Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Sep 1.
Published in final edited form as: Child Youth Serv Rev. 2014 Jun 18;44:243–248. doi: 10.1016/j.childyouth.2014.05.024

Improving Youth Mental Health through Family-Based Prevention In Family Homeless Shelters

Cynthia J Lynn 1, Mary C Acri 1, Leah Goldstein 1, William Bannon 1, Nisha Beharie 1, Mary M McKay 1
PMCID: PMC4141706  NIHMSID: NIHMS612960  PMID: 25157200

Abstract

This exploratory study examines changes in suicidal ideation among a sample (N = 28) of homeless youth, ages 11-14, residing within family shelters in a large metropolitan area. Changes in suicidal ideation from pretest to posttest are compared across two group approaches to delivering HIV prevention. Youth and their families participating in the HOPE Family Program, incorporating a family strengthening approach, are compared to those receiving a traditional health education-only approach. Multivariate analyses reveal that youth in the HOPE Family Program were 13 times more likely to report a decrease of suicidal ideation. These findings indicate that health education programs integrating a family strengthening approach hold promise for positively impacting mental health outcomes for vulnerable youth.

While there has been an increased focus upon suicidal risk among homeless youth (Rew, Taylor-Seehafer, & Fitzgerald, 2001; Yoder, Whitbeck, & Hoyt, 2008a; Yoder, Longley, Whitbeck, & Hoyt, 2008b), research focusing on self-harm among youth living in family shelters is rare. Further, little is known about effective interventions to decrease thoughts of self-harm among this subpopulation of homeless youth. The purpose of this exploratory study is to investigate the prevalence of suicidal ideation, the relationship between various risk factors, and the impact of participation in family-based HIV prevention programs upon self-harm among a sample of adolescents residing in urban homeless shelters with their families. Two family-based approaches are compared. Guided by Social Action Theory (Ewart, 1991) it is expected that the program including a family-strengthening approach aimed at bolstering key family and youth processes including communication, family decision-making, parent leadership and supervision, and youth problem-solving, will have a beneficial impact on youth suicidal ideation.

Family Homelessness

Families with children make up 38% (239,403) of the total homeless population (633,782) across the United States (National Alliance to End Homelessness, 2013). This number is on the rise, despite that the population of other subgroups has remained relatively stable. In fact from 2011 to 2012, families with children were the only subgroup of homeless persons to experience an increase, resulting in 162,246 homeless children in 2012 (National Alliance to End Homelessness, 2013). Homeless families face multiple adversities. They experience the longest stays in temporary housing (e.g., emergency shelters, transitional housing, and permanent supportive housing) comparative to single adults (U.S. Conference of Mayors, 2008), and are at risk for significant levels of poverty, domestic violence, and increased involvement with the child welfare system (Cutuli et al., 2010; HUD, 2009; Stainbrook et al., 2006).

Further, studies consistently find homeless parents, who are most likely to be single women (Gewirtz et al., 2009), have high rates of mental health and substance abuse difficulties (Gewirtz et al., 2009; Zima et al., 1996). And two decades of research show that their offspring have increased likelihood of evidencing developmental delays, mental health problems, substance abuse, academic and peer-related difficulties, and involvement in the juvenile justice system (Bassuk & Rubin, 1987; Buckner, 2008; Busen & Engebretson, 2007; Haldenby, Berman, & Forchuk, 2007; Obradovic, 2010; Rhode, Noell, Ochs & Seeley, 2001; Rotheram-Borus & Futterman, 2000; Rhule-Louie et al., 2008).

Adolescent Suicide Risk

Recently, there has been an increased focus upon suicidal risk among homeless adolescents. Thoughts of self-harm and suicidal acts are highly associated with mental health disorders including major depressive disorder, posttraumatic stress disorder, and substance abuse (Yoder, Longley, Whitbeck, & Hoyt, 2008b), and are the third cause of death among youth 12 to 18 years of age (National Center for Injury Prevention and Control, 2009). In fact, exposure to violence and victimization has been linked to an increased risk for suicidal ideation among adolescents (Chemtob et al., 2011; Turner et al., 2012; Wolitzky-Taylor et al., 2009), as has youth substance abuse (Gould et al., 1998; Vermeiren et al., 2003, Wolitzky-Taylor et al., 2009).

Homeless adolescents are at significant risk for suicide. One recent study found over two-thirds of youth endorsed one or more indicators of intent to self-harm (Yoder et al., 2008a, 2008b), and a second study found approximately 35% of adolescents had thought seriously about suicide, with 12% attempting suicide on one or more occasions (Rew et al., 2001). However, there is little knowledge about suicide risk among sheltered youth living with their families, as the samples in these studies tend to be runaways, youth who were forced out of their homes, or who live on the streets. The current study extends upon this knowledge base by investigating prevalence, as well as child and family-level correlates of suicidal ideation among family shelter-dwelling youth.

