Introduction
There is a scarcity of prevalence data in relation to accommodation needs among patients attending the Irish mental health services. The prevalence data that is known mostly emerged as a result of attempts to plan for the closure of old psychiatric hospitals and inappropriate community residences. These findings were that 45% of acute beds in the Eastern Area Health Board were occupied by non-acute patients (Keogh et al. Reference Keogh, Roche and Walsh1999) and over 3000 places in over 400 mental health community residences were occupied by persons whose housing needs should be the responsibility of the housing authorities (Report of the Expert Group on Mental Health Policy 2006). Tedstone Doherty et al. (Reference Tsemberis2007) evaluated Irish community residential mental health services and found that while residents generally stated they were happy living there, the internal environment was not ideal, the climate and culture reflected ‘mini-institutions’ rather than a home, with constricting rules and regulations and an excess of care in some cases. Further, this study highlighted the lack of independent housing as a significant constraint on the effective management of community residential facilities and consequently some were operating as a ‘home for life’ or ‘home for the homeless’.
In 2009, the Health Service Executive (HSE) published ‘The Efficiency and Effectiveness of Long-Stay Residential Care for Adults within the Mental Health Services’ (HSE, 2009). This report reviewed 4 709 long stay beds in the adult mental health services and found that about one-quarter of those in community residences were inappropriately placed and almost two-thirds of those in acute and long stay hospital beds would be more appropriately placed in community residences. The report concluded that the HSE should withdraw from the management of low and medium supported accommodations. Vision for Change (2006) recommended a limited number of high support residences be provided by the mental health services to meet the accommodation needs of those with enduring problems. For most, it was envisaged that housing requirement would be provided by the local housing authorities with appropriate support from the mental health community-based teams.
Other countries have attempted to measure the housing need of in-patients. In an analysis of 730 acute admissions in the United Kingdom, a quarter could have been supported effectively in the community in specialist accommodation if available and many patients might have benefited from alternative, community-based services (Bartlett et al. Reference Bartlett, Holloway, Evans, Ownes and Harrison2001). In a subsequent study by Commander & Rooprai (Reference Commander and Rooprai2008), almost 20% of acute psychiatric beds were occupied by patients for >6 months. Further, an average of 40% of patients on acute admission wards had accommodation needs at any one time (Corbluth, Reference Corbluth2011). In Canada, Durbin et al. (Reference Durbin, Cochrane, Goering and Macfarlane2001) reported 60% of in-patients could live independently with appropriate support.
The challenges are well highlighted in the rehabilitation literature. In the United Kingdom, Macpherson et al. (Reference Macpherson, Shepherd and Edwards2004) concluded that mental health services present a picture of extensive geographical variation, with typically fragmented and poorly organised systems for allocating special needs housing and a stark lack of awareness of unmet need within the system. Six years later, a report by the Royal Collage of Psychiatrists in Reference Srebnik, Livingston, Gordon and King2010 noted that housing is one of the main concerns of inpatients. Corbluth (Reference Corbluth2011) points out that the consequences of delayed discharges are vast and include institutionalisation, loss of coping skills, damage to or loss of relationships, reduced staff morale and retention rates, delays in admitting other at-risk people, and the premature discharge of others. In the United States, the programme directors of state psychiatric hospitals highlighted the importance of ensuring links with the community in order to seamlessly integrate individuals back into community settings (Parks & Radke, Reference Parks and Radke2014). The financial implications of delayed discharge are also large. Corbluth (Reference Corbluth2011) estimated that the cost of an inpatient psychiatric bed was £300/night and delayed discharges were estimated to cost the Department of Health up to £19 million each year. In 2012 in Ireland, the cost of an in-patient bed was estimated at €138 700/annum or €380/day (HSE casemix).
