PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Issues Ment Health Nurs. Author manuscript; available in PMC 2010 June 30.
Published in final edited form as:
PMCID: PMC2894524
NIHMSID: NIHMS200981

Enabling Older Homeless Minority Women to Overcome Homelessness by Using a Life Management Enhancement Group Intervention

Olivia G. M. Washington, PhD, PMHCNS-BC, NP, LPC
College of Nursing/Institute of Gerontology, Wayne State University, Detroit, Michigan, USA
David P. Moxley, PhD
School of Social Work, University of Oklahoma, Norman, Oklahoma, USA
Jacquelyn Y. Taylor, PhD, PNP-BC, RN

Abstract

This paper describes the importance of a life management enhancement (LME) group intervention for older minority women in developing personal control and self-confidence in social relationships as they overcome homelessness. Women in the treatment group showed significantly greater personal control and higher levels of self-confidence following the six-week intervention than women in the control group. Increasing personal control and developing self-confidence in social relationships can help individuals achieve desired outcomes as a result of their actions, efforts, and abilities. These attributes can help women increase and sustain appropriate coping methods and overcome homelessness.

As women become a greater presence among the homeless, African American women continue to be disproportionately represented (National Coalition for the Homeless, 2007). The number of older homeless people approaching 50 years of age is expected to increase dramatically as unmet demands for affordable housing continue unabated and increasing numbers of baby-boomers reach older adulthood (Cohen, 1999; Rosenheck, Bassuk, & Salomon, 1999). When these women are left alone either through the death of a spouse or divorce or experience financial problems from the loss of their jobs, negative consequences can accumulate and the loss of housing can ensue. This possibility is heightened for older African American women because they are more likely than their male counterparts to live alone or be unmarried (Bierman & Clancy, 2001).

A variety of factors (e.g., poverty, change in marital status, unanticipated circumstances, etc.) can affect older minority women's ability to overcome homelessness. The issues they face also can result from their diminished status, which is the association between risks of vulnerability and social forces (e.g., social status, race, and gender) that combine to produce marginality (Ryff, Keyes, & Hughes, 2003). Diminished status can produce negative life outcomes (e.g., lack of health care, adequate housing, appropriate employment, etc.) for older women who may lack the required supports to resolve them successfully. The accumulation of negative life outcomes can result in additional predicaments as the women lose control over health, appearance, and motivation (Washington, 2005), further diminishing their status. When women become vulnerable to negative life outcomes that emerge unexpectedly, the results can be overwhelming. The diminished status of vulnerable women suggests that they have few claims to necessary resources, and there is little societal obligation to supply these resources or make them readily available at an adequate level. According to Freddolino, Moxley, and Hyduk (2004), when a trigger event occurs (e.g., a house fire compounded by inadequate income or insurance), women may take refuge on the streets, move into homeless shelters, or reside temporarily with family members who may not have the means to sustain them. Vulnerable populations may possess social, physical, or biological characteristics that those in mainstream society do not value, leading to stigmatization, discrimination, or dehumanization. Because of the lack of social supports, benefits, and rights, Freddolino et al. suggested that vulnerable populations may become marginalized and may lack access to life-sustaining resources, such as housing and health care.

Interventions are needed to help homeless women develop skills in building social relationships and increasing personal control over their lives. Moreover, the effectiveness of these interventions needs to be examined through research. The purpose of this study is to compare a Life Management Enhancement (LME) group intervention and a non-LME group intervention on their effectiveness to help older African American homeless women develop personal control over their lives, become confident in social relationships, and overcome homelessness.

OVERCOMING HOMELESSNESS AS A RECOVERY PROCESS

Social class has been strongly related to the triangulation of race, poverty, and gender-based threats to African American women's well-being. African American women experience this triple threat more than men or women of other racial or ethnic groups (Brown, 2003). African American women live in a society that frequently makes negative social comparisons about them and emphasizes how unequal they are (especially those who have low income or are otherwise socioeconomically disadvantaged) (Belle & Doucet, 2003). Poverty in U.S society is deeply discrediting. Ross (2000) stressed that African Americans are the most impoverished ethnic group in U.S society, and Freddolino et al. (2004) asserted that this group generally is assigned a diminished social status within the larger society—a society that does not perceive itself as having to honor any obligation to help these women.

Some African American women's vulnerability to social forces reach a magnitude from which they cannot extricate, or insulate, themselves. When multiple inequities and stressors challenge them concomitantly (e.g., poverty, loss of a living wage, aging, lack of affordable housing, etc.) combined with being a member of a disadvantaged social class, they are at greater risk for losing their housing (Belle & Doucet, 2003; Radley, Hodgetts, & Cullen, 2006). Large numbers of African Americans also are exposed to the cumulative toll of segregation, exclusion, and various forms of discrimination and racism over their lifetimes. In a culture that promotes expectations of achievement relative to merit, discrimination that results in decreased economic and social status can increase frustration and anger (Belle & Doucet, 2003).

Reinforcement of diminished status imposes a considerable burden that can threaten African American women's overall health and functioning and must be addressed (Freddolino et al., 2004). In this paper, vulnerability is defined as having few environmental and personal resources, manifested by a lack of necessities, and supports the need for sustained advocacy. Vulnerability often leaves people exposed to conditions that can lead to negative outcomes, for example, damage to health and even death (Belle & Doucet, 2003; Gelles, 1996). Vulnerability can occur across all stages of the homeless condition, with impediments emerging at different times with different degrees of intensity (Freddolino et al., 2004).

Homeless women need to overcome circumstances that contributed to their loss of a permanent residence. Preconceived notions of recovery are changing and new perspectives that identify recovery as more than a set of outcomes are emerging. According to some investigators (Jacobson, 2001), recovery is a distinctive change process that is highly personal, subjective, and phenomenological. The medical concept of recovery has changed from management of symptoms and reduction of negative outcomes to a concept that supports the healing stages through which people move. The interaction of many internal and external factors, and reclaiming control over these factors or conditions, can contribute to the recovery process. This paper shares findings that reflect how 40 homeless, older, African American women, participating in a cognitive-behavioral LME group intervention, incorporated a compendium of intervention tools to develop new expectations, increase personal-control, and build self-confidence that could be used to improve functioning and productivity (Cipher, Clifford, & Roper, 2007; Pack-Brown, Whittington-Clark, & Parker, 1998; Washington & Moxley, 2001). The LME intervention may prove useful in helping women accelerate their movement out of homelessness.

