Homeless healthcare: raising the standards
Abstract
Over the past 3 years the number of homeless people in the UK has increased by 34%. Most will die young, largely due to treatable conditions. Secondary care can, and must, do more for the silent killer that homelessness is.
Introduction
Perhaps the true measure of greatness in a healthcare system is how it provides health for its weakest citizens. This is not only a moral responsibility but, through the 2012 Health and Social Care Act, also a legal responsibility. Almost 20 years on from the Royal College of Physicians’ (RCP’s) working party report Homelessness and ill health, this article reviews the health and illness of homeless individuals today and reviews the care initiatives that exist to provide services to this vulnerable population.1
Homelessness is not a single entity. The Department of Health's definition includes people sleeping rough (on the streets), those in the hostel system, those in squats and those dependent on insecure temporary accommodation with friends (‘sofa surfers’). There is also a significant health effect for people living in inadequate accommodation, because it is overcrowded, unsafe or not suitable for their needs, and those whose accommodation is temporary. Significant anxiety is associated with the imminent threat of homelessness.
In the UK in 2009–10, it was estimated that 40,500 people were homeless within the hostel system and, over the course of 1 year, another 100,000 individuals use night shelters.
Official statistics on numbers of rough sleepers are widely acknowledged as an underestimate. However, during 2011–12, 5,678 people were seen sleeping rough by outreach workers in Greater London alone – an increase of 43% on the previous year.2
Overview
Homelessness is associated with early death, with the average age at death of a homeless man consistently reported in the range 40–47 years.2 Tragically, homeless people still die of treatable medical conditions.
Homelessness is associated with ill health, probably as both a causative and a causal factor. A survey carried out by the homelessness charity St Mungo's in 2008 revealed that 43% of service users had a medical problem on the day of survey.3 The RCP's 1994 working party into homelessness and ill health found that around two-thirds of medical problems in homeless adults predate their homelessness.4 Homelessness is characterised by tri-morbidity: a combination of mental ill health, physical ill health, and drug or alcohol misuse.
In the National Health Service (NHS), ‘NFA’ – no fixed abode – is the current coding for homelessness. This is likely to under-record homelessness because patients will often give a hostel address or a friend's address to avoid being labelled homeless. With that caveat in mind, analysis of hospital episode statistics demonstrates that, when comparing adults aged 15–64 years who are documented NFA, 89% of secondary care usage by homeless patients is in emergency care, compared with 41% of emergency usage by the housed age-matched population. There is a specialty bias, with higher usage of emergency medicine and lower usage of ear, nose and throat, and oncology services. Length of stay is also significantly longer, at 6.2 days compared with 2.1 days in age-matched control individuals.2
When smaller cohorts of homeless patients registered with specialist services are compared with local averages, presentation to the emergency department (ED) is five times higher in the homeless population and admission three times as likely.5
Medical problems associated with homelessness
Homelessness is clearly associated with poor health and premature mortality owing to complex interactions between reasons for homelessness (eg poverty and substance misuse) and consequences of homelessness (eg cold environment and overcrowding).
Fig 1 shows the most frequent diseases found among homeless people. The most common health needs of homeless people relate to drug dependence, alcohol dependence and mental health problems. Multiple, coexisting, physical and mental morbidities are the norm. A survey, by the charity Crisis in 2002 of 389 homeless people in London, showed that 372 had used an illicit substance within the last month, with heroin and crack cocaine being the most common, although four out of five respondents had started taking a new drug after becoming homeless.6 However, estimates of substance misuse rates vary according to the population sampled; a Leicester series found that around half of the homeless patients had neither drug nor alcohol dependence.7
The Gloucester experience
Even in more rural areas the homeless population contributes significantly to the workload in the ED and to the acute medical takes. A review undertaken at Gloucester Royal Hospital found that, in 2012–13, approximately 70 homeless people were admitted to hospital on 125 occasions, with a third of these patients being readmitted within 30 days. The main reasons for admission were:
27% alcohol and drug related (including leg ulcers, deep vein thromboses and abscesses)
21% mental health issues (including suicide attempts, overdoses and deliberate self-harm)
9% chest pain
9% abdominal pain.