Family-level Variables

Additionally, we studied the impact of a family strengthening and HIV prevention program upon suicide risk. Evidence suggests parenting factors, including lack of parental support, poor parenting practices (e.g., low monitoring, lax rules, poor problem solving), and a discordant parent/child relationship are associated with suicide risk among youth (Brausch & Gutierrez, 2010; Garber et al., 1998; Johnson et al., 2002; King et al., 1993; King et al., 2001; Princstein, Boergers, Spirito, Little, & Grapentine, 2000), yet there is little information about this relationship among shelter-dwelling families. Homeless families living in the shelter system are at particular risk for disruptive parenting practices and impairments in the parent/child relationship. Evidence suggests the stressors that accompany homelessness interfere with important parenting processes, including monitoring and supervision, provision of support, and discipline (Howard, Cartwright, & Barajas, 2009). Families living in shelters also contend with lack of privacy, relinquishing control to shelter rules and regulations, and concerns that their parenting skills are being assessed by staff, which further impairs parenting and stresses the family unit (Fraenkel, Hameline, & Shannon, 2009; Paquette & Bassuk, 2009; Howard et al., 2009). This literature, coupled with the lack of knowledge about the relationship between these processes among homeless families, led to the second aim of the study; to determine whether participation in a family-focused prevention program that aimed to enhance parenting processes and the parent/child relationship would also impact suicidal ideation among these youth.

Methods

Description of the Intervention

Data used in the current study were gathered from participants in the HOPE (HIV Outreach for Parents and Early Adolescents) Program, a study that involved a sample of young adolescents (11-14 years of age) residing within family housing shelters in New York City. All necessary Institutional Review Board approvals were obtained before undertaking the larger study. Further, in regards to mental health safeguards, it should be noted that when a youth was identified as having suicidal ideation, she/he was clinically assessed by a Masters-prepared professional, parents were involved in supporting the child, and active attempts were made to secure clinical care as indicated.

The HOPE Program study contrasts two prevention strategies. The first is the HOPE Family Program, which offers an intensive family strengthening intervention meant to build communication, parental monitoring, and supervision skills, as well as assist parents to manage stressful situations both inside and outside of the shelter. Through strengthening family functioning and promoting youth mental health, subsequent youth risk-taking behaviors (e.g., suicide risk and HIV risk) are hoped to decrease. The HOPE Family Program consists of an eight-session, weekly one-hour intervention that includes both separate and conjoined sessions for parents and youth to engage together and separately with the content material. This is done in order to allow for both caregivers and youth to be able to discuss issues freely among their peers before coming together to discuss them as a family.

In contrast, the HOPE Health Educational Program provides informational sessions pertaining to methods of prevention of HIV/AIDS and sexually transmitted infections, the effects of the use of illicit substances, and normative adolescent changes (i.e., puberty). HOPE Health consists of three 2-hour generationally separate group sessions for caregivers and youth. A social worker is present for both the HOPE Family Program and HOPE Health Program to provide clinical support to youth and parents as needed. The primary facilitators for both programs are community members with five years or more experience in HIV prevention services.

Setting and Sample

The sample of the current study (N = 28) was drawn from a larger study of 204 urban parents and their school aged children residing in NYC family homeless shelters. Unlike traditional shelters, family shelters offer multiple services including case management services, housing assistance, employment services, childcare, and other resources to facilitate returning to permanent housing (Barnes, 2004). Of these 204 families residing in the shelter, 48.5% (n = 99) were assigned to the HOPE Family Program, and 51.5% (n = 105) were assigned to HOPE Health within randomly assigned shelters with one program offered in each shelter.

In terms of racial/ethnic background, 42% (n = 86) of adult caregivers described themselves as Hispanic/Latino, 47% (n = 96) described themselves as Black, and 11% (n = 22) described themselves as a Black/Hispanic racial/ethnic mix or other. Over ninety percent (92%, n = 188) of the sample of adult caregivers was female. The average age of adult caregivers was 38.4 (SD = 6.8) years old. Eighteen percent (n = 37) of adult caregivers reported having an 8th grade education or less, 27% (n = 56) had some high school, 26% (n = 54) had a high school/GED diploma, 18% (n = 37) had some college, and 5% (n = 11) reported that they had completed college. The remaining caregivers reported either completing post college courses (2%, n = 4) or did not respond (4%, n = 8). Adult caregivers reported their marital status as being 57% (n = 116) single, 25% (n = 51) married/common law marriage, 18% (n = 37) divorced/separated/widowed. Fifty-seven percent (n = 116) of adult caregivers reported that this was their first time staying in a shelter.

Children described their racial/ethnic background as 40% (n = 81) Hispanic/Latino, 43% (n = 88) Black, and 17% (n = 35) described themselves as a Black/Hispanic racial/ethnic mix or other. Fifty-five percent (n = 112) of the sample of children was male and 45% (n = 92) female. The average age of children was 12.8 (SD = 1.2) years old.

Of the 204 youth that provided data at baseline, 198 provided data regarding suicidal ideation. Of the 204 youth at baseline, 68% (n = 139) completed a posttest survey, of these 139, 135 provided posttest data regarding suicidal ideation. Among these groups, there were 131 youth that provided overlapping pretest and posttest scores describing their suicidal ideation. Overall, the rate of suicidal ideation at baseline for these 131 youth was 21.4% (n = 28). Chi-square analysis indicated that among these 131 youth, there was not a significant difference between rates of baseline suicidal ideation between the two study groups, X2(1) = .02, p = .89. Specifically, within this group of 131 youth, the youth in HOPE Health (n = 67) evidenced a rate of suicidal ideation at baseline of 21% (n = 14), and the youth in HOPE Family (n = 64) indicated a comparable rate 22% (n = 14). The current study focuses specifically on those youth from each of the intervention conditions reporting suicidal ideation at baseline (n = 28).