OVERVIEW OF HOMELESSNESS AMONG MINORITY WOMEN

Homelessness is a multifaceted problem, resulting from a variety of unmet needs. To fully understand homelessness, a need exists to identify and examine subtypes, such as older African American women, who may be over-represented in the homeless population. Their reasons for homelessness and its effects on their general well-being, have received little attention in research (Radley, Hodgetts, & Cullen, 2006; Rosenheck et al., 1999). Minority women (e.g., African American, Hispanic, and Native American women) may be impacted at earlier periods in their lives by antecedents of homelessness such as, unemployment; minimum wage work; sporadic employment; poverty; social, structural, and economic inequities (Proctor & Dalaker, 2002; Williams & Williams-Morris, 2000). However, African American women are affected disproportionately by these antecedents, resulting in a chronic lack of opportunities and limited access to resources (Belle & Doucet, 2003; Ross, 2000).

According to Ross (2000), the greatest challenge for older African Americans may be poverty, as they have the highest poverty rate of all races. More than 30% of older African Americans live in poverty compared with 10% of elderly Whites. Older women who are at risk for poverty often live alone and are more likely to become homeless (Hovland, 2001). Although beyond the scope of this paper, unless the conditions that set the stage for such negative outcomes as poverty, unemployment, inadequate education, and so forth are effectively addressed, poverty and all of its consequences are likely to continue among African American women. As a consequence, African American women tend to lag behind other groups in terms of health gains and life expectancy. This situation has led to a widening of the gap in health between minority and majority populations (U.S. Department of Health and Human Services, 2000).

Cohen and Crane (1996) observed that many women, regardless of race or ethnicity, reported being homeless for the first time in their mid-fifties, whereas men reported becoming homeless in their mid-forties. This difference is likely due to older women's life styles (e.g., maintaining a greater number of social ties, higher levels of family support, lower levels of alcoholism and drug abuse, as well as fewer incidents of criminal behavior). Compared to other ethnic groups, African American women approaching 50 years of age may be at greater risk for becoming homeless. At these ages, they generally do not qualify for entitlements (e.g., Social Security, etc.), particularly those available to people with disabilities, serious mental illness, or ones that are granted based on a person's employment history. The National Coalition for the Homeless (2005) reported that the increased cost of living and the declining availability of affordable housing contribute to this group's homelessness.

The number of homeless women 50 years of age and older is expected to increase dramatically as homelessness and the unmet demand for affordable housing continues unabated, and increasing numbers of baby-boomers reach older adulthood (Cohen, 1999; Rosenheck et al., 1999). Finding and keeping employment can be challenging for these women as they may lack the resources, skills, and education required for jobs that pay enough to allow older women to become self-sustaining. The help they require to obtain job training and education to qualify for these types of jobs may be elusive, because neither services for aging people nor programs for the homeless are designed with these women in mind (Kisor & Kendal-Wilson, 2002; Wenzel, 2006). Without required skills, only part-time, low-wage, sporadic employment may be available. Due to the harsh living conditions associated with their homeless situations, many African American homeless women may lose control over their health, which could result in their inability to work (Washington, 2005).

Their physical health, aggravated by poor nutrition and harsh living conditions, can induce premature aging and increase poor health outcomes, resulting in these women resembling those 8 to 20 years older (Washington, 2005). As these individuals experience declines in their health and resources, they may become vulnerable to changes in the housing market that can displace vulnerable individuals disproportionately (Belle & Doucet, 2003). This displacement can result in fragmented communities, with rising housing costs that result in a large homeless class.

Homelessness for older African American women may be the culmination of a process that involves exposure to gender-based disadvantages in family life, work, and welfare. In general women earn less, work fewer years, are more likely to work part-time, and are less inclined to participate in pension plans. A majority of the time spent out of the labor force is devoted to their role as caregivers for children and elderly relatives (Population Resource Center, 2005). Consequently, as they grow older, the impact of family dissolution through separation, widowhood, or divorce can have grave consequences for minority women. Women who were socialized to be homemakers may have been dependent upon their spouses' incomes. Their means of support may have been lost if the marriage was dissolved. Compared to white women, nearly twice as many (40%) single black and Hispanic women 65 years of age and older live below the poverty line (Population Resource Center, 2005). According to the Center for Disease Control (CDC, 2002), both marriage and cohabitation are likely to fail in poor neighborhoods, where black women are more likely to live. Black women also are less likely to get and remain married. After ten years of marriage, black women have the highest rate of divorce (47%) and the lowest rate of remarriage (32%) than either white or Hispanic women (Bramlett&Mosher, 2001; CDC, 2002). Higher divorce rates or separation frequently can lead to increased hardships for women because of poverty or a reduced standard of living.

Women also can find themselves homeless when confronted with various crises (e.g., eviction, sudden loss of employment, crime, lack of insurance, and/or costly illness (Belle & Doucet, 2003; Hovland, 2001). Other factors contributing to this group's homelessness include: substance abuse; physical illness; fragile and unstable support systems that cannot offer stable housing; and urban dispossession. Although many homeless people desire to work and actively seek work, data appear to support the assertion that Blacks suffer from unemployment more than Whites (Williams & Jackson, 2000). Lower socioeconomic levels that are experienced at each stage of an African American's life course are associated with lower wages. Large numbers of African Americans experience financial difficulties due to lower incomes related to undereducation and underemployment that produce less accumulated wealth and purchasing power, as well as diminished access to quality healthcare over the life course (Rooks & Whitfield, 2004). The stress associated with social inequities (e.g., racism and discrimination) that is pervasive in the lives of African Americans (Belle & Doucet, 2003) is often managed by using a coping mechanism described as “John Henryism.” This coping mechanism is manifested by a tendency to work harder and exert more effort in response to stress (Rooks & Whitfield, 2004), which can erode health and result in serious illness.

