Warm Front Better Health: health impact evaluation of the Warm Front scheme
Warm Front, Fuel Poverty, temperature, energy efficiency, indoor temperature, London, Paul Wilkinson, residents, homes, thermal comfort, conditions, Tadj Oreszczyn, Home Energy Efficiency Scheme, indoor temperatures, winter deaths, properties, fuel consumption, Warm Front Study Group, Jan Gilbertson, poor health, Cold Comfort, fuel bills, Physical health, cold conditions, National Audit Office, Michael Marmot, cause stress, levels of stress, stress, Ben Armstrong, physical function, English House Condition Survey, installation, World Health Organization, comfort levels, Nicol F., room temperatures, scientific evidence, intervention studies, living conditions, The Stationery Office, energy efficiency rating, Energy Retail Association, Rudge J
Warm Front Better Health: Health impact evaluation
of the Warm Front Scheme GREEN, Geoff and GILBERTSON, Janet Available from Sheffield Hallam University Research Archive (SHURA) at: http://shura.shu.ac.uk/18167/ This document is the author deposited version. You are advised to consult the publisher's version if you wish to cite from it. Published version GREEN, Geoff and GILBERTSON, Janet (2008). Warm Front Better Health: Health Impact Evaluation of the Warm Front Scheme. Project Report. Sheffield, CRESR. Copyright and re-use policy See http://shura.shu.ac.uk/information.html Sheffield Hallam University Research Archive http://shura.shu.ac.uk
Warm Front Better Health Health Impact Evaluation of the Warm Front Scheme Geoff Green and Jan Gilbertson for the Warm Front study group
Warm Front Better Health Health Impact Evaluation of the Warm Front Scheme.
Science and policy
The Warm Front Scheme 4
Body and soul
Targets and targeting
.Published by the Centre for Regional, Economic and Social Research · Sheffield Hallam University ISBN 978 1 843 87 275 7 May 2008
Far too many people are affected by the scourge of fuel poverty. It derives from the combination of the poor thermal efficiency of buildings, low household incomes and high energy prices. Over the past decade in particular, UK governments have committed themselves to tackling this problem. Significant public expenditure has been targeted to address fuel poverty, particularly via the Warm Front scheme, complemented by the activities of energy suppliers through the Energy Efficiency Commitment. But recent increases in energy prices have slowed progress and even reversed earlier reductions amongst those occupying expensive-to-heat homes. It is perhaps judged self-evident that a cold, damp home is most likely to have a deleterious impact on the health of its occupants. Indeed, there is a growing body of research on specific health dimensions, such as cold-related excess deaths in winter, asthma, cardiovascular disease, or the psycho-social benefits of a warm and comfortable home. But little of this earlier work has adopted a systematic approach, or sought to identify and analyse all direct and indirect health impacts. Funded by the Government Department for Environment, Food and Rural Affairs (Defra) and the Welsh Assembly Government, a major study of the health impacts of the Government's flagship Warm Front programme was undertaken between 2001 and 2006. Led by
Professor Geoff Green of Sheffield Hallam University, this research involved an experienced and multi-disciplinary team based at the London School of Hygiene and Tropical Medicine, the National Centre for Social Research, Sheffield Hallam University and University College London. Project finance and management was undertaken by the Energy Saving Trust; and the study was overseen by a Steering Group comprising representatives of several central and devolved government departments, leading medical academics and charities, and the Warm Front scheme manager Eaga. Much of the detailed research arising from this project has been published recently in peer-reviewed academic journals. But this final Overview Report summarises the key findings of a large and very ambitious research programme. It is, we all believe, essential reading for those engaged in tackling fuel poverty and in reducing its health impacts. Finally, I express sincere thanks to all of the researchers engaged in this vital project for their dedicated enthusiasm and sheer hard work; to those who assisted the research through their active engagement in the study; to the Energy Saving Trust for their project management; and to the members of the Steering Group for their invaluable research and policy insights.
Professor John Chesshire OBE Chair of the Evaluation Steering Group Chairman of the Energy Efficiency Partnership for Homes Deputy Chairman of the Fuel Poverty Advisory Group
Members of the Steering Group Prof. John Chesshire (chair) Elaine Waterson/Zoltan Zavody (secretaries) Pam Wynne (Defra) Dr. Noel Olsen (Public Health Physician) Prof. David Strachan (St. George's Hospital Medical School) Prof. Mark McCarthy (University College London) Prof. Mike Harmer (Welsh Assembly Government) Marina Pappa (Defra) Diana Lowe (Department of Health) Colleen Williams (Department of Health) Dr. Mike Gill (formerly South East Regional Director of Public Health) Philip Hurst (Age Concern England) Dr. Rebekah Widdowfield (Scottish Executive) Kristen Ormston (Office of the Deputy Prime Minister) Paul Lincoln (National Heart Forum) Dr. Bhash Naidoo (Health Development Agency) Annette Rowe (Eaga Partnership Ltd) Aodhan O'Donnell (Health Promotion Agency, Northern Ireland) Martin Hodges (Bristol City Council)
Acknowledgements I should like to add my thanks to all members of the Steering Group chaired so creatively by John Chesshire. His support and encouragement steered us through this `large and very ambitious research programme.' Special thanks to Dr. Noel Olsen who was one of the prime instigators of the project, to Pam Wynne, Head of Fuel Poverty Team and Sustainable Energy Policy for Defra, for her steadfast support throughout, and to Zoltan Zavody, Elaine Waterson and Brooke Flanagan from the Energy Saving Trust who successively managed and monitored the contract so positively. The study, the scientific papers and this summary are a collective effort. The research team listed below harnessed the building science of University College London led by Tadj Oreszczyn, the environmental epidemiology of the London School of Hygiene and Tropical Medicine led by Paul Wilkinson and our own social science at Sheffield Hallam University. There were great synergies in this multi-disciplinary team and working together was I think a rewarding experience for us all. The logistics of the project were demanding, with tight timelines and a very small window for surveys of people and property before they received Warm Front measures. Many thanks to Tony Burton and Martin France from Eaga, the managing agents for Warm Front who supplied addresses and works schedules; to Richard Boreham from the National Centre for Social Research who managed the surveys of Warm Front recipients; to Roger Critchley from First Report who organised the property schedules and Glyn Johnson from Managed Services and Consultancy who undertook the property surveys and to Bill Wilkinson from EAC Integrated Services who produced energy efficiency ratings. Thanks above all to my friend and colleague Jan Gilbertson who contributed greatly to the scientific papers and to managing the whole process. Geoff Green Professor of Urban Policy, Centre for Regional, Economic and Social Research, Sheffield Hallam University. May 2008
Members of the Research Team
University College London Sung Hong, Tadj Oreszczyn, Ian Ridley Sheffield Hallam University Roger Critchley, Jan Gilbertson, Geoff Green, Michael Grimsley, Bernadette Stiell
London School of Hygiene and Tropical Medicine Ben Armstrong, Zaid Chalabi, Jacke Dowie, Emma Hutchinson, S. Hajat, Megan Landon, Wendy MacDowell, Maryjane Stevens, Nicki Thorogood, Paul Wilkinson National Centre for Social Research Richard Boreham
1 Introduction Joined-up thinking encouraged the UK Government to commission this Health Impact Evaluation of the Warm Front Scheme. 2 Scientific context By providing robust scientific evidence, universities can assist the policy community in making cost-effective investment decisions. 3 Conundrums Though politicians like cross-cutting programmes which deliver `win-wins' across different policy domains, the science is nuanced. Our study illuminates a number of conundrums. 4 Warm Front Scheme The Warm Front Scheme is the main instrument of the UK Government's Fuel Poverty Strategy and has the potential to deliver significant health gains. 5 Research design Scientists from three universities and many disciplines developed an innovative and ethical method of evaluating the health impact of Warm Front. 6 Targeting Though a government objective is to target Warm Front to householders in fuel poverty, in practice such selectivity is difficult to achieve without intrusive checks on status. 7 Application process Warm Front recipients are generally positive about the application process, probably because they exercised choice by signing up. 8 The works Generally recipients are also positive about the work of installing Warm Front measures, quickly recovering from the disruption. 9 Temperature Installation of Warm Front measures insulation and better heating systems had exactly the intended effect. Energy efficiency improved and indoor temperatures increased. The coldest properties benefited most. However a significant minority of properties are still below the 18°C threshold, posing a health risk. 10 Humidity Though insulation can reduce ventilation and increase humidity, there is no evidence of Warm Front having this effect. On the contrary, higher indoor temperatures produced lower humidity and less dampness.
