SciELO - Scientific Electronic Library Online

 
vol.5 número1Development and initial validation of an infant rearing practices instrumentThe intention of having sexual relations during the adolescence: the role of the television, parental relationships and cognitions índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados

Journal

Artigo

Indicadores

Links relacionados

Compartilhar


Universitas Psychologica

versão impressa ISSN 1657-9267

Univ. Psychol. v.5 n.1 Bogotá abr. 2006

 

Youth homelessness: prevention and intervention efforts in Psychology

Jhon J. Sanabria*

University of Illinois, Chicago

Recibido: enero 19 de 2006 Revisado: febrero 23 de 2006 Aceptado: marzo 8 de 2006

* Correspondencia relacionada con este artículo por favor enviarla a: jsanab2@uic.edu

ABSTRACT

In this paper, I review the prevention and intervention efforts addressing youth homelessness in the field of psychology between 1994 and 2004. Analyses of the literature revealed that the majority of papers including homeless youth as a population for study have focused on issues other than homelessness. These issues include HIV/AIDS and substance abuse prevention. Eleven journal articles addressing youth homelessness were reviewed. These articles focused on outcomes, interventions, and recommendations for clinical practice. Literature findings revealed that demographic variables did not predict outcomes for homeless youth; youth returning home with their parents have more positive outcomes than youth moving into other locations, emergency shelter services improve youth’s mental health and social condition, and services should be comprehensive and move beyond the individuals. Implications for community psychology, policy makers, and shelters are discussed.

Key words: youth homelessness, shelters, community psychology, social policy, prevention, intervention, youth hostels, adolescent psychology, helplessness.

Juventud desesperada: esfuerzos en prevención e intervención en Psicología

RESUMEN

En este documento reviso los trabajos relacionados con la prevención e intervención en la juventud desamparada, en la literatura psicológica entre el año 1994 y el 2004. Los análisis revelan que la mayoría de los documentos que plantean a la juventud sin hogar como población de estudio se han centrado en tópicos diferentes al desamparo, tales como el VIH/SIDA y la prevención del abuso de sustancias. Once artículos de revistas que tratan el desamparo juvenil son revisadas en este texto. Dichos documentos enfatizan los productos, las intervenciones y las recomendaciones para la práctica clínica. La literatura revela que las variables demográficas no son predictoras de los efectos en la intervención realizada en juventud desamparada; aquellos jóvenes que regresan a casa con sus padres tienen mejores pronósticos que los jóvenes que se desplazan a otros sitios. Adicionalmente, los albergues de los servicios de emergencia mejoran la salud mental y condición social de los jóvenes. Se concluye que los servicios sociales deben ser más amplios y deben trascender a los individuos. Las implicaciones para la psicología comunitaria, las personas que diseñan la política social y los albergues son discutidas.

Palabras clave: juventud sin hogar, desamparo, albergues juveniles, psicología comunitaria, política social, prevención, intervención, psicología del adolescente, desamparo.

Youth who lack permanent housing, for any number of reasons, are considered homeless. In this paper I will use the definition provided by the National Coalition for the Homeless (1999) for youth homelessness which states that individuals under the age of eighteen who lack parental, foster, or institutional care are considered homeless. This definition of youth homelessness excludes children who are homeless with their parents and children who are wards of the state. Therefore homeless youth consist only of youth with no connections to either private or public care and/ or support.

However, even though definitions are useful and necessary to understand different phenomenon, providing a clear and inclusive definition of homeless youth is challenging because homeless youth are a diverse set of individuals including runaway youth (if they left their families), thrownaway youth (if their families forced them to leave), street youth (if they live in the streets), and shelter youth (if they stay in a shelter) (Thompson, Safyer, & Pollio, 2001). These categories provide important distinctions regarding the reasons why youth are homeless, and about the conditions they experience. However, these categories should be used carefully as they could also limit our understanding of the phenomenon by artificially splitting the problem of homelessness among youth into multiple smaller problems relevant to particular groups, such as middle class runaway youth, or low-income urban street youth. These categories could also be misleading because they are dynamic and can overlap. For example, an adolescent may decide to runaway from an abusive home and then being forced out after trying to return to it (Hammer, Finkelhor, & Sedlak, 2002).

In the United States the number of homeless youth is estimated to be between 575,000 and more than 1.6 million each year (Thompson et al., 2002, Hammer et al., 2002), which makes it a significant social problem worthy of prevention research and intervention. However, prevention efforts seem to be mostly oriented to limiting its impact rather than to preventing homelessness itself.

Relevance to community psychology

Two important emphases in community psychology are to promote health and empowerment and to prevent problems in communities, groups, and individuals (Society for Community Research and Action, 2004). Furthermore, community psychologists are also interested in promoting distributive and procedural social justice. Distributive justice refers to recognizing that all individuals have the right to a fair share of social resources (Prilleletensky & Nelson, 2002), and procedural justice refers to recognizing that all individuals have the right to an effective participation in decision-making processes (Fondacaro & Weinberg, 2002).

Youth homelessness reflects the very kind of social injustice that community psychologists aim to eradicate. As a social illness, youth homelessness compromises youths’ health and disempowers them by negatively impacting their individual and collective well-being (Shinn, 1992; Staller & Kirk, 1997). At the individual level, homelessness in adolescence hampers youth well-being because it disrupts their development by breaking up natural social networks and families, and by promoting the development of mental health problems related to instability, poverty and other stressors (Shinn, 1992; Smollar, 1999; Staller, & Kirk, 1997). Specifically, homeless youth are at increased risk of experiencing violence, substance abuse, HIV and other STDs, mental health problems, lack of regular health care, as well as physical, psychological, and sexual abuse (Thompson et al., 2002; Thompson et al., 2001). Compromised well-being, in turn, impedes collective well-being (Prilleltensky & Nelson, 2002) by obstructing the attachment of individuals to communities, diminishing their sense of belonging and care, and by bringing additional problems related to public health, violence, and poverty to communities (Shinn, 1992).

Purpose

The purpose of this paper is to conduct a literature review of the journal articles describing interventions aimed at preventing homelessness among adolescents, and to answer the following research questions:

1.  How is youth homelessness conceptualized in the prevention and intervention literature?

2.  What kind of prevention and intervention efforts are psychologists and other social scientists implementing to address the causes of youth homelessness?