John Henryism, as a coping mechanism, may be adaptive and can lead to positive outcomes when sufficient financial, interpersonal, and material resources are available (Rooks & Whitfield, 2004). Conversely, people who possess limited education, financial, and other resources are likely to experience negative consequences from John Henryism. High effort with a single-minded objective to succeed is more likely to result in anger, frustration, and negative social consequences (i.e., social and emotional disengagement and failure), as well as poor health for those individuals. Behavioral responses to experiences with racism are evident in people turning to alcohol or other substances to diminish their anger (Clark, Anderson, Clark, & Williams, 1999). Krieger's (1990) findings indicated that black women who generally accepted and repressed their feelings of unfair treatment tended to be at greater risk of high blood pressure, in contrast with those who either took action or disclosed their feeling to others.

THEORETICAL MODEL

The Comprehensive Health Seeking and Coping Paradigm (CHSCP; Nyamathi, 1989; Nyamathi, Wenzel, Keenan, Leake, & Gelberg, 1999) is a client-oriented, multidimensional framework that has shown promise with African American women. The CHSCP provides a framework for initial and ongoing assessment of factors that can contribute to homelessness, make people vulnerable, and negatively affect their health outcomes, as well as their subsequent use of services (Nyamathi et al., 1999). This model can be used to guide research and practice that focus on altering, supporting, or enhancing health-seeking and coping goals. These goals include the thoughts and behaviors that individuals engage in when confronting life's crises and overcoming health threats. The CHSCP model include situational and personal resources, as well as sociodemographic factors. In addition, nursing goals and strategies in CHSCP incorporate cognitive appraisal, client health goals, health-seeking and coping behaviors, perceived compliance, and perceived coping effectiveness to attain immediate and long-term health outcomes. Situational factors (i.e., environmental constraints, length and frequency of homelessness, sequence of events leading to homelessness, and past experiences or adverse circumstances) impact most participants' environments and influence their health-seeking and coping behaviors (Nyamathi, 1989; Nyamathi et al., 1999). Personal factors in the CHSCP include perceived needs and objectively evaluated needs of the target population related to health conditions. These factors can involve many additional elements (e.g., individual's beliefs, values, and commitments) that affect efficacy and well-being on a cognitive, emotional, and physical level (Nyamathi, 1989; Nyamathi et al., 1999).

Berrenberg (1987) asserted that personal control is a component of self-efficacy expectations that allows people to perceive that desired outcomes are the result of their actions, efforts, and abilities. Assessing the degree of perceived control can help clinicians identify people who are at risk for ineffective health-seeking and coping behaviors. Following this assessment, clinicians can develop interventions to enhance coping and reduce threats to health (Nyamathi, 1989). Goodman, Saxe, and Harvey (1991) noted that loss of personal control associated with homelessness can deplete people physically, psychologically, and socially resulting in an increase in their allostatic load. Loss of personal control can lead to increased feelings of powerlessness and helplessness, as well as a pervasive generalized sense of passivity. Macer (2006) asserted that low socioeconomic status can add to the allostatic load, and can have a negative influence on health outcomes (e.g., increased blood pressure, emotional instability, etc.). Conversely, personal control, a sense of being in command of one's life, can have a stronger influence on positive health outcomes than use of appropriate behavioral control actions (e.g., dieting and exercise). Using the CHSCP model as the basis for the present study, the researchers developed a cognitive-behavioral group psychotherapy intervention to increase older homeless African American women's sense of personal control. Making changes in beliefs regarding personal control also may facilitate the use of appropriate coping strategies (Epel, Bandura, & Zimbardo, 1999).

THE RELEVANCE OF GROUP WORK TO SUCCESSFUL EMERGENCE FROM HOMELESSNESS

Older homeless minority women encounter many challenges during their emergence from homelessness. Group work can help participants recognize the commonalities that they share. This arrangement creates opportunities to build alliances, helps the women to exercise reciprocal affirmations, and fosters mutual respect. Socially-supportive relationships reduce vulnerability to stress, depression, and physical illness (Bandura, 1997). These relationships can enhance personal efficacy and bolster feelings of personal control and mastery that may improve one's coping resources, sense of well-being, and subsequently reduce stress and its negative outcomes. The benefit of supportive relationships also can be manifested in perseverance and modeling effective attitudes and strategies for managing problems, as well as providing motivation and resources for effective coping. Social relationships also are governed by norms that regulate human behavior and alter practices associated with social systems, and can result in enhanced health outcomes (e.g., community-based health promotion programs focused on disease prevention and risk reduction).

While group work has existed for almost a century, women's groups have existed for substantially less time, with few empirical research studies published on the outcomes of these groups (Horne, 1999). However, women's groups (e.g., self-help, feminist therapy, support, Internet, member-led, spiritual, and time-limited groups) have made major contributions to the body of literature on group work. Some women's groups have focused on individual, interpersonal, community, and social change, and have been effective in decreasing women's isolation. Nevertheless, certain major limitations still exist. Understanding how mental and physical health care professionals systematically measure, understand, and document the efficacy of the impact of group work with women is important.

Goldberg and Simpson (1995) addressed methodological weaknesses in research on groups by employing a modified training group model that adapted small group process to a larger group of 30 to 40 African American and Hispanic adult male and female clients. They established structure; clarified roles, purpose, content and process; and focused on group members' strengths and positive qualities instead of pathology. The group psychotherapy experience explored in the current study used the small group process, which has been found to be effective with African American women (Carter, Sbrocco, Gore, Marin & Lewis, 2003). This women's group was designed to foster strong bonds, affirm personal qualities, and augment support. Through the use of a homogeneous group design, the unequal power dynamics inherent in traditional mixed groups was avoided and closeness and self-disclosure of personal feelings and strong emotions (e.g., anger, fear, etc.) were encouraged (Horne, 1999).

Addressing additional concerns, Seligman (1995) critiqued the effectiveness of psychotherapy, concluding that the ideal study should combine the best features of the survey method, be conducted in a naturalistic setting, and measure multiple outcomes, including both improvement and global gains (e.g., productivity, mood, growth, interpersonal relations, etc.). In addition to being methodologically stringent, an effective psychotherapy modality could produce clinically meaningful results and be cost effective.

