Review on Energy Resilience

IIndoorndoor andand Builtuilt EnvironmentSpecial Issue – Fuel Poverty

Mixing methods, maximising results: Use of mixed methods research to investigate policy solutions for fuel poverty and energy vulnerability

Kimberley O’Sullivan and Philippa Howden-Chapman

Abstract Mixed methods research is particularly well suited to investigate complex policy problems such as fuel poverty, which require empirical evidence to inform intersectoral solutions. This paper draws on our theory-driven interdisciplinary studies in the built environment to analyse the outcomes and possible policy pathways to address fuel poverty. Insights are drawn from our own fuel poverty research using community-based mixed methods research to explore both the outcomes of interventions and the behaviour of people in indoor environments. Finally, we suggest future directions for mixed methods policy-focused fuel poverty research to leverage growing national and international political interest and policy recognition of the problem.

Keywords Energy poverty, Mixed methods, Research translation, Energy policy, Public health, Multidisciplinary, Retrofit

Accepted: 6 April 2017

Introduction

Solving the complex problem of fuel poverty, increas- ingly understood to include broad definitions of energy vulnerability,

12 necessitates the development and use of

intersectoral policy solutions spanning siloed govern- ment departments and ministries that traditionally hold primary responsibility for relevant policy areas, such as energy, housing, building, and health. Fuel pov- erty, like most public policies, requires a strong case to be made that each of these departments has something to gain by contributing to the development and imple- mentation of a coordinated policy, which delivers co- benefits to all,

3 regardless of the leading department.

Mixed methods research offers opportunities for drawing on varied sources of evidence to better capture the multi-dimensional experiences and outcomes, as well as antecedents, of fuel poverty within communities and potential solutions to resolve household energy vulnerability. Furthermore, mixed method research approaches are typically born of pragmatist and trans- formative epistemologies. These understandings are

ideal for inter-disciplinary, multi-disciplinary and indeed transdisciplinary research investigations, which include researchers from different disciplinary back- grounds to produce new frameworks and research methods.

4 We describe practical examples of how to

create the evidence base to underpin the translation of research to policy. We site fuel poverty as a public health problem within the social, cultural and economic structural determinants of health and draw on relevant literature to provide suggestions for ongoing engage- ment to achieve research translation. We go on to out- line the advantages and disadvantages of mixed methods research to achieve these goals. Finally, we

He Kainga Oranga/Housing and Health Research Programme, Department of Public Health, University of Otago, Wellington, New Zealand

Corresponding author: Kimberley O’Sullivan, University of Otago, Wellington, PO Box 7343, Wellington 6242, New Zealand. Email: [email protected]

Indoor and Built Environment

2017, Vol. 26(7) 1009–1017

! The Author(s) 2017

Reprints and permissions:

sagepub.co.uk/

journalsPermissions.nav

DOI: 10.1177/1420326X17707327

journals.sagepub.com/home/ibe

suggest new roads forward for mixed methods research to contribute to the policy agenda and solutions to fuel poverty, using examples from New Zealand.

Research translation to policy

Public health research is concerned with the fundamen- tal determinants of population health and well-being, particularly those aspects that can be influenced by public policy.

5 This presents a significant challenge, as

even within one discipline, many different types of research are required to influence policy and tracing the influences of research on policy is not always easy.

6

However, an overarching goal of fuel poverty research is the translation of research to policy. This goal therefore provides the rationale for seeking research approaches and methods that are broadly based and can meet policy needs. Translating research into policy, or in an alternative framing – providing an evidence base to inform policy – has been widely dis- cussed in public health research and we focus on these examples in this paper.

7–12 These examples of research

translation can provide useful lessons to the broader fuel poverty research community.

Smith and Katikireddi 7 described several theories of

policymaking developed in the political sciences and based on empirical evidence, which emphasise the com- plexity of the policymaking processes.

7 They suggested

that public health advocates may require persistence over time to influence policy, and that as multiple actors and interactions are involved in policymaking, researchers ‘may therefore need to move beyond making singular lists of policy recommendations for generic ‘policymakers’ and instead consider how to effectively target key messages to multiple different

audiences’. 7 Several authors have also described the

importance of face-to-face relationships between researchers and policymakers,

9,10,12,13 especially in the

planning and dissemination stages of research, facilitat- ing what has been termed a ‘knowledge brokering’ role.

11 These ideas are consistent with the ideas of the

modern roles of public health and public health researchers, and the need for increased dissemination of research, particularly in politically and publicly accessible formats.

14–16 To avoid having useful research

passed over during the selection process for forming evidence-based policy, or having the findings selectively framed by politicians, wide dissemination of research findings through open access publishing, media, and social media is recommended, in addition to making and maintaining links between policymakers and researchers.

13

Qualitative studies investigating the influence of research evidence on health inequalities policy provide some suggestions for increasing that influence.

8,12

Policymakers emphasise the need for research to pro- vide simple and clear messages, with less priority given to methodological details and limitations, and to be timely and relevant, aligning with policy debates.

