The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 60:S152-S159 (2005)
© 2005 The Gerontological Society of America
The Causes of Homelessness in Later Life: Findings From a 3-Nation Study
Maureen Crane1,,
Kathleen Byrne2,
Ruby Fu1,
Bryan Lipmann3,
Frances Mirabelli3,
Alice Rota-Bartelink3,
Maureen Ryan4,
Robert Shea5,
Hope Watt4 and
Anthony M. Warnes1
1 Sheffield Institute for Studies on Ageing, University of Sheffield, UK.
2 Elders Living at Home Program, Boston Medical Center, Massachusetts.
3 Wintringham, Flemington, Victoria, Australia.
4 Committee to End Elder Homelessness, Boston, Massachusetts.
5 Pine Manor College, Chestnut Hill, Massachusetts.
Address correspondence to Maureen Crane, Sheffield Institute for Studies on Ageing, University of Sheffield, Elmfield, Northumberland Road, Sheffield, South Yorkshire S10 2TU, U.K. E-mail: m.a.crane{at}sheffield.ac.uk
 |
Abstract
|
|---|
Objectives. This article presents findings from a study of the causes of homelessness among newly homeless older people in selected urban areas of the United States, England, and Australia.
Methods. Interviews were conducted in each country with 
122 older people who had become homeless during the last 2 years. Information was also collected from the subjects' key workers about the circumstances and problems that contributed to homelessness.
Results. Two-thirds of the subjects had never been homeless before. Antecedent causes were the accommodation was sold or needed repair, rent arrears, death of a close relative, relationship breakdown, and disputes with other tenants and neighbors. Contributory factors were physical and mental health problems, alcohol abuse, and gambling problems.
Discussion. Most subjects became homeless through a combination of personal problems and incapacities, welfare policy gaps, and service delivery deficiencies. Whereas there are nation-specific variations, across the three countries, the principal causes and their interactions are similar.
Homelessness is an intractable problem in many affluent countries and affects people of all ages, although much research and service provision have concentrated on young adults. Since the late 1980s, a few studies have focused on older homeless people and have found that many become homeless for the first time in later life, raising questions about why this happens, the unmet support needs of older people, and how their homelessness can be prevented. Recently, a few specialist services have been developed to meet their needs (Warnes & Crane, 2000
). Cohen and Sokolovsky<--?1-->(1989) argued that many homeless people aged 50–59 years have chronic health problems and disabilities normally associated with old age and are unlikely to return to work. The age group may be particularly disadvantaged as many welfare services are available to people only when they reach the officially recognized thresholds of old age. This article reports a study of the causes of homelessness among newly homeless older people in Boston, Massachusetts, four English cities, and Melbourne, Australia. It begins with brief descriptions of the study sites and the local policies, services, and homeless populations.
|
POLICIES, SERVICES, AND HOMELESSNESS IN BOSTON, MASSACHUSETTS
|
|---|
Since the mid-1980s, the U.S. federal government and many city governments have promoted policies and service developments to prevent and alleviate homelessness. The Boston Housing Authority owns 14,000 units of public housing and provides rental assistance to around 11,000 low-income households in the private sector through the Section 8 Housing Choice Voucher Program and Massachusetts' rental voucher program. Social Security benefits are available for people who are disabled or have retired. Under the Older Americans Act 1965, the Area Agencies on Aging made grants available to the states for community support programs. The Boston Commission on Affairs of the Elderly manages the federal program in the city and through Aging Service Access Points serves 6,500 older people through case management, home support services, and nutrition programs (Boston Partnership for Older Adults, 2003).
The national stock of low-cost private rental and public housing has declined since the 1980s, with reduced new-build and maintenance of federally subsidized housing, the demolition or redevelopment of low-cost housing, changes in the federal tax structure, rising interest rates, and fewer incentives for private investors to create new low-cost housing (Koegel, Burnam, & Baumohl, 1996). Boston's housing market has also been affected by its growing economy. More than 42,000 jobs were created in 1999, and rents increased by 47% during 1995–1999, with the result that >50,000 of the city's residents spent over half their income on housing (Menino, 2000). There are 14,000 people on Boston's public housing waiting list, of whom 1,000 are aged 60 or older (Boston Partnership for Older Adults, 2003). The mayor launched a 3-year housing strategy, "Leading the Way," in October 2000 to expand and preserve the supply of housing in Boston and a second one in 2004. Since 2003, the Boston Housing Authority has intermittently refused new applications for its public housing and rental assistance program. The housing market changes have coincided with a rise in demand for low-cost housing from poor people. In 2003, 35.9 million people (12.5%) nationally were in official poverty, an increase from 34.6 million in 2002 (DeNavas-Walt, Proctor, & Mills, 2004). In Boston, >18% aged 65 or older have incomes below the poverty line, an increase of 15% in 10 years (Boston Partnership for Older Adults, 2003).