Measures

Youth and family demographic characteristics

Information was collected via a demographic measure developed by the authors to gather data about the child (e.g., age, gender, and race), parental factors (e.g., educational level and employment history), and family-level factors including household composition and structure, income, and residential moves.

Shelter related characteristics

A series of questions were presented to youth in order to examine their shelter related experiences. These included, “Do you have friends at the shelter?,” “Do you feel safe at the shelter?,” and “Are there things for people your age to do at the shelter?”

Youth and family variables

While the measures employed here have previously been applied to studies with poverty-impacted families, they have only recently begun to be used with homeless youth and their families (Bannon et al., 2012; Schwartz-McGuire, in press).

Within family support

The within family support subscale has 6 items measured on a 4-point Likert-type scale (1 = strongly disagree, 4 = strongly agree). An example of an item is “I listen to what other family members have to say, even when I disagree.” The possible range of this subscale is 6 to 24, with higher scores representing a greater degree of within family support. Scores are determined by adding the values of each contributing item. The alpha coefficient for the current sample is .88.

Parent-child communication regarding “hard to discuss” topics

Parent-child communication regarding “hard to discuss” topics were also measured using subscales from the Family Assessment Measure (Tolan, Florsheim, McKay, & Kohner, 1993). This subscale consists of 7 items measured on a 4 point Likert-type scale (1 = we don't talk about this, 4 = A lot). Each item asks “How often do you talk with your parents/child about” items such as: alcohol, drugs, HIV/AIDS, having sex, sexually transmitted infections, gangs and puberty. The possible range of this subscale is 7 to 28, with higher scores representing a greater frequency of family communication. Scores are determined by adding the values of each contributing item. The alpha coefficient for the current sample is .91.

Parental monitoring and supervision

Reports of parental monitoring and supervision were measured via the Parenting Skills Questionnaire (Tolan & Gorman-Smith, 1991). This 30-item parent report contains four subscales: discipline effectiveness, positive parenting, parent involvement and supervision, and child compliance. Only parent involvement and supervision is considered here. The parent involvement and supervision subscale assesses reports of the ways parents keep track of children's whereabouts, who they are with, household rules, and how often they are left in charge of other children. This instrument has 17 items with a 4-point scale (1 = Always true, 4 = Always false). The possible range of this instrument is 17 to 68 with higher scores representing a greater degree of parental involvement and supervision. Scores are determined by adding the values of each contributing item. The alpha coefficient for the current sample is .75.

Violence Exposure

Data provided at baseline was used to measure Violence Exposure via a 9-item risk subscale from the Family Stress Scale (Tolan, Miller, & Thomas, 1988). Caregivers and youth were asked to identify whether they or another member of their family experienced a series of potentially stressful events (yes/no). Items incorporated within this subscale reflect exposure to community violence, domestic violence, and substance abuse within a respondent's household. In the current study, caregivers and youth provided data via this subscale.

Youth Substance Use

The degree of youth substance use was measured via three items from Monitoring the Future survey (Johnston et al., 1993) that taps the onset and frequency of substance use. Specifically, youth were asked “During the past month, how many cigarettes did you smoke per week?,” “During the past month, how many times did you have a drink of alcohol other than a few sips?,” and “During the past month, how many times did you use marijuana?” For each item, if youth indicated any substance use, they were coded as having used that substance (yes/no).

Youth suicidal ideation

Youth suicidal ideation was measured via a single item on the Child Depression Inventory (Finch, Saylor, Edwards, and McIntosh, 1987) in order to detect any presence of suicidal ideation. The CDI contains 27 items, each of which consists of statements where youth indicate which best describes their feelings over the past two weeks. Among these items is an option for youth to rate their thoughts concerning suicide, with the available responses being: 1) “I do not think about killing myself,” 2) “I think about killing myself but I would not do it,” and 3) “I want to kill myself.”

Analysis

Changes in within family support were computed through producing difference scores via subtracting the composite scale score of parents at posttest from their scores on the identical scale at pretest. In order to generate odds ratios, the continuous difference scores were dichotomized into positive changes in within family support (=1) vs. no or negative changes within family support (=0). Scores are dichotomized for this measure and for additional measures below in order to maximize the usefulness of skewed data.

Changes in the parent-child communication variable were computed through producing difference scores via subtracting the composite scale score of parents at posttest from their scores on the identical scale at pretest. In order to generate odds ratios, the continuous difference scores were dichotomized into positive changes in parent-child communication regarding “hard to discuss” topics (=1) vs. no or negative changes in parent-child communication regarding “hard to discuss” topics (=0).

Changes in parental monitoring and supervision were computed through producing difference scores via subtracting the composite scale score of parental report at posttest from their scores on the identical scale at pretest. In order to generate odds ratios, the continuous difference scores were dichotomized into positive changes in parental monitoring and supervision (=1) vs. no or negative changes in parental monitoring and supervision (=0).