3.  What evidence exists regarding the effectiveness of those prevention and intervention efforts?

4.  Based on the existing prevention literature on youth homelessness, what should be the focus of future research projects?

To achieve this goal, I will first provide a context for the review by describing the history and current state of youth homelessness in the United States. Second, I will review all publications that directly focus on preventing homelessness among adolescents, ages 13 to 17, with respect to the four questions listed above. Finally, I will discuss the implications of the results of this review for policy, practice, and theory in community psychology.

History

Youth homelessness is not a recent phenomenon in the United States. On the contrary, homeless youth have always been part of U.S. history (Smollar, 1999). Homeless youth were present during the settlement of the original thirteen colonies when many adolescents left home seeking adventure and economic opportunity. However, the causes and characteristics of youth homelessness have experienced great transformations over time.

Homeless youth were also prevalent during the 1800s, particularly in large urban areas, where they were perceived as criminals representing an important social problem (Smollar, 1999). During the 1800s, homeless youth came from poor immigrant families and were forced into stealing and/or begging as they were neither wanted nor needed in the labor force (Smollar, 1999). Social workers interceded on behalf of the homeless youth and established “houses of refuge”, which later became known as reform schools, dedicated to providing youth with positive environments in hope of changing their “delinquent behavior” (Smollar, 1999).

During the Great Depression, many youth became homeless, but it was generally ignored as a specific problem because large segments of the population were homeless at the time (Smollar, 1999). It was not until the 1960s that youth homelessness again resurfaced as an important social problem due to the presence of a new group of homeless adolescents referred to as “runaways” (1999). Runaways were different from the homeless youth of the 1800s because they left middle and upper class homes as a statement of political protest, in an effort to reject their families’ values and explore and express themselves (Staller & Kirk, 1997; Smollar, 1999).

These rebellious adolescents of the 1960s were concentrated in counter-culture meccas like the East Village in New York City and Haight Ashbury in San Francisco (Staller & Kirk, 1997). During the 1960s, runaway youth came into contact with the juvenile justice system and the juvenile courts did not distinguish between rebellious youth and those committing serious offenses. Moreover, the juvenile justice system relied on incarceration as a way of controlling the behavior of runaways. As expected, the public sector criticized this approach, arguing that incarceration was an inappropriate remedy for controlling misbehavior (Staller & Kirk, 1997). As a result, federal laws in 1974 eliminated incarceration as a judicial sanction for rebellious behavior, leading to three different types of family court cases: cases of abuse and neglect, juvenile delinquency cases, and status offenses. The status offenses include troublesome (but not criminal) behavior and running away (Staller & Kirk, 1997). Of particular relevance to the statutory offences is the Runaway Act of 1974, which shielded shelters from charges of “custodial interference” lodged by parents, afforded youth a short period of immunity from any parental control, and provided funds for establishing shelters to temporarily house youth (Smollar, 1999; Staller & Kirk, 1997).

The main objective of the 1974 Runaway Act was to protect rebellious adolescents who were at risk of being victimized by adults (Smollar, 1999). The Act was revised in 1977 and renamed the Runaway and Homeless Youth Act. This revision was a response to reports from shelter personnel who “were seeing many adolescents who had not run from home, but had been pushed out of their homes, been abandoned by parents, or left home for life on the streets with their parents’ knowledge and consent” (Smollar, 1999, p. 49). The Act stressed family reunification but acknowledged that both youth and their families needed extensive services for reunifications to be successfully accomplished (Smollar, 1999). Despite the laws passed during the 1970s, during the 1980s and the 1990s the number of teens that had been pushed out or abandoned or remained away with parental consent continued to increase, and during the 1990s poverty emerged again as a contributing factor to youth homelessness (Smollar, 1999).

In summary, youth homelessness has been a significant problem throughout the history of the United States, and its causes and characteristics have changed through the times. In 1974, the federal government formulated laws and assigned funding to address this phenomenon, and these laws have determined the different intervention approaches. Since 1974 the emphasis on services has been temporary shelter (with stays limited to 30 days), and interventions oriented towards family reunification.

Current state

Estimating the number of homeless youth is not an exact science as different studies use different definitions of the population, and different sampling and estimation techniques. For example, Ringwalt et al. (1998) surveyed a nationwide random sample of 6,496 teenagers, 12 to 17 years of age, and found that 7.6% reported having experienced homelessness for at least one night. In their study Ringwalt et al. (1998) defined homelessness episodes as when “they [youth] spend the night in a youth or adult shelter or in any of several locations not intended to be dwelling places or where their safety would be compromised” (Ringwalt et al. 1998, p. 1325).

Other studies estimate the number of homeless youth to be between 575,000 and more than one million each year (Thompson et al., 2002), and even up to 1.3 to 1.7 million each year (Coco & Courtney, 1998; Cauce et al., 1994; Hammer et al., 2002). For example, Hammer and collaborators (2002) interview parents and adolescents nationwide about runaway and thrownaway episodes and estimated that in 1999 a total of 1,682,900 youth had a runaway/thrownaway episode. In summary, even though the exact numbers for homelessness among youth are unknown, there is broad consensus that these numbers are growing with no decline in sight (Cauce et al., 1994).

Considering that homelessness among youth is not a recent phenomenon in the United States, and that the number of cases is increasing, it is crucial for social and behavioral scientists to understand this phenomenon, to develop and implement appropriate prevention and intervention efforts and to inform policy makers. Following is a description of the known causes of youth homelessness in the United States, and the characteristics of youth experiencing it.

Current Causes and Characteristics of Youth Homelessness

There is no single cause for youth homelessness. However, youth consistently report family conflict as the primary reason for their homelessness (Whitbeck et al. 2002; Robertson & Toro, 1998; Smollar, 1999). Common causes of conflict with parents include youths’ relationships with stepparents, sexual activity and sexual orientation, pregnancy, school problems, and alcohol and drug use. Neglect and physical abuse in the home are also common experiences. Similarly, neglected and sexually abused youth are more likely to run away than their non-abused counterparts (Yoder, Whitbeck, & Hoyt, 2001). Moreover, family conflict, neglect, and abuse tend to be long-standing problems rather than single episodes occurring shortly before youth leave homes (Smollar, 1999).