8

Researchers identify five types of evidence as particu- larly persuasive: ‘observational evidence showing the existence of a problem; narrative accounts of the impacts of policies from the household perspective; controlled evaluations; natural policy experiments; and historical evidence’.

12 Both policymakers and

researchers agree with the concept of a ‘jigsaw of evi- dence’, by which is meant that ‘the most valuable policy relevant information was not one single piece of evi- dence, but rather many different bits, of varying qual- ity, creatively pieced together’.

12 Researchers and

policymakers also agree that using both quantitative and qualitative evidence, for example, the combination of observational studies that identify policy problems, descriptive studies that provide examples of human impact, and experimental and quasi-experimental stu- dies for guiding interventions, can provide information that is useful in shaping policy.

12

Fuel poverty and the resultant health problems asso- ciated with cold housing

17 fall into the category of what

is termed in the health field as ‘the social determinants of health’.

18 A lack of political will, vested corporate

interests with lobbying power, and continued focus on economic measures of success have prevented ending or even markedly reducing such problems. Crouch describes the political situation of the United States of America and much of Western Europe as a state of ‘post-democracy’, where politics is shifting on a par- abola away from maximal democracy.

19 He maintains

that ‘powerful minority interests have become far more active than the mass of ordinary people in making the

Poli�cal advocacy

Public health prac��oners provide advice to poli�cians only when asked

Public health prac��oners ac�vely disseminate research to media and poli�cians

Public health prac��oners ac�vely a�empt to influence the poli�cal process

Public health prac��oners engage in poli�cs and gain poli�cal posi�ons to achieve their aims

Poli�cal passivism

Poli�cal engagement

Poli�cal ac�vism

Figure 1. Ladder of political activism for public health professionals (adapted from Mackenbach,

16 p.183).

1010 Indoor and Built Environment 26(7)

political system work for them’ and ‘political elites have learned to manage and manipulate popular demands’.

19

Starfield similarly asserts that ‘Even the best evidence and most consistent public support fails to change poli- cies when interests in maintaining the status quo have the power to thwart change’.

20 These authors point to

the contributions of corporate entities that lobby polit- icians provide political funding, achieving policy agen- das that are not in the public interest. Broader measures of success, rather than simply gross domestic product or gross national product have been suggested as a means to shift the focus of governments to address inequalities affecting health.

21,22 For example, Stiglitz

et al. 22

recommend collecting data on measures of well- being and quality of life that include material living standards, health, education, environment, social activ- ities and connectedness, and economic as well as phys- ical security.

22

Various strategies for the public health community to reduce inequalities caused by social determinants such as fuel poverty have been put forward. Suggested stra- tegies include greater involvement of public health aca- demics and practitioners in the media, political, and corporate realms – areas they have traditionally largely avoided.

14–16 Reilly and McKee

14 put forward a

similar argument that modern western political systems are better viewed as ‘decipractic (Latin root word¼decepio, meaning to ‘deceive’)’, with democrat- ically elected leaders having unspoken agendas (see Reilly and McKee

14 , p. 305) They argue that the deci-

practic state requires public health to engage and advo- cate through media and social media to quickly counter propaganda, helping to reclaim ‘politics as a vehicle of change for the good’ (see Reilly and McKee

14 , p. 306).

Mackenbach 16 goes further as depicted in Figure 1, sug-

gesting a ‘ladder of political activism’ for public health professionals (p. 183).

While the examples of research translation discus- sions above are drawn from within the ‘‘silo’’ of our own discipline, Public Health, they are clearly relevant to fuel poverty research more generally, which also has translation of research to policy solutions as a desired outcome.

Parallels of the above examples are also to be found in broader discussion of public policy development. For example, Weiss

23 characterised seven models of

‘research utilization’ in policy. Of these, arguably the least likely ways in which research translation occurs are the ‘knowledge-driven’ model, where the result of basic science leads to further investigation and develop- ment of a policy and the ‘problem-solving’ model, where identification of a problem results in a search for existing social science knowledge on which research-based policy can be developed.

23 The ‘inter-

active model’, where policymakers seek knowledge

from a range of sources, including social scientists to develop policy; the ‘political model’ where social sci- ence to support the desired policy is selectively identi- fied and used; and the ‘tactical model’ where the support or commissioning of research is framed to jus- tify policy or a policy delay, are also potential routes. However, these are less likely to result in successful policy for problems such as fuel poverty. Instead, the ‘enlightenment model’ where social science findings become widely known and are incorporated into the dominant frame, so that these will be diffused into policy, is arguably more likely.

23 While the problem

of fuel poverty has been widely publicised through community organisations and research efforts and some remedial fuel poverty policy implemented – par- ticularly in the United Kingdom,

24 the United States,

25

and New Zealand 26

– this has not yet resulted in a successful and systematic policy drive to eliminate fuel poverty. Weiss’ call for researchers ‘to improve the contribution that research makes to the wisdom of social policy’ remains relevant.

23

Stevens 13 puts forward an evolutionary approach for

translating research into policy, which holds that biased selection of research supporting the ideas of policy- makers may undermine attempts to have evidence- based policy adopted.

