A public health approach to homelessness

Full text

143
9
A public health approach
to homelessness
Louise Marshall and Jo Bibby
In common with many population health challenges,
homelessness is a complex social problem that arises from a
system of multiple interrelated causes and consequences. Both
public health and homelessness require a preventive approach
that considers the complex systems of determinants that lead to
each issue. Both need a strong evidence base to inform policy
and practice.
A complex systems model of public health
conceptualises poor health and health inequalities as
outcomes of a multitude of interdependent elements
within a connected whole. These elements aect
each other in sometimes subtle ways, with changes
potentially reverberating throughout the system. A
complex systems approach uses a broad spectrum
of methods to design, implement, and evaluate
interventions for changing these systems to improve
public health (Rutter etal, 2017).
There is a long tradition of exploring the causes and
consequences of homelessness, and there is consensus in the
TEIXERA_Using evidence to end homelessness_text_3.6.indd 163TEIXERA_Using evidence to end homelessness_text_3.6.indd 163 02/04/2020 16:24:0902/04/2020 16:24:09

Using Evidence to End Homelessness
144
sector that population-level prevention is crucial. However,
a gap exists in translating this into action, and political and
practical action is too often focussed on rough sleeping and
individual-level interventions (MHCLG 2018a, 2018b).
Despite the focus on individual-level action, there is a lack
of good evidence and investment in the research required to
understand eective strategies to support people at risk of, or
experiencing, homelessness (Centre for Homelessness Impact,
2018).1 The causal evidence that does exist is largely from
the US and focuses on interventions around healthcare or
supported housing for people already experiencing or at
imminent risk of homelessness.
Given the impact of homelessness on people’s long-term life
chances and health outcomes, there is an urgent need for more
investment in research to support evidence-informed policy-
making and practice. In this chapter, we share our thoughts on
evidence and evidence-informed policy-making and practice
in the context of complexity from the field of public health,
considering how these relate to homelessness.
Perspectives from public health: evidence-informed policy
and practice in complex systems
In public health, there is a long history of evidence-informed
practice and policy-making. Starting with the control and
eradication of infectious diseases, evidence has been at the
centre of decision-making. More recently, the rise of non-
communicable diseases has required methods of building
evidence to evolve, to take better account of a complex system
of causes and consequences of health problems and inequalities
in health (see Figure9.1) (Rutter etal, 2017).
Major contemporary public health challenges (including
obesity, diabetes, depression, anxiety and many cancers) occur
in the population because of interrelated social, environmental,
economic and commercial determinants, known in public
health as the ‘wider determinants’ of health. They do not
have a single risk or causal factor, and so cannot generally
be tackled by simply changing or eliminating one aspect of
behaviour or environment.
TEIXERA_Using evidence to end homelessness_text_3.6.indd 164TEIXERA_Using evidence to end homelessness_text_3.6.indd 164 02/04/2020 16:24:0902/04/2020 16:24:09

A public health approach
145
Despite this, research into these health issues has tended to focus
on treating them once they have already occurred or preventing
them by changing the most proximal risk factors in isolation.
This generally includes individuals’ actions, such as smoking or
dietary intake. There has been far less research attention paid to
understanding the factors that shape these actions – the wider
determinants of health and the complex relationships between
them – and the strategies needed to eect change.
This mismatch between the focus of evidence and the action
needed to improve health and reduce health inequalities at the
population level is in part due to the predominant evidence
paradigm for public health research. Public health evidence
that is sought to inform policy and practice is generated largely
through research and translation methodologies that were
developed in the field of biomedical science to understand the
eectiveness of clinical interventions. These include: randomised
controlled trials (RCTs), systematic reviews and evidence-based
guidelines. They are based on cause and eect and are highly
eective in investigating the eect of a single intervention in a
controlled context, where nothing else changes.
Figure 9.1: Characteristics of complex systems
Definition Example
Emergence
Properties of a complex system
that cannot be directly predicted
from the elements within it, and
are more than the sum of their
parts
The changing distribution of obesity
across the population can be
conceptualised as an emergent property
of the food, employment, transport,
economic, and other systems that shape
the energy intake and expenditure of
individuals.
Feedback
Situations in which change
reinforces or balances further
change
If a smoking ban in public places
reduces the visibility and convenience
of smoking, and this makes it less
appealing, fewer young people might
then start smoking, further reducing its
visibility, and so on in a reinforcing loop.
Adaptation
Adjustments in behaviour in
response to interventions or
other changes
A tobacco company may lower the price
of cigarettes in response to a public
smoking ban.
Source: Adapted from Rutter et al (2017)
TEIXERA_Using evidence to end homelessness_text_3.6.indd 165TEIXERA_Using evidence to end homelessness_text_3.6.indd 165 02/04/2020 16:24:0902/04/2020 16:24:09