Other variables found to be causally associated with youth homelessness include: residential instability, broken ties in families or origin, and substance abuse. For many youth, homelessness appears to be part of a long pattern of residential instability. Also disrupted family histories may contribute to the risk for homelessness. Additionally, many homeless youth report substance abuse by themselves and by their parents, which may contribute to family conflict and to youth homelessness (Thompson et al., 2001; 2003).

Additionally, while studying the causes of youth homelessness researchers have identified variables that are not directly associated with youth homelessness. These variables include setting (i.e. rural, urban. Sub-urban), gender, race/ethnicity, sexual orientation, and poverty. In addition, youth with recent histories of homelessness come from urban, sub-urban and rural settings. Also, the gender distribution among homeless youth varies according to sampling procedures. Shelter samples tend to have more females, while street samples tend to have more males (Robertson & Toro, 1998).

Furthermore, it has been found that there are no differences in rates of recent homelessness by racial or ethnic groups (Hammer et al., 2002), sexual orientation or income. The rate of gay or bisexual orientation among homeless youth varies across studies (ranging from 3% to 38%), according to the sampling procedure. However, these studies suggest that homeless youth are no more likely than non-homeless youth to report gay or bisexual orientation. However, sexual orientation and sexual activity may be indirectly related to youth homelessness as they may trigger family conflict. Similarly, family poverty may not be related to homelessness among youth per se. However, family poverty may be related to more chronic or repeated homelessness. This is, among homeless youth, the ones from poor families are more likely to experience repeated episodes of homelessness than the ones from middle or high-income families (Robertson & Toro, 1998).

Researchers have also identified variables that are associated with youth homelessness, but are not considered causes of it. These variables include school difficulties and mental disorders. The contribution of school problems to homelessness is unclear. School problems can be a precipitant of family conflict that results in a runaway response. However, it is also suggested that school difficulties are merely symptoms of more pervasive family problems. Similarly, it is difficult to determine whether a homeless youth’s emotional disturbance at a given point in time is more casually associated with an underlying emotional or mental disorder, the exigencies of homelessness, chronic stress, the youth’s own use of alcohol or other drugs, or a combination of these (Cauce et al., 1994; Thompson et al., 2001).

Finally, homelessness among youth has been identified as a cause of several other problems for youth including suicide attempts, poor health, and increased risk of pregnancy and sexually transmitted infections. Suicide attempt rates are higher for homeless youth than for normative groups. Additionally, homeless youth appear to be at greater risk than their domiciled counterparts for a variety of medical problems, and their health often deteriorates while homeless. Additionally, the literature reveals high rates of sexual activity among homeless youth, but variable rates of protection against pregnancy or sexually transmitted diseases (Staller & Kirk, 1997; Cauce et al., 1994).

One of the many consequences of unprotected sex is pregnancy. Studies have identified as many as 10 to 20 percent of homeless young women who are currently pregnant. Young women who are pregnant while homeless are at risk for low-birth weight babies and high infant-mortality because they are unlikely to get prenatal care and may not have adequate health and dietary habits. Likewise, homeless youth present a high-risk profile for HIV infection. Specific high-risk sexual and drug use behaviors include multiple sex partners, high-risk sexual partners, survival sex, minimal condom use, injection drug use, sharing needles, and having sex while high (Robertson & Toro, 1998).

Given that the youth frequently report family conflict as the main reason for their homelessness it would be expected to see youth homelessness prevention efforts address the family dynamics and conflict. Likewise, knowing the impact of homelessness in youth, it would also be expected to see services for youth who are already homeless focus on reducing the risk for HIV/AIDS and addressing health and mental health conditions. A review of the prevention literature in psychology in the last ten years regarding youth homelessness is used to assess whether these causes and consequences are addressed.

Method

Literature Selection

Following is a review of the prevention efforts in psychology addressing youth homelessness in the last ten years. Selecting the literature for this review involved four steps. First, a series of literature searches were conducted on PsycINFO, ERIC, and Social Sciences Abstracts. Second, the results were classified by type of publication, location, and main focus. Third, journal articles of studies conducted in the United States, whose main focus was homelessness, were further classified according to the population they were studying. Fourth, of the remaining publications those studying youth were further classified by main focus, and eleven were selected for the review. Following is a detailed description of these four steps and how they ultimately led to the selection of the 11 articles included in the review.

Step One: Literature searches

A series of literature searches were made in PsycINFO using the following seven terms: homeless, homelessness, runaway behavior, homeless youth, runaway youth, street youth, and thrownaway youth. The results for each of those terms were limited to treatment and prevention, year of publication (between 1994 and 2004), and age group (adolescents between 13 and 17 years). Additional searches were made in ERIC and Social Services Abstracts. In ERIC, searches were made using the following terms: runaway, homeless, youth or adolescent, and prevention. The results of those searches were limited by year of publication (between 1994 and 2004). A final series of searches were made on Social Services Abstracts using the following terms: runaway prevention, homeless youth prevention, homeless youth intervention, and runaway intervention. All the searches were limited by year of publication (between 1994 and 2004). The differences in search terms for the three databases correspond to the options and limits set by their browsers.

Step Two: Classification by Type of Publication, Location and Main Focus

A total of 151 unique publications were found. The abstracts of those 151 publications were reviewed and classified according to the following variables: type of publication (i.e. journal article, dissertation, book, or chapter), location of the study (i.e. United States or other countries), and main focus of the publication (i.e. homelessness, HIV/AIDS, Sexually Transmitted Infections [STIs], substance abuse, or other).

Publications focusing mainly on homelessness include the following themes: services and interventions for homeless, their needs, their characteristics, and their physical and mental health. The publications in the HIV/AIDS, Sexually transmitted Infections (STIs), and Substance Abuse categories view homeless youth as the population of focus for such particular issues. Publications focusing on youth at risk or other populations that included homeless youth as a subgroup in their studies were considered “other” (e.g. guidelines for intervention with at risk populations, elopement, and inpatients leaving the hospital without authorization).