Of these patients, 15% self-discharged, possibly because of lack of appropriate support and tailored care during their stay. An anonymous feedback questionnaire from 12 homeless people admitted acutely to hospital identified that 11 were not asked about accommodation on discharge and most returned to the streets, with one returning to living in a car and another returning to a cave. Most felt that they had been let down by their hospital experience and were dependent on their care being resumed by the Vaughan Centre (local general practitioner [GP]-led homeless healthcare clinic) (Gayle Clay, personal communication).
Romero-Ortuno et al. recently reviewed the profiles of 1,460 homeless people admitted to St James’ Hospital, Dublin between 2002 and 2011 (Table 1). They found that the most frequent reasons for admission were related to alcohol and substance abuse, similar to the Gloucester cohort. Other common conditions requiring admission to secondary care were seizures, venous thromboses, cellulitis, abscesses and hepatitis.8 Other cohorts have shown similar distribution of illness; notably, in another Irish study, respiratory disease is prevalent owing to a combination of smoking (80%), overcrowding, poor nutritional status and HIV infection.9
Morbidity affecting homeless people extends beyond acute crisis episodes, with a significant impact on life expectancy and quality of life. A study of tuberculosis (TB) patients in London found TB rates of 788/100,000 population, with strong associations to smear-positive TB, poor adherence to treatment, and patients less likely to complete treatment and be ‘lost to follow-up’.10
Much morbidity may be unrecorded owing to a reticence among homeless people to seek medical help. The recent British Medical Association report Drugs of dependence: the role of medical professionals warns that drug users may be discouraged from seeking help because of a fear that they will be treated as criminals.11 A study in Glasgow found that a homeless patient admitted with a drug-related problem was seven times more likely to die over the next 5 years than a patient with the same drug-related problem who was not homeless.12
Healthcare utilisation by homeless people
The charity CRISIS reports the following: homeless people are almost 40 times more likely not to be registered with a GP and five times more likely to have difficulty getting onto or staying on a GP's list than the general public
Homeless people are four times more likely to turn to the ED when they are unable to speak to a GP.13 Overall, homeless people attend the ED five times more often than those who are not homeless, are admitted 3.2 times as often and stay in hospital three times as long. The study from Dublin found a mean length of stay of 12 days, with 7% having prolonged admissions.8 This results in unscheduled secondary care costs that are eight times higher than for patients who are not homeless.5
As described in the cohort from Gloucester, homeless patients may have a negative experience of care, particularly in the hospital environment.5 The NHS mandate to provide high-quality, cost-effective care for all must therefore include interventions that improve the care provided to patients who are homeless.
Interventions to improve the health of homeless people
A multifaceted approach is needed to tackle the extremely high levels of mortality and morbidity associated with being homeless. Particular attention must be given to substance abuse, mental health disorders, physical ill health and homelessness itself.
Complex interactions between each facet exist and many risk factors for homelessness such as poverty and substance abuse are also strong independent factors for ill health. The combination of stable housing, and psychiatric and medical treatment has shown superior outcomes to treating health problems in the absence of stable housing.14
Provision of housing for homeless people with mental illness
In 2008, in two randomised controlled trials (RCTs), Forchuk et al studied 14 homeless people admitted to a psychiatric ward.15 On discharge, seven returned to the streets and all but one remained homeless after 6 months. Of the seven discharged to shelters, all remained housed at 6 months after hospital discharge.
Provision of housing for homeless people with substance abuse
In 2009, Larimer et al looked at 134 homeless people who had been housed with severe alcohol problems, and high healthcare use and costs.16 The median number of drinks consumed progressively declined from 15.7 drinks per day to 10.6 drinks per day at 12 months, at a rate of approximately 2% per month.
In two RCTs, Milby et al showed that the combination of behavioural day therapy with abstinent-contingent housing (‘dry houses’) and work therapy was more effective than behavioural day therapy alone.17,18 Assessing 110 homeless people in 2000, and repeating the study on 141 homeless people in 2003, showed that at 6 months 15% attained abstinence with day therapy alone compared with 41% of the triple-therapy group.