DESIGN AND PURPOSE OF THE GROUP WORK INTERVENTION

The LME group process used in the present study was designed to alter behavior by changing perceptions and thinking, using systematic procedures. This process employed a focused, randomized, controlled cognitive-behavioral program that included didactic and interactional components. The intervention offered opportunities for learning a variety of strategies designed to address the pervasive effects of homelessness. Tangible, emotional, and informational support and assistance were provided to participants in the experimental group. Several instruments were used to measure the effectiveness of the intervention in reducing the harmful effects of homelessness on women's physical, mental/emotional health outcomes, and social relationships. The application of group strategies can provide participants with more energy to expend on coping and recovering from homelessness (Goldberg & Simpson, 1995).

The group work offered participants a safe environment and opportunities to try out and practice new behaviors with the support of the members and group leader (Bieling, McCabe & Antony, 2006). Each participant contributed, was heard, and experienced the impact of the group on their beliefs and behaviors. This activity allowed participants to increase available options and strategies to alter and enhance their health-seeking and coping behaviors (Nyamathi, 1989). Group therapy has demonstrated usefulness in helping reduce isolation and loneliness in vulnerable populations. Social support provided by the group was beneficial in enabling individuals to perceive their situations as less overwhelming, while enhancing reassurance and security that could foster the use of adaptive coping (Thompson et al., 2000).

The purpose of the LME intervention in this study was to help participants increase personal control over their lives and build confidence in social relationships that could facilitate their overcoming homelessness. The research design enabled a comparison between older African American women who participated in the LME intervention and those who participated in an alternative group experience.

The hypotheses that were addressed in this study were:

  1. Older African American homeless women who participated in a life management enhancement group intervention to help them overcome homelessness differ in their beliefs of their personal control over their lives than women in the untreated control group.
  2. Older African American homeless women who participated in a life management enhancement group intervention to help them overcome homelessness differ in their perceptions of benefits of social relationships than women in the untreated control group.

METHOD

Participants

Homeless African American women who were 50 years of age and older served as the target population for the present study. Additional criteria for participation in the groups included being able to understand and respond to interview questions; undiagnosed with a psychotic disorder; oriented to time, place, and person; and being homeless for at least one month at the time of the study. The Mini Mental Status (Exam MMSE; Crum, Anthony, Bassett, & Folstein, 1993) examination was used as a screening instrument to assess respondents for orientation to time, place, recall ability, memory, and simple verbal and written communication ability.

Participants were contacted using advertisements left at eight community sites, including three homeless shelters, one health clinic for homeless people, two warming centers, one community mental health center, and one feeding center offering services to homeless people; direct person-to-person solicitation was also used. Prior to contacting potential volunteers, and following approval by the Wayne State University Institutional Review Board, administrators of these facilities were contacted and the PI met with appropriate personnel to gain support for the study. Following these meetings, flyers advertising the study were posted in suggested locations. Once all protocol requirements were satisfied, baseline data were collected using a semi-structured 35-item demographic interview and three additional instruments.

Potential participants were screened for eligibility at the eight sites. Interviews were conducted by five baccalaureate-prepared health care professionals (four nurses and one social worker) as well as one senior level bachelor of science nursing student and one upper level student majoring in psychology at each community site until recruitment goals were met. Of the seven interviewers, six were African American and one was Caucasian. Before beginning the interview, the interviewers received a detailed training manual containing the protocol, study forms, and decision rules. They participated in an 8-hour training program conducted by the principal investigator to promote standardization and reduce interviewer effect. To test for inter-rater reliability as a means of assuring consistency across the interviewers, all of them watched a video-taped interview and recorded their responses on the interview protocol. The results of the Kendall's coefficient of concordance (.98) used to examine the responses provided support that the raters were recording data reliably.

A volunteer sample of 76 African American women aged 50 years and older, drawn from selected urban community sites, met the study criteria. Forty were randomly assigned to the treatment condition and 36 to the control condition. Homeless women in the treatment condition participated in 12 group intervention sessions (12 sessions) where they learned to apply cognitive and behavioral techniques to improve personal agency and perceived control, as well as to increase confidence in social relationships and to reduce the effects of psychological trauma attributable to homelessness. Data were collected at baseline, post-treatment, and 3-month follow-up.

Procedures

Each of the 76 women in the study were assigned to one of seven interviewers. The interviewers completed the Interview Protocol and the three study instruments with the women individually prior to beginning their participation. The women were randomly assigned to either the treatment or control group after completing the initial data collection. At the end of the treatment intervention, the same interviewers completed the posttest study instruments with the women to whom they had been initially assigned. The researcher who acted as the group facilitator was not involved with either the pre- or posttesting and interviewers were blinded as to which women were in the treatment or control groups.

Each of the 40 women in the treatment group participated in one of six LME groups, which formed the treatment condition of the randomized experiment. Each group met for 90 minutes of cognitive-behavioral group therapy twice a week for six weeks (12 sessions). The cognitive/behavioral intervention concluded in the sixth week, with Self-Image Enhancers providing one session of service (i.e., hand and feet soaks, massage, and a mini back massage) for each participant. This noninvasive care of the participants' back and appendages was intended to reduce stress, relieve pain, and return energy to the body increasing their physical well-being. Touch provided through massage therapy can be therapeutic for people with limited opportunities for physical contact (i.e., homeless people, and patients without intimate family or friends). Potential effects of this treatment included: feelings that someone cared for them, enhanced self image, and higher levels of self-efficacy. In addition, this intervention has a low risk of adverse effects (Vickers & Zollman, 1999). Following completion of each group program, the participants of that particular treatment group also attended a formal end of program session. The cumulative effects of these therapeutic interventions were intended to increase personal efficacy and control, reduce effects of psychological trauma resulting from circumstances associated with their homelessness, and prepare them for possible reintegration into their communities.