Step Three: Classification according to Population

For the present study only journal articles1 of studies conducted in the United States, focusing on homelessness were selected. A total of 34 publications met these criteria (See Table 1). Those 34 publications were then reviewed and classified by population (i.e. individuals experiencing homelessness on their own, families experiencing homelessness), and by age group2 (i.e. children, youth, adults). Out of these 34 publications 16 were removed from the review at this step. Of the 16 removed, seven publications were removed because they studied families, and 9 because they studied children or adults. This selection process yielded a total of 18 publications studying adolescents (13 to 17 years of age) experiencing homelessness on their own.

1. The dissertations, books and chapters focusing on homelessness were interested in the impact of homelessness on health, education, and development. These publications were not addressing the prevention of youth homelessness and therefore were not included in the subsequent analyses.

2. Although the literature searches were limited to population between 13 and 17 years old, some studies involving children and adults were included by PsycINFO. This was the case for publications studying children up to 13, or adults 17 or older.

Step Four: Main Focus of the publications studying youth 13 to 17

A detailed review of the 18 remaining publications revealed the following foci: outcomes for youth using homeless shelters, health and mental health issues of homeless youth, general interventions for homeless, interventions for aggressive behavior in youth living at shelters, relationship between foster care and homelessness, and general guidelines for clinical intervention with homeless youth (See Table 2).

Only three publications describe general interventions that may prevent future episodes of youth homelessness, which is the focus of the present study. Therefore, I decided to review those three publications describing general interventions, plus five describing outcomes for homeless youth using shelters, and two providing general guidelines for clinical intervention with homeless youth. These last seven publications were selected because they don’t have a particular focus and their results may be applicable to general prevention efforts addressing youth homelessness.

At this point, a quick inspection of the list of original publications revealed some researchers appeared several times. The names of such authors were used in an additional series of searches on PsychINFO. Authors’ names used include: Toro, P. A., Thompson, S. J., Pollio, D. E., and Teare, J. F. Results were reviewed and only one new publication fitting the criteria applied at steps two through four. The focus of this last publication is on outcomes for youth using shelters. Thus a final sample of 11 publications served as a basis for review in this study: three describing general interventions, six describing outcomes for youth using shelters, and two describing general guidelines for clinical intervention with homeless youth.

Review of the Literature

The purpose of this review is to identify the prevention and intervention efforts in psychology regarding youth homelessness, as well as to describe their conceptualization of the phenomenon. More specifically, with this review I will answer the following questions:

1.  How is youth homelessness conceptualized in the prevention and intervention literature?

2.  What kind of prevention and intervention efforts are psychologists and other social scientists implementing to address the causes of youth homelessness?

3.  What evidence exists regarding the effectiveness of those prevention and intervention efforts?

4.  Based on the existing prevention literature on youth homelessness, what should be the focus of future research projects?

Conceptualization of Youth Homelessness

According to Thompson and collaborators (2001) homeless youth have been frequently defined and grouped into three categories. These categories had been empirically confirmed recently and include:

· Runaway/homeless youths are youth who stay away from home at least overnight without the permission or knowledge of their parents or guardians.

· Thrownaway youths are the ones that leave home because their parents have encourage them to leave or have locked them out of the house.

· Independent youths are youth who feel that they have no home to return to because of irreconcilable conflicts with their families, have lost contact with their families, or have families that are homeless (Thompson et al., 2001).

Thompson and collaborators (2001) recommend future research to focus on developing services and interventions specific to these unique groups. In addition to these categories, Coco and collaborators (1998), subdivided runaways into two other groups: “running from” and “running to”. Running-from youth are escaping family turmoil caused by a crisis, excessive parental expectations and control, or physical or sexual abuse. Running-to youth are often searching for excitement. Running-to youth normally come from a more stable family and run away for existential reasons. Finally, Bronstein (1996) suggests the main difference between runaways and homeless youth is that runaways have a family to return to, while homeless youth do not. Bronstein also makes another classification within the homeless youth group: youth on their own and teenage mothers. Although youth on their own and teenage mothers have similar reasons for being homeless, Bronstein (1996) emphasizes the unique situation and needs of the teenage mothers, as they are youth and adults at the same time.

It becomes clear then that there are many ways in which homeless youth can be categorized. However, there are two main groups that overlap many of the existing classifications: youth experiencing conflict with their families and youth who have no home to return to. The group of youth that are experiencing conflict with their family would encompass runaway and thrownaway youth. In these cases either the family or the youth are escaping from each other. The group of youth having no home to return to include youth who are no longer in contact with their families and youth whose families are homeless. In either case, the proliferation of labels suggests the complexity and diversity of the problem (Staller & Kirk, 1997). These labels should be further explored and validated in future research because their use can be as beneficial to future programs as it can be dangerous. Accurate distinctions between different types of homeless youth would be useful to tailor programs. However, factitious categories can limit our understanding of the phenomenon and the dissemination of effective practices.

Intervention and Prevention Efforts in Psychology

My review of the literature indicates that when the terms prevention, intervention, and youth homelessness are present in a psychological journal article they normally refer to HIV/AIDS, STIs, or substance abuse prevention programs targeting homeless youth, mental health needs assessments, or use of health services by homeless youth. These results indicate researchers are directly addressing the main negative consequences of homelessness among youth. However, researchers are not preventing homelessness, as only a minority of the publications on prevention and youth homelessness (n = 18, see Table 2) refer to youth homelessness as the targeted issue. The three interventions that did targeting homeless youth addressed three different issues, using three different approaches, as follows:

· Recidivism for runaway female teenagers: addressed with clinical counseling for the family, using a family systems approach (Coco & Courtney, 1998).

· Social skills for youth staying in a shelter: addressed through the teaching of social skills using role-plays and a token-based economy as motivational support (Teare et al., 1994).

· Mental health and social adjustment: addressed with intensive case management for homeless youth (Cauce et al., 1994).

These three studies are not doing primary prevention as they are addressing homelessness after youth have left home. The study by Coco and Courtney (1998) is the only one directly aiming to prevent recidivism. In their intervention they targeted the whole family, aiming to reduce conflict and increase family satisfaction. These objectives are congruent with the known causes of youth homelessness, which lay mostly with family conflict.

