About
Our work
Evidence and data tools
Join the movement
Some reliable evidence
Insufficient evidence available
At least one high quality study shows negative impact
Homeless shelters are a basic form of temporary accommodation where a bed is provided in a shared space overnight. One of the key features of a homeless shelter is that it is transitional and an option for those homeless who are not yet eligible for more stable accommodation. Shelters are not usually seen as stable forms of accommodation as the individual must vacate the space during daytime hours with their belongings. Homeless shelters often place additional requirements on potential users including night time curfews.
Homeless shelters are not always free, and some shelters will charge the individuals a fee for the bed, while others require that users be claiming benefits from the government. Additional services that may or may not be provided by the homeless shelter are warm meals for dinner and breakfast or support from volunteers who help individuals make connections to other services.
Day shelters for homeless individuals act as a drop-in centre, often aimed towards those homeless with additional needs such as substance abuse, or mental illness. Services may include access to case workers, meals, access to laundry facilities, or support groups. The obvious difference between night and day shelters is that a day shelter will not offer a bed to the individuals who use the services.
Homeless families will also utilise homeless shelters. Although authorities have increased responsibility to rehouse families with children into stable accommodation more efficiently, families are sometimes be placed in a shelter as an emergency measure. When children are residing in a shelter other child services will usually become involved to ensure their safety and welfare. Homeless families will often be provided with additional support such as a case management worker. Homeless families will usually be kept together, so their accommodation will often be a bedsit room or apartment.
Some criticisms of homeless shelters relate to overcrowding, physical altercations, theft, substance abuse, and unhygienic sleeping conditions.
The goal of homeless shelters is to provide accommodation for people who have nowhere to sleep. Homeless shelters are used to quickly deal with emergency situations such as adverse weather conditions or domestic abuse.
Homeless shelters are viewed as an adequate option for those homeless people who do not have the option of safe, stable housing.
Homeless shelters provide emergency accommodation and refuge for those at the point of crisis. This provision of a bed seeks to protect the individual from adverse environmental conditions and physical or psychological illnesses associated with spending the nights on the streets.
Homeless shelters appear to produce positive effects relating to employment and cost effectiveness but this should be interpreted cautiously as these findings are based on a single study.
Sheltered accommodation does not appear to be useful for general health issues, substance abuse issues, or social functioning of homeless individuals.
Of the six studies included in this summary, five were conducted in North America and one was conducted in Holland. Four studies were randomised control trials and two were cohort studies.
Three studies measured two outcomes in their evaluation of shelters, while the remaining three studies measured three outcomes. Of the six included studies, five measured outcomes relating to health and substance abuse, three looked at the effects of shelters on employment and income, two measured outcomes related to housing stability, two looked at capabilities and wellbeing, two were interested in access to services, and one measured the cost of the shelter.
The effectiveness of shelters is measured across this range of outcomes at various time points. The longest follow-up of included participants was two years.
The participants included were varied. One RCT included 251 young people, one RCT included 330 families with children, two studies were interested in homeless people with alcohol and drug issues one of which was a RCT including 106 individuals and the other was a cohort including a sample of 74. Finally, participants included people with complex needs including a dual diagnosis, this population was tested in a cohort design containing 205 individuals, and in a RCT (sample size not reported).
For outcomes related to health and substance abuse, the benefits of sheltered accommodation are negligible. In a RCT that looked at the effects of sheltered accommodation on young people at six-month follow-up, alcohol use had increased while drug use decreased with time spent in a shelter. In a cohort study on individuals with substance abuse issues (n=74), being placed in a shelter did not improve outcomes related to overall distress, psychiatric symptoms, or substance abuse at the three-month follow-up. Finally, in a cohort study (n=205) that looked at health outcomes they found that 37% of all shelter participants had been admitted to hospital, a further 10% were admitted to intensive care facilities where more than 3% died. This study demonstrated that a large majority of participants (70%) could not provide informed consent when admitted to hospital and highlights the usefulness of planning advance care with local hospitals when individuals first present themselves to a shelter.
The two studies that measured outcomes relating to the participants’ housing stability demonstrate different effects of sheltered accommodation. In a large RCT on 330 families the researchers found that those who were in the shelter only group (no services added) had spent more days in sheltered accommodation, and less time between their exit from sheltered accommodation and subsequent readmittance to sheltered accommodation at the two-year follow-up. However, in an RCT on 106 individuals with alcohol and drug issues, the sheltered accommodation participants demonstrated improvements in housing stability at the one-year follow-up.
The studies that measured outcomes relating to the participants’ capabilities and wellbeing were inconclusive. In a RCT that looked at the effects of sheltered accommodation on young people, satisfaction with social relationships decreased, this finding is supported in a cohort study on individuals with substance abuse issues (n=74) who after being placed in a shelter did not improve outcomes related to social functioning at the three month follow-up. However, in the aforementioned RCT positive effects relating to depression, anxiety, and competence were observed.
In an RCT on 106 individuals with alcohol and drug issues, the sheltered accommodation participants demonstrated improvements in employment outcomes at the one year follow-up.
Cost data are available, but there are no cost effectiveness studies. Only one study included information about the cost of sheltered accommodation. This study calculated that the cost of an individual to stay in sheltered accommodation was $25 USD per unit, per person, per day (2010 figure).
If you are implementing a shelter service, create a welcoming environment with clear rules. To build trust, ensure that your staff have excellent people skills and make service users feel listened to and cared for.
Staff should focus on helping service users to move on to more permanent housing. A shortage of appropriate move-on accommodation may mean that people remain in short-term hostels for longer than necessary, taking up spaces needed by new arrivals.
Partnerships and joint-working protocols with other agencies can help service users move on to more permanent accommodation and access multi-agency support.
If you’re involved with funding for shelters, offer consistent and long-term funding to promote stable service provision. You should also think about how to ensure funding and provision is seamless across accommodation and support services.
Ensure you offer flexible support that can be tailored to meet the needs of specific groups, e.g. young people, LGBT people and people who have experienced trauma. Staff may need additional training to be able to meet different needs.