13 To avoid having useful

research passed over during the selection process, or having the findings selectively framed by politicians, wide dissemination of research findings through open access publishing, media, and social media is recom- mended, in addition to making and maintaining links between policymakers and researchers. Furthermore, Head

27 acknowledges that not only is there more than

one ‘evidence base’ for social policy development, but the broader social and political context of policy devel- opment require consideration of three ‘lenses’ of know- ledge and evidence that are important for the translation of research to policy. The lens of ‘political knowledge’, similarly to the other ideas introduced above, acknowledges that scientific evidence is expected to support the desired policy and not necessarily shape it. The ‘scientific knowledge’ lens, prioritising system- atic methodological analysis of policy problems, pro- vides further justification for multidisciplinary studies to bring both rigour and context to the policy develop- ment table. The ‘professional practice’ lens refers to the viewpoint of those in programme implementation of policy within government and non-government organ- isations. In terms of fuel poverty research and policy development, this is particularly relevant as much of the fuel poverty research has been born out of the experi- ences of community advocates and organisations. Fuel poverty researchers, generally speaking, have relatively good access to this practical implementation know- ledge, as reflected by the literature. All of these

O’Sullivan and Howden-Chapman 1011

examples stress the need to prioritise networking rela- tionships between researchers, community organisa- tions and policymakers, while engaging with media and the broader public to disseminate findings of fuel poverty research and advance a policy agenda to reduce the problem.

As acknowledged, much of the current fuel poverty research being undertaken has underlying pragmatic, emancipatory, and translational goals, which require consideration of policy uptake at the outset. Having described models of research translation to policy in public health, and broader public policy, we highlight here how mixed methods research can offer fuel poverty researchers credible forms of evidence, as well as oppor- tunities for engagement and dissemination for enhan- cing research translation.

Mixed methods research designs

Mixed methods research arises from the knowledge that complex system and interdisciplinary research questions, including fuel poverty, require pluralistic methods to answer them.

28,29 Incorporating a range of quantitative

and qualitative techniques into mixed methods research allows exploration of causality, effect pathways and pos- sible problem areas for future.

30,31

Key considerations in mixed methods research designs are: how qualitative and quantitative mixing occurs, either within or across the research; whether quantitative and qualitative components of the study are considered equal, or one is prioritised; and the time ordering of qualitative and quantitative phases.

30,31 Six major strategies for data collection

used in mixed-methods research have been identified as: questionnaires, interviews, focus groups, tests, observation, and secondary data.

32

Basic data analysis methods in mixed methods include both qualitative and quantitative data analysis, as appropriate. For example, qualitative data analysis methods typically employed within mixed methods research include the use of thematic analysis and the use of either categorical analysis strategies, that focus on sections of data, or contextual analysis strategies of the entire narratives or texts.

32 Basic quantitative data

analyses include descriptive and inferential statistical methods. More sophisticated methods, such as multi- variate analysis and parametric testing, may also be used.

32

Further, many mixed methods research designs have been characterised and used. For example, Creswell and Plano Clark

30 give examples of 15 different mixed

methods design classifications across different disciplines. They describe six major mixed methods designs, including four that use quantitative and quali- tative studies that are differently ordered, prioritised,

and integrated: the concurrent; the explanatory; the exploratory, and the embedded mixed methods designs.

30 They also include two designs that include

multiple design elements: the transformative design, which is guided by a transformative theoretical frame- work; and the multiphase design, using multiple meth- ods over three or more phases of a study programme.

30

Advantages and disadvantages of mixed methods research

When methods are carefully combined in a mixed meth- ods study, the strengths of the different methods can contribute to overall outcomes, while weaknesses can be overcome.

31 Triangulation between the different

methods making up a mixed methods investigation pro- vides better understanding of the phenomenon studied, adding rigour and depth, and strengthening the conclu- sions that can be drawn.

31–33 Mixed methods research

also allows for a greater range of divergent viewpoints of a phenomenon to be explored, or divergent findings, which is useful because it can prompt transformation of data, investigation of the quality of the inference, or further research.

32

However, some researchers have questioned whether even when mixed methods are nominally used, post- positivist quantitative aspects are privileged over qualitative aspects.

28,31,33 Furthermore, mixed methods

research is not without practical limitations; it requires a broad understanding of methods, is time-consuming, and can be expensive. Mixed methods can also require the contributing expertise of a research team rather than being carried out by an individual researcher, par- ticularly when the research design uses concurrent study phases.

31 Buswell et al.

34 provide an in-depth

examination of the cost and time involved with running a thorough multidisciplinary, mixed methods investiga- tion of energy use within 20 family homes. In this exam- ple, over the four-year study, the researchers reported that 429 researcher hours and 217 participant hours per house were required, at a cost of £22,000–27,000 per house using established detailed socio-technical investi- gative methods and up to £46,000 per house when rela- tively untested detailed socio-technical methods are used.

34 Despite these challenges, we believe that multi-

disciplinary, mixed methods research is key to solving the problem of fuel poverty, and we outline our own case report of some of the techniques we have used below.

Case report: He Kainga Oranga’s mixed methods fuel poverty research

During the 1990s, Brenda Boardman’s seminal work stimulated an awareness that improving the structure

1012 Indoor and Built Environment 26(7)

of housing was only part of the solution for making houses warmer.

