Minimising the impact of COVID-19 on people sleeping rough: an overview of UK and global responses
People who experience long-term or repeated rough sleeping are particularly vulnerable to COVID-19. While policy innovations in many countries – mainly providing emergency accommodation on a large scale – have reduced the immediate risks, these interventions are typically short-term in nature. So what is the evidence for what’s working in the UK and around the world, and how might short-term interventions be translated into longer-term answers to the issue of homelessness?
By Professor Nicholas Pleace, Director of Centre for Housing Policy at the University of York
- The international evidence is sobering and shows the potential risks to people sleeping rough during the COVID-19 pandemic. People sleeping rough cannot socially distance. People who experience long-term or repeated rough sleeping are likely to have underlying conditions that increase the risks from COVID-19.
- Internationally, ‘shared-air’ emergency shelters offering basic services with shared sleeping and living spaces are widely used as a response to rough sleeping. Some of these services have proven highly vulnerable to COVID-19 infection.
- However, across the world, policy innovation is reducing the immediate risks from the pandemic both for people sleeping rough and for wider public health. Short-run interventions have prioritised provision of emergency accommodation. Longer-term measures emphasise providing ordinary housing quickly, and then providing a package of support to deal with other needs.
- Finnish policy and practice, driven by a tightly integrated, multi-level national strategy combining a housing-led/Housing First approach with preventative services, has reduced rough sleeping to very low levels. Sustained reductions in rough sleeping are the best response to managing the potential public health risks around people sleeping rough for the current and future pandemics.
- The ‘Everyone In’ initiative in England – and similar programmes in Scotland, Wales and Northern Ireland – very rapidly deployed hotel rooms and other temporary accommodation that enabled people sleeping rough to leave the street. These same initiatives allowed the shared-air emergency shelters that operate in the UK to close or reduce capacity to safer levels.
- Fixed-site UK homelessness services for people sleeping rough are more likely to offer self-contained rooms which avoid the problems of emergency shelters. Use of housing-led or Housing First services that centre on using ordinary housing is also increasing.
- Everyone In is estimated to have prevented 34% of England’s homeless population (in shared-air services and living rough) getting COVID-19 between March and August 2020, with 364 deaths, 4,074 hospital admissions, and 572 critical care admissions being prevented.
- All UK nations have also introduced eviction bans, reducing the potential increases in all forms of homelessness that may stem from increasing unemployment as a result of the pandemic.
- International evidence highlights some of the potential risks if structural challenges facing the UK homelessness sector are not addressed, and if current short-term initiatives, such as Everyone In and the eviction bans, are withdrawn too early, or in an unplanned way.
The risks around COVID-19 and people living rough
Living rough can be both the result of other problems, from mental health to addiction, and a factor making them worse. Rates of severe mental illness can be high among rough sleepers, as can levels of addiction, often combined with poor physical health, limiting long-term illness and disability. The risks of some forms of infection, including tuberculous, can be higher than for the general population. If someone has been living rough for some time, or on a repeated basis, they are likely to have one or more underlying conditions that increase the risk of serious illness and mortality from contracting COVID-19.
People sleeping rough cannot self-isolate because they have no adequate space in which to do so. Alongside the risks to anyone living rough with COVID-19, there is a wider public health risk if someone cannot self-isolate when they need to. Canadian research reports that people with a recent history of homelessness are over 20 times more likely to be admitted to hospital for COVID-19, more than 10 times more likely to require intensive care, and over five times more likely to die within 21 days of their first positive test result.
Throughout much of the world, the main response to rough sleeping has been the provision of emergency shelters. These services can vary markedly in the nature and quality of service they offer. However, internationally, a high proportion of this sector is ‘shared-air’ services, in which people live and sleep in shared spaces that can contain dozens, even hundreds of people.
In France, infection rates in some shared-air emergency shelters have been reported to be between 23% and 62%, while rates of 17%, 36% and 66% have been reported in some US emergency shelters. In Brussels, a high rate of hospitalisation for COVID-19 has been found among a homeless population that often uses shared-air shelters and services, and that has high rates of serious underlying conditions.