The ‘housing crisis’ is a topic of much political debate at present and the lack of data in Ireland on the housing needs of patients with mental illness has implications for service planning. Unless we know basic data like prevalence and type of housing needs, we cannot develop service responses like voluntary housing agencies, dedicated local authority rental options or specialised tenancy support services. With this in mind, we studied the prevalence of accommodation related needs among inpatients in the acute psychiatric unit in Tallaght Hospital.
Methods
The approved psychiatric unit in Tallaght Hospital provides acute admission beds for the Dublin West and Dublin South West catchment area. At the time of this study, it served a population of 273 419 making it the second largest psychiatric unit and hospital catchment population in the country (Daly & Walsh, Reference Daly and Walsh2013). The unit has 52 beds in total across three wards: Rowan ward (23 male beds), Cedar ward (23 female beds) and Aspen ward (a mixed high observation ward with six beds). In 2012 there were 633 admissions and 628 discharges and a total of 38 patients (6.1%) stayed for >3 months (Daly & Walsh, Reference Daly and Walsh2013).
This study set out to determine the prevalence of housing need in individuals on the approved acute psychiatric unit in Tallaght Hospital. It did not establish the cause or type of housing need nor did it illicit the views of patients and carers or develop plans to meet the housing needs. It was concerned only with the numerical data on housing need over a 1-year period. The data was collected weekly from 1 October 2012 for 52 consecutive weeks. For reporting purposes, in the graphs below, each group of 4 weeks was averaged, resulting in 13 points to show the trends over the 52-week period. Nursing managers and/or key nurses who were ‘on the ground’ in each ward and knew the patients well, were asked the following six questions:
-
1. How many inpatients are on the ward?
-
2. How many have accommodation related needs?
-
3. How many ‘long stay’/‘delayed discharge’/‘non-acute’ inpatients are on the ward?
(Delayed discharge was defined as clinically fit for discharge as determined by the patients treating team and discharge not imminent within the next 48 hours. A guiding question was: ‘Would this patient be discharged if suitable accommodation and supports were available in the community?’).
-
4. How many of these delayed discharge patients have accommodation related needs?
-
5. How many inpatients were discharged to the homeless services in the past week?
-
6. How many inpatients on the ward are currently waiting for a nursing home placement?
Results
Inpatients and accommodation needs
The total number of inpatients ranged between 41 and 52 each week, with an average of 47 over the 12-month period. In-patients with accommodation needs ranged between 12 and 25 at any one time and 18 on average. A total of 38% of in-patients had accommodation needs at any one time over the 12 months of the study (Fig. 1).
Fig. 1 Total numbers of inpatients, inpatients with accommodation needs, long stay inpatients with accommodation needs and inpatients waiting for a nursing home placement. The 13 points represent the average numbers for each 4-week block (4-week block×13=52 weeks).
Long stay/delayed discharge in-patients with accommodation needs
Figure 2 shows the number of long stay patients in the approved centre and the proportion that had accommodation needs. Most delayed discharge patients had accommodation needs (98%). Further, this patient group accounted for 28% of all inpatients and for 72% of all inpatients with accommodation related needs. Data was also collected in relation to inpatients waiting for nursing home placement. On average three individuals (6%) were waiting for a nursing home placement at any one time. Figure 1 shows these trends over the study period.
Fig. 2 Numbers of long stay inpatients and long stay inpatients with accommodation needs.
Inpatients discharged to homeless services
Data relating to the frequency of discharge to homeless services was also sought. A total of 39 patients (male=36, female=3) were discharged to the homeless services over the course of the study. The large gender difference resulted in more males discharged to homeless services at a ratio of 12:1. In other words, the average frequency of discharges to homeless services was every 10 days for a male and every 4 months for a female. On average, an individual was discharged to the homeless services every 9 to 10 days.
Discussion
Over one-third of inpatients had accommodation needs over the course of our 1 year study. From reviewing the literature, there is no similar Irish data by which to compare these findings of housing need among inpatients in approved centres. Current inpatient data systems like Mental Health Information System, Hospital In-Patient Enquiry system and National Psychiatric Inpatient Reporting System do not gather data on housing need. Similarly, annual reports from the Mental Health Commission and Health Research Board do not publish this information either.