35 International comparisons of admin-

istrative data on fuel poverty highlighted that more countries than had been previously realised had a prob- lem with fuel poverty.

36 Our research has subsequently

shown that not only did New Zealand have a high rate of fuel poverty, estimated at 25% of households, but that there were inequalities in those who suffered from fuel poverty.

26

He Kainga Oranga/Housing Health Research Programme was publically funded almost two decades ago, explicitly to reduce inequalities in the social, eco- nomic and environmental determinants of housing and health.

37,38 This intervention area was chosen because

New Zealand housing was recognised to be of relatively poor quality, being old, large, in relatively poor quality and cold, but there was little solution-focused research and scant awareness of fuel poverty. A multidisciplin- ary research team consisting of social scientists, econo- mists, epidemiologists, building scientists, physicists, architects, and mycologists was assembled to frame problem areas with end-users in central and local gov- ernment as well as community and iwi (M�aori indigen- ous tribes) groups. Overall, the Programme’s investigation into energy issues and fuel poverty is structured as a multiphase mixed methods design with each subsequent phase informed by and building on previous phases of research. Integrative studies weaving results of individual quantitative and quantitative stu- dies within the mixed methods programme as well as incorporating analyses of secondary data sources are undertaken frequently.

The first study undertaken as part of the Programme in 2001–2002 was a large-scale, randomised community trial involving almost 5000 people. The Housing Insulation and Health Programme was set up to evalu- ate the effectiveness of a retrofitted insulation pro- gramme for 1400 households, where at least some family members had chronic respiratory problems.

39,40

The study included both independent and self-reported measures and outcomes of energy use and health, along with self-reported measures of feelings of physical and mental health and health care utilisation. The positive effects in reducing energy consumption and improving health were statistically significant. However, while the independent measures of temperature and relative humidity showed that the indoor temperature was raised, the temperatures were still on average below 18�C, the temperature recommended by the World Health Organization.

41

This study design was able to establish causation between retrofitted insulation and indoor temperature, but was not able to establish why households did not maintain their indoor environments at a healthy level. To explore whether these temperatures were due to the

cultural practice of heating only one room, 42

we under- took a second randomised community trial of 400 chil- dren, with doctor-diagnosed asthma. We ensured all their houses were insulated and then following the stag- gered intervention design we had used previously, par- ticipating households in the control group, after the end of the study also received the intervention – their choice of effective, non-polluting heaters.

43 The results demon-

strated that retrofitting insulation and installing these heaters increased the temperature of the house to almost the recommended level, reduced asthma symp- toms and children having days off school.

44

Aware that further public investments to reduce fuel poverty would require cost-benefit evaluations if we were to gain the support of cabinet ministers holding relevant portfolios (energy, housing and health), we undertook economic evaluations of our community trials, which consistently showed that subsidised inter- ventions, such as retrofitting insulation and installing heaters were cost-effective.

45,46

However, associated qualitative research carried out by the first author showed that the ability to afford electricity, which is the main form of energy used for heating in New Zealand meant that people were often colder than they would have liked and in an older population, low income was often given as the reason.

47 Participants in this study spoke about the

high cost of electricity in the context that they would like to be warmer, ‘But unfortunately power s so expen- sive that, pfhh! It will just take all your money away hahahah!. . . And of course I m only on just enough to barely live as it is [Sigh]’

47 On the other hand, some

participants felt that the only option was to accrue debt on their electricity account to keep warm enough to protect their fragile health, ‘I worry about the cost alright, but I don’t care I just don’t want to be cold.’

47

The accumulation of debt had, in one case, resulted in one household having a prepayment meter installed, which had serious consequences for the health of the occupant suffering from Chronic Obstructive Pulmonary Disease, who had previously been admitted to hospital, when they were unable to purchase more electricity to credit their meter and she could not use her nebuliser to help her breathe at home.

48

We are currently carrying out two further commu- nity trials, which involve giving electricity vouchers to households, a discretionary benefit currently available to those receiving government Social Welfare pay- ments. These mixed methods studies both include a range of quantitative and qualitative data collection, and the analyses will blend these results to form an integrated analysis of whether these interventions are beneficial in terms of maintaining health and providing economic savings. In the first of these community trials, the Warm Homes for Elderly New Zealanders, people

O’Sullivan and Howden-Chapman 1013

aged over 55 with Chronic Obstructive Pulmonary Disease were given NZ$500 in their electricity account (an amount calculated to be sufficient to heat their houses during the winter when they were in the intervention group) and instructed that ‘heat was their medicine’.

49 For this study, a range of data have

been collected during home visits, including physical health measurements; self-reported questionnaires from patients and, where applicable, healthy adults living in the home; and summer and winter activity, symptoms, and medications diaries. Temperature in the homes has been objectively monitored. Hospitalisation data were collected from the central registry and electricity usage data collected from the electricity companies of each household. The results of this study will be available soon.