These levels of infection, hospitalisation rates and mortality have not been paralleled among people living rough in the UK. As the challenges presented by the pandemic shift and change, the nature of any future risks we face is uncertain. Nevertheless, in early 2021, excess risk from COVID-19 to people sleeping rough and other people experiencing homelessness has, so far, been largely contained in the UK.
The strengths of UK homelessness services
Over the past three decades, the UK has been moving away from using a shared-air emergency shelter approach that offers only shared dormitory accommodation, reflecting a broader trend away from basic emergency accommodation as a response to rough sleeping. This approach – led in part by the homelessness sector itself – was pursued as evidence mounted that emergency shelters with only limited support tended to ‘warehouse’ people experiencing homelessness with complex needs; i.e. they would stay in emergency shelter and never move on into settled housing. In consequence, a lot of UK homelessness services offer better-standard accommodation, including individual bedrooms and higher levels of support.
The UK has also begun to move towards the use of housing-led and, particularly, Housing First services as a response to rough sleeping and wider homelessness. Housing First is most developed as a strategic response in Scotland but has also been widely advocated by the homelessness sector in England, and is the subject of a central government pilot programme in Liverpool, Manchester and the West Midlands. Scottish, English, Welsh and Northern Ireland homelessness policy is placing more and more emphasis on Housing First as a primary means to reduce rough sleeping.
Housing First is an approach first developed in the United States in the 1990s, in response to evidence that existing services were not reducing overall homelessness. Housing First rapidly places someone living rough straight into ordinary housing which is intended to become their settled home, providing support to enable them to sustainably exit homelessness.
Housing First was designed to replace US homelessness services that tried to make someone living rough ‘housing ready’ while staying in a shelter or temporary supported accommodation, before moving them into their own home. Randomised control trials in Canada, France and the US have all reported higher levels of success in reducing homelessness for Housing First, compared with orthodox ‘housing ready’ services.
Over a similar period (1998-2010), policy in England achieved a 60-80% reduction in rough sleeping using a more holistic strategy that aimed to tackle health and other needs alongside provision of housing, including stemming the flow of people on to the streets, dealing with the range of factors that kept them there (including mental health, drugs, alcohol etc), and providing a route out to housing and a job.
While Housing First is not perfect – the rates of social integration and improvements in mental and physical health are not consistent – global evidence indicates that around eight out of every 10 people with high and complex needs, including people sleeping rough on a recurrent or sustained basis, stay out of homelessness for at least one year when using Housing First.
International interest has focused on the Finnish national homelessness strategy. This pursues a Finnish version of Housing First, which emphasises that responses to homelessness should be focused on ordinary housing. By building an integrated national strategy within a Housing First framework, Finland has seen unusual levels of success in reducing rough sleeping and long-term and repeated homelessness. Interest in Finnish policy has been expressed at pan-European and global levels, as well as within the UK.
The ‘Everyone In’ initiative in England, which uses hotel rooms and other temporary accommodation to enable people sleeping rough to leave the street, was deployed very rapidly after the outbreak of COVID-19. This and similar initiatives throughout the UK allowed shared-air emergency shelters that were operating to close, or at least to reduce their capacity to safer levels.
Everyone In is estimated to have prevented 34% of England’s homeless population (in shared-air services and living rough) getting COVID-19 between March and August 2020, with 364 deaths, 4,074 hospital admissions and 572 critical care admissions being prevented. Very similar initiatives were pursued by governments in Scotland, Wales and Northern Ireland.
Throughout the world, similar steps have also been taken – centring on the provision of emergency shelter to stop people having to live on the streets, being unable to self-isolate, and to avoid infection or spreading infection. There has been comment in Australia that the hitherto ‘complex’ problem of ending rough sleeping has been rapidly – albeit temporarily – solved in the face of a public health emergency.
However, many of these policies are interim measures, designed to last only as long as the risk from the virus is present. Without further changes, when these measures are switched off, levels of people sleeping rough will rise again. In England, the Next Steps Accommodation Programme has been announced, drawing on the Housing First model to ensure that people accommodated under Everyone In will not return to homelessness.