The items measuring youth substance use were summed to create a composite variable that indicates the frequency of youth substance use over the past month (possible range 0-3). Due to low youth reports of substance use over the past 30 days, this variable was dichotomized into a factor representing a youth having used of a substance within the past 30 days (yes/no).

Youth who indicated either item 2 or 3 (i.e., item 2 “I think about killing myself but I would not do it,” and item 3 “I want to kill myself”) at baseline or posttest were coded as evidencing suicidal ideation at that time point. Youth were then coded into those who had ideation at baseline, but not at posttest (=1); and those who had ideation at baseline and at posttest (=0).

There were two steps involved in the data analysis plan. First, all independent variables were entered into single predictor logistic regression equations in order to examine which independent variables were associated with the outcome variable, approaching significance (p < .10). Next, a multivariate logistic regression model was used to examine the relationship between multiple covariates and changes in youth suicidal ideation from baseline to posttest. All variables related to the outcome variable at a probability level no greater than .10 at the bivariate level were entered into the final multivariate model, along with the variable indicating study group membership. Youth gender and reports of a first stay in the shelter system were controlled for in the final model. Checks for multicollinearity among study variables revealed no significant problems (Menard, 1995).

Results

Description of Study Variables

All independent variables are presented descriptively in Table 1, which also describes how these variables differ in relation to the outcome variable, changes in suicidal ideation. Of the 28 youth with suicidal ideation at baseline who also participated in the posttest phase of the HOPE study, 64% (n = 18) indicated changes in suicidal ideation and then subsequently indicated no ideation at posttest. The remaining 36% (n = 10) of youth that reported suicidal ideation at baseline, also reported suicidal ideation at posttest.

Table 1. Single-predictor logistic regression analysis of youth, family, and youth risk factor variables on youth changes in suicidal ideation (N = 28).

Variable Youth with Ideation
(n = 10)		 Youth Changes in Ideation
(n = 18)		 B (SE) OR (95%CI)
n % n %
Positive change in parent-child communication -.18 (.80) .83 (.17-4.01)
 Yes 4 40.0% 8 44.4%
 No 6 60.0% 10 55.6%
Positive change in within family support -1.54 (.85) .21(.04-1.14)†
 Yes 7 70.0% 6 33.3%
 No 3 30.0% 12 66.7%
Positive change in parental monitoring/supervision -.41 (.80) .68 (.14-3.19)
 Yes 6 60.0% 9 50.0%
 No 4 40.0% 9 50.0%
High violence exposure .00 (.79) 1.00 (1.00-1.00)
 Yes 5 50.0% 9 50.0%
 No 5 50.0% 9 50.0%
Substance use -2.43 (1.22) .09 (.01-.95)*
 Yes 4 40.0% 1 3.2%
 No 6 60.0% 17 94.4%
Open in a new tab
*

= (p < .05),

†

= (p < .10).

Bivariate Analysis

Predictors of changes in suicidal ideation

Table 1 presents the single predictor logistic regression analyses that examine independent variables significantly related to youth changes in suicidal ideation. Parent reports of positive changes in parent-child communication, within family support, and parental monitoring/supervision were not significantly related to the outcome variable. However, a non-significant trend is noted regarding changes in within family support as associated with changes in suicidal ideation [X2 (1, n = 28) = 3.27, p < .10]. Data indicate that parent reports of positive changes in within family support from baseline to posttest were associated with a greater likelihood (OR =.21; 95% CI=.04-1.14) of youth reporting no changes in suicidal ideation.

Additionally, data within Table 1 describe that substance use was significantly associated with changes in suicidal ideation [X2 (1, n = 28) = 4.00, p < .05]. Specifically, relative to youth who had not used a substance within the past 30 days, youth that reported at least one substance were over 11 times (OR = .09; 1/.09=11.11; 95% CI = .01-.95) more likely to have reported no changes in suicidal ideation. High violence exposure failed to reach a level of statistical significance in relation to the outcome variable.

Multivariate Analysis

Logistic regression analysis of covariates explaining changes in thoughts of suicide

The results of the multivariate logistic regression model are summarized in Table 2. Results indicated a statistically significant model [X2 (5, n = 28) = 13.71, p < .05].

Table 2. Logistic regression analysis of covariates on youth changes in suicidal ideation (N=28).
Variable B (SE) OR (95% CI)
Positive change in within family support -2.26 (1.27) .11 (.01-1.26)
Substance use -2.48 (1.57) .08 (.00-1.80)
Study group 2.61 (1.30) 13.57 (1.06-173.79)*
Open in a new tab

Model. X2 = 13.71, df = 5,

*

= (p < .05).

The strength of the association of the individual covariates with changes in suicidal ideation was examined by evaluating the statistical significance of individual coefficients and their odds ratios within the context of the full-hypothesized model. Within the full model, only study group membership (participation in either HOPE Health or HOPE Family) was significantly associated with the dependent variable [X2 (1, n = 28) = 4.02, p < .05], while substance use (significant at the bivariate level) and positive changes in within family support (included because of trend noted at the bivariate level) were not significantly related. Specifically, youth in the HOPE Family Program were over 13 times (OR = 13.57; 95% CI = 1.06-173.79) more likely to report reductions from baseline to posttest of suicidal ideation.