We are conducting the NEST study, a community trial in houses with 400 new-born babies, where we are retrofitting insulation into the houses and providing heaters with thermostats in the babies’ bedrooms, as well as NZ$200 to pay for the estimated cost of electri- city for using them during the babies first year of life. Our hypothesis is that this will reduce the current high rate of hospitalisation in infants during the first year of life.

As well as these community trials, which are well publicised in the media, other qualitative research has highlighted that while there was a cultural component of stoicism to coping with cold houses,

50 if it was not a

matter of cost or ability to effectively heat their homes, many New Zealanders would prefer to be warmer at home. Thus, fuel poverty seems to be a consequence of the quality of housing, lack of effective heaters, lack of disposable income to afford energy and policies to increase energy security.

We analysed the official Household Economic Survey, which showed an association between the de- regulation of the wholesale electricity market, which meant that residential electricity prices were no longer subsidised by commercial electricity prices, and increas- ing relative inequality in the percentage of household expenditure that the lowest income quintile spent on energy compared to the higher income quintiles.

51

Moreover, as home-ownership rates dropped, by 2013, half the population of New Zealand were ren- ters.

42 Landlords are not required to provide any

source of heating, but can require tenants to have pre-payment meters, which means that they pay higher tariffs. Consequently, significant numbers of ten- ants manage their budgets by ‘self-disconnecting’.

52

These secondary analyses of available datasets have contributed to the overall ‘jigsaw of evidence’

12 the

mixed methods research programme is building to strengthen our policy recommendations to reduce fuel poverty.

Researchers in the Programme have taken every opportunity to remonstrate that on the basis of hous- ing-preventable hospital admissions and other social costs, money is better spent on making houses warmer than on hospital expenditure.

53 In particular,

researchers of He Kainga Oranga have initiated and responded to media requests of the public health impli- cations of two widely publicised cases of fuel poverty- related deaths in New Zealand,

54,55 and highlighted the

causes and problems of fuel poverty. The first case was the death of an adult with respiratory problems in her home following disconnection of electricity for non- payment,

56 and the second when a coroners’ report

on the death of a toddler was released with the finding that cold housing could not be discounted as a con- tributory factor in her death.

57

As well as national policy our approach has informed international policies. In their report Spreading the Net, the International Energy Agency policy recommendations are that stabilising energy demand and highlighting links between heating and health, as we showed in our studies, enhance the finan- cial position of companies and improve their standing with their customers.

58 The World Health Organization

also drew on our studies to recommend that the link between carbon mitigation, energy and health could and should be a policy driver of the Green Economy.

59 We have also been careful to measure

co-benefits when we were undertaking our evalu- ations.

3 For example in our Heating study,

43 we

showed that installing effective heaters reduced fuel poverty and the number of days the children were absent from school due to ill-health.

60 In part due to

our concerted campaign on several fronts, successive central governments have supported a large national programme the Warm-Up NZ; Heat Smart Programme, which received continued supported after a change of government. We were part of a consortium that evaluated the first 45,000 houses that were insu- lated and again found significant reductions in electri- city use,

61 hospitalisation for children,

62 but also with

the larger sample, a drop in mortality for those over 65 who had been hospitalised the previous year. The bene- fit/cost ratio of the intervention overall was 4:1 and for children was 6:1.

61

Nonetheless, while this state-subsidised programme is a critical keystone policy to reducing fuel poverty nationally, the current government announced in the 2016 budget that they were winding the programme down, although only a third of existing houses have had retrofitted insulation installed. This is a clear indication that while policy to address fuel poverty can be evidence based, the rational model of decision- making is clearly bounded by political priorities and influences.

1014 Indoor and Built Environment 26(7)

To increase the breadth of our research to inform policies to address fuel poverty, we are continuing to extend our use of mixed methods research to include more innovative approaches within individual studies that make up the He Kainga Oranga/Housing and Health Research Programme. For example, our recent work exploring the experiences of fuel poverty and cold indoor environments among young people has used mixed methods participatory action research to find that fuel poverty is a widespread youth issue in New Zealand.

63 This study used classroom-based workshops

to allow a group of young co-researchers to act as investigators throughout the research process, from survey questionnaire development to research dissem- ination.

62 We have recently completed a pilot study,

using participatory action research, to explore whether classrooms provide thermal comfort. Our young collab- orators undertook the primary data collection, install- ing dataloggers to measure temperature and relative humidity, and running youth-led (and designed) quali- tative classroom workshops with younger children to explore their experiences of thermal comfort within the classroom.

Discussion

In this paper, we have described the utility of mixed methods approaches to facilitate translation of research into policy to improve fuel poverty. Mixed methods research offers tools for gathering evidence available within the broader social and political context, which can be useful in advocating for evidence-based policy to improve population health. We have provided a case report of the research of He Kainga Oranga/Housing and Health Research Programme as an example of an interdisciplinary programme of research, which consist- ently uses mixed methods research over time to achieve policy-shifts towards addressing fuel poverty in New Zealand. We have used the concept of ‘framing’, where advocates deliberately position issues such as fuel poverty as a population health problem, requir- ing broad policy responses.