Using Evidence to End Homelessness
146
In contrast, the identification, implementation and evaluation
of eective responses to population health challenges require a
dierent approach to evidence, focused on the complex systems
from which they emerge (Rutter, 2017). Ensuring research
is translated into eective practice and policy also requires a
focus on building understanding of complex systems among
researchers, practitioners and policy-makers, and the need for a
complementary set of methods.
Case study: obesity
There is robust evidence to support the eectiveness of bariatric
surgery in treating obesity and reversing some of the health
consequences in obese individuals (NICE, 2015). However, no
one would argue that this should be our sole approach to tackling
the problem or a means of achieving a population of healthy
weight. We also know that, under controlled experimental
conditions, changes to individuals’ dietary intakes and activity
levels can lead to weight loss. In contrast, much less is known
about the eectiveness of actions to reshape the complex
system of environmental and societal factors that drive obesity
in the population (Government Oce for Science, 2007),
including those determining the food people eat or how active
they are able to be.
More than a decade ago, the UK government’s Foresight
project looked at how a sustainable response to obesity could
be implemented in the UK (Government Oce for Science,
2007). The project drew on scientific evidence from a wide
range of disciplines and involved a large number of diverse
stakeholders, to identify the broad range of factors that drive
obesity. The project aimed to create a shared understanding
of the relationships between factors, their relative importance
and to design eective interventions to address rising rates
of obesity. The resultant system map has been instrumental
in depicting the complexity of obesity and the need for a
whole system approach to tackle it. It illustrates how what
we eat and our individual levels of activity are not just based
on individual choice, but are a result of the food system and
activity environment.
TEIXERA_Using evidence to end homelessness_text_3.6.indd 166TEIXERA_Using evidence to end homelessness_text_3.6.indd 166 02/04/2020 16:24:0902/04/2020 16:24:09

A public health approach
147
Until recently, research, policy and practice have largely
focussed on single initiatives, delivered at an individual level
(Jebb, 2017). However, there are signs that understanding systems
is starting to drive a wider set of system-level actions, such as
the reformulation of food products, advertising restrictions and
active transport systems.
Recently, there has been success in reducing smoking rates
in the UK. In 2011, 19.8per cent of adults in England were
smokers; by 2017, this figure was down to 14.9per cent (Selbie,
2018). This came about through a series of population-level
interventions, including increasing levels of taxation, the
plain packaging of tobacco products, larger and more graphic
warnings on tobacco packaging and bans on smoking in cars and
public places. The synergy between these separate interventions
has made it increasingly expensive and inconvenient to smoke
and has changed the social norms around smoking. As fewer
people smoke, smoking is less visible and it becomes less
attractive to take up. In addition, with the introduction of
each measure, and as social norms progressively changed, the
introduction of further policy measures that might once have
been unthinkable became increasingly acceptable to the public
and politicians.
This experience shows that it is possible to reshape the
interacting factors within a system to achieve a desirable
outcome. The policy interventions were not planned or
implemented simultaneously, but gradually as the system adapted
and was reshaped. Valuable lessons can be learned from this and
applied to other complex challenges.
A remaining challenge is to address the rising inequality
in smoking rates. While rates in the population as a whole
are at an all-time low, there are striking dierences in rates
between socioeconomic groups. For example: people with
no qualifications are four times as likely to smoke as those
with a degree; one in four people in routine and manual
occupations smoke, compared with one in ten in managerial
and professional roles; and a three-fold dierence exists
between geographical areas with the highest and lowest rates. If
not addressed, this will contribute to rising health inequalities
in the future. A systems perspective is needed to understand
TEIXERA_Using evidence to end homelessness_text_3.6.indd 167TEIXERA_Using evidence to end homelessness_text_3.6.indd 167 02/04/2020 16:24:0902/04/2020 16:24:09