11 Fuel Our evidence on fuel consumption is contradictory. Though fewer residents reported difficulty paying fuel bills after Warm Front, their overall fuel consumption increased. This unresolved conundrum bears on the climate change and fuel poverty agendas. 12 Comfort Residents reported greater thermal comfort after Warm Front, feeling most comfortable at only 19.1°C. Nearly 2°C below the Government's recommended threshold, this average comfort vote has a positive bearing on fuel consumption and climate change targets. 13 Stress By focusing on physical health outcomes, the Fuel Poverty Strategy neglects the major psychosocial benefits of Warm Front, including the alleviation of stress. 14 Mental health Relief from financial pressures is associated with a reduction in anxiety and depression. Reducing fuel poverty is a major route to improving mental health. 15 Body and soul Resident satisfaction with living conditions is linked to better general health. Better mental health may over time lead to better physical health. 16 Physical health Though we have no direct evidence of Warm Front having an impact on physiological health, there is evidence of an indirect pathway via the alleviation of fuel poverty and stress. 17 Death Cold indoor temperatures are significantly linked to excess winter deaths in England and Wales, primarily because of heart failure. 18 Health impacts Overall we assess the Warm Front Scheme as having a positive impact on (a) improving mental health (b) alleviating respiratory problems in children and (c) reducing deaths of older people. The investment is cost-effective in extending years of life. 19 Conclusion Our study broadly confirms `wins' in the two policy domains of health and fuel poverty, plus a possible `win' slowing climate change.
Independent Health Impact Evaluation of Warm Front
Main message: Joined-up thinking encouraged the UK Government to commission this Health Impact Evaluation of the Warm Front Scheme.
`The first priority is therefore to ensure that by 2010 no
energy efficiency of homes (b) the cost of energy and (c) low incomes.
older householder, no family with children, and no
Launched in 2000, Warm Front is the Government's main tool for
householder who is disabled or has
tackling fuel poverty in England via grants
long-term illness need risk ill-health Figure 1: Potential pathways to health to improve home energy efficiency.
due to a cold home.' The UK Fuel Poverty
Strategy, Department of Trade & Industry. 2001, page 10.
higher indoor temperature
greater thermal comfort
This Health Impact Evaluation of the Warm Front Scheme is a world leading research project which reflects
less fuel poverty
better health less stress
Essentially, the evaluation
explored the impact of
less use of health services
Warm Front on a number of possible pathways to health,
shown schematically in
`joined up thinking' by the UK Government.
Commissioned by the Energy Saving Trust on
Though links between poor quality housing, fuel poverty and
behalf of the Department for Environment, Food and Rural Affairs
health are widely recognised, there are few robust studies showing
responsible for energy efficiency it is also supported by the Welsh
positive links between better housing and better health2, and fewer
Assembly, the Department for Business, Enterprise and Regulatory
still address the extent to which energy efficiency improvements
Reform which leads on fuel poverty and the Department of Health
promote better health. This makes it difficult for many in the health
which supports `upstream' measures to promote health and reduce
sector to fully engage with the issue. So over the past 5 years a
demands on the National Health Service.
research team 3 drawn from three universities has monitored thousands
of houses and surveyed their residents to test the hypothesised impact
Compelling evidence1 from the UK and beyond links cold housing with
of Warm Front. The results of our enquiry appear as a series of
poor health and gave impetus to the UK Fuel Poverty Strategy
scientific papers published (and forthcoming) in academic journals.
published in 2001. It aims to remove the health risks of cold homes
This report distils some of the key findings for the wider policy
with measures addressing the main causes of fuel poverty (a) poor
London School of Hygiene and Tropical Medicine · Sheffield Hallam University · University College London
1 Acheson D. Independent Inquiry into Inequalities in Health Report. The Stationery Office (London 1998) Available from www.archive.official-documents.co.uk/document/doh/ih/part2c.htm#'4 2 Thomson H, Pettigrew M and Morrison D (2002) Housing Improvement and Health Gain: Summary and Systematic Review. Occasional Paper 5. MRC Social and Public Sciences Group. Glasgow. 3 London School of Hygiene and Tropical Medicine, Sheffield Hallam University, University College London.
Science and policy
Main message: By providing robust scientific evidence, universities can assist the policy community in making cost-effective investment decisions.
In Millennium year 2000, scientists and politicians
finally `shook hands.' Two concepts originating in 400
the academic world had assumed political salience.
First, Dr. Brenda Boardman's book on `Fuel Poverty,'
published a decade earlier 4, resonated with a
renewed government commitment to address all
aspects of poverty.
Figure 2: Excess winter deaths in London
Daily deaths in London
Excess winter deaths
Second, the concept of `Excess Winter Deaths' (EWD) already enumerated in a series of academic papers as a significant problem in the UK
The figure shows seasonal fluctuations. Individual deaths are represented by red points.
The solid line shows a two-month moving average.
01 Jan 1994
01 Jan 1996
had just been linked definitively to poor housing
by Paul Wilkinson's team from the London School of Hygiene and Tropical Medicine.5 Figure 2 shows the trend of excess winter deaths
in London between 1986 and 1996.
But third, a systematic review of the Health effects of housing
improvement,8 reflected scientific concerns about the exact nature and
Three other sources of evidence helped set the scene for publication of
strength of causal chains, about both the attribution of poor health to
the UK Fuel Poverty Strategy in 2001. First, in Cutting the Cost of
cold homes and especially better health to improved homes. A parallel
Cold: Affordable Warmth for Healthier Homes,6 a group of leading
can be drawn with early science on the link between smoking and lung
academics summarised the state of scientific evidence linking poor
cancer. Like Sir Richard Peto in the 1950's, scientists and policy-
health with cold housing conditions and offered similar prescriptions for
makers were faced with some biological evidence, some statistical
improvement. Second, the extent of the problem was revealed by the
evidence from a number of small `intervention' studies (assessing the
long awaited, and much trailed Energy Supplement to the 1996
impact of housing investment) but not yet a definitive connexion which
English House Condition Survey (EHCS),7 thanks to the persistence of
took account of all the complex influences on peoples' lives. The
government scientist, Richard Moore.
Government view was that because of its large scale, our Health
Impact Evaluation of Warm Front would add to the weight of scientific
4 Brenda Boardman (1991) Fuel Poverty: from cold homes to affordable warmth. Bellhaven Press. London. 5 Results were known a year prior to formal publication in Wilkinson P, Landon M, Armstrong B, Stevenson S, McKee M. (2001) Cold Comfort: the social and environmental determinants of excess winter deaths in England, 1986-1996. Joseph Rowntree Foundation. York.
evidence, helping legitimate its substantial investment in the UK housing stock.
6 Rudge J. & Nicol F. (eds) Cutting the Cost of Cold: Affordable Warmth for Healthier Homes. E&FN Spon. 2000.,
7 Department of Environment, Transport and the Regions. (2000). English House Condition Survey 1996: Energy Report. DETR. London.
8 Hilary Thomson, Mark Petticrew, David Morrison. (2001) Health effects of housing improvement:systematic review of intervention studies. British Medical Journal 2001; 323:187-190.
Warm homes : better health? 4 Conundrums
Main message: Though politicians like cross-cutting programmes which deliver `win-wins' across different policy domains, the science is nuanced. Our study illuminates
a number of conundrums.