64 Framing of building

regulation as a population health problem, rather than individualistic ideas of fuel poverty such as a nationwide cultural preference for cold housing, or a ‘rite of passage’ for students to live in poor quality buildings, has helped to shift the media and policy focus to the ill-health outcomes of cold housing and fuel poverty.

The authors and the research team have conducted a number of studies to provide a ‘jigsaw of evidence’

12 to

improve residential building quality, reduce fuel pov- erty and improve public health. Alongside this research, there has been a deliberate approach of stra- tegically engaging in media advocacy,

65 with the

purpose of framing fuel poverty as a problem of inad- equate building policy and regulation, which causes negative externalities for population health. We have also engaged in ‘knowledge brokering’ activities; mem- bers of the team regularly attend meetings with and respond to information requests from local and gov- ernment policymakers.

Largely as a consequence of the earlier studies we have described, successive governments have supported funding insulation retrofit programmes, although most recently the programme has effectively faced a budget cut.

66 This political decision is related to government

reluctance to usefully increase insulation standards for rental housing, when changes were recently made to the Residential Tenancies Act.

42,62 While the retrofit pro-

grammes have been targeted to low-income and rental properties, we have not been successful to date in our efforts to advocate for targeting to households with children up to 18 years of age, and we are continuing to build a policy case for this with our current research. More positively, although we have been unsuccessful in getting the reduction of fuel poverty to be an official government policy, in 2016, Statistics New Zealand for the first time is undertaking an official report of the trends in fuel poverty, a year before a national election, which is likely to raise national attention on this important public health issue. These outcomes suggest that our mixed methods research investigating the residential built environment in New Zealand is to some extent informing policy development and implementation.

Conclusions

The complexity of fuel poverty spanning building, health, economic, and environmental sectors continues to require research to support the broad policy neces- sary to eradicate this wide-reaching problem. Mixed methods research techniques allow exploration and explanation of complex phenomena such as fuel pov- erty, while also providing opportunities to suggest pathways for remedial policies. Crucially for policy translation, mixed methods research also offers mul- tiple opportunities for researchers to engage with pol- icymakers and disseminate results throughout a research programme, while also allowing for framing, media, and public engagement to advance fuel poverty on the policy agenda. The continuing integration of results from mixed methods research undertaken by multidisciplinary teams, such as He Kainga Oranga/ Housing and Health Research Programme, suggests that evaluating policy aimed at improving the built environment, despite a changing energy landscape, will continue to be a critical activity to ensure that fuel poverty diminishes.

O’Sullivan and Howden-Chapman 1015

Authors’ contribution

All authors contributed equally in the preparation of this

manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) declared the following potential conflicts of

interest with respect to the research, authorship, and/or pub- lication of this article: The first author was supported by a Lotteries Health Postdoctoral Fellowship while preparing this article. He Kainga Oranga/Housing and Health Research

Programme is funded by the New Zealand Health Research Council.

References

1. Bouzarovski S and Petrova S. A global perspective on domestic

energy deprivation: overcoming the energy poverty–fuel poverty

binary. Energ Res Soc Sci 2015; 10: 31–40.

2. Middlemiss L and Gillard R. Fuel poverty from the bottom-up:

characterising household energy vulnerability through the lived

experience of the fuel poor. Energ Res Soc Sci 2015; 6: 146–154.

3. Howden-Chapman P and Chapman R. Health co-benefits from

housing-related policies. Curr Opin Environ Sustain 2012; 4: 414–

419.

4. Lawrence RJ. Housing and health: from interdisciplinary prin-

ciples to transdisciplinary research and practice. Futures 2004; 36:

487–502.

5. Marmot Review Team. Fair society, healthy lives (The Marmot

Review). London: Author, 2010.

6. Marris P. Witnesses, engineers, and storytellers: using research for

social policy and action. College Park, MD: University of

Maryland, 1997.

7. Smith KE and Katikireddi SV. A glossary of theories for under-

standing policymaking. J Epidemiol Commun Health 2013; 67:

198–202.

8. Petticrew M, Whitehead M, Macintyre SJ, Graham H and Egan

M. Evidence for public health policy on inequalities: 1: the reality

according to policymakers. J Epidemiol CommunHealth 2004; 58:

811–816.

9. Davis P and Howden-Chapman P. Translating research findings

into health policy. Soc SciMed 1996; 43: 865–872.

10. Jansen MWJ, van Oers HAM, Kok G and de Vries NK. Public

health: disconnections between policy, practice and research.

Health ResPolicy Syst 2010; 8: 37–50.

11. Lomas J. The in-between world of knowledge brokering. BMJ

2007; 334: 129–132.

12. Whitehead M, Petticrew M, Graham H, Macintyre SJ, Bambra C

and Egan M. Evidence for public health policy on inequalities: 2:

assembling the evidence jigsaw. J Epidemiol Commun Health

2004; 58: 817–821.

13. Stevens A. Survival of the ideas that fit: an evolutionary analogy

for the use of evidence in policy. Soc Policy Soc 2007; 6: 25–35.

14. Reilly RG and McKee M. ‘Decipio’: examining virchow in the

context of modern ‘democracy’. Pub Health 2012; 126: 303–307.