To provide consistency across the separate six treatment conditions, the facilitator adhered to plans for each group session as outlined in the treatment manual. Strategies and activities detailed in the manual had been used in previous group interventions and have been found to be effective for use with older homeless African American women (Washington, 2001; Washington & Moxley, 2001). The intervention goals were intended to alter health-seeking behaviors of older homeless African American women and improve their coping responses to negative events (Nyamathi, 1989). Each of the 36 women in the control group participated in one of six non-LME groups that were held at the same time during the six-week period as the LME groups.

Women in both groups completed the study instruments three times: prior to beginning the treatment intervention, at the completion of the intervention, and three months following the intervention. The women in both groups received $30.00 at each data collection point for a maximum of $90.00.

Instruments

Four instruments were used in this study. A 35-item Demographic Profile Survey developed specifically for this study provided information on participants' personal characteristics and some of their experiences with homelessness. The Mini Mental Status Exam (MMSE; Crum et al., 1993) was used as a screening instrument to determine the cognitive ability of the women to participate in the semi-structured interviews. The Belief in Personal Control Scale (BPC; Berrenberg, 1987) was used to determine participants' sense of being in command of their lives. The Interpersonal Dependency Inventory (IDI; Hirschfeld et al., 1977) was used to measure participants' social efficacy, or effectiveness, in social relationships.

The 35-item Demographic Profile Survey was developed to obtain information directly from the older, homeless African American women using a semi-structured interview format. The items were categorized into three groups: personal characteristics, physical and mental health information, and drug and alcohol information. The response formats for the questions included forced-choice, fill-in, and short answer. Prior to beginning the study, the survey was reviewed by a gerontologist, as well as shelter and warming center administrators who worked with homeless people. A suggestion to add additional space for elaborating on participants' responses was incorporated into the final survey.

The MMSE (Crum et al., 1993) is considered to be a “gold standard” screening tool for dementia. Although it is not used in the present study to diagnose dementia, it can be used to assess the degree of memory and cognitive patterns. The test is comprised of a number of questions that measure: (a) orientation, (b) calculation, (c) recall, and (d) language skills and can be completed in 15 to 20 minutes. The maximum score is 30. Scores less than 23 indicate mild cognitive impairment; scores less than 18 indicating severe cognitive impairment. Women with scores 18 or higher were included in the study (Tombaugh & McIntyre, 1992). The MMSE has been tested for internal consistency, with an alpha coefficient of .96 obtained by a mixed group of medical patients. Test-retest reliability results ranged from .80 to .95 for both impaired and non-impaired individuals who completed the MMSE at two months or shorter intervals. To determine the validity of the MMSE, Tombaugh and McIntyre reported on the sensitivity and specificity of the MMSE, with 87% of participants correctly classified, supporting the sensitivity of the MMSE. Specificity for elderly participants without a clinical diagnosis was 62%. In addition, correlations were obtained between scores on the MMSE and other cognitive screening tests. The correlations ranged from .70 to .90 using the tests administered to a diverse group of participants. This instrument has been used with many ethnic groups, including African Americans.

The BPC (Berrenberg, 1987) is a 45-item scale of perceived sense of personal, exaggerated, and God-mediated control. These subscales measure: (a) general external sense of control—the extent to which a person believes his or her outcomes are self-produced (internal) or produced by fate or others (externally), (b) God-mediated control—the belief that God can be enlisted in the achievement of outcomes (distinguishing among people who believe that they have no control over their outcomes and those who believe they control outcomes through God), (c) exaggerated sense of control—an extreme and unrealistic belief in personal control (Berrenberg, 1987). The items are rated using a 5-point Likert-type scale that ranges from 1 for “never true” to 5 for “always true.” The scores for each subscale are obtained by summing items on the subscale and dividing by the number of items to obtain a mean score that reflects the original unit of measure. Higher scores indicate an internalized sense of control, a more exaggerated belief in control, and less belief that God or a powerful agent is a mediator of control. Alpha coefficients, as a measure of reliability, ranged from .85 for general external control to .97 for God-mediated control. Test-retest correlations at four-week intervals ranged from .81 for general external control to .93 for God-mediated control. According to Berrenberg (1987), validity was established by correlating scores on the BPC with other measures (i.e., Internal-External Locus of Control Scale [Rotter, 1966] Taylor Manifest Anxiety Scale [Taylor, 1953], Janis-Fields Feelings of Inadequacy Scale [Janis et al., 1959]) that were considered to be psychometrically sound measures of constructs that were related to personal control. The correlations with these scales supported the validity of the BPS.

The IDI (Hirschfield et al., 1977) is a 48-item scale used to measure thoughts, behaviors, and feelings that focus on the need to associate with valued people using three subscales: “emotional reliance on others,” “lack of self-confidence,” and “assertion of autonomy.” Interpersonal dependency was based on three theories: psychoanalytic theory of object relations, social learning theory of dependency, and ethological theory of attachment. Split-half reliabilities were obtained for each of the three subscales. The resultant coefficients were .87 for emotional reliance on others, .78 for lack of social self-confidence, and .72 for assertion of autonomy. The IDI has good concurrent validity, with emotional reliance on others and lack of social self-confidence significantly related to the three subscales on the Symptom Checklist 90 (Derogatis, 1999): anxiety, interpersonal sensitivity, and depression. The IDI is able to distinguish between people with psychiatric disorders and a normal control group. However, the statistically significant correlations between the IDI and the MMPI social desirability scale may be indicative of response bias, with respondents answering in ways they feel are socially desirable (Hirschfield et al., 1977).

FINDINGS

Women in the experimental group had a mean age of 52.90 (SD = 3.46) years, with a range from 50 to 68 years. In comparison, the ages of women in the control group (M = 54.75, SD = 5.62) ranged from 50 to 80. The largest group of women in the experimental group had less than a high school education (n = 15, 37.5%), while those in the control group had completed high school (n = 15, 41.7%). Most women in the treatment (n = 37, 92.5%) and control groups (n = 36, 100.0%) were not married. The majority of the women in both groups (n = 65, 85.5% reported they had children, with 8 (22.9%) women in the experimental group and 8 (24.2%) women in the control group indicating they had dependent children for whom they were caring. See Table 1.