Homelessness is an increasing problem in Massachusetts. In the city of Boston, there were 6,210 homeless adults and children on the night of the homeless census in December 2002, an increase of 41% since 1992 (Anderson et al., 2003). The number of older homeless people in Massachusetts has also increased. A local census in 2000 enumerated 1,228 homeless people aged 50 or older: Six hundred ten were in the city of Boston, a 39% increase since 1993 (Boston Partnership for Older Adults, 2003). Among 15,609 people admitted to emergency shelters in Massachusetts in 2003, 43% were aged 45 or older (compared with 37% in 2001 and 28% in 1999) (Meschede et al., 2004).
|
POLICIES, SERVICES, AND HOMELESSNESS IN ENGLAND
|
|---|
In England, local authority housing and social services departments have a statutory duty to provide housing and personal support services for vulnerable people through the Housing Act 1996 and the National Health Service and Community Care Act 1990. Until the late 1980s, local authorities were the main providers of social (not-for-profit) housing. Incorporated Housing Associations have since become the sole suppliers of new social housing and have taken ownership of much public housing. Housing subsidies are available for low-income households and social security benefits for the unemployed, disabled, or retired. The National Health Service provides free health care to people in need. The Homelessness Act 2002 placed a duty on local authorities to develop strategies for the prevention and alleviation of homelessness. Schemes to help people sustain tenancies have since developed rapidly, including housing advice services, debt management schemes, and tenancy support teams.
Changes in housing policy, such as the "right to buy" for local authority tenants and a reduction in new-build, have led to a shortage of affordable rented housing, which Hawtin and Kettle (2000) estimated at a half million. Access to social housing is now regulated by need. Local authorities are required to obtain housing for people who are "unintentionally" homeless or threatened with homelessness, in specified "priority need" categories, and have a "local connection." The "priority needs" include being "elderly" (customarily 60 years or older for women and 65 years or older for men) and having serious health problems. People in their 50s and men in their early 60s are generally excluded.
Possession orders and evictions from social housing more than doubled from the mid-1990s to almost 25,000 in 2000 (Warnes, Crane, Whitehead, & Fu, 2003). Most such actions are for rent arrears with a few for antisocial behavior. The rise reflects the scarcity of support for people who cannot manage independently, the increased pressure on social housing providers to reduce rent arrears, and endemic problems with the administration of housing benefit (a state rent-subsidy paid to landlords). One-third of older people do not claim the social security income benefits to which they are entitled and one-tenth the housing subsidies. Some are unaware of their entitlement; others find the application procedure too complex.
There are no comprehensive statistics of homeless people in England. In 2002, 195,590 households were accepted by local authority housing departments as homeless compared with 165,390 in 1997 (Office of the Deputy Prime Minister, 2003). Of the former, 4,450 were accepted as in "priority need" of rehousing on the grounds of old age. Many others sleep on the streets or stay in homeless hostels, and among them 15–20% are aged 50 or older (Warnes et al., 2003). In London, 700 people of this age were in hostels on one night in 2000, and 527 slept on the streets in 2003 (Broadway, 2004; Crane & Warnes, 2001a).
|
POLICIES, SERVICES, AND HOMELESSNESS IN VICTORIA, AUSTRALIA
|
|---|
In Australia, the state governments are responsible for public housing and for health care services. In Victoria, the Office of Housing in 2001 managed 74,773 social housing properties for low-income households (Department of Human Services, 2002a). The housing market in Victoria is, however, dominated by private owners (73% of dwellings). A universal health insurance scheme, Medicare, provides free public hospital treatment and free or subsidized community-based treatment. The social security system provides a "safety net" for people in financial need who are unable to work or have retired. Households in the private-rented sector whose rents exceed 20% of their income are entitled to Commonwealth Rent Assistance. Home- and community-based care services have grown rapidly since the 1980s. The Supported Accommodation Assistance Program (SAAP) has, since 1985, funded local governments and not-for-profit agencies to develop support services and transitional housing for homeless people and those at risk (details: http://www.facs.gov.au). By 2000–2001, 8,580 people aged 50 or older used SAAP services (2,840 in Victoria). They were 9% of all the clients (Lai, 2003).