(a) Biology versus thresholds The Fuel Poverty Strategy refers to ample scientific evidence of a direct physiological link between cold conditions and increased risk of both heart and respiratory disease. Yet it was not absolutely clear how this human biology 9 10 translates into minimum domestic temperatures. The Strategy recommends 18°C in living rooms to avoid risk to health, yet the EHCS Energy Report defines unhealthy living room temperatures as below 16°C and various reviews by the World Health Organization are inconclusive.11 A human comfort zone above 21°C is also contentious, with `objective' laboratory generated thresholds increasingly challenged by `adaptive' 12 models of `subjective' comfort a `malleable construct' of the human psyche. Or put another way, comfort can depend on whether you wear a cardigan. (b) Damp, mould and mites Dust mites thrive in warm humid conditions. Mould grows in cold damp conditions. Both are linked to respiratory problems, including asthma. The challenge for Warm Front is to create an indoor environment which 9 James Goodwin. Cold stress, circulatory illness and the elderly. In: Rudge J, Nicol F, eds. Cutting the cost of cold: affordable warmth for healthier homes. London: E&FN Spon 2000. 10 Ken Collins. Cold, cold housing and respiratory illnesses. In Rudge J, Nicol F, eds. Cutting the cost of cold: affordable warmth for healthier homes. London: E&FN Spon 2000. 11 World Health Organization. Health impact of low temperatures. WHO Regional Office for Europe. Copenhagen, 1987. 12 Brager GS, De Dear RJ. Thermal adaptation in the built environment: a literature review. Energ Buildings 1998; 27:83-96. 13 Tadj Oreszczyn and Stephen Pretlove. Mould Index. In Rudge J, Nicol F, eds. Cutting the cost of cold: affordable warmth for healthier homes. London: E&FN Spon 2000. 14 William Keatinge and Gavin Donalson. Cold weather, cold homes and winter mortality. In Rudge J, Nicol F, eds. Cutting the cost of cold: affordable warmth for healthier homes. London: E&FN Spon 2000. 15 Geoff Green and Jan Gilbertson. Housing, poverty and health: the impact of housing investment on the health and quality of life of low income residents. Open House International 1999; 24:41-53. 16 Paul Wilkinson, Sam Pattenden, Ben Armstrong, Astrid Fletcher, R Sari Kovats, Punam Mangtani and Antony McMichael. Vulnerability to winter mortality in elderly people in Britain: population based study. British Medical Journal 2004; 329: 647-0.
discourages both mould and mites. Some energy efficiency measures will raise temperatures and reduce Relative Humidity, condensation and damp.13 But there is a danger that draught-proofing may reduce ventilation and increase humidity. How can Warm Front strike a balance between these competing elements? Our health impact evaluation was designed to address the issue with innovative building science. (c) Poverty versus lifestyle Prior to the Strategy, scientists argued about the causes of between 20 and 40,000 excess winter deaths (EWD) in England and Wales. Do they reflect the pattern of environmental and economic inequalities in our society or do individuals bear responsibility? Pointing to societal causes, Cold Comfort highlighted housing conditions as a significant correlate of EWD in the UK. Pointing to contrasting European lifestyles, the Eurowinter Group maintained that winter illness and death is also caused by people unprepared when venturing outside in cold weather.14 Scientists have yet to determine which are the most important. If lifestyle influences do predominate then the scope for making an impact with Warm Front is limited. (d) Poverty versus fuel poverty Logically, fuel poverty and poverty are connected. The Strategy identified low incomes as one of three determinants of fuel poverty and Breadline Britain identified the lack of warm living conditions as a key component of poverty.15 Yet a large scale population study by Paul Wilkinson concluded that `Elderly people are more likely to die during the winter, but being poor does not affect the risk of dying.' 16 Though other studies appear to confirm that EWD is not a function of socioeconomic status, certainly the risk to health is increased by low temperatures. The issue to be resolved is the precise pathways between poverty, fuel poverty, inadequate heating and poor health.
The Warm Front Scheme
Main message: The Warm Front Scheme is the main instrument of the UK Government's Fuel Poverty Strategy and has the potential to deliver significant health gains.
We set out to evaluate the health impact of the Warm Front Scheme operating from June 2000 until it was upgraded in 2005 following
publication of The Government's Plan for Action.17 The National Audit
The target population is those in fuel poverty but only forensic
Office provided a good review and critique of the Scheme as we were
examination of individual household accounts would reveal who they
finishing our field work in 2003.18 In brief, Warm Front's aim is to improve
Figure 3: Figure title
are and where they live. Because this test is difficult to apply in practice, eligibility is
energy efficiency for vulnerable households in fuel poverty in the private
Before Warm Front : spatial shrink
instead based on receipt of specific state income benefits. Filtered through this
rented and owner-occupier sectors. It
process are the groups most vulnerable to
provided grants of up to Ј1500 for
the effects (primarily the health effects)
insulation, energy efficiency measures
of fuel poverty; low income households
and heating improvements. Warm Front
with children; disabled people or those
Plus is an extension of the Scheme
with long-term illness; and older (over
available to the over 60s which provides
60) low income households.
central heating and had then a grant
maxima of Ј2500.
There are two main types of energy
Warm Front, overseen and funded by
After Warm Front package
efficiency measures made available to
DEFRA, was administered by two scheme
Warm Front recipients. First are
managers, Eaga Partnership Ltd and TXU Warm Front Ltd. Expenditure of around Ј150 million a year during the period of the study was increased following the Government's 2004 Spending Review. Between
Insulation cЈ1500 max Heating repair or replacement cЈ2500 max
improvements in the heating system. At
the top end of the scale, the Ј2500 grant
Draught proofing cЈ100
under Warm Front Plus facilitated the installation of a new central heating system. Less costly and
within the Ј1500 grant limit for Warm
Front recipients aged below 60, were
2001 and 2004 the
measures to repair systems, install
Scheme assisted more than 900,000
individual room heaters or replace
vulnerable households by spending more
defective central heating boilers. Second
than Ј600 million.
is insulation, to lofts, cavity walls, hot water tanks and via draught
proofing, to doors and windows. Other measures include energy
17 Department for Environment Food and Rural Affairs. (2004) Fuel Poverty in England: The Government's Plan for Action. DEFRA. London. 18 National Audit Office (2003) Warm Front: Helping Combat Fuel Poverty. The Stationery Office. London.
advice, energy efficient light bulbs, a thermal jacket to the hot water tank and timer controls for electric space and water heaters. If all goes well, homes will be transformed from (a) to (b) in figure 3.
Main message: Scientists from three universities and many disciplines developed an innovative and ethical method of evaluating the health impact of Warm Front.
Evaluating the health impact of Warm Front was a complex business. How living conditions changed, how residents responded, how far sustained improvements in health and
Figure 4: Measuring instruments
Building Social Environment Science Science Epidemiology
well-being replaced the stresses of the improvement process, the implications for winter mortality all these issues
1. Property survey
property characteristics; energy efficiency
required a research input from a combination of building science, social science and environmental epidemiology.
2. Electronic data loggers 3. Resident interview
indoor temperature and relative humidity income, subjective health, well-being, comfort, stress
Measurement and modelling
comfort, temperature, daily activities
In order to measure all these dimensions, the team deployed a battery of instruments shown in figure 4.
5. In-depth interviews
process, opinion, impact
The research team planned to survey about 4000 properties
air infiltration, insulation
building survey fuel consumption
and interview one resident in each household, asking them
also to keep daily diaries for a couple of weeks afterwards. Electronic
the Winter Fuel Allowance. However, ethically it was difficult to justify
data loggers recorded temperature and humidity in the living rooms
forming a control group of eligible residents by deferring their receipt
and bedrooms of about half of these properties. There were 49 in-
of Warm Front measures another winter. So the team devised an
depth interviews and intensive surveys of 191 properties.
innovative survey sequence to compare pre- and post- intervention
groups at two points in time. Technically this facilitated both a `cross-
Realising from the onset that it would be difficult to evaluate some
sectional' and `longitudinal' analysis.'
health impacts directly (and almost impossible to
detect a reduction in deaths) the study also used modelling techniques. Better health
Figure 5: Survey sequence
and fewer winter deaths were estimated by `triangulating' evidence from other scientific studies linking health to rising domestic temperatures.
Post-Improvement N = 477 Pre-Improvement N = 373
WAVE 2 SURVEY Winter 2002-3 Post-Improvement N = 231
Before and after
Post-Improvement N = 785
Ideally, a robust study would (a) compare living
WAVE 1 SURVEY Winter 2001-2
conditions and health `before' and `after' the Warm
Front measures were installed, and (b) for the
same period deploy a `control group' of `non-intervention' properties to
account for changes resulting from other factors, such as an increase in
Pre-Improvement N = 1050
Targets and targeting
Main message: Though a government objective is to target Warm Front to householders in fuel poverty, in practice such selectivity is difficult to achieve without intrusive
checks on status.