On the contrary, the study by Teare and collaborators (1994) addresses social skills for youth using shelters. According to the authors this approach will prevent conflict in the youth’s life during and after their stay in the shelter. The authors in their literature review highlight the issue of conflict while staying in the shelter as they report high incidences of victimization of adults at shelters. This victimization is carried out by shelter personnel and by other residents, and includes neglect, robbery and physical abuse.

Finally, in the study of Cauce and collaborators (1994) the authors addressed mental health and social adjustment. Although these are important issues for homeless youth the study can’t be considered a primary prevention effort as they are targeting youth who are homeless, and the causal relationship between mental health and social adjustment and homelessness is unclear. Currently there is not enough information to determine what comes or constitutes a cause for the other: homelessness, mental health, and social adjustment.

In summary, according to the psychological literature of the last ten years, psychologists and other social scientist are not developing or conducting programs directly aimed at preventing homelessness among youth. The few existing interventions target youth after they ran away or became homeless, and address different issues related to youth homelessness. Over the last ten years the psychological literature also indicates homeless youth are being recognized as population at risk and have been considered a target of prevention programs for other conditions, including HIV/AIDS and substance abuse.

Evidence regarding the interventions’ effectiveness

There is little evidence supporting the effectiveness of the intervention efforts described in the literature. In the case of the family systems approach to prevent recidivism of female runaway teenagers (Coco & Courtney, 1998), data collected by the researches indicate that participants were more satisfied with the family after the intervention. However, there is no information assessing recidivism directly. In other words, the authors indicate their intervention is effective in restructuring family relationships and increasing satisfaction with the family, but they do not provide long-term information regarding the presence or absence of runaway episodes among participants.

In the case of the social skills intervention (Teare et al., 1994), the researchers provided information regarding the implementation of the program, the participants’ satisfaction with it, the number of participants who ran away from the shelter, number of participants who injured themselves, and participants’ suicidal thoughts and attempts. However, there is no information regarding the change in the participants’ behavior before and after the program. According to the data, the program provided more than 24 social skills teaching interactions to each participant per day. Participants were satisfied with the program and none reported abuse while at the shelter. This result about abuse is important for the authors as they believe adults at shelters are frequently neglected by shelters’ staff and abused by other residents.

Nevertheless, Teare and collaborators (1994) did not relate the program activities with the fact that out of a 100 participants, 22 ran away from the shelter, one attempted suicide, 11 verbalized suicidal intent, and one had a self-inflicted injury. In other words, the authors reported these incidents but did not describe whether they may or may not be related to the intervention activities. Authors only indicated youth reporting suicidal ideation were administered a lethality risk assessment scale and, when necessary, provided with additional assessment and intervention by a community professional. In summary, the authors provided data regarding the implementation of the program and the participants’ satisfaction. However there was no outcome data, and some important behaviors (i.e. running away, suicidal intent) were reported and not discussed in relationship with the intervention being provided.

Finally, for the mental health and social adjustment intervention (Cauce et al. 1994), the data show improvements but the facts are not conclusive. Results indicate mental health and social adjustment of homeless youth improved with case management, whether traditional or intensive. However, given the design of the study the authors had no way to explain the mechanisms for change; nor could they eliminate the possibility of change without intervention. In this study researchers compared two interventions. However, the lack of a control group with no intervention, as well as the lack of qualitative information about the participants’ experiences limits the researcher’s ability to make solid conclusions about the causes of the improvements observed.

In summary, the evaluations conducted are mostly focused on process and consumer satisfaction. While necessary, those elements are not sufficient to assess the effectiveness of the interventions. In addition, two of the evaluations did not include a pre-post measure of relevant behavior. For the one intervention where outcome data was available, results were not conclusive and further research is needed. There is a clear need for outcome evaluation. These results reflect youth homelessness is currently being under studied and the little that is known about interventions with homeless youth is more anecdotal than conclusive.

In addition, in the six outcomes studies for youth using shelters, researchers explored the relationship between a set of predictor variables and outcomes. The predictor variables considered were: a) number of runaway episodes in the past, b) demographic information, c) personal and family characteristics, d) type and severity of individual’s problems, and e) type of shelter used. The outcomes considered included: a) recidivism (Baker et al., 2003), b) family reunification (Thompson, Kost & Pollio, 2003; Thompson, Safyer & Pollio, 2001), c) post shelter placement (Teare, Authier & Peterson, 1994), d) long-term individual change (Thompson, Pollio & Bitner, 2000), and e) short-term individual change (Thompson et al., 2002).

Results of these outcome studies indicate that first time runaways are more likely to recidivate if they face family conflict (Baker et al., 2003). Also, no demographic variable can predict family reunification (Thompson, Kost & Pollio, 2003). Youth with serious mental health and behavioral crisis are discharged from the shelters into more restrictive environments, all other youth returned to their families of origin and have better outcomes overall (Teare, Authier & Peterson, 1994). Finally, results also indicate short-term and emergency shelters are equally effective in improving youth’s mental health, behavior and social adjustment in the short term (Thompson et al., 2002).

Results of these studies suggest that: 1) recidivism correlates with family change for first-time runaways, and with youth’s emotional problems for repeat runaways (Baker et al., 2003); 2) ethnicity does not predict family reunification (Thompson et al., 2003); 3) youth at suicidal risk, and youth from dysfunctional families are discharged from shelters into settings more restrictive than those where they were living before entering the shelter (Teare et al., 1994); 4) no demographic information can predict outcomes for runaway and homeless youth using shelter services, 5) youth returning home with their parents have better outcomes than youth discharged to other locations, which is a reflection of the fact that youth with dysfunctional families are being discharged to other locations (Thompson et al., 2000); and 6) emergency shelters are effective in improving outcomes for runaway and homeless youth six weeks after discharge, which is not enough time to assess long term outcomes (Thompson et al., 2002).

It is important to notice that only one of these six studies examined ser vice effectiveness: Thompson and collaborators (2002) compared the outcomes of youth using emergency shelters with the outcomes of youth using regular long-term shelters. The other five studies focused on the relationship between demographic information and outcomes. This focus reflects little interest in the impact of services on outcomes, and provides little information about the services available for homeless youth. It is also worth noting that two researchers (Sanna J. Thompson and David E. Pollio) were co-authors in four of these six studies, which indicates there are not many researchers addressing youth homelessness directly.