Discussion

Families with children constitute almost half of the rural and suburban homeless population, and more than 70% of the homeless living in urban areas (HUD, 2007). Consequently, there is a great need to understand the issues confronting this understudied population and direct programmatic efforts to reduce risks to youth wellbeing. This study attempts to close this knowledge gap by exploring suicide risk among adolescents living with their families in a shelter system, youth and familial risks associated with thoughts of self-harm, and the potential benefits of family-based health prevention programs among these youth. Social Action Theory (Ewart, 1991) guides the understanding of study results and the identification of targets of change to impact homeless youth mental health.

Results of this study showed one-fifth of all 204 youth participating in either treatment condition (n = 28, 21%) reported some level of suicidal ideation at baseline. While this number is lower than estimates found in existing studies of homeless adolescents (Rew et al., 2001; Yoder et al., 2008a; 2008b), a sizable number of youth reported psychological distress nonetheless. These findings confirm previous research involving homeless youth, and suggest there is a specific need for prevention and intervention efforts to target youth living with their families in temporary housing.

Additionally, results showed youth who participated in the HOPE Family Program evidenced a significant decrease in suicidal ideation comparative to youth who received the HOPE Health Educational Program. Although more research is necessary in order to discern which components of the HOPE Family Program were associated with decreased thoughts of self-harm, one speculation is that the intense focus on family processes, communication and coping skills may have had some impact upon the youth mental health. Prior studies (Cleverley & Kidd, 2011; Kidd & Shahar, 2008) suggest youth perceived resilience and self-esteem appear to act as protective factors against suicidal ideation among homeless youth. It is possible that family-strengthening interventions improve these characteristics in youth. As noted earlier, the type and quality of parenting, which may be stressed and negatively impacted by homelessness and the shelter experience, plays a substantial role in risk among youth. Therefore, HOPE Family's focus upon building and enhancing these parental processes may have been why youth benefited from this program and not the informational component.

Yet contrary to our expectations, there was no relationship between positive changes in suicidal ideation when family support improved. One possible explanation of this finding is that family support was assessed by the parent's self-report, not the child's perception. Therefore, perhaps youth were not concordant in their beliefs that support improved from pretest to posttest. Or, it is possible that youth were already feeling supported. The sample generally felt positively toward and supported by the shelter and staff. For example, over two-thirds (67%) reported liking staff, almost 84% had friends at the shelter, and 81% reported that staff at the shelter helped them and their family. Youth may have already felt a strong sense of support through these various venues, and were experiencing distress due to exposure to other vulnerabilities, such as the instability and strain associated with homelessness, difficulties in school or with peers, additional mental health difficulties, or parental factors such as substance use or interpersonal violence.

A final result worth noting is that youth who reported using at least one substance within the prior 30 days were 11 times more likely to report no changes in suicidal ideation relative to baseline. This finding is concordant with a larger literature that suggests the strong association between substance use and psychological need (Chan, Godley, Godley, & Dennis, 2007; Garland, Lau, Yeh, McCabe, Hough, & Landsverk, 2005), especially among the homeless adolescent population (Rhule-Louie et al., 2008; Whitbeck et al., 2000). Further, in a study by Rhode et al. (2001) examining suicidal ideation and sexually transmitted disease risk among homeless youth, results revealed youth depression was associated with STDs, and specifically that suicidal ideation was linked to high rates of intra-venous drug use. Taken together, this body of literature suggests the need to target multiple domains (e.g., mental health, substance abuse prevention, HIV/STD prevention,) in order to provide the greatest benefit to the adolescent's health.

Limitations

Several limitations must be considered when interpreting these findings. First, this sample was limited to 28 youth and represents a fairly homogenous sample within one urban city and family shelter system. This small sample size limits the number and type of statistical tests that can be performed, as well as limiting the power of the analyses to detect differences. Therefore, generalizing the findings beyond this sample must be done cautiously. Secondly, a single item measured the primary dependent variable, suicidal ideation. The inclusion of additional measures of suicide risk will strengthen future studies of this type. Additionally, as parents completed some measures without overlapping child or staff reports, it is unclear whether the youth shared the same beliefs about changes in family processes. And of note, family shelters are somewhat unique in their provision of multiple services. Thus, it is unclear whether these findings would be similar among homeless youth residing in less supportive environments. Despite these limitations, however, the results of this study offer promise to uncovering the preliminary impact of a family-focused HIV prevention program upon suicidal ideation among adolescents living in the family shelter system.