Using Evidence to End Homelessness
148
why approaches have or have not worked in dierent areas
and eective strategies to reduce rates in these segments of
the population.
Health and homelessness: products of complex systems
As with many other population health challenges, homelessness
is a product of a complex system of multiple and interrelated
factors. There is no single causal factor and, as such, simple,
linear models of cause and eect are insucient to create
solutions for prevention. This is widely understood in the
homelessness sector, and the Centre for Homelessness Impact
have carried out system mapping. As in public health, however,
a significant gap exists in the use of this evidence to inform
whole systems approaches to policy and practice.
Poverty, especially during childhood, is the biggest single
predictor of homelessness, accounting for 25–50per cent of
the chance of experiencing homelessness as an adult (Boswell
etal, 2018). This is compounded by issues in the housing
market. There has been a reduction in social housing stock,
accompanied by an increase in private rental housing – much of
which excludes people on housing benefits. Housing benefits are
also inadequate to cover the cost of rent in most of the country.
These elements, together with many other personal, social,
economic and cultural risk and protective factors, determine
a person’s risk of being or becoming homeless. Many of these
risk factors are closely related, occurring together and aecting
each other (Bramley etal, 2015).
The impact of clustering and accumulation of risk factors
over time and the resulting risk of someone experiencing
homelessness is starkly illustrated later in this chapter, comparing
two individuals with very dierent life stories (adapted from
Boswell etal, 2018). This example highlights that homelessness
does not ‘just happen’ to anyone: it is determined by events that
occur throughout our lives (see Figure7.1).
This evidence about the determinants of homelessness must
be applied in its prevention, both addressing the risk factors
and targeting support at a far earlier stage to individuals at high
risk. Building this understanding into policy and practice can
TEIXERA_Using evidence to end homelessness_text_3.6.indd 168TEIXERA_Using evidence to end homelessness_text_3.6.indd 168 02/04/2020 16:24:0902/04/2020 16:24:09

A public health approach
149
also help promote a longer-term, population-level approach.
For example, child poverty has risen at an alarming rate in
many parts of the UK in recent years; what does this mean
for future prevalence of homelessness and how can risks be
mitigated now?
Understanding the risk factors that put and keep people in
poverty and increase their chances of becoming homeless can
help identify multiple points of action and intervention to
reshape the system. This understanding can also help identify
appropriate measures to monitor progress and help predict
whether positive outcomes are likely to be achieved in the
longer term.
Evidence-informed policy-making and practice for
homelessness
Better evidence-informed policy-making and practice is
urgently needed to end homelessness sustainably. A culture
of experimentation must be adopted, with appropriately
designed evaluation consistently embedded to increase
understanding about what is eective, or is not, and why.
Critically, mechanisms must be developed for feeding back
learning into policy and practice.
There is a lack of good evidence from well-designed
studies examining what is eective in preventing and tackling
homelessness (Culhane, Fitzpatrick and Treglia, this volume).
There is not a tradition of rigorous evaluation in the homeless
sector. Most eectiveness studies are from the US; only 12 have
been conducted in the UK, and all focus on healthcare action
to support people who are already homeless. This evidence
is mostly about relatively straightforward, single interventions,
delivered at an individual level to small subpopulations
experiencing specific conditions (Centre for Homelessness
Impact, 2018).2
In most cases, very little is known about the quality and
eectiveness of some of the most common interventions used
in homelessness (Centre for Homelessness Impact, 2018).3
There is huge scope to better understand how to help people
who are homeless or at risk of homelessness. Understanding
TEIXERA_Using evidence to end homelessness_text_3.6.indd 169TEIXERA_Using evidence to end homelessness_text_3.6.indd 169 02/04/2020 16:24:0902/04/2020 16:24:09