15. Dew K. The cult and science of public health: a sociological inves-

tigation. New York: Berghahn Books, 2012.

16. Mackenbach JP. Politics is nothing but medicine at a larger scale:

reflections on public health’s biggest idea. J Epidemiol

CommunHealth 2009; 63: 181–184.

17. Marmot Review Team. The health impacts of cold homes and fuel

poverty. London: Friends of the Earth and the Marmot Review

Team, 2011.

18. Wilkinson R and Marmot M. Social determinants of health: the

solid facts. 2nd ed: World Health Organisation, 2003.

19. Crouch C. Post-democracy. Cambridge: Polity Press, 2004, p.19.

20. Starfield B. Politics, primary healthcare and health: was Virchow

right? J Epidemiol Commun Health 2011; 65: 653–655.

21. Marmot M. Policy making with health equity at its heart. J Am

Med Assoc 2012; 307: 2033–2034.

22. Stiglitz JE, Sen A and Fitoussi J-P. Report by the commission on

the measurement of economic performance and social progress,

2009.

23. Weiss CH. The many meanings of research utilization. Pub

Administrat Rev 1979; 39: 426–431.

24. Sovacool BK. Fuel poverty, affordability, and energy justice in

England: policy insights from the warm front program. Energy

2015; 93(Part 1): 361–371.

25. Kwon HJ and Jang M. Housing quality and health of the U.S.

Senior Residents experiencing fuel poverty. Indoor Built Environ

(in press).

26. Howden-Chapman P, Viggers H, Chapman R, O’Sullivan K,

Telfar Barnard L and Lloyd B. Tackling cold housing and fuel

poverty in New Zealand: a review of policies, research, and

health impacts. Energ Policy 2012; 49: 134–142.

27. Head BW. Three lenses of evidence-based policy. Aust J Pub

Administrat 2008; 67: 1–11.

28. Creswell JW. Controversies in mixed methods research.

In: Denzin NK and Lincoln YS (eds) Handbook of qualitative

research. 4th ed. Thousand Oaks, CA: Sage, 2011, pp.269–284.

29. Tashakkori A and Teddlie C. Handbook of mixed methods in

social science and behavioural research. Thousand Oaks, CA:

Sage, 2003.

30. Creswell JW and Plano Clark VL. Designing and conducting

mixed methods research. 2nd ed. Thousand Oaks, CA: Sage, 2011.

31. Johnson RB and Onwuegbuzie AJ. Mixed methods research: a

research paradigm whose time has come. Edu Res 2004; 33:

14–26.

32. Tashakkori A and Teddlie C. Foundations of mixed methods

research: integrating quantitative and qualitative approaches in

the social and behavioral sciences. Thousand Oaks, CA: Sage,

2009.

33. Denzin NK. Triangulation 2.0. J Mixed Meth Res 2012; 6: 80–88.

34. Buswell R, Webb L, Mitchell V and Leder Mackley K.

Multidisciplinary research: should effort be the measure of suc-

cess? Build Res Inform 2016; 1–17. DOI: http://dx.doi.org/

10.1080/09613218.2016.1194601 (accessed 18 April 2017).

35. Boardman B. Fuel poverty: from cold homes to affordable warmth.

London: Belhaven Press, 1991.

36. Healy JD. Housing, fuel poverty and health: a pan-european ana-

lysis. London: Ashgate Publishing Limited, 2004.

37. Howden-Chapman P and Carroll P. Housing and health:

research, policy, and innovation. Wellington: Roger Steele, 2004.

38. He Kainga Oranga/Housing and Health Research Programme.

www.healthyhousing.org.nz (2017, accessed 31 March 2017).

39. Howden-Chapman P, Crane J, Matheson A, Viggers H,

Cunningham M, Blakely T, O’Dea D, Cunningham C,

Woodward A, Saville-Smith K, Baker M and Waipara N.

Retrofitting houses with insulation to reduce health inequalities:

aims and methods of a clustered, randomised, community-based

trial. Soc Sci Med 2005; 61: 2600–2610.

40. Howden-Chapman P, Matheson A, Crane J, Viggers H,

Cunningham M, Blakely T, Cunningham C, Woodward A,

1016 Indoor and Built Environment 26(7)

Saville-Smith K, O’Dea D, Kennedy M, Baker M, Waipara N,

Chapman R and Davie G. Effect of insulating existing houses on

health inequality: cluster randomised study in the community. Br

Med J 2007; 334: 460–464.

41. World Health Organization. Health impact of low indoor tempera-

tures: report on a WHO meeting. Copenhagen: WHO, 1987.

42. Howden-Chapman P. Home truths: confronting new zealand’s

housing crisis. Wellington, New Zealand: Bridget Williams

Books Limited, 2015.

43. Howden-Chapman P, Pierce N, Nichols S, Gillespie-Bennett J,

Viggers H, Cunningham M, Phipps R, Boulic M, Fjallstrom P,

Free S, Chapman R, Lloyd B, Wickens K, Shields D, Baker M,

Cunningham C, Woodward A, Bullen C and Crane J. Effects of

improved home heating on asthma in community dwelling chil-

dren: randomised controlled trial. Br Med J 2008; 337:

1411–1420.