Future Research: Recommendations from the existing literature

Based on their understanding of youth homelessness and their results, the studies reviewed suggest the following steps for future research: conduct qualitative studies, focus on families and systems; study youth living in the streets; conduct longitudinal studies, and conduct pre and post comparisons. Qualitative studies are necessary to understand youth’s history and plan services accordingly. Qualitative studies should aim at understanding the conditions of homeless youth and their families, before and after the youth stay in a shelter. Similarly, interviewing youth who recidivate, and conducting a qualitative analysis of the their experience at discharge, would provide a deeper understanding of their experience and allow the youth to have a voice in the development of knowledge regarding recidivism (Baker et al., 2003).

Also, future research should focus on families and their context because youth who use shelters normally come from, and return to, their homes. Causal studies and evaluations of new interventions should address issues at multiple levels, including the youth, their families, and the environment where they live (Teare et al., 1994; Thompson et al., 2000). Similarly, Bronstein (1996) suggest an ecological systems approach as an overarching paradigm to guide work with homeless youth.

It is also necessary to study youth living in the streets that may not use shelter services. Existing studies are limited to data provided by shelters. This limitation implies we do not know if the current findings in literature apply to all homeless youth. Additionally, we do not know what prevents some youth from using shelters. Therefore, studying youth living in the streets would help to either validate or modify what is known about homeless youth and will give service providers guidelines to improve outreach and increase service utilization (Thompson et al., 2001, 2003).

Additionally, there is a clear need for longitudinal studies in addition to the existing cross-sectional studies. Cross-sectional studies provide important information regarding personal and family problems related to youth homelessness. However, such problems can’t be organized in a clear timeline, limiting the researchers’ capacity to provide causal explanations. Multivariate models and longitudinal studies would enhance the ability to control alternate hypothesis and increase the credibility of any causal inference drawn from that data (Cauce et al., 1994; Thompson et al., 2000, 2001, 2002).

Finally, it is also necessary to conduct pre and post comparisons for youth using shelters including control groups. Most of the existing studies collect data about youth only once, and the few that make pre and post comparisons lack a control group. In the future, researchers could conduct pre and post intervention studies including a comparison group. Such studies would provide valid information regarding the effectiveness of the existing intervention and services. This information can then be used to develop interventions and improve existing services (Cauce et al., 1994; Thompson et al., 2000).

Conclusion

In the last ten years psychologists and other social scientist did not develop or conduct programs directly aiming to prevent homelessness among youth. The few available interventions target youth after they have become homeless, while the prevention efforts available for homeless youth are aiming to prevent other problems possibly caused by homelessness, like substance abuse and HIV/ AIDS. In addition, there is not enough information about the effectiveness of the existing services and interventions. Literature also suggest future research should focus on qualitative studies to better understand the population, assessing not only individuals but families and their environment, studying homeless youth that may not use shelters, conducting longitudinal rather than cross-sectional studies, and conducting pre and post comparisons to assess service effectiveness.

The limited existing literature indicates that homeless youth have been frequently grouped into three categories: a) runaway/homeless youth, b) thrownaway youth, and c) independent youth (Thompson et al., 2001), and the first two of these categories refer to youth having conflict with their families. Therefore, according to the literature, a prototypical homeless youth is an adolescent facing considerable family conflict, and in many cases some form of abuse, whether physical or sexual, and neglect (Cauce et al., 1994; Teare et al., 1994; Thompson et al., 2002). Moreover, whether youth run from home or are thrown out by their parents, both parties use the homelessness episode as a temporal refuge from the family turmoil (Coco & Courtney, 1998; Thompson et al., 2000). A typical homeless youth normally comes from, and return to, home with its parents. The literature also tells us that there is no particular demographic group that is more likely to become homeless because of their racial, ethnic, socio-economic, or family structure background (Hammer et al., 2002; Robertson & Toro, 1998).

Consequently, homeless youth prevention efforts should focus on identifying and intervening with families experiencing crisis. It is important to consider that any type of crisis can lead an adolescent to run/be thrown out, including crisis due to the normal developmental issues of adolescence (like establishing independence from their parents), crisis due to sudden family changes (like the death of a parent, a divorce), and crisis due to serious behavioral problems from either the parents or the adolescent (like substance abuse, inappropriate sexual activity, parental abuse of the adolescent).

Discussion

The findings of the literature review indicate most of the publications are addressing the consequences of youth homelessness rather than homelessness itself. In a way, according to the literature, psychologists and other social scientists view youth homelessness as a cause not a consequence, a demographic variable rather than a social problem. Known consequences of homelessness among youth include decreased health and mental health, increased risk of substance abuse, HIV/AIDS and other sexually transmitted infections. Current prevention and intervention efforts are addressing these issues. Although it is necessary to prevent these consequences, it is more cost effective to prevent the cause rather than each one of the many consequences. Additionally, most of the few studies that focus on youth homelessness directly are looking at individual level variables to predict outcomes. This focus on individual factors implies victim blaming, as they are looking for something about the individual that will determine positive or negative outcomes rather than looking at the social conditions that contributed to homelessness. Following are some of the implications of these findings.

Implications for community psychology

As mentioned before, current studies of outcomes for youth using shelters focus exclusively on youth’s variables to predict their outcomes. This emphasis on individual-level predictors is victim blaming and provides no feedback for agencies providing services. Our responsibility as social science researchers is to find influencing factors and solutions at all levels (i.e. individual, family, community, and society), and focus our attention on those variables that link structural and individual levels of analysis (Shinn, 1992). Psychologists (especially community psychologists) must move beyond the individual level predictors and start looking at the impact of prevention and intervention programs, as well as the impact of social variables (Shinn, 1992).

To move beyond individual level variables, psychologists should recognize and work at all levels of human interaction, using theories and methods that recognize individual, family, school and community influences on the development of individuals (e.g. Bronfenbrenner’s systems theory of human development, 1979). In particular, at the family level, psychologists could focus on the development and evaluation of programs for parents to understand adolescent development, and for all family members to develop skills for communication, conflict resolution, and management of stressful conditions. These interventions could increase family cohesion, reduce conflict and prevent runaway/thrownaway episodes.