Housing need data is collected on a periodic basis by the Department of the Environment and Local Government. The purpose of this is to plan to meet housing needs nationally and is based on examining the Local Authority waiting lists at a moment in time. The 2013 snapshot was taken on 7 May 2013. Examining the figures showed that 1034 or 1% of applicant households nationally, sought social housing for mental health reasons. Specifically, South Dublin County Council (SDCC) had five applications on mental health grounds and Dublin City Council had 16 (The Housing Agency, 2013). These figures are vastly under representative of the actual need. Stigma and reluctance to disclose need, difficulty getting onto housing lists, individuals not understanding the purpose of disclosing, penalising disclosure as proof of mental health difficulty are all factors in the underrepresentation of mental health difficulties on housing need databases. Similar difficulties are noted for people with other disabilities (Browne, Reference Carling2007). In many ways, patients with mental health problems and housing needs are ‘invisible’ to local authorities and housing providers due to this lack of data. Similarly, the ‘Happy Living Here’ study interviewed 138 community residents across three Mental Health Services in Ireland and found that many residents were not registered with on their local housing list. Mental health services could help highlight the housing needs of their patients by regularly publishing such figures. This could form the basis of discussion between the mental health service and housing providers aimed at solving some of the problems in our health service. Each service could also develop a database of housing need and establish which patients are not on housing waiting lists. Developing links with local authorities and other housing providers like Approved Housing Bodies, and familiarity with the local Housing Strategy for People with Disabilities would also help. Knowledge of relevant guidelines like National Guidelines for the Assessment and Allocation Process for Housing Provision for People with a Disability [Department of the Environment, Community and Local Government (DECLG), 2014] and the developments with the local authority hosted Housing Steering Group for People with a Disability could also be beneficial. A ‘person by person’ approach can ensure subjective preferences for housing and support drive care planning.
Accessing accommodation suitable to the needs of those identified as not capable of independent living is going to be a challenge as national mental health plans aim to discontinue low and medium support residences. Similarly, accessing suitable accommodation (social and rented) for those who can live independently will also be very challenging as no systems are in place with local authorities, the private rented sector or supported housing bodies to facilitate timely discharges to suitable, affordable, desirable and secure accommodation. Clearly, assessing inpatient’s placement, housing and support needs are the responsibility of the mental health service. Creating a database of this assessed need for planning purposes with local authorities and supported housing bodies should also be a key task of the mental health service. Otherwise suitable accommodation and support will only be accessed on an individual case by case basis and a range of alternatives will be slower to be developed.
Discharge to homeless services was male dominated. There is very limited emergency homeless accommodation available outside of city centre hostels. In our study, relatively low numbers of females were discharged to homeless services which could reflect clinician bias and a more paternalistic approach to female patients. Kennedy (Reference Kennedy1985) coined the term ‘hidden homeless’ to describe thousands homeless women in Dublin who try not to enter the official homeless services for reasons of self respect, safety, security, stigma, fear of getting stuck in the system, and loss of control and influence over their lives. Further, the number of homeless discharges in our study is not representative of the all the inpatients who were concerned about homelessness. For example, the homeless inpatients who recover sufficiently to be able to stay with friends or family members following discharge were not counted. Recent figures available from SDCC homeless show that very few individuals state ‘mental illness’ as the reason for presenting as homeless. Consequently, mental health issues remain out of sight and are not a priority consideration for planning.
More recently, national housing policies have provided an unprecedented opportunity for joint working between the local authorities and the mental health services. In 2011, the DECLG produced the ‘National Housing Strategy for People with a Disability 2011-2016’ which discussed the specific housing issues relating to people with a mental health disability (DECLG, 2011). This was the first time a housing strategy document specifically addressed mental health. While the major topic discussed was the transition plan for 1180 residential places from the HSE medium and low support accommodation to the local authorities, the National Housing Strategy also provided a dedicated ‘Individual Assessment of Need Protocol-Mental Health’. This protocol outlines the process for applications and the responsibilities of both agencies. The principle underlying the protocol is that housing is the responsibility of local authorities and the mental health services should work in liaison with the local authorities. With this in mind, Dublin South Central Mental Health Service and SDCC have developed our own liaison protocol (available on request).
