings i.e. how does ward A compare with ward B in terms of incident likelihood This gives at pan-London or borough level assessment of relative risk Wards ranked highest should receive the greatest focus of our prevention activities Within the areas of highest likelihood, priority should be given to those who experience a disproportionate number of fires

Prioritising households using Mosaic

What is Mosaic? A commercially available consumer profiling database Its comprehensive analysis of citizens at postcode and household level provides deep insight into the socio- demographics, lifestyles, culture and behaviour of UK citizens Detailed lifestyle profiles together with preferences for marketing communications Additionally, it provides a common currency that enables the same citizen to be viewed in the same way by all public bodies, supporting joined-up government and partnership working

Comparing Mosaic groups with accidental fires in the home Previous targeting has focused on what we have learnt about fire death and injury But this targets the consequence of fire and not necessarily those who are most at risk of a fire occurring Less is known about those who experience fire but are not harmed by it Approximation using consumer profiling More complete profile of people that does not focus on one single factor Age or ethnicity or deprivation

Analysing accidental dwelling fires using Mosaic ADFs compared with Mosaic Groups Mosaic GroupNo. of HHNo. of AFDsProp. of HHProp. of ADFsADFs per HHIndex A 458, %8.79% B 87, %1.41% C 505, %9.20% D 499, %14.32% E 947, %26.04% F 470, %17.82% G 24, %0.78% H 163, %3.65% I 56, %2.28% J 103, %2.80% K 1, %0.02% n/a % Totals 3,318, %

Analysing accidental dwelling fires using Mosaic Over representationUnder representation

Comparing Mosaic groups Group E - Young, single and mostly well-educated, these people are cosmopolitan in tastes and liberal in attitudes Key Features - Young singles, few children; Full time students/Professionals; Open-minded Receptive to - Internet; Leaflets, Posters; Telephone advice lines; Magazines, Broadsheets Group F - People who are struggling to achieve rewards and are mostly reliant on the council for accommodation and benefits Key Features - Families, many young children; Low incomes; Heavy watchers of TV Receptive to - TV; Telemarketing; Leaflets, Posters; Red top newspapers

Taking the work forward Accessibility of mapped data Priority postcode data Mistrust of statistics Confusion between at risk, prevalence vs. proportions and risk reduction Need to refine Mosaic by using type analysis Applying Mosaic to other incident types

Questions?

Contact details London Fire Brigade Risk Information Team Andy Mobbs Paul Eady Riana Smit

Modelling explained

Statistical approach Multi-variate regression modelling Standard industry technique 70 data sources used Includes, historic incidents, demographic and geographic data Data used to explain pattern of historic incidents, then rolled forward to give a predictive element to the modelling Best model selected after many iterations and evaluation of outlying data Each ward then ranked based on predicted incident count iRAT model outputs give better match to incidents than any single variable (incident trends, deprivation, rented accommodation etc) Approach validated by independent consultants

Data sources National Statistics 2001 Census Demographics, dwelling types, health, transport Ordnance Survey Geographic area, road types and length Mosaic consumer profiles 11 groups Indices of multiple deprivation Income, employment, health, education, housing, crime, living environment Regulatory fire safety site inspections High, medium and low risk grading Historic incident information 8 years of incident data Home fire safety visits 3 years of visit data

Risk analysis aligned to performance management Incident types linked to performance management BVPIs, local indicators Cause considered Prevention activities are significantly different Accidental or deliberate Risk analysis shows where to target Performances analysis shows impact 17 incident types modelled