TABLE 1
Demographic Characteristics of Women in the Study (N = 76)

The length of time homeless (in months) ranged from 1 to 120 months, with a mean of 24.95 (SD = 51.49) months for women in the experimental group. Women in the control group had been homeless for a mean of 17.44 (SD = 27.23) months, with a range from 1 to 120 months. The largest number of women in the treatment (n = 22, 55.0%) and control (n = 16, 44.4%) groups had been homeless one time.

After covarying out the effects of the pretest measurements for the three subscales measuring “belief in personal control,” a statistically significant result was obtained on the multivariate analysis of covariance (MANCOVA) for the posttest measures of “belief in personal control,” F (3, 67) = 5.21, p .003. In examining the univariate Ftests, one subscale, “exaggerated sense of control” differed significantly between the treatment (M = 2.13, SE = .09) and control groups, (M = 2.63, SE = .10); F (1, 69) = 13.49, p < .001. The two subscales, “general external control” and “God mediated control” did not differ significantly between the two groups. As lower scores on “exaggerated sense of control” reflected a more moderate sense of control over their lives, this finding provided support that women in the intervention group were more likely to have an appropriate sense of control over their lives. See Tables 2 and and33.

TABLE 2
Multivariate and Univariate Analysis of Covariance for Belief in Personal Control
TABLE 3
Means and Standard Errors for Belief in Personal Control as a Function of Group Assignment

A second MANCOVA was used to determine if differences existed between the experimental and control groups on “interpersonal dependency.” The pretest scores on the three subscales (“emotional reliance on others,” “lack of self-confidence,” and “assertion of autonomy”) were used as covariates in this analysis. The results of the MANCOVA provided no evidence of a statistically significant difference when the three subscales were taken as a group F (3, 67) = 1.90, p = .138. However, when the three subscales were examined separately, a statistically significant difference was found for “lack of self-confidence,” F (1, 69) = 5.40, p = .023. Women in the experimental group (M = 1.99, SE = .07) had significantly lower scores than women in the control group (M = 2.21, SE = .07). Lower scores on this subscale indicated higher levels of confidence. See Tables 4 and and55.

TABLE 4
Multivariate and Univariate Analysis of Covariance for Interpersonal Dependency
TABLE 5
Means and Standard Errors for Interpersonal Dependency as a Function of Group Assignment

Pearson product moment correlations were used to determine if a relationship existed between the subscales measuring “belief in personal control” and “interpersonal dependency.” Using the posttest scores, statistically significant correlations in a negative direction were found between “general external control” and “emotional reliance on others,” r (38) = −.35, p = .031; lack of self-confidence, r (38) = −.35, p = .029. Similar results were obtained for the control group, with the correlations between “general external control” and “emotional reliance on others,” r (36)= −.55, p = .001 and “lack of self-confidence,” r (36) = −.65, p < .001. The remaining correlations were not statistically significant. See Table 6.

TABLE 6
Pearson Product Moment Correlations—Belief in Personal Control and Interpersonal Dependency as a Function of Group Membership

DISCUSSION

Homelessness is a growing concern for all elements of society. When particularly vulnerable groups, such as older African American women, become homeless, resources generally are not available to help them. In becoming homeless, their sense of personal control over their lives and their interpersonal relations, including those with significant others, may be negatively affected, resulting in diminished self-care. When combined with the degraded conditions to which homeless women are exposed, serious negative health outcomes can ensue.

Older homeless women's ability to care for themselves during periods of homelessness and traumas associated with homelessness suggested that the quality of social relationships may be an important factor in helping them cope with adverse circumstances. Confidence in quality social relationships along with the social interactions and support that they produce may impact individual's coping resources and coping strategies as well as alleviate negative reactions to stress (Bandura, 1997; Macer, 2006). A sense of personal control and mastery can assist people in exercising control over problems by giving them the impetus to take effective action, which can be important for women negotiating an exit out of homelessness.

Women who participated in the cognitive-behavioral group intervention improved their self-confidence and gained more control over their personal lives. These outcomes were associated with the possibility of being able to move forward with overcoming homelessness and becoming domiciled. The researchers chose a cognitive-behavioral therapy (CBT) group intervention to guide this study because previous research supported its efficacy in helping distressed people learn effective coping strategies in a short period of time (Bieling et al., 2006; Cipher et al., 2007; Thompson, et al., 2000). CBT is appropriate for individual or group therapy, is highly active and participatory, and focuses on activating and empowering its participants. This type of therapy uses an active, time-limited, structured approach that has been successful with special populations (e.g., older adults and low-income minority women coping with chemical dependency) (Thompson et al., 2000; Washington & Moxley, 2001). Thompson et al. (2000) contended that CBT is particularly suited to older adults who experience less social contact and may feel less stigmatized when their issues are addressed in a group context. The CBT approach is active instead of passive and structured rather than abstract. Social scientists perceive that CBT is useful in the treatment of African American women because of the focus on changing behavior and cognition (Pack-Brown et al., 1998; Washington & Moxley, 2001). These researchers asserted that although group counseling approaches in general are appropriate for African American women, the CBT approach might be more useful in meeting their therapeutic needs. Many social scientists believe that a link exists between emotions, behavior, and the cognitive process. Consequently, how people think (cognition) about their conflicting issues can greatly determine how they feel and then act on their issues.

Many women become homeless as they endeavor to escape physical, sexual, and emotional abuse. Fleeing distressful situations and seeking refuge on the streets and in shelters frequently exposes them to more trauma and displacement (Vandemark, 2007). Repeated exposure to these conditions (e.g., loss of home, belongings, and place in society) can result in loss of control over their lives and reduced personal confidence in the benefits of social relationships. For these reasons, focused and structured interventions, such as the LME group process, show promise in reducing harmful effects of homelessness on women's physical, mental/emotional health outcomes, and social relationships. This type of group therapy has been found to be helpful for people exposed to prolonged or repeated trauma (Bieling et al., 2006; Cipher et al., 2007). Helping older homeless women develop higher levels of confidence and heightened sense of control over their lives could enhance their health-seeking behaviors (e.g., social interactions, seeking treatment, and self-management skills), aid in reconnecting with healthy and meaningful social roles, and subsequently improve their coping responses and ability to overcome their homelessness.