There has been a shift away from the development of public housing and an increase in the payment of "rent assistance" to private renters. Reduced commonwealth government funding has meant that Victoria has been unable to develop or upgrade its housing. The waiting list increased from 41,000 in 2000 to 44,500 in 2001, with 12% aged 65 or older (Department of Human Services, 2002b; Ronaldson, 1999). Gentrification and "up-market" housing developments in Melbourne have led to the closure of rooming houses, the conversion of private hotels for tourist accommodation, and spiraling rents. Many low-income older people and single men who used to occupy rooming houses have been forced to move. Approximately 6,000 people aged 60 or older in Victoria pay rents that exceed 30% of their income (Ronaldson, 1999). The number of homeless people in Australia decreased from 105,304 in 1996 to 99,000 in 2001, but in Victoria, the number increased by 14% to 20,305. Of these, 9% were aged 45–54 years, and 11% were older than 55 (Chamberlain & Mackenzie, 2004).
|
CURRENT UNDERSTANDING OF THE CAUSES OF HOMELESSNESS
|
|---|
Theoretical contributions and empirical research findings have supported two broad explanations of homelessness: one associated with structural economic and policy conditions, such as poverty, unemployment, and a shortage of affordable rented housing, and the other featuring personal incapacity, vulnerability, and behavior. Many theorists perceive homelessness as the result of interacting structural and individual factors and occurring when people experience negative or major life events and lack the ability to cope or the resources to compete in the housing and employment markets (Lee, Price-Spratlen, & Kanan, 2003; Main, 1998; Rossi, 1989; Sullivan, Burnam, & Koegel, 2000). Homelessness has also been associated with deficient or inaccessible services (Sosin, 2003). Elliott and Krivo (1991) showed that in U.S. metropolitan areas, relatively high expenditure on residential mental health care was associated with lower rates of homelessness. Less attention has been paid to the contribution of service delivery factors to homelessness.
Empirical studies have identified high rates of mental illness, substance abuse, and disruptive childhood experiences among homeless people (Caton et al., 2000; Herman, Susser, Struening, & Link, 1997; Koegel, Melamid, & Burnam, 1995). Susser, Moore, and Link (1993) developed a model of causal pathways that incorporated personal risk factors at different stages of the life course, the most influential in later life being deficient economic and social resources, early-acquired personal characteristics, and poor health. Cohen (1999) proposed that the risk of homelessness accumulates over time and that the event occurs when several risk factors co-present. The most influential risks during middle and later adulthood are imprisonment, substance abuse, mental and physical health problems, victimization, lack of family and social networks, and low income. Among people aged 50–59 years, enforced unemployment, income decline, and the age group's few entitlements to social security benefits and support services were also factors.
The transitions that commonly precede homelessness in later life are widowhood, the death of a parent, marital breakdown or household disputes, stopping work, the loss of accommodation tied to a job, evictions for rent arrears, and the onset or increased severity of a mental illness (Cohen & Sokolovsky, 1989; Crane, 1999; Crane & Warnes, 2001b; Keigher, 1992; Wilson, 1995). There has, however, been little rigorous or longitudinal research into the causes of homelessness among older people. Many people experience changes in later life that create vulnerability, such as widowhood and retirement, but do not become homeless. This raises the question of why some people who have been conventionally housed for decades become homeless for the first time in old age. Which attributes, states, and events are implicated, and why does the welfare "safety net" not prevent the problem? Repeatedly identified risk factors for homelessness, such as disturbed childhoods, are likely to have less influence on the entry into homelessness in old age than in young adulthood.
Building on the debate about the interactions of structural and individual factors, this study applied two theoretical conceptions. It was hypothesized that many entries into homelessness are associated with structural or welfare policy factors, with personal problems and behavior and with deficiencies in the delivery of health and welfare services. The operational form of a structural factor is a "policy gap," defined as an entitlement that is unavailable in the country of interest but available in one or more others. It may refer to a state-funded or subsidized benefit or service that is lacking, to the restricted resources made available to provide the entitlement or service, or to a condition or restriction upon an entitlement.
A "service deficiency" was defined as a failure to deliver a benefit or service to a client who is entitled and in contact with the provider agency. Both previous research by the British authors, including a 2-year longitudinal study of the outcomes of the resettlement of older homeless people (Crane & Warnes, 2002), and the innumerable contacts with homeless people of the Boston and Australian authors had demonstrated that service delivery failings were frequently implicated in pathways into homelessness. The most apparent service delivery problem is a failure of a responsible agency to deliver a benefit or service to a client who has an entitlement and has requested the service. Common examples in the United Kingdom are the failure of local authority housing departments to award an applicant with low income and assets a "housing benefit" (a social security benefit that meets all or a percentage of rent charges). Cases arise through both inefficiency and the client's failure to complete and return the application and renewal forms. Many cases can be described as a "service deficiency," whereas some are more clearly associated with the client's behavior.