Using Evidence to End Homelessness
150
the eectiveness of an intervention is a key step in evidence-
informed decision-making, and more robust evidence from
well-designed research studies is needed (Craig etal, 2018;
Culhane, Fitzpatrick and Treglia, this volume).
The aim must be to end, rather than manage, homelessness,
and greater attention also needs to be paid to prevention by
researchers, policy-makers and practitioners. This means
building evidence about what is eective to favourably reshape
the complex system of determinants of homelessness and
ensure this informs policy and practice. This requires a wider
approach to generating and translating evidence, and a better
understanding of how to interpret and use this in policy-making
and practice. Complexity is not a reason not to build and use
good evidence. Rather, it means that a wide range of methods
needs to be deployed.
Taking a complex systems approach
The public health and homelessness sectors both recognise that
a system-wide perspective is needed to eectively prevent their
respective problems. This is a sound foundation on which to
build. Now, systems thinking needs to become integral to the
design of research and to the translation of the resulting evidence
to policy and practice.
While there is good understanding of the major drivers
of homelessness, less is known about how these cluster or
interact and how they influence the risk of homelessness. Less
still is known about how to address these broad factors in an
aligned way, as part of a system-wide approach. Developing
a comprehensive understanding of the system and system
dynamics can enable better design of research into what works
in tackling homelessness.
Taking a systems approach involves building a shared
understanding of the system of causes and consequences,
using appropriate research methods including ongoing
monitoring and evaluation, and – importantly – building an
understanding of these among practitioners and policy-makers
to support translation.
TEIXERA_Using evidence to end homelessness_text_3.6.indd 170TEIXERA_Using evidence to end homelessness_text_3.6.indd 170 02/04/2020 16:24:0902/04/2020 16:24:09

A public health approach
151
System mapping
The Centre for Homelessness Impact recently convened a broad
range of stakeholders to map the system of factors influencing
homelessness. System mapping involves identifying as many
factors as possible from all perspectives within a sector, and then
mapping these out alongside the connections between them.
This is typically done in workshops involving as many relevant
stakeholders as possible.
When carefully constructed, system mapping workshops are
commonly characterised by a positive energy and openness
among participants, who find value in stepping away from their
day-to-day focus to take a birds’ eye view of a problem and
discover where their work links – or should link – with that of
others. The value to be gained from convening people working
with a shared goal, who may have never met or spoken before, is
not to be underestimated. This can be an important intervention
in and of itself. Involving those who may not have realised the
important role their part of the system can potentially play in an
issue can eectively act to engage them with the issue.
Mapping the system in this way can be an important step
in developing a system-informed approach to action and
evaluation. Building a visual representation of a system around
an issue, with interconnections, pathways and feedback loops,
can provide insight to help policymakers, practitioners and
researchers make better informed decisions. The quality and
usefulness of such maps, however, depends on the process of
their creation. A system map reflects only the perspectives
of those involved in its development. Getting this right and
involving relevant organisations and people, including those
with lived experience of an issue, is therefore crucial to provide
views from all parts of the system.
A system map can enable identification of areas where
evidence exists or where there are gaps, against where there is
currently action or investment, or indeed, a lack of it. It can
also be enlightening that evidence, practice, and perspectives
about what might work, are not always entirely aligned (NIHR
School for Public Health Research, 2019). The system map
can also help in understanding the contextual factors that need
TEIXERA_Using evidence to end homelessness_text_3.6.indd 171TEIXERA_Using evidence to end homelessness_text_3.6.indd 171 02/04/2020 16:24:0902/04/2020 16:24:09

Using Evidence to End Homelessness
152
to be in place for an action to be as eective as possible, as
well as the potential wider consequences – both intended and
unintended. This can help identify measures for evaluation, to
help understand processes and pathways, and ensure no harm is
caused as a consequence of unanticipated eects.
System mapping can identify a broad range of system-wide
process, output and outcome measures that need to be measured
in the evaluation of any action to reshape the system, in order
to fully understand whether, why and how it is eective. This
is important whether evaluating a system-wide programme
of action, or a single, individual-level intervention. Taking
this wider perspective even to very specific interventions is
important for their application in practice. The system view can
identify contextual factors that should be measured in evaluation
to understand the context within which the intervention is
eective and any potential barriers to its eectiveness.
Building evidence in complex systems
Models of evidence to understand and evaluate action in
complex systems are currently the subject of much interest and
development in the field of public health. There are, however,
sources of guidance and well-established, robust methods that
already exist and could be applied to homelessness.
The NIHR School of Public Health Research has published
guidance on systems approaches to local public health evaluation
(Egan et al, 2019a), providing an accessible introduction
to thinking about systems and the benefits of this wider
perspective, plus considerations for planning and adopting a
systems-informed approach to evaluation.
It does not cost anything to think about activities from
a systems perspective, nor does the incorporation of
a systems perspective into an evaluation need to be
dicult, laborious or expensive. Bringing a systems
approach to an evaluation may merely involve
thinking slightly dierently about the kinds of ways in
which an intervention may exert its eects, and how
those eects might be assessed (Egan etal, 2019b).
TEIXERA_Using evidence to end homelessness_text_3.6.indd 172TEIXERA_Using evidence to end homelessness_text_3.6.indd 172 02/04/2020 16:24:0902/04/2020 16:24:09