44. Free S, Howden-Chapman P, Pierse N, Viggers H and Housing

Heating and Health Study Team. Does more effective home heat-

ing reduce school absences for children with asthma? J Epidemiol

Commun Health 2010; 64: 379–386.

45. Preval N, Chapman R, Pierse N and Howden-Chapman P.

Evaluating energy, health and carbon co-benefits from improved

domestic space heating: a randomised community trial. Energ Pol

2010; 38: 3965–3972.

46. Chapman R, Howden-Chapman P, Viggers H, O’Dea D and

Kennedy M. Retrofitting housing with insulation: a cost-benefit

analysis of a randomised community trial. J Epidemiol Commun

Health 2009; 63: 271–277.

47. O’Sullivan KC. ‘‘Gee my account is in credit!’’ Qualitative com-

ponent of the warm homes pilot study. Unpublished Masters

Thesis, http://ourarchive.otago.ac.nz/handle/10523/367 (accessed

13 april 2017).

48. O’Sullivan KC, Howden-Chapman P and Fougere G. Making

the connection: the relationship between fuel poverty, electricity

disconnection and prepayment metering. Energ Policy 2011; 39:

733–741.

49. Viggers H, Howden-Chapman P, Ingham T, Chapman R, Pene

G, Davies C, Currie A, Pierse N, Wilson H, Zhang J, Baker M

and Crane J. Warm homes for older people: aims and methods of

a randomised community-based trial for people with COPD.

BMC Pub Health 2013; 13: 176. DOI: 10.1186/1471-2458-13-176.

50. Cupples J, Guyatt V and Pearce J. ‘‘Put on a jacket, you wuss’’:

cultural identities, home heating and air pollution in christch-

urch. Environ Plann A 2007; 39: 2883–2898.

51. Howden-Chapman P, Viggers H, Chapman R, O’Dea D, Free S

and O’Sullivan K. Warm homes: drivers of the demand for heat-

ing in the residential sector in New Zealand. Energ Policy 2009;

37: 3387–3399.

52. O’Sullivan KC, Howden-Chapman P and Fougere G. Fuel pov-

erty, policy, and equity in New Zealand: the promise of prepay-

ment metering. Energ Res Soc Sci 2015; 7: 99–107.

53. Howden-Chapman P. We should cap the health budget and

spend more money on housing and food – the ‘yes’ case.

J Primary Health Care 2012; 4: 337–339.

54. McFie R. Cold homes. Listener 2015; http://www.noted.co.nz/

archive/listener-nz-2015/cold-comfort/ (accessed 18 April 2017).

55. Akel T. The high price of keeping warm. The Nation. TV3, 2015..

56. O’Sullivan KC, Howden-Chapman P and Fougere G.

Death by disconnection: the missing public health voice in

newspaper coverage of a fuel poverty-related death. Kotuitui

2012; 7: 51–60.

57. Shortland HB. Findings of Coroner H. B. Shortland: an inquiry

into the death of Emma-Lita Pepe Quintanella Bourne.

Whangarei 2015; 1–11. Available at https://static.stuff.co.nz/

files/Coronersreport.pdf (accessed 18 April 2017).

58. Ryan L and Campbell N. Spreading the net: the multiple benefits

of energy efficiency improvement. Paris: International Energy

Agency, 2012.

59. World Health Organization. Health in the Green Economy: health

co-benefits of climate change mitigation – housing sector. Geneva:

World Health Organization, 2011.

60. Free S, Howden-Chapman P, Pierce N, Viggers H and the

Housing Heating and Health Study Team. More effective home

heating reduces school absences for children with asthma.

J EpidemiolCommun Health 2010; 64: 379–369.

61. Grimes A, Preval N, Young C, Arnold R, Denne T, Howden-

Chapman P and Telfar Barnard L. Does retrofitted insulation

reduce household energy use? Theory and practice. Energ J

2016; 37(4). DOI: http://dx.doi.org/10.5547/01956574.37.4.agri.

62. O’Sullivan KC, Telfar Barnard L, Viggers H and Howden-

Chapman P. Child and youth fuel poverty: assessing the known

and unknown. People Place Policy 2016; 10: 77–87.

63. O’Sullivan K, Howden-Chapman P, Sim D, Stanley J, Rowan R,

Harris-Clark I, Morris L and the Waiopehu Colleg 2015

Research Team. Cool? Young people investigate living in cold

housing and fuel poverty. A mixed methods action research

study. Soc SciMed Popul Health 2017; 3: 66–74.

64. Katikireddi SV, Bond L and Hilton S. Changing policy framing

as a deliberate strategy for public health advocacy: a qualitative

policy case study of minimum unit pricing of alcohol. Millbank

Quarter 2014; 92: 250–283.

65. Chapman S and Lupton D. The fight for public health: principles

and practice of media advocacy. London: BMJ Publishing Group,

1994.

66. Thomas F. ‘As close to a chop as you can get’: Philippa Howden-

Chapman on warm home scheme. New Zealand Doctor.

Auckland, 2016.

O’Sullivan and Howden-Chapman 1017