The second causal concept was that in many cases, homelessness results from a combination of predisposing or risk factors (e.g., a housing shortage or an individual's mental health problem) and antecedent causes or "triggers" (e.g., withdrawal of a social security benefit or bereavement). Apart from natural disasters or armed conflicts, few "events" are the sole cause of homelessness. The triggers or precipitants such as widowhood or redundancy destabilize a vulnerable person. When combined with their poverty, addiction problems, mental illness, or poor living skills, the disadvantaged person who becomes homeless lacks the resources, skills, or support to prevent the negative event, leading to ramifying consequences that culminate in homelessness. Many of the likely risk factors and precipitating events can be specified a priori; the problem is to determine their relative prevalence and independent and interactive effects.
|
METHODS
|
|---|
The aims of the three-nation study were (a) to increase understanding of the causes of homelessness among older people and (b) to contribute to prevention practice. The rationale was that by studying in contrasting welfare and philanthropic regimes a relatively homogeneous category of homeless incidence, that is, recent cases among late-middle-aged and older people, valuable insights into the relative contributions of the policy, service, and personal factors would be obtained. Evidence of unusually prevalent pathways into homelessness in one country might be explained by its distinctive welfare policies and the presence or absence of services or alternatively by atypical features of its social pathologies.
Population of Interest
The study focused on newly homeless older people purposively to gather detailed and reliable information about the prior and contextual circumstances. The inclusion criteria were that the person became homeless during the previous 2 years and was aged 50 years or older at the time. To have included people who had been homeless for several years would have reduced the quality of the data, as a subject's recall of events several years before would be less reliable. The agreed definition of homelessness for the three countries was (a) sleeping on the streets or in temporary accommodation such as shelters or hostels, (b) being without accommodation following eviction or discharge from prison or hospital, and (c) living temporarily with relatives or friends because of lack of accommodation. The latter applied if the stay had not exceeded 6 months and the person did not pay rent and was required to leave. People who had previously been homeless were included if they had been housed for at least 12 months prior to the current episode of homelessness. The target was 125 in each country.
Instruments
Accounts of the passage into homelessness were collected through a semistructured questionnaire completed with the subjects during a face-to-face interview and a self-completion questionnaire by the "key worker." The subject questionnaire collected the circumstances prior to homelessness, including housing during the previous 3 years, previous homelessness, employment history, income, health and addiction problems, and contacts with family, friends, and formal services. The respondents were also asked to rate whether specified factors were implicated in becoming homeless "not at all," "a little," or "a lot." When appropriate, a following open-ended question sought elaboration. The specified factors were bereavement, relationship breakdown, work-related problems, financial difficulties, physical health, mental health, alcohol, drug, and gambling problems, and criminality.
Shelters or other projects assign key workers to assess a client's problems and to advise and support them. All subjects in Boston and Melbourne and all except seven in England had key workers. The key worker self-completion questionnaire focused on their understanding of the events and states that led to the subject's homelessness. It also had direct factor assessment and open-ended questions. Both instruments were developed collaboratively by the partners. To maximize validity across the three countries, close attention to the underlying concepts and the terminology was required. Consensus taxonomies of types of housing, home support, and health care services were developed. The instruments were piloted in each country and revised twice. The final schedules and the coding scheme were identical in all countries apart from country-specific categories for ethnicity.
Sampling and Interviewing
No study area had a sample frame of all newly homeless older people. Data on the number, age, gender, and ethnicity of older homeless people in London indicated the population's characteristics (Crane & Warnes, 2001a). Similar data were available on older people admitted to Boston's shelters (Meschede et al., 2003). The samples were recruited through referrals to the research team from service providers and represent a large (but precisely unknown) percentage of newly homeless older people who were in contact with service providers during the study period. In Boston and Melbourne, a majority were clients of the organizations conducting the research. To increase the representation of women in Melbourne and England, during the final months, they were selectively recruited. The interviews were conducted between July 2001 and August 2003.
Data Recording and Derived Scores
Each partner entered the precoded responses into a database, and the open-ended response categories were agreed collaboratively. Data quality-control procedures included blind checks of the data coding and keying. The three-country database has 290 variables and 377 cases. The scores for individual factors were aggregated into four constructs: personal factors, service deficiencies, policy gaps, and "unattributable or other" factors. The scores reflect the semantically differentiated reported influence of the factors, that is, 25 for a "little" contribution, 50 when identified but "unweighted," and 75 when rated as having "a lot" to do with becoming homeless. If it was impossible to assign a declared influence to the policy, service, or personal groups, it was scored to "unknown." As some accounts were sketchy, a minimum score at half the average was imposed by increasing the "unknown" score. This was required for 62 (16%) of the subjects' accounts and 47 (12%) of the key workers'.
|
RESULTS
|
|---|
Profiles of the Samples
The achieved samples comprise 122 subjects in Boston, 131 in England, and 124 in Melbourne. Most were men and stayed in hostels or shelters, but 5% in Boston, 9% in Melbourne, and 42% in England had slept on the streets since being homeless. Seventy-seven percent became homeless between the ages of 50 and 64 years, and only 9% were aged 70 years or older (Table 1). The gender and age distributions in England replicate those of London's older street and hostel homeless populations (Crane & Warnes, 2001a). The Boston sample overrepresents women when compared with those aged 55 years or older admitted to Massachusetts shelters in 2003 (37% vs 22%), but the age and marital status distributions are similar (Meschede et al., 2003).