LIMITATIONS

Homeless women included in the present study generally were living in shelters or warming centers in an urban area. They may not be representative of all homeless African American women. To participate in the study, the women had to be at least 50 years of age. The findings may not be generalizable to women who were less than 50 years of age. Because the study focused on urban African American homeless women, generalizations of the findings to men or women of other ethnic/cultural groups or other settings (i.e., rural) may not be appropriate.

CONCLUSION

The social conditions and circumstances that result in homelessness can substantially reduce a person's sense of control and capacity to develop interpersonal relationships. If appropriate interventions are not available or not provided in time, coping resources can be compromised and vulnerability to psychological stress can become a risk factor for various health problems and potential earlier mortality (Belle & Doucet, 2003; Gelles, 1996). Rather than permit older homeless women to remain victims of their homeless situation, they can be taught to bolster their feelings of personal control, increase their satisfaction in personal relationships, and take responsibility for their actions. The provision of such an intervention during the homeless experience can prepare women to engage in recovery activities and enhance their motivation to locate resources to maximize their potential for finding appropriate housing.

Evidence supports the belief that quality relationships and a sense of personal control may also contribute to emotional and biological resilience in the face of life challenges (Goodman et al., 1991; Macer, 2006). The program of group intervention used to reduce the effects of homelessness on the older African American women identified in this study may hold promise for helping these women thrive and overcome their homeless situations. This CBT life management enhancement program may provide an initial structure for strengthening their tenacity; acquiring the determination needed to persevere in the face of adversity; setting useful and attainable goals; evaluating strategies for reversing their situations; identifying needed adjustments; and remaining focused on their goals. In addition, the group format facilitated development of personal relationships. The investigators noted that considerable contact among participants continued after termination of the group experience. Expanding CBT life management enhancement programs among larger samples of homeless African American women, children, and people from other cultural and ethnic groups holds considerable promise in enhancing recovery from homelessness and reducing health challenges.

Acknowledgments

This research is supported by a grant from the National Institute on Aging RO3 AG203300-01.

Footnotes

Publisher's Disclaimer: The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

This article is not subject to U.S. copyright law.