Additionally, at the community level psychologists can focus on awareness, capacity building and program development and evaluation. Awareness campaigns can target all community members: individuals, families, school personnel, business leaders, and staff from community-based agencies. These awareness programs can increase knowledge about causes of youth homelessness, as well as about the needs of youth experiencing homelessness. Capacity building programs could target staff from schools and community-based agencies. These programs could develop and improve staff ’s knowledge and skills to address the needs of homeless youth and their families. Finally, psychologists can also develop and evaluate programs for schools and community-based agencies to support adolescents and families under stress. These programs can help individuals and families cope with family and community stressors (e.g. high rates of substance abuse, violence, poverty and poor housing conditions).

Finally, at a more global level psychologist can inform policy makers on issues related to youth homelessness, and advocate for more efficient prevention and intervention services. Psychologists can also work in multidisciplinary teams (e.g. with social workers, nurses, educators, public health professionals and administrators) to delineate goals and indicators for comprehensive, evidence-based services and programs to be implemented in multiple communities.

Implications for policy makers

Policy makers should recognize youth homelessness as a unique social problem (different from homelessness among adults and families). Policy makers must also recognize that two types of interventions are needed for homeless youth. One type of intervention should address the immediate needs of youth experiencing homelessness, and the other should prevent future youth homelessness (Robertson and Toro, 1998). Although funding is already available for shelters and other emergency services for homeless youth there is little information about their effectiveness, or even their reach. More support is needed for research on the incidence and prevalence of youth homelessness as well as for research on effectiveness of services and programs.

Policy makers must also encourage and support more research and interventions that address the issue of youth homelessness directly; in the same way it has been addressing some of its consequences (e.g. HIV/AIDS, substance abuse, mental health conditions). Encouragement and support from policy makers means should be traduced into more funding to prevention and intervention strategies, with emphasis on comprehensive community-based programs (Cauce et al., 1994). Policy makers should also consider the social and economical cost of not intervening to eliminate youth homelessness. Not addressing youth homelessness will decrease quality of life for all. Ignoring youth homelessness will cost money in lost productivity, and funding for prevention of and care for associated personal and social problems (e.g. HIV/ AIDS, substance abuse, mental health problems, robbery, youth prostitution).

Implications for shelters

The little information available about the effectiveness of services indicates youth can be satisfied with their experiences at shelters and improve their mental health and social conditions (Teare et al., 1994; Thompson et al., 2002). We also know that shelters should keep their general goal of family reunification, as youth returning home to live with their parents have more positive outcomes than other youth (Thompson et al., 2000). On the other hand, there is no information about how these intervention efforts move beyond the context of the shelter experience. That is, interventions at shelters and other social service agencies could go beyond the individuals and involve the family, the schools, and the community.

At the family level, shelters could try to improve communication, problem solving strategies, rules and expectations. At schools, teachers and counselors could provide youth with information about youth homelessness (in order to make it a less attractive alternative) while informing them about resources and alternatives for common problems related to their developmental stage as well as their family and social conditions. Finally, for community involvement, shelter staff could become advocates and case managers to facilitate service coordination and service provision for youth because youth need many services that can be difficult to find and obtain. Even when services are sought lack of coordination can make services ineffective. In this context, if would be very helpful if shelters’ staff could work as a bridge between the youth in need and the services available (Bronstein, 1996).

Implications for Prevention science

The main single predictor of youth homelessness is family conflict (Whitbeck et al. 2002; Robertson & Toro, 1998; Smollar, 1999). Therefore prevention efforts should focus on improving family dynamics, and increasing parents’ skills on conflict resolution, communication, non-violent discipline, and understanding of adolescent development. Such programs should also acknowledge the role that societal and environmental factors play on family relationships. In other words, such programs should acknowledge and incorporate the context in which youth and family live. These contextual programs addressing family dynamics would very likely have an impact on youth homelessness and other behavioral and health issues in both parents and youth. Examples of these issues are anxiety, depression, substance abuse, unprotected sex, and academic/work performance.

Currently there are programs addressing family dynamics and its impact on the development of youth. These programs include support groups for parents, parenting skills classes, and conflict resolution skills for youth and adults among others. If these programs are effective in improving the family relationships they are likely to prevent youth homelessness. Therefore a possibility worth exploring for social scientists interested in preventing youth homelessness is to partner with such programs. These partnerships would eliminate any chances of duplicating services (i.e. conducting the same programs aiming to prevent different issues) and promote a better use of limited resources. These partnerships should then highlight the importance of youth homelessness prevention and outcome evaluation.

Outcome evaluation of prevention programs addressing family dynamics is a key research issue for prevention programs with youth because these programs have the potential to impact multiple outcomes on youth. Therefore, a relevant research question for prevention science would be: which prevention approach is more effective for youth, a universal approach where programs promote general positive development, or a targeted approach where each program focus on a particular outcome? The answer to this question could guide the development and modification of prevention programs, interagency collaborations, and community coalitions.

Implications for future research

In order to conduct prevention and intervention efforts targeting youth homelessness directly it is necessary to clearly understand the experiences of homeless youth before and after they leave their homes and before and after use shelters and other similar services. This understanding would require a combination of research strategies including longitudinal studies, use of qualitative data, pre and post behavioral comparisons, and comparison between youth that use shelters and youth that do not. Longitudinal and qualitative studies would help us create timelines and causal relationships between environmental, family and individual variables and homelessness. Some of the relevant variables associated with youth homelessness that would benefit from further clarification through longitudinal and/or qualitative studies are: school problems, depression and other mental health issues, and substance abuse. In the same manner we could explore more about family conflict to identify ways in which families can deal with conflict in a constructive way and prevent runaway/ thrownaway episodes.

Additionally, to have a complete understanding of the experiences of homeless youth, researchers must also study youth not using shelters and other services. Currently we don’t know if the demographic, behavioral, and family characteristics of youth using shelters are representative of all homeless youth or if there are any important variables preventing some youth from using services available to them. Also, future research should continue to assess current services. According to the literature some shelters and programs are trying to increase social skills, improve conflict resolution skills, improve mood, reduce mental health conditions, and improve quality of family dynamics. Future evaluations of services would benefit from including pre and post comparisons of any of these behavioral or health variables they are aiming to change.