As to ethnicity, 47% in Boston were Black (mainly African American), 45% White, and 8% Latino or Asian; in England, 89% were White British or Irish; and in Melbourne, 62% were White Australian born and most others (37%) born overseas. The Australian composition is consistent with the ethnic diversity of the national older population and reflects post-1945 immigration to the country from Europe (Australian Bureau of Statistics, 2002). Two thirds of the subjects had worked for most of their adult lives (see Table 1). The lower percentages in Boston and Melbourne reflect the relatively high number of women, many of whom stopped work in early adulthood to raise children. In Boston, 23% were employed when they became homeless. For the majority, homelessness was a new experience: Across the three countries, 68% had never been homeless before, including 79% in Boston.
Prior to becoming homeless, just less than one fifth in each country had been owner occupiers. Around one half in England had rented from nonprofit housing providers, but in Melbourne and Boston, the respondents were more likely to have rented from private landlords (Table 2). One half had been living alone, and the household composition varied little by country. Those in their 50s (27%) were more likely than the older subjects (13%) to have been living with a spouse or cohabiting partner (
2 = 10.7, df = 1, p <.001). In all three countries, around one-fifth had lived at their last address for at least 10 years. Most subjects had weak informal and formal support networks. Although most had relatives, 30% had had no contact with them for years, and another 19% saw a relative or close friend less than once a month. Only 30% received financial assistance or help with household tasks from informal supporters, with men (26%) less likely than women (42%) to have received this help (2 = 8.3, df = 1, p <.004). As for formal home support or social services, 11% in England, 14% in Boston, and 48% in Melbourne had received this help. The subjects who had been homeless before (34%) were more likely than the others (20%) to have been receiving formal support (2 = 7.5, df = 1, p <.006).
Antecedent Causes or Triggers
The subjects described many events and states that they believed precipitated their entry into homelessness (Table 3). One fifth had to leave because the accommodation was sold or was to be converted or needed repair (28% in Melbourne). A few Boston and English subjects and 16% in Melbourne left because they had problems accessing or maintaining their housing when their health deteriorated. Twenty-seven percent said that difficulty with paying rent or mortgage repayments triggered homelessness. In England and Melbourne, a common sequence was that the subject accumulated rent arrears but remained in the housing until they were evicted, whereas in Boston, several gave up the tenancy before arrears accrued.
The death of a relative or close friend precipitated homelessness for one-tenth of the subjects. Some abandoned the accommodation because they found it too distressing to remain. Others had been living with a parent or spouse who was responsible for the household and financial tasks but could not manage when he or she died and were evicted for rent arrears. The breakdown of a marital or cohabiting relationship triggered homelessness for one-fifth of the subjects. Some immediately became homeless, whereas others moved but did not settle and left after a few months. Disputes with landlords, co-tenants, relatives, and neighbors triggered homelessness for 23% and was most frequently reported in England and Melbourne. Some in private-rented accommodation complained that other tenants were noisy or difficult and provoked their departure. In England, nine heavy drinkers admitted that they and their friends were noisy and disruptive, which led to complaints from neighbors and eviction.
Predisposing or Contributory Factors
The subjects nominated various problems that they believed contributed to them becoming homeless but were not the antecedent cause, and the key workers broadly corroborated their accounts. Seventy-seven percent reported physical health problems, and 28% believed that these problems were implicated. Sixty-four percent reported depression or other mental health problems, and 23% said that they contributed to homelessness. Several stopped work through ill health, which led to financial problems, whereas nearly one tenth said that health problems contributed to family and marital breakdown or affected their ability to cope at home. Most with physical illnesses had treatment prior to becoming homeless, but among the 242 who reported mental health problems, only 45% received treatment and only 21% from mental health specialists. Most who had not had treatment said that they had not asked for help.