Implications for Practice and Future Research

This preliminary study yields several important programmatic and research implications, beginning with the evident need to address suicidal ideation among family shelter-dwelling youth. Although the sample was small, a concerning number of youth reported suicidal ideation. These findings, coupled with evidence that homeless youth are unlikely to utilize mental health services (DeRosa et al., 1999; Solorio, Milburn, Andersen, Trifskin, & Rodriguez, 2006), suggest the need to either deliver mental health interventions within the family shelter system, or remove barriers (e.g., cost, lack of insurance, location, transportation, low access to services) that impede homeless youth from utilizing mental health care. As a case in point, California developed a network of programs and services specifically for homeless adolescents including crisis and long-term shelters, medical and mental health treatment, employment, and educational services. Several of the agencies offered multiple services and services in multiple locations (or mobile units), thus reducing logistical and access barriers. Results of this model showed a substantial number of youth (78%) used the shelter system, and while mental health usage was low (9%), using the shelter was linked to use of the other services (DeRosa et al., 1999). These findings indicate that the shelter system may be an important factor in service utilization among homeless youth.

The findings of this study also speak to the need to consider the entire family unit in reducing the risk of suicidal harm. Indeed, homeless families are faced with multiple stressors and vulnerabilities, which are exacerbated by living in a shelter system (Fraenkel et al., 2009; Paquette & Bassuk, 2009; Howard et al., 2009). Consequently, these risks may adversely impact the youth's mental health. Yet, membership in a prevention program that focused on enhancing familial processes was associated with reduced suicidal risk. Although this was a preliminary study, these results lend support that targeting the entire family may have a beneficial impact upon risks to homeless youth.

Future research is necessary to understand which family components are most useful and directly applicable to adolescent outcomes. As noted previously, improvements in family support did not appear to impact youth psychological distress, yet involvement in the HOPE Family Program did. While it is possible and likely that other factors such as the stressors these youth experience or supports provided through the shelter influenced these findings, subsequent research would be strengthened by identifying which child, family, and external factors buffer and exacerbate psychological health among family shelter-dwelling youth. In addition, future research could examine which components of HOPE Family were protective for the youth's emotional wellbeing in order to best support the mental health of homeless adolescents. Overall, despite the small sample size and other limitations, the current findings provide beginning evidence pointing to the implementation of an education intervention that imbeds family-processes strategies, as improving the suicidality of adolescents impacted by family homelessness.

Highlights.

  • Study examines changes in suicidal ideation among homeless youth (N = 28).

  • Two groups of youth and their families are compared pretest and posttest.

  • HOPE Program youth were 13 times more likely to report decreased suicidal ideation.