Who and where are the vulnerable households in fuel poverty? Finding them has been a major challenge for Warm Front; highlighted by the
Modelling and measuring
National Audit Office (NAO) in 2003,19 criticised by the Public Accounts
Instead, Warm Front identifies low incomes as proxy for fuel poverty,
Committee, then addressed by the
Figure 6: Mismatch
Government's Plan for Action in 2004. Warm Front recipients Fuel poor
retaining a `state benefit passport' approach to eligibility. A complex model estimates both the extent of fuel poverty and the impact of Warm Front. The key outcome of `adequate
warmth' is predicted by modelling income, fuel costs and
The problem is a mismatch between fuel poor households
little improvement made non-vulnerable
already energy efficient
Warm Front reaching not eligible the right people and for benefits
energy efficiency. Our study throws new light on targeting by identifying
and eligibility for Warm Front measures. The result is (a) many
non-means improving the energy not claiming
tested benefits efficiency of their
cold homes.21 If low temperatures of below 16°C in either living room or bedroom are signifiers of fuel
fuel poor households not claiming or ineligible for Warm Front, and (b)
middle homes hard to treat
poverty, then what characterises these households? Can these common characteristics help locate fuel poverty?
many households eligible for Warm Front
but not in fuel poverty. The NAO identified causes (figure 6) and Tom
Our results are only modestly encouraging. Figure 7 shows three
Sefton 20 identified numbers 82% of 2.8 million eligible households not in fuel poverty and 62% of 1.4 million fuel poor 100
Figure 7: Targeting cold homes
models of increasing complexity. Model 1 includes three variables available to a local authority without a
households not eligible for Warm Front.
% of cold homes captured
40 At the heart of the problem is a dilemma. More precision could be achieved by 20
Model 1 Model 2 Model 3
visit property age and type plus the neighbourhood index of multiple deprivation. On this basis it would be necessary to target 73% of homes in England to capture 80% of the households with living room
thoroughly investigating the financial
temperatures of less than 16°C. A
circumstances and housing conditions of
80 100 refined model 2 (adding a short
potential beneficiaries. It's an option
interview with the householder to
favoured by the Energy Retail Association, but `checking up' on
ascertain age, sex, educational attainment, household size and
applicants or trawling records to identify potential clients is intrusive
satisfaction with the heating system) targets 57% of households to
and costly to administer.
capture 80% of cold homes. Model 3, adding in an energy efficiency
rating, focuses on only 50% of households for the same result.
19 Report by the Comptroller and Auditor General (2003). Warm Front: Helping to Combat Fuel Poverty. National Audit Office. The Stationery Office. London. 20 Tom Sefton. (2004) Hitting the Target: An evaluation of the effectiveness of Warm Front in helping to meet the Government's fuel poverty target. London School of Economics. London. 21 Emma Hutchinson, Paul Wilkinson, Sung Hong, Tadj Oreszczyn & the Warm Front Study Group. Can we improve the identification of cold homes for targeted home energy-efficiency improvements? Applied Energy 83 (2006) 1198-1209.
Even the most refined of our models has limited potential for identifying cold homes, presumably because of multiple influences on home heating. But the approach may help to target fuel poverty more effectively than passport eligibility.
Main message: Warm Front recipients are generally positive about the application process, probably because they exercised choice by signing up.
Shadows Though family death and divorce cause most stress in life, the process of renovating a house is high on the list: wherever you live and in almost all circumstances. So we anticipated that the process of planning and installing Warm Front measures might cast a shadow over residents' lives, possibly negating positive health impacts of warmer, more comfortable homes. Our earlier research in Liverpool had highlighted the problem, concluding: `levels of stress associated with the redevelopment process are significantly associated with poorer health and in the short term appear to counterbalance the benefits of improved living conditions.'22 Our Liverpool findings were confirmed by Terry Allen's investigation of a council estate near Bradford. In his article 23 Housing Renewal Doesn't it make you sick? he highlights how personal tenant control over the process is linked to health, but also how difficult this is to achieve with `monolithic' programmes on large council estates. Application process Warm Front is a big but not monolithic Scheme. Indeed some critics say the Scheme `pepper pots' properties rather than systematically improving all the housing in a neighbourhood like its sister scheme 22 Critchley R, Gilbertson J, Green G, Grimsley M. (2004) Housing Investment and Health in Liverpool. CRESR. Sheffield Hallam University. 23 Terry Allen. Housing Renewal Doesn't it make you sick? Housing Studies, (2000) Vol 15, No 3, 443-461. 24 Gilbertson J, Stevens M, Stiell B, Thorogood N. Home is where the hearth is: Grant recipients' views of England's Home Energy Efficiency Scheme (Warm Front). Social Science and Medicine 63 (2006), 946-956.
Warm Zones. The upside is individual choice to join the Scheme or not. And choice implies a degree of control. Our 49 `in-depth' qualitative interviews of Warm Front recipients24 reveals the process of applying to EAGA, the scheme manager, was generally straightforward "no bother" and "relatively painless". But then there was a delay of weeks, even months before Warm Front measures were installed, as contractors surveyed the property to determine the exact specification for heating systems and insulation. In some instances delays could cause confusion and anxiety "so many people came it was unreal" but most people were phlegmatic: "Yes it was a bit long drawn out but obviously there were different channels that had to be gone down, grants and what have you, but I think it is well worth the wait and it was put in before the next winter". "..Yes it seemed a bit lengthy... Yeah, between four and six months but of course there's a lot of people making applications so it is going to take a time, sit back and you know as long as it came before winter came that was alright that was the important thing". The final result of a warmer, more comfortable home appears to have compensated for earlier anxieties about delay and uncertainty. And the speedier process introduced by DEFRA in 2004 will have helped minimise this downside.
Main message: Generally recipients are also positive about the work of installing Warm Front measures, quickly recovering from the disruption.
Installation of measures Most residents of the qualitative study found the installation of energy efficiency measures relatively proficient. Over a third thought the contractors were "friendly", "polite" and "cleaned up afterwards". A quarter were impressed with their efficiency and speed in fitting boilers and other heating measures, and with the quality of the equipment. Typically, a single pensioner from Manchester said: "There was nothing wrong with the process; they were just very, very nice. They just came in and did the work; they never made a mess and they told me how to use it and everything else." For an older couple living with an elderly parent in Newcastle: "The boiler is absolutely exceptional, really is...just the quality of it, instant hot water, instant heat you switch it on and the radiators [are] hot in three, four minutes easy". There were a few cases where work was not carried out to residents' satisfaction: indeed one recipient reported a flooded cellar and another was taking the contractor to court. Typically, however, most accepted a degree of inconvenience during installation. On the whole the results were well worth it. Feelings about Warm Front Overall most people we talked to were very satisfied with Warm Front. Some talked about feeling "grateful" that such a Scheme existed and being able to benefit from having the work done. Some were surprised
to get government money for improvements to their own homes. Being a recipient of a government scheme did concern some householders. Around a fifth felt they didn't have much choice over the type of boiler or control over how the work was undertaken. In part because they felt they did not have the sense of authority that a `paying customer' would have. "I accepted simply because we was going to get a better installation than the existing one and `half a loaf is better than none' and that was the way I looked at it. I was willing to compromise on my own little faddishness about design etc. for what I could get for no cost so that was all I did." Although this was usually seen as an inevitable aspect of getting "something for nothing", a small minority did find the feelings of powerlessness difficult. "I think I've felt really bad, I've felt like I was asking for something, do you know, like begging? I've felt like I was begging, and I didn't want to be in that situation at all." Pointers Some participants in our qualitative study experienced problems with their applications and/or installations which could potentially influence their well being and quality of life. Overall there was little qualitative evidence of the Warm Front process casting a shadow over the health of recipients.
Main message: Installation of Warm Front measures insulation and better heating systems had exactly the intended effect. Energy efficiency improved and indoor temperatures
increased. The coldest properties benefited most. However a significant minority of
properties are still below the 18°C threshold, posing a health risk.