A global comparison
Globally, not having the resources or infrastructure to manage the risks of COVID-19 among people sleeping rough and other people experiencing homelessness represents a real challenge. As much of the existing system is still ‘shared air’, widespread emergency measures are necessary. US researchers estimate that, given approximately 200,000 single adults were in shared-air shelters on a typical night at the last count, reducing emergency shelter density by 50% – to allow some social distancing – while maintaining current capacity would require an additional 100,000 units.
The US is not alone in this; much Canadian service provision and many homelessness services in Eastern and Southern Europe are shared-air emergency shelters. This creates the same two challenges: having to thin-out occupation of existing services so there is the chance to socially distance; while at the same time, having to find yet more emergency accommodation for anyone not already in a homelessness service. South Africa had to respond to this crisis by converting an entire stadium into emergency accommodation.
The contrast with the UK is marked. In the UK, many fixed-site homelessness services have been able to manage COVID-19 more successfully because people have their own rooms, and relative levels of housing-led services (including Housing First) are also probably higher. Everyone In and the other initiatives worked very well, in large part because of the way the UK homelessness sector works. Existing capacity did not need to be suddenly doubled to allow for social distancing, because the services were not, in large part, shared-air models.
Everyone In has not been an unqualified success, however. Rough sleeping did not end, and some early research led by King’s College London has suggested mixed results from the attempts to provide settled housing. Related research from St Mungo’s indicates there have been some health gains linked to properly supported hotel stays, but also concerns about the viability of creating high-quality services that enable sustainable exits from homelessness. Nevertheless, the combination of: (i) a homelessness sector that could generally cope very well with COVID-19; (ii) the political will and resources to clear the streets and close down or restructure shared-air services that were running; and (iii) an increasingly housing-led response to rough sleeping, which reduces long-term and recurrent living rough, has – for now – stopped COVID-19 being a much greater threat to people experiencing homelessness than to the rest of the UK population.
It is not the case that most people who are evicted become homeless, and they certainly do not all become rough sleepers. However, the risk of mass eviction following the economic impacts of the pandemic might reasonably be expected to mean some increase in homelessness and, alongside that, some increase in rough sleeping. What such an increase in homelessness may look like is difficult to know because, as with much else in this pandemic, we are on uncertain ground.
Globally, almost all 37 of the OECD countries have introduced some form of emergency eviction ban, as have almost all EU member states, with different versions existing in England, Scotland, Wales and Northern Ireland. Every eviction ban is temporary – and almost every such ban has been characterised by seeing the original period it was designed for extended, then extended again, and sometimes extended for a third time. The inability to contain the virus after apparent gains in the summer of 2020, with the subsequent emergence of variants that (in the UK and elsewhere) spiralled rapidly out of control, saw measures that were never supposed to last a year or more being rolled forward.
It is, as previously stated, difficult to know quite what will happen next. Internationally, we will see some countries take careful steps to – very slowly – close down their eviction bans. Others will effectively hit an ‘off switch’ without making any contingency plans. Instinct tells us the former approach is likely to be the more sensible one, but this is not an experiment we have run before. The effect could be to open the floodgates on street homelessness – or the impact might be much less pronounced.
There is a broader point here. Various potential flows into homelessness are being slowed by these multiple interventions. Globally, millions have been spent on emergency accommodation to reduce rough sleeping, and a great deal more on various interventions to ban evictions, maintain employment through furlough arrangements, and additional supports when unemployment is occurring.
These arrangements are almost certainly stopping some homelessness, and may well be stopping very significant increases. But none of these interventions are permanent at present, and the next set of challenges will centre on how to manage what may be an increase in homelessness resulting from major economic shifts following the pandemic.
Homelessness prevention is an area where the UK has been a global pioneer over the last 20 years, with significant recent innovation being led by the Welsh Government. Here, a combination of services is used both to try to stop eviction from happening where possible (such as negotiating deals with landlords to pay off rent arrears over time), and to ensure rapid re-housing (i.e. getting someone into an alternative settled home before an unstoppable eviction potentially makes them homeless).