Thirty-two percent of the subjects described heavy drinking or alcohol problems, with a significant gender differential (men 38%, women 14%; 2 = 18.6, df = 1, p <.001). In all study areas, the key workers reported higher rates of known or suspected alcohol problems (44% in England, 50% in Boston, 65% in Melbourne). Twenty-one percent of the subjects believed that alcohol problems contributed to them becoming homeless, as a result of either marital breakdown or eviction for rent arrears. Illegal drug use was reported by 9% and gambling problems by 15%. The latter were exceptional in Melbourne, being reported by 39% of the subjects, of whom 23% said that gambling problems had been an instrumental factor in their homelessness, mainly through irresponsible spending and rent arrears. Few with a gambling problem had sought help.
One in two subjects said that financial problems contributed to them becoming homeless. As described earlier, difficulties with paying rent or mortgage repayments precipitated homelessness for 27%. Most others said that financial problems had led to relationship problems and breakdown. Many associated financial difficulties with the end of a job, rent increases, or problems with social security benefit and housing subsidy payments. Twenty-five percent of the subjects (52% in Melbourne) reported poor money management skills and budgeting difficulties, and most of them had mental health or addiction problems.
National Variations in the Prevalence of the Reported Causes
The national and aggregate average scores for the 10 most prevalent contributory causes are shown in Table 4. Housing difficulties have not been included as present by definition. The rank order of the causes was similar in all countries and replicates the patterns reported by previous studies and by British local government homelessness statistics (Warnes et al., 2003). The highest factor coefficient of variation was for gambling problems and the lowest for mental health problems (Table 5<--?3-->). Drug, alcohol, physical health, and criminality problems had high variability, whereas financial, work-related, personal relationship, and bereavement problems had low variability. The exceptionally low physical health problems score in England is plausibly explained by the National Health Service and its dense network of primary care health centers. The respondents' accounts produced significantly low scores for alcohol problems in Boston (2 = 13.1, df = 2, p <.01) and for work-related problems in Melbourne (2 = 9.6, df = 2, p <.01).
Over the three countries, the aggregate score of the reported reasons for homelessness was 453.5 from the respondents and 198.2 from the key workers (Table 6). The adjustments for "unknown causes" had little effect on the mean scores, which increased to 458.2 and 199.3, respectively. The greater the age of the respondent, the slighter was their report of why they became homeless: Those aged 65 years or older generated a mean score (before adjustment for unknown) of 370, only 73% of that given by those aged 50–54 years. On the other hand, there was no relationship between the age of the subject and the fullness of the key workers' account.
Although the key workers' accounts were less fulsome than the respondents', the two produced broadly similar allocations among the personal, policy gap, service delivery, and unknown factors (see Table 6). Personal circumstances, events, and actions accounted for one-third of the respondents' scores and 27% of the key workers'. Both sets of reports described "unattributable factors" that accounted for one fourth or more of the aggregate score, but they disagreed on the relative importance of policy gaps and service delivery deficiencies. The subjects' ratings gave more weight to service defects than policy gaps, but the key workers the reverse. It is understandable that a lack of support may be seen by a subject as a service delivery failure but recognized by a key worker to be a policy or funding gap.
|
DISCUSSION
|
|---|
The majority of the subjects had reached later life without ever previously being homeless. Diverse pathways and multiple reasons were evinced, and, as hypothesized, most cases involved personal problems and incapacities, policy gaps, and service delivery defects. Of the contributing factors reported to us, around one fourth could not be confidently allocated to the three sets of factors and were deemed "unattributable." Some subjects lacked the skills or resources to cope with changes or stresses experienced at older ages, whereas policy gaps and changes meant that some services and resources were unavailable for people in need or they were intentionally excluded. In many cases, extant welfare services did not effectively respond to people who were vulnerable.
In all three countries, recent changes in housing markets and housing management practices were implicated in many of the transitions to homelessness. The diminished stock of affordable public and social (or subsidized) housing has produced long waiting lists and intentionally or inadvertently excluded many low-income older people. The situation of many of the Boston and Melbourne subjects living in privately rented housing had been made insecure by "up-market" housing developments. In England, comparable insecurities derived from assertive rent arrears management by social housing providers. Such housing market circumstances interacted with low income, the lack of financial reserves and social support, and poor money management skills to lead the subjects into financial difficulties, rent arrears, and homelessness.
Particular deficiencies of the welfare safety net were exposed by the many subjects in all three countries who reported untreated mental health problems or who were unable to manage everyday tasks after their main "carer" died. Community mental health services target people with severe disorders, and few welfare services are required to seek out those who are isolated and have unmet needs. In England, even the universal-access health services assume that people in need will ask for help. Similar problems have been reported in Australia, where few older people use SAAP (Lai, 2003). Homelessness therefore occurs because health and welfare services do not have the responsibility or resources to search for people with unmet treatment or support needs and weak informal support networks.