Calculated using a Standard Assessment Procedure (SAP)25 the energy efficiency of `post-intervention' properties averaged 62 (on a scale 10 to +120) compared with 41 for those awaiting Warm Front measures. This new SAP was above the English average in 2001 but below the Government benchmark of 65 for `Decent Homes.' Figure 8 compares the range of SAPs before and after Warm Front measures. The `tail' is explained mainly by older properties not amenable to cavity wall insulation. Also, younger or disabled householders did not receive a complete upgrade of their heating systems because of prevailing grant limitations.
Proportion of properties 0 10 20 30 40 50 60 70 80 90 100
Figure 8: Energy efficiency: `before and after'
30% before Warm Front 25% 20% 15% 10% 5% 0% SAP
after Warm Front
From perhaps the most comprehensive collection of dwelling-related temperature data for English homes,26 we report a significant rise in temperature in post-intervention properties. The SAP rating of a property was the most significant predictor of temperature. In the least energy efficient (SAP < 41) living room temperatures averaged
25 BRECSU (for DEFRA) 2001. The Government's Standard Assessment Procedure for Energy Rating of Dwellings, Version 9.70. ed., Garston, Watford, BRE, 2001. 26 Tadj Oreszczyn, Sung Hong, Ian Ridley, Paul Wilkinson and the Warm Front Study Group. Determinants of winter indoor temperatures in low income households in England. Energy and Buildings. (March 2003) Vol.38 Issue 3: 245-252. 27 Roger Critchley, Jan Gilbertson, Michael Grimsley, Geoff Green and the Warm Front Study Group. Living in cold homes after heating improvements: Evidence from Warm Front, England's Home Energy Efficiency Scheme. Applied Energy 84 (2007) 147-158.
17.5°C whereas in the most energy efficient they averaged 19.8°C. Figure 9 shows, insulation and heating measures both had an impact. Combined, they raised living room temperatures by 1.64°C and bedroom temperatures by 2.82°C.
Increase in degrees Celcius
Figure 9: Change in standardised temperatures with Warm Front intervention
0 Insulation only
+ 2.51 °C + 1.60 °C Heating only
+ 2.82 °C + 1.64 °C Both
Though the headline increases are modest from 17.9°C to 19.6°C in living rooms and from 15.9°C to 18.3°C in bedrooms many properties are lifted above the 18°C threshold which theoretically avoids a risk to health. Indeed one of the features illustrated by figure 10 is shortening the `tail' of coldest properties where the rise in living room temperatures was around 2.5°C compared with 1°C at the top end.
Figure 10: Range of living room temperatures
after heating plus
15 18 21 25
A minority of post-intervention properties are still below the 18°C threshold households either remain in fuel poverty or prefer cooler living conditions.27
Main message: Though insulation can reduce ventilation and increase humidity, there is no evidence of Warm Front having this effect. On the contrary, higher indoor
temperatures produced lower humidity and less dampness.
Building scientists point to a possible downside to the Warm
Figure 11: Relative humidity: causes and consequences
Front Scheme. Better insulation could reduce ventilation,
tending to increase relative humidity inside the home.
Figure 11 charts consequences for damp, mould and
respiratory problems. House mites also thrive in humid
conditions and are linked to asthma and eczema.
When relative humidity is high, moisture condenses on external walls,
providing ideal conditions for mould growth. Figure 13 shows how
Warm Front measures have cut the prevalence of severe mould from an average of 12% to 8%. Where Warm Front has increased energy
In theory retrofit measures, combining cavity wall and loft insulation
efficiency up to and beyond a SAP of 70, then the proportion with
plus draught proofing, should
severe mould falls by half to 6.8%
reduce the air infiltrating into the property by a quarter. In practice,
Figure 12: Relative humidity before and after Warm Front measures
The question remains; `why does even a
the gaps opened up by retrofitting 60 gas central heating increased air 50 infiltration. When we pressure 40 tested a sample of 191 houses, background ventilation rates in 30 post-intervention houses were on 20
small percentage of energy efficient properties suffer from mould?' The answer lies in the great variation in how occupants live their lives. For example, some properties have cold spots behind wardrobes where relative humidity is
average almost as high as in those 10
high and mould will grow. In others,
awaiting Warm Front measures.28 Relative humidity
Heating + occupants may produce a lot of moisture
Insulation or may not adequately ventilate their
homes. So their lifestyles contribute to
high humidities and mould growth. However, fabric of the property is
In a larger sample of 1095 properties monitored for relative humidity,
of paramount importance.
there was a small but significant difference between pre- and post-
intervention properties, the effect of raising temperatures.29 Figure 12
Figure 13: Severe mould
shows that insulation and heating measures both make a contribution.
When combined they reduce relative humidity from 46.4% to 42.8%
in living rooms and from 52.5% to 47.2% in bedrooms, though these
are averages and mask quite a variation.
Percentage of properties
28 Sung Hong, Ian Ridley, Tadj Oreszczyn,The Warm Front Study Group. The impact of energy efficient refurbishment on the airtightness of English dwellings. 29 Tadj Oreszczyn, Ian Ridley, Sung Hong, Paul Wilkinson, Warm Front Study Group. Mould and Winter Indoor Relative Humidity in Low Income Households in England. Indoor Built Environment 2006; 125-135.
2 0 Before
Mould Severity Index derived from the 1996 English Housing Condition Survey
Heating + Insulation
Main message: Our evidence on fuel consumption is contradictory. Though fewer residents reported difficulty paying fuel bills after Warm Front, their overall fuel consumption increased.
This unresolved conundrum bears on the climate change and fuel poverty agendas.
Warm Front is designed to reduce fuel poverty. In theory, improved energy efficiency gives recipients two options. Either they (a) reduce fuel consumption whilst maintaining adequate temperatures or (b) boost temperatures using the same amount of fuel as before. The second option is commonly referred to as `take back.' Both alleviate fuel poverty.
Figure 15: Increase in fuel consumption after Warm Front measures 30% Pre-intervention (mean=72, n=1497) Post-intervention (mean=83, n=1404) 20% 10%
Percentage Proportion 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300
Positive perceptions 0% Asked about difficulty paying their fuel bills, 2171 surveyed residents indicated a third "win-win" scenario of reduced fuel costs alongside the confirmed increase in temperatures. Heating measures (figure 14) tended to have the most significant effect in reducing the percentage reporting difficulty. Since our analysis accounts for income and fuel price variations, the logical implication is that fuel consumption has fallen, reducing the prevalence fuel poverty.
kWh/day This conundrum was first highlighted in 2005 (published in 2006)31 and explanations sought in a further report commissioned by DEFRA.32 The big question is `why does fuel consumption remain fairly constant when the government model predicts up to a 60% reduction after energy improvements?'
Figure 14: Difficulty in paying fuel bills
60 50 40 30 20 10 0 Pre-intervention Insulation only
very easy fairly easy fairly difficult
Both heating and insulation very difficult
Conundrum A closer `objective' investigation of metered fuel consumption revealed an apparent contradiction with these `subjective' resident perceptions. Fuel consumption actually rose on average after Warm Front measures (figure 15) against the 60% reduction predicted by the government's preferred model.30 30 BR Anderson et al (2002) BREDEM-12 (BRE Domestic Energy Model). BRE Press. Bracknell. 31 Sung Hong, Tadj Oreszczyn, Ian Ridley, Warm Front Study Group. The impact of energy efficiency refurbishment on the space heating fuel consumption of English dwellings. Energy and Buildings 38 (2006): 1171-1181. 32 Sung Hong, Tadj Oreszczyn, Warm Front Study Group. (2006) Additional analysis of the Health Impact Evaluation of Warm Front energy efficiency data. University College London.
Explanation Explanations fall into two categories. First, we question the theoretical assumptions of the government model, which are based on comparatively modern properties. Second, there are three uncomfortable discoveries which reduce energy efficiency below its theoretical value. Our intensive infrared images of 85 Warm Front dwellings revealed missing areas of both loft (13%) and cavity wall (20%) insulation. Monitored ventilation rates were higher than predicted because of the gaps opened up by retrofitting central heating and because occupants of warmer homes are more likely to open windows. Residents may not know how to use new central heating systems effectively; some may even resort to using their old fires. These only partly explain the discrepancies between `modelled' and `monitored' fuel consumption. Though the issue is of vital importance to both the Climate Change and Fuel Poverty agendas, the conundrum remains largely unresolved. More focused research is required to match theory with practical realities.