A public health approach
153
The NIHR guidance outlines the broad uses of systems
approaches in understanding problems (mapping the system
as described earlier), identifying and assessing potential levers
of change and comparing hypothetical scenarios that involve
changing part, or parts, of a system (Egan etal, 2019b). It also
identifies the potential value of systems approaches whether
evaluating actions to change a single point in a system, ‘whole
system’ approaches to changing many points, or those to change
relationships within a system (for example that encourage joint
working across sectors). Six main types of systems evaluation
methods are described that are currently in use in public health,
acknowledging the methodological innovation taking place in
the sector.
The NIHR guidance describes three levels of complexity that
can characterise public health issues and the approaches to them,
that may need to be considered by evaluators:
• Complex interventions: comprising a number of dierent
activities, flexible forms of delivery and requiring input from
dierent people or organisations. Reshaping complex systems
will often require a complex intervention.
• Complex environments: made up of people, activities,
organisations, rules and places, that all interact as part of
a system. Regardless of the level of complexity of the
intervention, the environment it is delivered in is highly
complex and constantly changing.
• Complex consequences: in individuals – who may be aected
in several different ways – populations, political and
economic conditions, and in the way dierent agencies
interact. Feedback loops, when consequences influence the
intervention itself, add further to this complexity.
The Medical Research Council (MRC) has published draft
guidance on the planning, development, feasibility testing,
evaluation and implementation of complex interventions,
aiming to support more ‘complexity-informed’ research
(MRC, 2019). This describes complexity of an intervention
as being not only a property of the intervention characteristics,
but also the context in which the intervention is located, and
TEIXERA_Using evidence to end homelessness_text_3.6.indd 173TEIXERA_Using evidence to end homelessness_text_3.6.indd 173 02/04/2020 16:24:0902/04/2020 16:24:09

Using Evidence to End Homelessness
154
the research perspective taken – and of the interaction between
these factors. The characteristics of interventions are described
as lying along a spectrum from ‘simple’ to ‘complicated’, and
the research perspective on a continuum from ecacy to
systems, including:
• Ecacy perspective: to what extent does the intervention
produce the intended outcome(s) in experimental settings?
• Eectiveness perspective: to what extent does the intervention
produce the intended outcome in real world settings?
• Realist perspective: what works, for whom, under which
circumstances, and why?
• Systems perspective: how does the intervention interact with
the system to produce change?
Complexity-informed research, with an awareness of system(s)
can encourage: (i)researchers to develop research questions that
take into account the wider contextual factors that influence
an intervention; and (ii)researchers, funders, practitioners and
policy makers to develop, evaluate, and implement interventions
using the most appropriate tools and methods.
This MRC draft guidance discusses the range of methods
available to researchers and selection of the best available
method for the circumstances. Randomised experimental
methods are a means of eliminating certain biases in research
and should be considered where appropriate. RCTs have been
the predominant research method in evidence-based medicine,
and widely adopted in public health research since. They are
the ‘gold standard’ method for evaluating the eectiveness of
single, individual-level interventions, independent of wider
changes or context. For actions delivered at the population level,
this may not be possible; in those cases, other methods can be
used to build a robust evidence base about what works. This
includes other experimental randomised designs, which may
be used where it is not appropriate or possible to carry out a
conventional RCT, but also natural experimental designs and
systems designs, including modelling and case studies.
TEIXERA_Using evidence to end homelessness_text_3.6.indd 174TEIXERA_Using evidence to end homelessness_text_3.6.indd 174 02/04/2020 16:24:0902/04/2020 16:24:09