Acknowledgments

Author Note: Funding for this study was provided by a grant from the National Institute on Drug Abuse, Family-based HIV & Drug Abuse Prevention for Homeless Youth (R01 DA018574). The authors gratefully acknowledge all participating families, as well as the following members of the Community Collaborative Board who had a significant role in organizing the HOPE project: Nisha Beharie, Angela Paulino, Ervin Torres, Rita Lawrence, Anna Miranda and Anita Rodriquez-Rivera.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Bannon W, Beharie N, Olshtain-Mann O, McKay M, Goldstein L, Cavaleri M, LoIacono M, Elwyn L, Kalogerogiannis L, Torres E, Paulino A, Lawrence R. Youth substance use in a context of family homelessness. Children and Youth Services Review. 2012;34:1–7. doi: 10.1016/j.childyouth.2010.12.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Barnes J. New York State's tier II family shelter system. Poster session presented to The U.S. Health Resources and Services Administration. 2004 Slides available at http://hrsa.gov/homeless/pa_materials/…/ny_barnes_presentation_ppt.doc.
  3. Bassuk E, Rubin L. Homeless children: A neglected population. American Journal of Orthopsychiatry. 1987;57(2):279–286. doi: 10.1111/j.1939-0025.1987.tb03538.x. [DOI] [PubMed] [Google Scholar]
  4. Brausch AM, Gutierrez PM. Differences in non-suicidal self-injury and suicide attempts in adolescents. Journal of Youth and Adolescence. 2010;39:233–242. doi: 10.1007/s10964-009-9482-0. [DOI] [PubMed] [Google Scholar]
  5. Buckner JC. Understanding the impact of homelessness on children: Challenges and future research directions. American Behavioral Scientist. 2008;51:721–736. [Google Scholar]
  6. Busen NH, Engebretson JC. Facilitating risk reduction among homeless and street-involved youth. Journal of the American Academy of Nurse Practitioners. 2007;20:567–575. doi: 10.1111/j.1745-7599.2008.00358.x. [DOI] [PubMed] [Google Scholar]
  7. Chan Y, Godley MD, Godley SH, Dennis ML. Utilization of mental health services among adolescents in community-based substance abuse outpatient clinics. Journal of Behavioral Health Services & Research. 2007;36(1):35–51. doi: 10.1007/s11414-007-9100-4. [DOI] [PubMed] [Google Scholar]
  8. Cleverley K, Kidd SA. Resilience and suicidality among homeless youth. Journal of Adolescence. 2011;34(5):1049–1054. doi: 10.1016/j.adolescence.2010.11.003. [DOI] [PubMed] [Google Scholar]
  9. Chemtob CM, Madan A, Berger P, Abramovitz R. Adolescent exposure to the World Trade Center attacks, PTSD symptomatology, and suicidal ideation. Journal of Traumatic Stress. 2011;24(5):526–529. doi: 10.1002/jts.20670. [DOI] [PubMed] [Google Scholar]
  10. Cutuli JJ, Wiik KL, Herbers JE, Gunnar MR, Masten AS. Cortisol function among early school-aged homeless children. Psychoneuroendocrinology. 2010;35:833–845. doi: 10.1016/j.psyneuen.2009.11.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Ewart CK. Social Action Theory for a public health psychology. American Psychologist. 1991;46:931–946. doi: 10.1037//0003-066x.46.9.931. [DOI] [PubMed] [Google Scholar]
  12. Fraenkel P, Hameline T, Shannon M. Narrative and collaborative practices in work with families that are homeless. Journal of Marital and Family Therapy. 2009;35(3):325–342. doi: 10.1111/j.1752-0606.2009.00119.x. [DOI] [PubMed] [Google Scholar]
  13. Garber J, Little S, Hilsman R, Weaver KR. Family predictors of suicidal symptoms in young adolescents. Journal of Adolescence. 1998;21:445–457. doi: 10.1006/jado.1998.0161. [DOI] [PubMed] [Google Scholar]
  14. Garland AF, Lau AS, Yeh M, McCabe KM, Hough RL, Landsverk JA. Racial and ethnic differences in utilization of mental health services among high-risk youths. The American Journal of Psychiatry. 2005;162(7):1336–1343. doi: 10.1176/appi.ajp.162.7.1336. [DOI] [PubMed] [Google Scholar]
  15. Gewirtz AH. Promoting children's mental health in family supportive housing: A community-university partnership for formerly homeless children and families. Journal of Primary Prevention. 2007;28:359–374. doi: 10.1007/s10935-007-0102-z. [DOI] [PubMed] [Google Scholar]
  16. Gould MS, King R, Greewald S, Fisher P, Schwab-Stone M, Kramer R, Flisher AJ, Goodman S, Canino G, Shaffer D. Journal of the American Academy of Child & Adolescent Psychiatry. 1998;37(9):915–923. doi: 10.1097/00004583-199809000-00011. [DOI] [PubMed] [Google Scholar]
  17. Haldenby AM, Berman H, Forchuk C. Homelessness and health in adolescents. Qualitative Health Research. 2007;17:1232–1244. doi: 10.1177/1049732307307550. [DOI] [PubMed] [Google Scholar]
  18. Howard K, Cartwright S, Barajas RG. Examining the impact of parental risk on family functioning among homeless and housed families. American Journal of Orthopsychiatry. 2009;79(3):326–335. doi: 10.1037/a0016599. [DOI] [PubMed] [Google Scholar]
  19. Johnston LD, O'Malley PM, Bachman JG National Institute on Drug Abuse. National survey results on drug use from the Monitoring the Future study, 1975-1992. Volume I: Secondary school students (NIH Publication 93-3597) 1993 [Google Scholar]
  20. Kidd S, Shahar G. Resilience in homeless youth: The key role of self-esteem. American Journal of Orthopsychiatry. 2008;78(2):163–172. doi: 10.1037/0002-9432.78.2.163. [DOI] [PubMed] [Google Scholar]
  21. King CA, Segal HG, Naylor M, Evans T. Family functioning and suicidal behavior in adolescent inpatients with mood disorders. Journal of the American Academy of Child and Adolescent Psychiatry. 1993;32:1198–1206. doi: 10.1097/00004583-199311000-00013. [DOI] [PubMed] [Google Scholar]
  22. Menard S. Applied logistic regression analysis. Thousand Oaks, CA: Sage; 1995. [Google Scholar]
  23. Miklowitz DJ, Taylor DO. Family-focused treatment of the suicidal bipolar Patient. Bipolar Disorders. 2006;8(5 pt 2):640–651. doi: 10.1111/j.1399-5618.2006.00320.x. [DOI] [PubMed] [Google Scholar]