Perhaps we can learn from other countries in implementing a change in our approach to the housing needs of our patients. In North America the emphasis is on autonomy (choice and preference), separation of housing and services, individualised, independent secure accommodation with appropriate wrap around supports and services. The housing models most referenced are the evidenced-based models; ‘permanent supportive housing’ (Substance Abuse and Mental Health Services Administration, Reference Tanzman2010) and ‘housing first’ (Tsemberis, 2010). The values and principles of these models exemplify the recovery approach sought by the Irish mental health services. These models are based on the consistent findings of consumer preference studies (Tanzman, Reference Tedstone Doherty, Walsh and Moran1993), the supported housing approach (Carling, Reference Clifford, Charman, Webb, Craig and Cowan1993) and the positive outcomes associated with choice over housing. Perceived choice plays a major role in housing stability, housing satisfaction, psychological well being and decreasing psychiatric symptoms (Srebnik et al. 1995; Greenwood et al. Reference Greenwood, Schaefer-McDaniel, Winkel and Tsemberis2005).
Clearly this study has some methodological limitations such as staff bias and the use of non-standardized questions in assessing housing need. The collection of data relied on consultation with nurse managers and/or key nurses. It is common for interview schedules in studies in this area to include these key personnel (Clifford et al. Reference Brooke1991; Beck et al. Reference Beck, Croudace, Singh and Harrison1997; Farragher et al. Reference Farragher, Fahy, Carey and Ownes1999; Keogh et al. Reference Keogh, Roche and Walsh1999; Tedstone Doherty et al. Reference Tsemberis2007). However, the limitations of relying on the opinion of just one particular group of providers, limits the findings to the subjective judgments of those providers while missing a more comprehensive collection of important opinions, namely, service users, carers and other members of the multidisciplinary team.
This study employed a broad interpretation of housing need. It is commonly accepted that local authorities distinguish between ‘housing want’ (applicants for housing) and ‘housing need’ (applicants who are successfully placed on the waiting list). Most critical commentators accept that an expressed need (e.g. waiting list numbers) is just one way of defining need (Bradshaw, Reference Bradshaw1972). Such a narrow definition of housing need does not reflect the complexity of housing related needs evident for individual patients or for the service. Housing need is not an easy concept to define, assess or measure. This is acknowledged among housing professionals like Brooke (Reference Browne2014) who describes it as a slippery concept often vaguely defined because to a significant extent it is a subjective evaluation, which does not lend itself easily to the objective prioritisation required for allocation.
Conclusion
The main finding of this study, as reported by key nursing personnel, was that 38% of acute in-patients and 98% of non-acute in-patients had housing related needs. At the moment, housing needs among patients attending the mental health services is underreported patients as mental health services do not routinely collect or publish the housing needs of this population. To be concerned about something as fundamental as secure shelter while acutely unwell and distressed is a significant stressor. It is also significant for staff who may feel powerless to address housing needs, and to the service which carries the cost of delayed discharges and resulting negative impact on availability of acute care beds. Housing is a key component of the care plan for many patients and future planning for local mental health services will depend on good co-working with local authorities and the availability of affordable, suitable and safe accommodation options. Clear working alliances are crucial in this regard. Data collection regarding the levels of housing and support needs is fundamental to planning a functioning mental health service that is responsive to the needs of the individuals we treat.
Acknowledgements
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committee on human experimentation with the Helsinki Declaration of 1975, as revised in 2008. The authors assert that ethical approval for publication of this audit was not required by their local REC.
Financial Support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Conflicts of Interest
None.