Main message: Residents reported greater thermal comfort after Warm Front, feeling most comfortable at only 19.1°C. Nearly 2°C below the Government's recommended
threshold, this average comfort vote has a positive bearing on fuel consumption
and climate change targets.
Patient and painstaking, 2402 residents recorded their perceptions of warmth and comfort with over 80,000 diary entries. Thermal comfort
is assessed on a seven point Bedford Scale, ranging from `much too
There is wide variation in temperatures and comfort even after Warm
cool' (1) via `too cool' (2) and `comfortably cool' (3) to a neutral
Front measures. Though some residents are clearly constrained by
`comfortable' (4); then by two stages, `comfortably warm' (5) `too
difficulties paying bills,33 a quarter say they are comfortable at
warm' (6) to `much too warm' (7). Figure 16 shows how Warm Front
temperatures lower than recommended in the Government's Fuel
works significantly raised comfort levels from an average `too cool'
Poverty Strategy. Individual metabolism accounts for some of these
Diary temperature, Td (°C) Thermal comfort score
to `comfortably cool' in bedrooms, and in living rooms from `comfortably cool' to `comfortable.' Comfort and temperature Predictably, occupants' perceptions of comfort are correlated with their readings of indoor temperature. Figure 17 shows the scatter of recordings32 both before (in blue) and after (in red) Warm Front measures. Here the Bedford Scale is
Figure 16: Thermal comfort 4.5 4 living room bedroom
preferences; as do more layers of clothing. Others choose lower temperatures, especially in bedrooms, because they
believe ventilation is good for health.
Below living room temperatures of 18°C
there is a potential conflict between
subjective feelings of thermal comfort and
scientific evidence (section 3) of an
No works Insulation Heating Insulation increased risk to health.34
rebased from minus 3 to plus 3, with zero as neutral or `comfortable'. Note the average `Comfort Vote' shifts from 18.66°C to 19.16°C, with
a reduced `tail' of residents saying their living room is too cool.
The bad news on fuel consumption (highlighted previously) is matched
Figure 17: Living room thermal comfort and temperature 30
by good news on the temperature which corresponds to the neutral `comfort vote.' Occupants saying they are `comfortable' (in the middle of the Bedford Scale) record average room temperatures of 19.0°C.
This average is below the Government's recommended comfort
temperature of 21°C, with positive implications for fuel consumption
Td =1.68CV + 18.66
and Climate Change targets.
Td =1.47CV + 19.16
As with fuel consumption, the reality of 80,000 recorded comfort votes
does not match predictions made by using Government's preferred
Comfort vote, CV
32 Sung Hong, Tadj Oreszczyn, Jan Gilbertson. (Forthcoming) Field study of thermal comfort in low-income dwellings before and after energy efficiency refurbishment. 33 Tadj Oreszczyn, Sung H. Hong, Ian Ridley, Paul Wilkinson. Determinants of winter indoor temperatures in low income households in England. Energy and Buildings 38 (2006) 245-252.
model. We inserted the property characteristics of our study sample into the official model and accounted for how occupants live their lives including the clothes they wear and the activity they undertake. The result was a 'Predicted Vote' for `comfortable' at a mean temperature of 20.4°C, 1.4°C higher than the `Comfort' Vote revealed by our study. Maybe the difference is accounted for by the lower than average
34 Roger Critchley, Jan Gilbertson, Michael Grimsley, Geoff Green. Living in cold homes after heating improvements: evidence from Warm Front, England's Home Energy Efficiency Scheme. Applied Energy 84 (2007) 147-158.
incomes and modest aspirations of our study population. Further investigation is needed.
Main message: By focusing on physical health outcomes, the Fuel Poverty Strategy neglects the major psychosocial benefits of Warm Front, including the alleviation of stress.
Living conditions profoundly influence a person's
state of mind. In a series of studies around
Glasgow, Ade Kearnes and his team
reveal how improving the physical
condition of a dwelling confers psychosocial benefits on their occupants35,36. So it proved
with Warm Front. Raised
temperatures, alleviation of fuel
poverty and better thermal comfort all tend to
reduce stress (figure 18).
Figure 18: Pathways to stress
difficulty paying their fuel bills. Figure
20 gives a baseline value of 1 to the
health services use
average stress level of residents who said they could pay their bills `easily.' Higher stress levels were
reported by residents with difficulty paying fuel
bills, with those finding it `very difficult' 2.5 times more
likely to report high or moderate stress.
Comfort and stress Low indoor temperatures are linked to stress but not as strongly as occupants' own assessment of thermal comfort. For those reporting `satisfactory' thermal comfort (mid-point on the 7 point scale in the earlier Comfort section of this report) the average stress level is given a
Moderate to high stress
Figure 19: Comfort and stress
Much too cool
Complex picture We have already shown how Warm Front has a positive effect on these psychosocial factors and will show next how they link forward to the health of occupants. The picture is of complex interrelationships. Though discomfort and fuel poverty are clearly connected, statistical analysis
baseline value of 1 (figure 19). Higher
shows both contribute independently to
stress levels were reported by residents saying their bedroom and / or
stress. Evidence of these psychosocial benefits of Warm Front adds an
living room were `too cool.' Those reporting conditions as `much too
important new outcome to the government's Fuel Poverty Strategy,
cool' were over 75% more likely to report high or moderate stress.37
which has hitherto focused on physiological benefits.
Fuel poverty and stress There is an even stronger link between stress and fuel poverty. We assessed fuel poverty by asking householders whether they had 35 Ade Kearnes, Rosemary Hiscock, Ann Ellaway and Sally Macintyre. (2000). `Beyond Four Walls.' The psychosocial benefits from the home: Evidence from West Central Scotland. Housing Studies, Vol 15, No.3, 387-410. 36 Rosemary Hiscock, Ade Kearnes, Sally Macintyre, Ann Ellaway. (2001) Ontological security and psychosocial benefits from the home. Qualitative evidence on the issues of tenure. Housing Theory and Society, 18, 50-66. 37 Michael Grimsley, Jan Gilbertson, Geoff Green. (Forthcoming) Psychosocial routes to health gain.
Moderate or high stress
Figure 20: Fuel poverty and stress
0 Very easy
Fairly easy Fairly difficult Very difficult
Paying fuel bills
Main message: Relief from financial pressures is associated with a reduction in anxiety and depression. Reducing fuel poverty is a major route to improving mental health.
We deployed a battery of health measures to gauge the impact of Warm Front. But first impressions were that only mental health showed significant improvement. And only in response to the General Health Questionnaire (GHQ12) which assesses levels of depression and anxiety.38 Compared with the responses of residents before Warm Front intervention, those in receipt of heating and insulation measures were nearly 40% less likely to report a high level of psychological distress (figure 21).39 Other mental health indicators reinforced this trend despite the disruption caused by the process of installation referred to in section 8 of this report.
Psychological distress (GHQ12 >4)
Figure 21: Warm Front reduces psychological distress
Heating and insulation
Fuel poverty Raised temperatures and thermal comfort are also associated with better mental health. Occupants maintaining bedroom temperatures at 21°C were 50% less likely to suffer high levels of psychological distress than those with temperatures less than 15°C. Fuel poverty was even more significant. Compared with householders who paid fuel bills easily (baseline value 1) those with great difficulty paying were over 4 times more likely (figure 22) to suffer anxiety or depression (on the EQ5D measure) or psychological distress (on the GHQ12 measure). 38 Kerry Sprotson & Paola Primatesta (editors) Health Survey for England, 2003: Volume 3; Methodology and Documentation. The Stationery Office. 39 Michael Grimsley, Jan Gilbertson, Geoff Green. (Forthcoming) Psychosocial routes to health gain. 40 Mark Taylor, David Pevalin and Jennifer Todd. (2006) The Psychological Costs of Unsustainable Housing Commitments. ISER Working Paper. University of Essex.
Anxiety or depression (EQ5D)
Figure 22: Fuel poverty increases depression
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0 Very easy
Fairly easy Fairly difficult Very difficult Paying fuel bills
Stress was the most significant predictor of poor mental health (figure 23). Compared with stress-free householders (baseline value 1) those with high stress levels were almost 25 times more likely to report anxiety or depression (EQ5D measure) and 21 times more likely to suffer psychological distress (on the GHQ12 measure).