Relatively low variability was found by country among the causal factors that are intrinsic to the human condition (bereavement and mental health problems), but relatively high variability characterized the social pathologies that are culturally influenced and time specific (namely, gambling and drug problems), a finding that partially validates the scoring system. The most distinctive reports were about the role of gambling problems in Australia. Electronic gaming machines were legalized in Victoria in 1992; since then, gambling debts have proliferated. The findings corroborate other evidence of the social pathologies and increased homelessness that have resulted from the recent rapid growth of gambling in Australia (Antonetti & Horn, 2001; Productivity Commission, 1999). In such cases, homelessness occurs among people who are weak and prone to addictive behaviors when social control is lifted.
The limitations of this study include its focus on newly homeless older people known to service providers. Older people with chronic histories of homelessness and housing instability and those who sleep on the streets or stay with friends and are not in contact with agencies have different needs. Some avoid services, whereas some have mental health problems that reduce their capacity to seek help. The subjects in the three countries differed, with more women and more aged 60 or older in Boston and Melbourne, whereas a higher proportion of the English subjects had slept on the streets. Nonetheless, the results indicate that the reasons why older people become homeless are similar in the three study areas. Although the instruments sought a detailed description of the events and states that preceded homelessness, it was impossible to collect comprehensive retrospective information about the causes. The primary informant was the homeless person, and their accounts were subjective and selective. In many cases, estranged relationships were implicated, and rarely was it possible to interview others. Moreover, most subjects had but a partial comprehension of the role of policy and service delivery factors. The key workers' assessments supplemented and partially verified the subjects' accounts, but a few in England had limited knowledge of the subjects' circumstances before they became homeless.
There are still many questions to research. Even the most strongly associated risk factors such as relationship breakdowns or low income do not predict homelessness in the absence of supplementary and reinforcing problems. Given the substantial contribution of service deficiencies, it is probable that more can be done to anticipate and monitor vulnerability and to deploy targeted services to those at high risk of homelessness. The authors have presented the findings to and held workshop discussions with health, housing, and social service agency staff in England, from which preliminary recommendations for improved prevention practice have been developed (Crane, Fu, & Warnes, 2004).
|
Acknowledgments
|
|---|
The research was funded by the Commonwealth of Australia Government's Department of Family and Community Services, the U.K. Economic and Social Research Council (Grant R000239675) and The Leverhulme Trust, and in Boston through the general funds of the partner organizations.
The authors thank the many service provider organizations, homeless people, and the Advisory Group members who made the study possible.
|
Footnotes
|
|---|
Decision Editor: Charles F. Longino, Jr.
Received for publication April 7, 2004.
Accepted for publication December 17, 2004.
|
References
|
|---|
- Anderson, D., Fernandes, K., Green, C., Greene, J., Greenberg, E., Mays, C., et al. (2003). Homelessness in the city of Boston: Winter 2002–2003. Annual census report, December 9, 2002. Boston: City of Boston.
- Antonetti, E., Horn, M. (2001). Gambling the home away: A study of the impact of gambling on homelessness. Melbourne: Hanover Welfare Services, Victoria, Australia.
- Australian Bureau of Statistics. (2002). Population composition: Older overseas-born Australians. In Australian social trends 2002: Population. Canberra: Australian Bureau of Statistics.
- Boston Partnership for Older Adults. (2003). 100,000 voices on growing older in Boston. Boston: Boston Partnership for Older Adults.
- Broadway. (2004). Annual rough sleeping summary report for London 2003. London: Broadway.
- Caton, C., Hasin, D., Shrout, P., Opler, L., Hirshfield, S., Dominguez, B., et al. (2000). Risk factors for homelessness among indigent urban adults with no history of psychotic illness: A case-control study. American Journal of Public Health, 90,258-263.[Abstract/Free Full Text]
- Chamberlain, C., Mackenzie, D. (2004). Counting the homeless 2001: Victoria. Hawthorn: Institute for Social Research, Swinburne University, Victoria, Australia.
- Cohen, C. (1999). Aging and homelessness. The Gerontologist, 39,5-14.[Abstract]
- Cohen, C., Sokolovsky, J. (1989). Old men of the Bowery: Strategies for survival among the homeless. New York: Guilford Press.
- Crane, M. (1999). Understanding older homeless people: Their circumstances, problems and needs. Buckingham: Open University Press.
- Crane, M., Fu, R., Warnes, A. M. (2004). Building homelessness prevention practice: Combining research evidence and professional knowledge. Sheffield: Sheffield Institute for Studies on Ageing, University of Sheffield.
- Crane, M., Warnes, A. M. (2001a). Single homeless people in London: Profiles of service users and perceptions of needs. Sheffield: Sheffield Institute for Studies on Ageing, University of Sheffield.