  24. National Alliance to End Homelessness. National facts on homelessness. 2013 Downloaded April 28, 2014 from: http://www.endhomelessness.org/library/entry/the-state-of-homelessness2013.
  25. Nunez R. Family homelessness in New York City: A case study. Political Science Quarterly. 2001;116(3):367–379. [Google Scholar]
  26. Obradovic J. Effortful control and adaptive functioning of homeless children: Variable focused and person focused analyses. Journal of Applied Developmental Psychology. 2010;31:109–117. doi: 10.1016/j.appdev.2009.09.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Paquette K, Bassuk EL. Parenting and homelessness: Overview and introduction to the special section. American Journal of Orthopsychiatry. 2009;79(3):292–298. doi: 10.1037/a0017245. [DOI] [PubMed] [Google Scholar]
  28. Rew L, Taylor-Seehafer M, Fitzgerald ML. Sexual abuse, alcohol and other drug use, and suicidal behaviors in homeless adolescents. Issues in Comprehensive Pediatric Nursing. 2001;24:225–240. doi: 10.1080/014608601753260326. [DOI] [PubMed] [Google Scholar]
  29. Rew L, Whittaker T, Smith L, Taylor-Seehafer M. Sexual health-risks and protective resources in gay, lesbian, bisexual, and heterosexual homeless youth. Journal of Specialists in Pediatric Nursing. 2005;10(1):11–19. doi: 10.1111/j.1088-145x.2005.00003.x. [DOI] [PubMed] [Google Scholar]
  30. Rhode P, Noell J, Ochs L, Seeley JR. Depression, suicidal ideation and STD-related risk in homeless older adolescents. Journal of Adolescence. 2001;24(4):447–460. doi: 10.1006/jado.2001.0382. [DOI] [PubMed] [Google Scholar]
  31. Rhule-Louie DM, Baer JS, Peterson PL. Substance use and health and safety among homeless youth. Journal of Child and Family Studies. 2008;17:306–319. [Google Scholar]
  32. Rotheram-Borus MJ, Futterman D. Promoting early detection of human immunodeficiency virus infection among adolescents. Archives in Pediatric and Adolescent Medicine. 2000;154:435–439. doi: 10.1001/archpedi.154.5.435. [DOI] [PubMed] [Google Scholar]
  33. Schwartz-McGuire M, Parker G, Small L, Kim P, McKay M. Relationships between Caregiver Violence Exposure, Caregiver Depression, and Youth Behavioral Health among Homeless Families. Research on Social Work Practice. doi: 10.1177/1049731514553921. in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Solorio MR, Milburn NG, Andersen RM, Trifskin S, Rodriguez MA. Emotional distress and mental health service use among urban homeless adolescents. The Journal of Behavioral Health Services & Research. 2006;33(4):381–393. doi: 10.1007/s11414-006-9037-z. [DOI] [PubMed] [Google Scholar]
  35. Stainbrook KA, Hornik J. Similarities in the characteristics and needs of women with children in homeless family and domestic violence shelters. Families in Society. 2006;87(1):53–62. [Google Scholar]
  36. Tolan P, Florsheim P, McKay M, Kohner K. Metropolitan Area Child Study Family Intervention Manual. University of Illinois; Chicago: 1993. [Google Scholar]
  37. Tolan P, Gorman-Smith D. Metropolitan area child study parent questionnaire. University of Illinois; Chicago: 1991. Chicago. [Google Scholar]
  38. Tolan P, Miller L, Thomas P. Perception and experience of types of social stress and self-image among adolescents. Journal of Youth and Adolescence. 1988;17(2):147–163. doi: 10.1007/BF01537964. [DOI] [PubMed] [Google Scholar]
  39. Turner HA, Finkelhor D, Shattuck A, Hamby S. Recent victimization exposure and suicidal ideation in adolescents. Archives of Pediatric and Adolescent Medicine. 2012;166(12):1149–1154. doi: 10.1001/archpediatrics.2012.1549. [DOI] [PubMed] [Google Scholar]
  40. The United States Conference of Mayors. Hunger and homelessness survey: A status report on hunger and homelessness in America's Cities. Washington, D.C.: ABT Associates; 2008. [Google Scholar]
  41. U.S. Department of Housing and Human Services (HUD) The 2008 Annual Homeless Assessment Report to Congress (2008) 2009 Downloaded December 15, 2009, from http://portal.hud.gov/portal/page/portal/HUD/press/press_releases_media_advisories/2009/HUDNo.09-108.
  42. U.S. Department of Housing and Urban Development. The annual homeless assessment report to congress. Washington, D.C.: Office of Community Planning and Development; 2007. [Google Scholar]
  43. Vermeiren R, Schwab-Stone M, Ruchkin VV, King RA, Van Heeringen C, Deboutte D. Suicidal behavior and violence in male adolescents: A school-based study. Journal of the American Academy of Child & Adolescent Psychiatry. 2003;42(1):41–48. doi: 10.1097/00004583-200301000-00009. [DOI] [PubMed] [Google Scholar]
  44. Whitbeck L, Hoyt D, Wa-Ning B. Depressive symptoms and co-occurring depressive symptoms, substance abuse, and conduct problems among runaway and homeless adolescents. Child Development. 2000;71(3):721–732. doi: 10.1111/1467-8624.00181. [DOI] [PubMed] [Google Scholar]
  45. Wolitzky-Taylor KB, Ruggiero KJ, McCart MR, Smith DW, Hanson RF, Resnick HS, de Arellano MA, Saunders BE, Kilpatrick DG. Has adolescent suicidality decreased in the United States? Data from two national samples of adolescents interviewed in 1995 and 2005. Journal of Clinical Child and Adolescent Psychology. 2010;39(1):64–76. doi: 10.1080/15374410903401146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Yoder KA, Longley SL, Whitbeck LB, Hoyt DR. A dimensional model of psychopathology among homeless adolescents: Suicidality, internalizing, and externalizing disorders. Journal of Abnormal Child Psychology. 2008b;36:95–104. doi: 10.1007/s10802-007-9163-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Yoder KA, Whitbeck LB, Hoyt DR. Dimensionality of thoughts of death and suicide: Evidence from a study of homeless adolescents. Social Indicators Research. 2008a;86:83–100. [Google Scholar]
  48. Zima BT, Wells KB, Benjamin B, Duan N. Mental health problems among homeless mothers: Relationship to service use and child mental health problems. Archives of General Psychiatry. 1996;53:332–338. doi: 10.1001/archpsyc.1996.01830040068011. [DOI] [PubMed] [Google Scholar]

RESOURCES