A public health approach
155
There is growing interest in natural experiments in the field
of public health research. These include events not under the
control of a researcher that divide a population into exposed
and unexposed groups. The naturally occurring variation in
exposure is exploited to identify the impact of the event on
outcomes of interest. The evaluation of natural experiments
requires careful consideration and design, but a growing range
of methods are available, which are described in detail elsewhere
(Craig etal, 2017).
Natural experiments are seen as key to evaluating large-scale
population health interventions – for example the introduction
of a policy or other population-level action – that are not
amenable to experimental manipulation (Rutter etal, 2017).
One example is the introduction of the soft drinks industry
levy (SDIL) in 2018 in the UK. Public health researchers
saw the opportunity here, and planned – well in advance to
allow collection of baseline data – a comprehensive, system-
level natural experiment to evaluate its impact (CEDAR,
2017). The evaluation aims to examine not only whether, how
and for whom the levy has an impact on health, but also the
process by which the levy came about and the wider changes
in public, political, societal and industry attitudes to sugar and
the SDIL over the four years before and four years following its
introduction. System mapping identified potential mechanisms
for action, and thereby the data collection and methodologies
necessary to explore them across a broad range of areas. These
diverse data sources include: purchasing data of soft drinks and
confectionary to examine whether people switch from sugary
drinks to sugary foods; market research data; surveys on attitudes
to sugar and the levy; and government data on health outcomes
including tooth decay and obesity.
This demonstrates the value of a systems approach in
identifying ways of assessing and understanding the broader
consequences of actions – both intended and unintended – and
the processes that may lead to them. Similar approaches can be
used to accelerate progress in building evidence for what works
in homelessness.
TEIXERA_Using evidence to end homelessness_text_3.6.indd 175TEIXERA_Using evidence to end homelessness_text_3.6.indd 175 02/04/2020 16:24:0902/04/2020 16:24:09

Using Evidence to End Homelessness
156
Translating evidence into action
Understanding the system, and careful and appropriate design of
research, will only get us so far, and barriers remain to putting
the insights generated into practice. These can include political
will, cognitive bias, separate budgets, organisational or sectoral
performance targets that do not reflect the value of these ways
of working, and the immediate benefits of action falling in
dierent parts of the system to those that need to take, or pay
for, that action.
The development of a deeper understanding of the
principles discussed here among policy-makers, practitioners,
commissioners and researchers is of fundamental importance
to developing an evidence infrastructure about what it takes to
end homelessness sustainably and to translate this knowledge
into policy and practice.
It is important to eectively communicate with policy-
makers and commissioners to build understanding of systems
approaches. This will enable them to judge the quality of
evidence produced from a systems approach to research
and evaluation.
In Canada, systems planning at a local level is being promoted
for the development of system-wide approaches to preventing
and ending homelessness. The Systems Planning Collective
(SPC), a group of Canadian organisations that have joined
forces to support local areas in systems planning, define this
as follows:
Systems planning at the local level is the process of
strategically mapping, coordinating and delivering
services, supports, and programmes with the
rights, needs, desires of the client/user at the centre.
Its aim is to create an integrated system of care, in
which various actors and systems work together
towards solutions to complex social problems. When
applied to the issue of homelessness, the unifying
high-level goal of systems planning work is to
prevent and end homelessness. (Systems Planning
Collective, 2019)
TEIXERA_Using evidence to end homelessness_text_3.6.indd 176TEIXERA_Using evidence to end homelessness_text_3.6.indd 176 02/04/2020 16:24:0902/04/2020 16:24:09