- Crane, M., Warnes, A. M. (2001b). Older people and homelessness: Prevalence and causes. Topics in Geriatric Rehabilitation, 16,1-14.
- Crane, M., Warnes, A. M. (2002). Resettling older homeless people: A longitudinal study of outcomes. Sheffield: Sheffield Institute for Studies on Ageing, University of Sheffield.
- DeNavas-Walt, C., Proctor, B., Mills, R. (2004). Income, poverty and health insurance coverage in the United States: 2003. Washington, DC: U.S. Census Bureau.
- Department of Human Services. (2002a). Summary of housing assistance programs 2000–01. Melbourne: Housing Analysis and Performance, Office of Housing, Victorian Government, Department of Human Services, Victoria, Australia.
- Department of Human Services. (2002b). Directions for change: A collaborative approach to improving our response to homelessness. Melbourne: Office of Housing, Victorian Government, Department of Human Services, Victoria, Australia.
- Elliott, M., Krivo, L. (1991). Structural determinants of homelessness in the United States. Social Problems, 38,113-131.
- Hawtin, M., Kettle, J. (2000). Housing and social exclusion. In J. Percy-Smith (Ed.), Policy responses to social exclusion. Towards inclusion? (pp. 107–129). Buckingham: Open University Press.
- Herman, D., Susser, E., Struening, E., Link, B. (1997). Adverse childhood experiences: Are they risk factors for homelessness? American Journal of Public Health, 87,249-255.[Abstract/Free Full Text]
- Keigher, S. (1992). A preliminary study of elderly emergency service clients in Chicago and their housing-related problems. Journal of Applied Gerontology, 11,4-21.[Abstract/Free Full Text]
- Koegel, P., Burnam, M. A., Baumohl, J. (1996). The causes of homelessness. In J. Baumohl (Ed.), Homelessness in America (pp. 24–33). Phoenix: Oryx.
- Koegel, P., Melamid, E., Burnam, M. A. (1995). Childhood risk factors for homelessness among homeless adults. American Journal of Public Health, 85,1642-1649.[Abstract/Free Full Text]
- Lai, R. (2003). SAAP monograph: Older SAAP clients. Canberra: Commonwealth Department of Family and Community Services.
- Lee, B., Price-Spratlen, T., Kanan, J. (2003). Determinants of homelessness in metropolitan areas. Journal of Urban Affairs, 25,335-355.
- Main, T. (1998). How to think about homelessness: Balancing structural and individual causes. Journal of Social Distress and the Homeless, 7,41-54.
- Menino, T. (2000). Leading the way: A report on Boston's housing strategy FY 2001–2003. Boston: City of Boston.
- Meschede, T., Raymond, J., Sokol, B., Friedmann, D. H., Canavan, D., Faris, M., et al. (2003). Characteristics of homeless individuals accessing Massachusetts emergency shelters 1999–2002. Boston: Center for Social Policy, University of Massachusetts.
- Meschede, T., Sokol, B., Raymond, J., Friedmann, D. H., Sullivan, N., Kahan, M., et al. (2004). Hard numbers, hard times: Homeless individuals in Massachusetts emergency shelters, 1999–2003. Boston: Center for Social Policy, University of Massachusetts.
- Office of the Deputy Prime Minister. (2003). Statutory homelessness: England fourth quarter 2002. News release 2003/0037. London: Office of the Deputy Prime Minister.
- Productivity Commission. (1999). Australia's gambling industries. Report 10. Canberra: Productivity Commission, Commonwealth of Australia.
- Ronaldson, H. (1999). Housing options for older Victorians in need of housing assistance. Paper presented at Aged Care Conference, Department of Human Services, Victoria, Australia.
- Rossi, P. (1989). Down and out in America: The origins of homelessness. Chicago: University of Chicago Press.
- Sosin, M. R. (2003). Explaining adult homelessness in the US by stratification or situation. Journal of Community and Applied Social Psychology, 13,91-104.
- Sullivan, G., Burnam, A., Koegel, P. (2000). Pathways to homelessness among the mentally ill. Social Psychiatry and Psychiatric Epidemiology, 35,444-450.[Medline]
- Susser, E., Moore, R., Link, B. (1993). Risk factors for homelessness. American Journal of Epidemiology, 15,546-556.
- Warnes, A. M., Crane, M. (2000). Meeting homeless people's needs: Service development and practice for the older excluded. London: King's Fund.
- Warnes, A. M., Crane, M., Whitehead, N., Fu, R. (2003). Homelessness factfile. London: Crisis.
- Wilson, D. (1995). "We will need to take you in": The experience of homelessness in old age. Edinburgh: Scottish Council for Single Homeless.
TOP
Abstract
Policies, Services, and...