Depression or anxiety (EQ5D)
Figure 23: Stress and depression
30 25 20 15 10 5 0 No stress
Low stress Moderate stress High stress
Policy pathways Of course there is an intimate and predictable relationship between stress and distress. Both are towards the end of a causal chain linking mental health back into the material circumstances of a household. Warm Front improves living conditions and comfort, but it is the Scheme's alleviation of fuel poverty which has the greater impact on improving mental health. This confirmation 40 that relief from financial pressures is associated with better psychosocial health, adds an extra dimension to the government's Fuel Poverty Strategy.
Body and soul
Main message: Resident satisfaction with living conditions is linked to better general health. Better mental health may over time lead to better physical health.
The Government's Fuel Poverty Strategy highlights how cold homes
Figure 24: Heating satisfaction and general health
mind and body
can damage physical health. Over 1.9
Poor general health (SF-36)
time, measures to improve energy
efficiency should help alleviate
these problems. Yet our study 1.4
found no direct link between Warm
Front measures and better physical 1.1 1.0
health, possibly because their full 0.9
impact was delayed beyond the 0.8
There is scientific evidence most famously in the Whitehall studies 42 43 that poor mental health is a prelude to poor physical health. Insofar as Warm Front improves mental health (see previous section) then in time, improvements in physical health should follow. Despite
the limited time frame, the results of
dissatisfied dissatisfied our study signalled a strong
association between the two. Figure 25 shows the relationship between mental health (using the GHQ12
However, these neutral headlines mask a more positive picture. Not all
measure) and the four physical dimensions of health (on the SF-36
Warm Front measures lead to significant improvements in living
measure) on a scale 1-100. For example, occupants reporting
conditions, but where they do, there is evidence of better physical as
moderate or good mental health had an average pain score of 67, a
well as mental health. Our survey of 2640 occupants shows fewer
full 24 points higher than those suffering psychological distress.
draughts are linked to better `General Health' (using the Short Form
36 measure) as are perceptions of living conditions.41
Figure 25: Physical and mental health
Compared with those satisfied with their accommodation, those who were very dissatisfied were 80% more likely to report poor General Health, even after adjusting for other influences. Similarly, compared with those who were satisfied with their heating system (baseline value 1), those very dissatisfied were on average nearly 50% more likely to report poor `General Health'. This statistically significant average is shown in figure 24 within 95 per cent confidence limits.
Self-reported health (SF-36)
Moderate to good mental health (GHQ12<4)
Physical function Physical role Pain General health
Poor mental health (GHQ12 = 4 or more)
41 Michael Grimsley, Jan Gilbertson, Geoff Green. (Forthcoming) Psychosocial routes to health gain. 42 Michael Marmot & Eric Brunner. (2004) Cohort profile: The Whitehall II study. International Journal of Epidemiology. 43 Nicholson A., Fuhrer R., (2005) Marmot M Psychological distress as a predictor of CHD events in men. Psychosomatic medicine. 67. 44 Evans R.G. , Hodge, M. and Pless, IB (1994) If not genetics, then what? Biological pathways and population health. In Evans, R.G., Barer, M.L. and Marmor, T.R. (editors) Why are some people healthy and others not? The Determinants of Health of Populations? Berlin/New York: de Gruyter, 161-88.
Policy pathways Though the timeframe of our study prevented a full assessment of health impact, we nevertheless signal probable improvements in physical health further down the timeline. Our study supports the `existence of a complex web of linkages, having important implications for health, between the nervous system and other body systems.' 44
Main message: Though we have no direct evidence of Warm Front having an impact on physiological health, there is evidence of an indirect pathway via the alleviation
of fuel poverty and stress.
Though the Government's Fuel Poverty
Figure 26: Fuel poverty, stress and physical health
measure. On a scale of bodily pain
Strategy highlights how cold homes can
ranging from 0-100, residents
damage physical health, perhaps
more damage is inflicted by the
stress of fuel poverty. Addressing these psychosocial determinants may be the
more important route
(highlighted a and b in figure 26) to
indoor temperature fuel poverty a
thermal comfort stress
physical health b
health services use
who said it was easy to pay fuel bills were on average 13 points higher than those who reported great difficulty. And the differentials for stress were
better physical health. Earlier sections have
even greater. Those with no stress were on average 23
shown how Warm Front reduces fuel poverty. What
points higher than those reporting high levels of stress. The
is the evidence on the subsequent pathway to physical health?
pattern was repeated for general health (with its large physical
component) and for physical role and physical functioning (figure 27).
Differences in physical function correspond to the scale of fuel poverty and stress; with every point of statistical significance (as illustrated by
Our study revealed45 both fuel poverty and stress are linked to all four
the 95% confidence limits).
dimensions of physical health covered by the Short Form 36 (SF-36)
Though the logical sequence is for difficulty paying fuel bills to
Figure 27: Fuel poverty and physical health 25
cause stress, the outcomes in figure 27 reflect the independent contribution of both, with stress the more important.
0 Very Fairly Fairly Very difficult difficult easy easy Paying fuel bills
High Moderate Low Stress
45 Michael Grimsley, Jan Gilbertson, Geoff Green. (Forthcoming) Psychosocial routes to health gain. 46 John Ware. (2006) SF-36 Health Survey Update. Quality Metric Incorporated. 47 Richard Wilkinson and Michael Marmot (2003) editors. Social Determinants of Health: the Solid Facts. WHO Regional Office for Health. Copenhagen.
Undoubtedly there is a two way relationship between physical function and stress. Those at the bottom of the scale `very limited in performing all physical activities including bathing or dressing' 46 are less likely because of this incapacity to be economically active and more likely to be fuel poor. On the other hand, our study results are consistent with the causal model developed by the Whitehall Studies. "Social and None psychosocial circumstances can cause long term stress... lack of control over work and home life have powerful effects on health... turning on the stress response diverts energy and resources away from many psychosocial processes important to long term health maintenance. Both the cardiovascular and immune systems are affected.'47 In time benefits from the Warm Front Scheme may help to reverse this process.
Physical function (SF-36)
Main message: Cold indoor temperatures are significantly linked to excess winter deaths in England, primarily because of heart failure.
Ratio of observed to expected deaths Deaths, winter:non-winter ratio
Death from cold is headline
Figure 28: Mortality drops as outdoor temperature rises
utilising national rather than
news and is indeed the grim 1.5 reality for many older people.
Warm Front data, we found that the trend was for the winter: non-
Mortality peaks in the coldest 1.25 winter months (see sections 2
winter mortality ratio to decrease by 1.3% per degree Celsius rise in
& 3) and in England and Wales as a whole there are 1.0
temperature. This gradient did not apply to deaths from respiratory
between 20,000 40,000
disease, but the trend for
excess winter deaths a year.
Influenza, coughs and colds 0.75
account for some of these but perhaps half are attributable to
Maximum daily temperature °C
Note: mortality rises slightly above 28C. Source: Cold Comfort
cold conditions. In their
cardiovascular (heart) deaths was highly significant. Figure 29 shows how the ratio falls as indoor temperatures increase. Put another way, for every degree
influential report Cold Comfort,48 Paul Wilkinson and his team plotted
Celsius increase in indoor temperature, the ratio falls by 2.9%.
daily deaths against temperatures, concluding that mortality falls with warmer weather (figure 28). That is until temperatures reach the mid20s°C, when heat-related illnesses start to take their toll.
Figure 29: Indoor temperatures by winter ratio of cardiovascular deaths 1.5
Cold or cold homes A key question for the Fuel Poverty Strategy is whether these coldrelated deaths are caused (a) by cold housing conditions or (b) people unprepared when venturing outside in cold weather. There is a big scientific debate about the balance between these two explanations (see also section 3) with the EuroWinter Group pointing outdoors.49 50 The Warm Front study provided an opportunity to go deeper into indoor housing determinants as an alternative explanation.51 In this case 52 48 Paul Wilkinson, Megan Landon, Ben Armstrong, Simon Stevenson, Sam Pattenden, Martin McKee M & Tony Fletcher (2001) Cold Comfort: The Social and Environmental Determinants of Excess Winter Deaths in England, 1986-1996. Report for the Joseph Rowntree Foundation. 49 William Keating (2001) Winter Deaths: Warm Housing is not enough. British Medical Journal: 323: 166 50 William Keatinge (2004) Winter mortality in elderly people in Britain: Action on outdoor cold stress is needed to reduce winter mortality. British Medical Journal: 323: 166 51 Paul Wilkinson, Sam Pattenden, Ben Armstrong, Astrid Fletcher, R Sari Kovats, Punam Mangtani and Antony McMichael. Vulnerability to winter mortality in elderly people in Britain: population based study. British Medical Journal: 2004; 329: 647-0. 52 Paul Wilkinson, Ben Armstrong, Zaid Chalabi (Forthcoming) Dependence of winter- and coldrelated mortality on indoor temperature.