A public health approach
157
A needs assessment for operationalising systems planning for
homelessness, carried out by the SPC, found that capabilities for
this vary greatly between local areas (Systems Planning Collective,
2019). This highlights a further challenge to implementing
evidence-informed, system-wide approaches to homelessness.
Understanding how these approaches can be put into
practice is critical to achieving impact and requires evidence
on how systems approaches can be implemented in practice.
An important aspect of this is evaluation, with a rapid feedback
cycle for making changes when an approach is not working, or
flexibility and responsiveness is required. The implementation of
local area systems approaches to homelessness in Canada creates
an important opportunity, to build understanding about this key
step of translation into action, if subjected to detailed evaluation.
Conclusions
Homelessness is the result of a complex system of interacting
determinants. The prevention of homelessness therefore requires
a system-wide perspective and upstream focus in both action
and research. This is not to suggest an alternative to learning
what works for people already experiencing homelessness, but
a more holistic approach to the problem that includes support
for these individuals.
Like public health, homelessness needs to embrace a
movement for evidence-informed practice to ensure that action
is based on the best available evidence. To facilitate this, we
need to build stronger evidence of eective strategies to support
people who are at risk of, or experiencing, homelessness and
to prevent homelessness further upstream. Evaluation must
consider the whole system to understand why or how individual
or population level interventions work, and the context
and conditions necessary for success. It is vital that once an
intervention or policy is put into practice, ongoing evaluation
and feedback becomes an integral part of learning for evidence-
informed approaches to the problem.
A key challenge for the movement will be to explicate the
need for evidence-informed, whole system approaches to
homelessness among researchers, policy-makers, commissioners
TEIXERA_Using evidence to end homelessness_text_3.6.indd 177TEIXERA_Using evidence to end homelessness_text_3.6.indd 177 02/04/2020 16:24:0902/04/2020 16:24:09

Using Evidence to End Homelessness
158
and practitioners. Widespread adoption of these principles
and research methods will ensure that we achieve the
change urgently required to support individuals, and prevent
homelessness for good.
Notes
1 Centre for Homelessness Impact. 2019. Evidence and Gap Maps of
Eectiveness and Implementation Studies. CHI: London: https://www.
homelessnessimpact.org/gap-maps
2 Centre for Homelessness Impact. 2019. Evidence and Gap Maps of
Eectiveness and Implementation Studies. CHI: London: https://www.
homelessnessimpact.org/gap-maps
3 Teixeira, L. etal. 2019. The Share Framework: a smarter way to end
homelessness. CHI: London: https://www.homelessnessimpact.org/share
References
Boswell, K., Tait, R. and Eisenstein, C. 2018. Tackling the
Homelessness Crisis: Why and How You Should Fund Systemically.
NPC. https://www.thinknpc.org/resource-hub/tackling-
the-homelessness-crisis-why-and-how-you-should-fund-
systemically/ (accessed 24April 2019).
Bramley, G., Fitzpatrick S., Edwards J., Ford D., Johnsen S.,
Sosenko F. and Watkins D. 2015. Hard Edges: Mapping Severe
and Multiple Disadvantage in England. Lankelly Chase. https://
lankellychase.org.uk/resources/publications/hard-edges/
(accessed 24April 2019).
CEDAR. 2017. Soft drinks industry levy evaluation. www.cedar.
iph.cam.ac.uk/research/dietary-public-health/food-systems-
public-health/sdil/ (accessed 24April 2019).
Centre for Homelessness Impact. Evidence and gap maps.
https://www.homelessnessimpact.org/gap-maps (accessed
27November 2019)
Craig, P., Gibson, M., Campbell, M., Popham, F. and Katikireddi,
S.V. 2018. Making the most of natural experiments: what can
studies of the withdrawal of public health interventions oer?
Preventive Medicine, 108: 17–22. https://reader.elsevier.com/
reader/sd/pii/S0091743517305133?token=49FF6804CE9F6
B4E29785CC3F59B20628D9F46D303B124189451F51F34
6C1885A8A45E3D0FC79487313471B446D4F101 (accessed
24April 2019).
TEIXERA_Using evidence to end homelessness_text_3.6.indd 178TEIXERA_Using evidence to end homelessness_text_3.6.indd 178 02/04/2020 16:24:0902/04/2020 16:24:09