Provision of abstinent-contingent housing was significantly more effective than not providing housing or providing non-abstinent-contingent housing in achieving abstinence in 196 homeless people with alcohol problems.19
In summary, for homeless people with substance abuse issues, provision of housing was associated with decreased substance abuse, improved health service utilisation and improved health outcomes. Abstinent-contingent housing was more effective in supporting stable housing status, substance abstinence and improved psychiatric outcomes than non-abstinence housing or no housing.
Health services designed to meet the needs of homeless people
There is no one standardised pathway of care for homeless people, but independent local care pathways are being increasingly developed, responding to the location-specific needs of the homeless populations. In 2010 a third of primary care trusts (PCTs) did not provide any specialist services for homeless people whereas the other two-thirds of PCTS provided services that can be loosely categorised into the four models shown in Fig 2.2
Models A and B are appropriate for clinical commissioning groups (CCGs) with small homeless populations. Model C is a full primary care specialist team justifiable in larger urban areas. Primary care services are crucial to the provision not only of access to drug treatment, alcohol treatment and mental health services but also of health promotion and screening services. Specialised practices tend to find it easier to work with specialised social workers who can access the social care services needed by the population.20
Model D is based on the fully integrated services provided in Boston, USA. Although this does not yet fully exist in England, pilots are under way to increase integration between primary and secondary care. Innovative work on ward rounds for homeless people in a London hospital has recently been published.5 This Pathway service worked to overcome the common perception among homeless people that they were not treated similarly to other patients, while facilitating multiagency care coordination. Particular clinical problems encountered included delays in following protocols for methadone treatment, inadequate alcohol withdrawal treatment and poor pain control. The evaluation of this project also demonstrated cost savings to the hospital of £300,000. Such a service would be applicable to trusts in the major cities of the UK.
Dedicated secondary care provision
With increasing rates of homelessness a tiered approach should be developed across all secondary care providers, depending on the size and need of the population served.
Level 1
All hospitals should have a system for identifying potentially vulnerable adults, including homeless individuals and hostel dwellers, and have available a list of relevant contacts and agencies for support.
Level 2
A locally negotiated in-reach housing adviser should work in collaboration with an identified person or team within the hospital such as the discharge team.
Level 3
For hospitals with significant numbers of homeless patients, there should be a Pathway-type model, with a team to support multiagency care coordination.
Conclusions
Improving the health of homeless people requires activity across a range of sectors. Within the NHS more needs to be done to ensure that the sporadic pockets of good practice extend to all areas where there is a need.
The Faculty of Homeless and Inclusion Health has been working with the National Inclusion Health Board at the Department of Health and with NHS England to improve the commissioning of homeless health services.21 These services will need to be delivered by a coordinated team with expertise in substance misuse, mental health, primary care, social care and housing. Hospitals have an important role in identifying and quantifying the population in need who may not present to other health services. Only by the whole health system investing in services designed for these specific populations will acute trusts reduce demand on unscheduled services, and by ensuring that appropriate follow-up and preventive care are available to patients who are homeless will we see a true reduction in one of the most striking morbidity and mortality inequalities in this population.
- © 2014 Royal College of Physicians
References
- ↵
- Royal College of Physicians
- ↵
- Department of Health, Office of the Chief Analyst
- ↵
- Bilton H.
- ↵
- Royal College of Physicians
- ↵
- Hewett N,
- Halligan A,
- Boyce T
- ↵
- Fountain J,
- Howes S,
- National Addiction Centre
- ↵
- Hewitt N,
- Hilley A,
- Gray J
- ↵
- ↵
- Holohan TW
- ↵
- Murad S,
- Roberts W,
- Verheyen M,
- Hayward AC,
- London Tuberculosis Nurses’ Network
- ↵
- BMA Board of Science
- ↵
- Morrison DS
- ↵
- Crisis
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- Milby JB,
- Schumacher JE,
- Wallace D,
- et al.
- ↵
- Wright NMJ,
- Tompkins CNE,
- Oldham NC,
- Kay DJ
- ↵
- Faculty for Homeless and Inclusion Health
Article Tools
Citation Manager Formats
Jump to section
Related Articles
- No related articles found.