There is some evidence that this is a much more cost-effective response than dealing with the consequences of homelessness or living rough – avoiding significant spending for the state and (more importantly) the human costs of experiencing homelessness or living rough. Prevention is also at the core of the highly effective, integrated national homelessness strategy in Finland.
COVID conclusions for the UK: uneven progress on homelessness
Much has gone wrong in how the UK has responded to the pandemic. Prevalence and mortality have been higher in the general population than in many comparable countries. The Office for National Statistics has reported clear associations between area deprivation (which includes poor-quality and overcrowded housing and increased risk of mortality) and deaths involving COVID-19. At 3.1 deaths per 100,000 population for the most-deprived areas in England, this is (statistically) significantly higher than the 1.4 deaths per 100,000 in the least-deprived areas. Black and minority ethnic populations are also more likely to be relatively higher in more-deprived areas.
However, in respect of containing and managing the pandemic among people sleeping rough, the UK has fared comparatively well. The emergency measures, centring on the eviction ban and variations of Everyone In, while not perfect, have almost certainly reduced potential infections among people sleeping rough and prevented some deaths. Throughout the UK, policy and practice draws on the most significant innovations from around the world, underpinned by increasingly housing-led and Housing First responses. In one area, homelessness prevention, the UK has been leading the way for some time.
Progress has not been even, though. Scotland is outperforming the other nations in terms of its rapid re-housing response, and Wales has set the pace around prevention. Major UK cities also tend to have more coordinated and effective responses to rough sleeping, which is the case in many comparable countries too.
Different visions of the future from the rest of the world
Across the world, shared-air homelessness services – centred on provision of emergency shelters and day facilities for people sleeping rough, in which sleeping quarters and other living spaces are shared – have shown very low resilience in the face of COVID-19. When the virus has taken hold in these services, infection rates have been astronomical. Containment of the virus has often necessitated halving or more-than-halving capacity, or even total closure of these services. This has not happened in the UK.
Nevertheless, evidence from the rest of the world suggests the UK can do better in some areas. Finland is something of a shining light, although there are things to learn from Denmark and Norway too. Rough sleeping can be not only contained but reduced when governments invest political capital and resources into an integrated strategy – i.e. a network of interconnected services working together.
Prevention services stop eviction or other triggers of homelessness – for example, using family mediation to prevent a young person leaving home in an unplanned way (if they are not at risk) because of a relationship breakdown; or stopping and removing a perpetrator of domestic abuse so someone else can remain safely at home. Where prevention cannot work, rapid re-housing systems step in. An integrated strategy means that when someone is at risk of homelessness because of support or treatment needs (e.g. around mental health), those services are coordinated with the systems designed to prevent and rapidly re-house.
If someone is homeless, the response should be integrated and scalable: ranging from low-level, temporary support – a bit of help with a rent deposit and settling in – to a full, multi-agency response involving housing, health, social care and other services facilitated through something like Housing First. The successes achieved by these systems illustrate the need for action by the state – and (as in the case of Finland) that no amount of innovation, prevention and coordination will in itself solve rough sleeping and homelessness without doing what the Finns also did, which was increase affordable, social housing supply.
But evidence from around the world also offers visions of what could be a troubling future for the UK with regard to homelessness. According to one study of local authority spending, more than £1 billion less was spent supporting single homeless people in England in 2017/18, compared with 10 years earlier. Endless uncertainty over unpredictable (and ever-decreasing) local authority homelessness budgets hampers strategic planning by both local government and the homelessness sector. This makes a negative scenario like that currently experienced by many other countries become more and more possible.
Everyone In was a positive response to the pandemic – but also a sign of weakness at the edges of a UK homelessness system that is stronger and more effective when it has more of the resources it requires to operate. Removing resources from homelessness services to the point where they start to break, rather than investing the money needed to make prevention work and, above all, addressing the seemingly endless shortfalls in affordable housing supply, can only have one consequence. Many other countries have a far less-developed, less-effective homelessness sector than the UK. They have shared air.
Professor Nicholas Pleace is Director of Centre for Housing Policy at the University of York