1.4 1.3 1.2 1.1 1.0 >14.8°C 14.8°C 16.6°C 18.4°C 19.4°C 16.5°C 18.3°C 19.3°C 27.0°C Indoor temperature (quintile bands) Room for improvement To reduce winter deaths the spotlight must be on those caused by cardiovascular disease. A theoretical maximum of around 18,000 is attributable to cold conditions and only a proportion to cold housing. Theoretically, most of these might be prevented if the housing stock were somehow transformed to the standard of the very warmest houses. Up to 10,000 deaths in the UK might be prevented if indoor temperatures were raised to 21°C. Or up to a 1000 deaths might be prevented if temperatures in the coldest properties were raised by 2°C.
Main message: Overall we assess the Warm Front Scheme as having a positive impact on (a) improving mental health (b) alleviating respiratory problems in children and
(c) reducing deaths of older people. The investment is cost-effective in
extending years of life.
Benefits Overall we assess the Warm Front Scheme as having a positive impact on (a) improving mental health (b) alleviating respiratory problems in children and (c) reducing deaths of older people. Of these three benefits (figure 30), our study directly detected an improvement only in mental
Figure 30: Summary of Health Impacts
(1) indoor cold
Winter low indoor
(2) psychosocial routes temperature
Main at risk group
(per 1000 a year)
Change following heating + insulation
Dwellings with mould severity index >1 Reduced from12% to 8.2%
Depression and anxiety reduced by 48%
c3 children with
c150 people with
depression and anxiety deaths per year
Children are most vulnerable to damp conditions but were not interviewed directly. Again, modelling techniques produced estimates. For those under 14 the underlying rate of respiratory symptoms leading to contact with the health service is about 11% 55, giving a baseline rate of 110 per 1000. A fall in exposure to
well-being (see section 14).
mould from 12% to 8%
Prevalence of anxiety or depression (`common mental disorder') fell
following Warm Front will reduce the estimated prevalence of
from 300 to about 150 per 1000 occupants after Warm Front
respiratory symptoms by 3 cases per 1000 children.
measures. This is a significant impact. For every 10,000 properties
(with two adults) improved by Warm Front about 3000 occupants will be relieved of anxiety or depression.
How do costs and benefits compare? Both are modest (figure 31) if
Death caused by cold is too rare an event for our study to detect a
we simply focus on death. The combined cost of insulation and
reduction in death rates. So the Scheme's impact was estimated
upgrading the heating system averaged Ј1410.56 The resulting
indirectly using modelling techniques.53 For the vulnerable group of
increase in temperature added an extra 0.56 months to the lives of a
elderly occupants, the underlying rate of cardiovascular death is around
65 year old couple living together 0.33 months for the man and
27 per 1000 a year.54 An average increase in indoor temperature of
0.22 for the woman.57 These extra few days may seem negligible, but
2.2°C, following combined heating and insulation measures, will
grossed up over many beneficiaries, the impact is thousands of life
reduce annual winter deaths by 0.4 per 1000 occupants. For typical
years saved each year. Over ten years the average cost of extending a
pensioner couple households this means an estimated annual reduction
recipient's life by one year is Ј12,905 if only insulation is installed.
of 80 deaths per 100,000 dwellings improved.
Upgrading the heating system results in higher temperatures but is less
cost-effective (at Ј26,629) in saving lives.
53 Zaid Chalabi & Paul Wilkinson. (Forthcoming) Mortality and non-mortality impacts of the Warm Front Home Energy Efficiency Scheme. 54 Office of National Statistics (2000) Mortality Statistics. Review of the Registrar General on deaths in England and Wales, 1998. The Stationery Office. London. 55 David Ormandy, Stephen Battersby, Megan Landon, Richard Moore and Paul Wilkinson (2003). Statistical evidence to support the Housing Health and Safety Rating System. Vol 2 Summary of Results. Office of the Deputy Prime Minister. London. 56 Comptroller and Auditor General. (2003) Warm Front: helping to combat fuel poverty. Report No:HC 769. The Stationery Office. London. 57 Zaid Chalabi, Jack Dowie, Ben Armstrong, Paul Wilkinson. (Forthcoming) Analysis of the health impact of England's Home Energy Efficiency Scheme.
Figure 31: Costs and years of life saved
Months of life saved Average cost per
life year saved
Insulation only Ј280
Insulation and heating
Main message: Our study broadly confirms Warm Front `wins' across the two policy domains of health and fuel poverty, plus a possible `win' slowing climate change.
Win: win: win?
Improving the energy efficiency of English homes is a demonstration, par excellence, of `joined-up' government policy.58 Warm Front, the UK Government's flagship programme to reduce fuel poverty in England,
Our study is positive but less conclusive about a `win' for fuel poverty which compares (a) income against (b) fuel expenditure in achieving (c) warmth. Contested definitions suggest caution. We used
also has the potential to improve health and help slow climate change.
households' self-reported `Difficulty paying fuel bills' as a proxy for fuel
Commissioned to evaluate health
poverty and showed how prevalence
impacts, our study broadly confirms `wins' across all three domains whilst also revealing some uncomfortable truths and
Figure 32: Projected effects of Warm Front
Impact of Warm Front measures detected by study
Probable effects Health Fuel poverty Climate
was reduced (but not eliminated) by Warm Front. An `objective' ratio of fuel expenditure to income was difficult to pin down.
(+) Indoor temperature
More than any other, our study
An executive summary at () Difficulty paying fuel bills (++)
illuminated warmth as the third element
the beginning of this report ( ?) Fuel consumption
(c) in the fuel poverty equation.
summarises the key messages from each of 18 sections. Footnotes refer to supporting
(+) Thermal comfort () Stress
Residents' self assessment of thermal
comfort (12) increased after Warm
Front, when they felt most comfortable
evidence in 15 scientific papers
at a living room temperature of 19.2єC.
written by the research team. The impact of Warm Front across the
Still nearly 2єC below the Government's recommended threshold, this
three policy domains is summarised in Figure 32.
average `comfort' vote (c) implies lower demand for fuel (b) and
therefore less fuel poverty than in the official formulation.
Better living conditions have a significant impact on health.
Increased temperatures (section 6) are linked to better health (13,
There are mixed messages on climate change. Prima facia, less
15) and fewer winter deaths (17,18). Less mould (10) reduces
reported difficulty paying fuel bills implies less fuel consumption to
respiratory problems (18). Psycho-social pathways to health are
achieve improved temperature and thermal comfort. Better energy
even more significant. The main route to health gain is via the
ratings using Standard Assessment Procedure points in the same
alleviation of fuel poverty. Warm Front recipients were less stressed
direction. Yet our objective measures also show an increase in fuel
(13) because it was easier to pay fuel bills (11). Less stress was strongly associated with better mental (14) and physical health
consumption (11) after Warm Front measures.
(16). We conclude that Warm Front has a significant impact on the
It is a conundrum we have yet to resolve by building science. Our
health of recipient households.
qualitative interviews may provide an answer. Recipient's said `Now
we can use the whole house instead of huddling around a living room
58 Geoff Green (forthcoming) Warm Front on Climate, Poverty and Health: Evaluation of a UK Government Programme to improve domestic energy efficiency.
fire.' Fuel takes a bigger share of their household budget because they now put more value on their home as a `haven.'
May 2008 Printed copies of this report are available from: Centre for Regional Economic and Social Research Sheffield Hallam University Unit 10 Science Park · Howard Street · Sheffield S1 2LX tel 0114 225 3073 · fax 0114 225 2197 ISBN 978 1 843 87 275 7 Design: www.paulpugh.co.uk