Declaration of Interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

REFERENCES

  • Bandura A. Self-efficacy: The exercise of control. W. H. Freeman and Company; New York: 1997.
  • Belle D, Doucet J. Poverty, inequality, and discrimination as sources of depression among U.S. women. Psychology of Women Quarterly. 2003;27:101–113.
  • Berrenberg JL. The Belief in Personal Control Scale: A measure of God-mediated and exaggerated control. Journal of Personality Assessment. 1987;51:194–206. [PubMed]
  • Bieling PJ, McCabe RE, Antony MM. Cognitive-behavioral therapy in groups. Guilford; New York: 2006.
  • Bierman AS, Clancy CM. Health disparities among older women: Identifying opportunities to improve quality of care and functional health outcomes. Journal of the American Medical Women's Association. 2001;58(4):155–160. [PubMed]
  • Bramlett MD, Mosher WD. First marriage dissolution, divorce, and remarriage: United States. Advance Data Number. 2001;323:1–19. Retrieved from www.cdc.gov/nchs/data/ad/ad323.pdf.
  • Brown DJ. A conceptual model of mental well-being for African American women. In: Brown DR, Keith VM, editors. In and out of our right minds: The mental health of African American Women. Columbia University Press; Chichester West Sussex, New York: 2003. pp. 1–19.
  • Carter MM, Sbrocco T, Gore KL, Marin NW, Lewis EL. Cognitive-behavioral group therapy versus a wait-list control in the treatment of African American women with panic disorder. Cognitive Therapy and Research. 2003;27(5):505–518.
  • Center for Disease Control . Cohabitation, marriage, divorce, and remarriage in the United States. The National Survey of Family Growth. Department of Health and Human Services; Washington, DC: 2002. Available: www.psychpage.com/family/mod_couples_thx/cdc.html.
  • Cipher D, Clifford P, Roper K. The effectiveness of geropsychological treatment in improving pain, depression, behavioral disturbances, functional disability, and health care utilization in long term care. Clinical Gerontologist. 2007;30(3):23–40.
  • Clark R, Anderson NB, Clark VR, Williams DR. Racism as a stressor for African Americans: A biopsychosocial model. American Psychologist. 1999;54(10):805–816. [PubMed]
  • Cohen CI. The forum: Aging and homelessness. The Gerontologist. 1999;39(1):5–14. [PubMed]
  • Cohen CI, Crane M. Old and homeless in London and New York City: A cross-national comparison. In: Bhugra D, editor. Homelessness and mental health. Cambridge University Press; London: 1996. pp. 150–169.
  • Crum RM, Anthony JC, Bassett SS, Folstein MF. Population-based norms for the Mini-Mental State Examination by age and educational level. Journal of the American Medical Association. 1993;269(18):2388–2391. [PubMed]
  • Derogatis MF. The SCL-90-R, Brief symptom Inventory, and matching clinical rating scales. In: Maruish ME, editor. The use of psychological testing for treatment planning and outcomes assessment. 2nd Ed. Lawrence Erlbaum; Mahwah, NJ: 1999.
  • Epel ES, Bandura A, Zimbardo PG. Escaping homelessness: The influences of self-efficacy and time perspective on coping with homelessness. Journal of Applied Social Psychology. 1999;29(3):575–596.
  • Freddolino PP, Moxley DP, Hyduk CA. A differential model of advocacy in social work practice. Families in Society: The Journal of Contemporary Social Services. 2004;85(1):119–128.
  • Gelles R. The book of David: How preserving families can cost children's lives. Basic; New York: 1996.
  • Goldberg EV, Simpson T. Challenging stereotypes in treatment of the homeless alcoholic and addict: Creating freedom through structure in large groups. Social Work with Groups. 1995;18(2/3):79–93.
  • Goodman L, Saxe L, Harvey M. Homelessness as a psychological trauma: Broadening perspectives. American Psychologist. 1991;46(11):1219–1225. [PubMed]
  • Hirschfeld RMA, Klerman GL, Gough HG, Barrett J, Korchin SJ, Chodoff P. A measure of interpersonal dependency. Journal of Personality Assessment. 1977;41(6):610–618. [PubMed]
  • Horne S. From coping to creating change: The evolution of women's groups. Journal for Specialist in Group Work. 1999 September;24(3):231–245.
  • Hovland K. Annotated bibliography for poverty and the lack of insurance. Homeless women: A review of literature. 2001. pp. 32–35. Available: http://academic.regis.edu/epasseri/cbr/homelesswomen.htm.
  • Jacobson N. Experiencing recovery: A dimensional analysis of recovery narratives. Psychiatric Rehabilitation Journal. 2001;24(3):248–256. [PubMed]
  • Janis IL, Field PB. Sex differences and personal factors related to persuasibility. In: Hovland CI, Janis IL, editors. Personality and Persuasability. Yale University Press; New Haven, CT: 1959. pp. 55–68.
  • Kisor AJ, Kendal-Wilson L. Older homeless women: Beneath the safety net. Age in Action. 2002;17(2):1–2. Electronic article. Available: http://www.vcu.edu/vcoa/ageaction/agesp02.htm.
  • Krieger N. Racial and gender discrimination: Risk factors for high blood pressure? Social Science and Medicine. 1990;30(12):1273–1281. [PubMed]
  • Macer DRJ, editor. A cross-cultural introduction to bioethics. Eubios Ethics Institute; UNESCO; Prakanong, Bangkok, Thailand: Bangkok: 2006. 2006. Available: http://www.unesocobkk.org/index.php?id=2508.
  • National Coalition for the Homeless Homelessness among elderly persons. 2005. Available: http://72.14.203.104/search?q=cache:5001 YOZqzgIJ:www.nationalhomeless.org/publication.
  • National Coalition for the Homeless How many people experience homelessness? 2007. Retrieved from www.nationalhomelessness.org.
  • Nyamathi A. Comprehensive health seeking and coping paradigm. Journal of Advanced Nursing. 1989;14:281–290. [PubMed]
  • Nyamathi A, Wenzel S, Keenan C, Leake B, Gelberg L. Associations between homeless women's intimate relationships and their health and well-being. Research in Nursing & Health. 1999;22(6):486–495. [PubMed]
  • Pack-Brown SP, Whittington-Clark LE, Parker WM. Images of me: A guide to group work with African American women. Allyn & Bacon; Needham Heights, MA: 1998.
  • Population Resource Center Older women and poverty: A demographic profile. 2005. Available: http://www.prcdc.org/sumaries/AgingSeries05/womenandpoverty.html.
  • Proctor BD, Dalaker J. Poverty in the United States; 2001. Current Population Reports. Washington, DC: U. S. Bureau of the Census; U. S. Government Printing Office; 2002. pp. 60–219.
  • Radley A, Hodgetts D, Cullen A. Fear, romance, and transience in the lives of homeless women. Social & Cultural Geography. 2006;7(3):437–461.
  • Rooks RN, Whitfield KE. Health disparities among older African Americans: Past, present, and future perspectives. In: Whitfield KE, editor. Closing the gap: Improving the health of minority elders in the new millennium. The Gerontological Society of America; Washington, DC: 2004. pp. 45–54.
  • Rosenheck R, Bassuk E, Salomon A. Special populations of homeless Americans. In: Fosburg LB, Dennis DL, editors. Practical lessons from the 1998 National Symposium on Homeless Research. 1999. pp. 1–27. Available: http://aspe.hhs.gov/progsys/homeless/symposium/toc.htm.
  • Ross H. Growing older: Health issues for minorities. Closing the Gap: A Newsletter of the Office of Minority Health. U. S. Department of Health and Human Services; May, 2000. pp. 1–2.
  • Rotter JB. Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs: General and Applied. 1966;80(1) Whole No. 609. [PubMed]
  • Ryff CD, Keyes CL, Hughes DL. Status inequalities, perceived discrimination, and eudaimonic well-being: Do the challenges of minority life hone purpose and growth? Journal of Health and Social Behavior. 2003;44:275–291. [PubMed]
  • Seligman MEP. The effectiveness of psychotherapy. American Psychologist. 1995;50(12):965–974. [PubMed]
  • Taylor JA. A personality scale of manifest anxiety. Journal of Abnormal and Social Psychology. 1953;48:285–290. [PubMed]
  • Thompson L, Powers DV, Coon DV, Takagi K, McKibbin C, Gallagher-Thompson D. Older Adults. In: White JR, Freeman AS, editors. Cognitive-behavioral group therapy for specific populations. American Psychological Association; Washington, DC: 2000. pp. 235–261.
  • Tombaugh TN, McIntyre NJ. The Mini-Mental State Examination: A comprehensive review. Journal of the American Geriatric Society. 1992;40:922–935. [PubMed]
  • U.S. Department of Health and Human Services . Healthy people 2010 conference edition. Author; Washington, DC: 2000.
  • Vandermark LM. Promoting the sense of self, place and belonging in displaced persons. The example of nonelessness. Archives in Psychiatric Nursing. 2007;21(5):241–248. [PubMed]
  • Vickers A, Zollman C. ABC of complementary medicine: Massage therapies. British Journal of Medicine. 1999;319(6):1254–1257. [PMC free article] [PubMed]
  • Washington OGM. Developing a sense of self in chemically dependent women. Archives in Psychiatric Nursing. 2001;25(1):32–40.
  • Washington OGM. Identification and characteristics of older homeless African American women. Issues in Mental Health Nursing. 2005;26(2):117–136. [PubMed]
  • Washington OGM, Moxley DP. A model of group treatment to facilitate recovery from chemical dependency. Journal of Psychosocial Nursing and Mental Health Services. 2001;39(7):30–41. [PubMed]
  • Wenzel SL. Help for homeless women. Rand Organization; 2006. Available: http://www.rand.org/commentary/061606WP.html.
  • Williams DR, Jackson JS. Race/ethnicity and the 2000 census: Recommendations for African American and other black populations in the United States. American Journal of Public Health. 2000;90(11):1728–1730. [PubMed]
  • Williams DR, Willliams-Morris R. Racism and mental health: the African American experience. Ethnicity & Health. 2000;5(3/4):243–268. [PubMed]