A public health approach
159
Craig P., Katikireddi, S.V., Leyland, A. and Popham, F. 2017.
Natural experiments: an overview of methods, approaches,
and contributions to public health intervention research.
Annual Review of Public Health, 38, 39–56. www.annualreviews.
org/doi/full/10.1146/annurev-publhealth-031816-044327
(accessed 24April 2019).
Egan, M., McGill, E., Penney, T., Anderson de Cuevas, R., Er,
V., Orton, L., Lock, K., Popay, J., Savona, N., Cummins, S.,
Rutter, H., Whitehead, M., De Vocht, F., White, M., Smith,
R., Andreeva, M., Meier, P., Marks, D. and Petticrew, M.
(2019a). NIHR SPHR Guidance on Systems Approaches to Local
Public Health Evaluation. Part 1: Introducing Systems Thinking.
https://sphr.nihr.ac.uk/wp-content/uploads/2018/08/
NIHR-SPHR-SYSTEM-GUIDANCE-PART-1-FINAL_
SBnavy.pdf (accessed 24April 2019).
Egan, M., McGill, E., Penney, T., Anderson de Cuevas, R., Er,
V., Orton, L., White, M., Lock, K., Cummins, S., Savona,
N., Whitehead, M., Popay, J., Smith, R., Meier, P., De Vocht,
F., Marks, D., Andreeva, M., Rutter, H. and Petticrew, M.
(2019b) NIHR SPHR Guidance on Systems Approaches to Local
Public Health Evaluation. Part 2: What to Consider When Planning
a Systems Evaluation. https://sphr.nihr.ac.uk/wp-content/
uploads/2018/08/NIHR-SPHR-SYSTEM-GUIDANCE-
PART-2-v2-FINALSBnavy.pdf (accessed 24April 2019).
Government Oce for Science. 2007. Tackling Obesities: Future
Choices. Foresight. https://assets.publishing.service.gov.uk/
government/uploads/system/uploads/attachment_data/
file/287937/07-1184x-tackling-obesities-future-choices-
report.pdf (accessed 27November 2019).
Jebb, S. 2017. Dusting O Foresight’s Obesity Report. https://
foresightprojects.blog.gov.uk/2017/10/04/dusting-off-
foresights-obesity-report/ (accessed: 24April 2019).
MHCLG (Ministry of Housing, Communities & Local
Government). 2018a. The Rough Sleeping Strategy. www.gov.
uk/government/publications/the-rough-sleeping-strategy
(accessed: 24April 2019).
TEIXERA_Using evidence to end homelessness_text_3.6.indd 179TEIXERA_Using evidence to end homelessness_text_3.6.indd 179 02/04/2020 16:24:0902/04/2020 16:24:09

Using Evidence to End Homelessness
160
MHCLG (Ministry of Housing, Communities & Local
Government). 2018b. Homelessness Reduction Act: Policy
FactSheets. www.gov.uk/government/publications/
homelessness-reduction-bill-policy-factsheets (accessed:
24April 2019).
MRC. 2019. Updated Guidance: Developing and Evaluating
Complex Interventions (draft for consultation). www.sphsu.
gla.ac.uk/stakeholder-survey-2019/Full%20complex%20
guidance%20draft%20for%20consultation%20v1.1%20
26.03.19.pdf (accessed 24April 2019).
NICE. 2015. Obesity Prevention CG43. Supporting Evidence.
www.nice.org.uk/guidance/cg43/evidence (accessed: 24April
2019).
NIHR School for Public Health Research. 2019. Developing
a Systems Perspective for the Evaluation of Local Public Health
Interventions. https://sphr.nihr.ac.uk/research/developing-
a-systems-perspective-for-the-evaluation-of-local-public-
health-interventions-theory-methods-and-practice/ (accessed
24April 2019).
Rutter, H., Savona, N., Glonti, K., Bibby, J., Cummins, S.,
Finegood, D.T., Greaves, F., Harper, L., Hawe, P., Moore, L.,
Petticrew, M., Rehfuess, E., Shiell, A., Thomas, J. and White,
M. 2017. The need for a complex systems model of evidence
for public health. Lancet, 390 (10112): 2602–4.
Selbie, D. 2018. Turning the Tide on Tobacco: Smoking in England
Hits a New Low. https://publichealthmatters.blog.gov.
uk/2018/07/03/turning-the-tide-on-tobacco-smoking-in-
england-hits-a-new-low/ (accessed: 24April 2019).
Systems Planning Collective. 2019. Systems Planning to Prevent
and End Homelessness. www.homelesshub.ca/sites/default/files/
attachments/SPC-HPS-NeedsAssessmentReport.pdf (accessed
24April 2019).
TEIXERA_Using evidence to end homelessness_text_3.6.indd 180TEIXERA_Using evidence to end homelessness_text_3.6.indd 180 02/04/2020 16:24:0902/04/2020 16:24:09