Conclusion

In conclusion, psychologists could partner with shelters and other providers to conduct long-term qualitative research to inform the development of programs, and evaluate the impact of services and social factors on outcomes. Similarly, psychologist could try to identify, partner with, and evaluate intervention strategies to prevent homelessness among youth. Intervention programs could be family, school, or community-based, and should involve multi-level structural changes (i.e. policy changes, regulations, service creation and coordination). Future prevention studies could also look at different venues to broadly disseminate information and prevention strategies. This is, future research on youth homelessness could focus on identifying which ways are more effective to inform and intervene with adolescents, families, and communities. In other words, future research could try to answer the basic questions about youth homelessness that psychologists seem to be avoiding: What can we do to prevent homelessness among adolescents? Who should do prevention? Where? How?

References

1. Baker, A. J. L., McKay, M. M., Lynn, C. J., Schlange, H. & Auville, A. (2003). Recidivism at ashelter for adolescents: First-time versus repeat runaways. Social Work Research, 27(2), 84-93.         [ Links ]

2. Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press.         [ Links ]

3. Bronstein, L. R. (1996). Intervening with homeless youths: Direct practice without blaming the victim. Child & Adolescent Social Work Journal, 13(2), 127-138.         [ Links ]

4. Cauce, A. M., Morgan, C. J., Wagner, V., & Moore, E. (1994) Effectiveness of intensive case management for homeless adolescents: Results of a 3-month follow-up. Journal of Emotional & Behavioral Disorders, 2(4), 219-227.         [ Links ]

5. Coco, E. L., & Courtney, L. J. (1998). A family systems approach for preventing adolescent runaway behavior. Adolescence, 33(130), 485-496.         [ Links ]

6. Fondacaro, M. R., & Weinberg, D. (2002). Concepts of social justice in community psychology: Towards a social ecological epistemology. American Journal of Community psychology, 30, 473-492.         [ Links ]

7. Hammer, H., Finkelhor, D. & Sedlak, A. (2002). Runaway/Thrownaway children: national estimates and characteristics. National Incidence Studies of Missing, Abducted, Runaway, and Thrownaway Children (NISMART), October, 2002. Washington, DC: Office of Juvenile Justice and Delinquency Prevention.         [ Links ]

8. National Coalition for the Homeless (1999). Homeless youth. NCH Fact Sheet # 11. Retrieved December 19, 2003 from the World Wide Web at: http://www.nationalhomeless.org/publications/facts/youth.pdf         [ Links ]

9. Prilleltensky, I. & Nelson, G. (2002). Doing psychology critically. Making a difference in diverse settings. New York: Palgrave Macmillan.         [ Links ]

10. Ringwalt, C. L., Greene, J. M., Robertson, M. & McPheeters, M. (1998). The prevalence of homelessness among adolescents in the United States. American Journal of Public Health, 88, 1325-1329.         [ Links ]

11. Robertson, M. J. & Toro, P. A. (1998). Homeless youth: research, intervention, and policy. In L. B. Fosburg, & D. L. Dennis (Eds.), National Symposium on Homelessness Research (pp. 77-108). Washington, DC: Department of Housing and Urban Development, Office of Policy Development and Research, and Department of Health and Human Services.         [ Links ]

12. Shinn, M. (1992). Homelessness: What is a psychologist to do? American Journal of Community Psychology, 20, 1-24.         [ Links ]

13. Smollar, J. (1999). Homeless youth in the United States: description and developmental issues. In Raffaelli, M. (Ed.); Larson, R. W. (Ed.), Homeless and working youth around the world: Exploring developmental issues. New Directions for Child and Adolescent Development (pp. 47-58).         [ Links ]

14. Society for Community Research and Action (2004). The SCRA mission. (2004, February 2). Retrieved April 26, 2004, from http://www.apa.org/about/division/div27.html         [ Links ]

15. Staller, K. M., & Kirk, Stuart A. (1997). Unjust freedom: The ethics of client self-determination in runaway youth shelters. Child & Adolescent Social Work Journal, 14(3), 223-242.         [ Links ]

16. Teare, J. F., Authier, K. & Peterson, R. (1994). Differential patterns of post-shelter placement as a function of problem type and severity. Journal of Child & Family Studies, 3(1), 7-22.         [ Links ]

17. Teare, J. F., Peterson, R. W., Furst, D., Authier, K., Baker, G. & Daly, D. L. (1994). Treatment implementation in a short-term emergency shelter program. Child Welfare, 73(3), 271-281.         [ Links ]

18. Thompson, S. J., Kost, K. A. & Pollio, D. E. (2003). Examining risk factors associated with family reunification for runaway youth: Does ethnicity matter? Family Relations, 52, 296-304.         [ Links ]

19. Thompson, S. J., Pollio, D. E. & Bitner, L. (2000). Outcomes for adolescents using runaway and homeless youth services. Journal of Human Behavior in the Social Environment, 3(1), 79-97.         [ Links ]

20. Thompson, S. J., Pollio, D. E., Constantine, J., Reid, D. & Nebbitt, V. (2002). Short-term outcomes for youth receiving runaway and homeless shelter services. Research on Social Work Practice, 12(5), 589-603.         [ Links ]

21. Thompson, S. J., Safyer, A. J. & Pollio, D. E. (2001). Differences and predictors of family reunification among subgroups of runaway youths using shelter services. Social Work Research, 25(3), 163-172.         [ Links ]

22. Toro, P. A., Trickett, E. J., Wall, D. D. & Salem, D. A. (1991). Homelessness in the UnitedStates. An ecological perspective. American Psychologist, 46, 1208-1218.         [ Links ]

23. Whitbeck, L. B., Hoyt, D. R., Johnson, K. D., Berdahl, T. A. & Whiteford, S. W. (2002). Midwest longitudinal study of homeless adolescents. Baseline report for all participating agencies. Lincoln, NE: University of Nebraska, Department of Sociology.         [ Links ]

24. Yoder, K. A., Whitbeck, L. B. & Hoyt, D. R. (2001). Event history analysis of antecedents to running away from home and being on the streets. American Behavioral Scientist, 45, 51-65.         [ Links ]

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons