Predictors of Hope Among Members of Mental Health Self-Help Agencies

Abstract

Hope is a key construct for successful community adjustment among those with severe mental illnesses, particularly given the strengths-based recovery model increasingly prevalent in mental health services and in social work. Consumer-run mental health self-help agencies (SHAs) are well suited to fostering hope via their supportive program environments. This study examines factors associated with hope among members of four self-help agencies via a two-stage least squares regression model (N = 310). Findings provide evidence of both individual and program-level associations with hope. These findings mirror social work ethical values of focusing on consumer strengths and self-determination, which are primary goals of SHAs.

As the largest mental health discipline, social work focuses upon the strengths of consumers (Cowger & Snively, 2002). This focus is particularly critical working with those with severe mental illnesses, as the strengths of these individuals are often overlooked. Nationwide, a strengths-based recovery model for mental health is emerging as a way of viewing mental illness not as an inevitable deterioration, but as a condition from which recovery is possible (Anthony, 2000).

Influenced by changing paradigms regarding the etiology, nature, and prognosis of mental illness, recovery is “a deeply personal, unique process of changing one’s attitudes, values, feelings, goals, skills and/or roles. It is a way of living a satisfying, hopeful, and contributing life” (Anthony, 2000, p. 159). Recovery has also been the guiding vision for both general mental health providers and the mental health consumers, self-help movements (Anthony, 2000; Russinova, 1999; Center for Psychiatric Rehabilitation, 1999). Many states are adopting recovery language and practice principles in their mental health systems (Jacobson & Curtis, 2000). While definitions of psychiatric recovery and its constituent components vary, hope is generally recognized as a necessary piece in the puzzle. Ralph (2000) found that consumers in two states rated “the ability to have hope” as the top indicator of recovery.

Hope is an essential but understudied component of the strengths-based recovery model in mental health. Described as an essential piece of a focus upon “human strengths” (Kashdan et al., 2002), hope is critically related to improved mental health functioning; in fact, it has been referred to as “the core process in facilitating positive psychotherapeutic change” (Snyder, IIardi, Michael, & Cheavens, 2000, p. 128).

While much of the hope research has been conducted in professional mental health agencies, there is evidence that one can become empowered, efficacious, and hopeful without professional involvement (Zimmerman, 1990). Hope can be the bridge between empowerment-fostering circumstances and positive personal outcomes (i.e., Snyder, 1991). Increasingly, consumer-run programs such as self-help agencies (SHAs) have emerged as a viable locus for the delivery of empowering mental health services (Segal, Hardiman, & Hodges, 2002). We suggest here that as a critical construct fostering empowerment of consumers, hope among self-help agencies warrants further examination.

These agencies are formally incorporated, serving individuals with severe and persistent mental illnesses. They include little or no professional involvement and tend to be nonmedically oriented (Zinman, Harp, & Budd, 1987). Service delivery is grounded in an egalitarian philosophy that focuses on decreasing stigma associated with mental illness, restoring dignity and self-esteem to members, encouraging empowerment and independent functioning, and supporting members in their pursuit of instrumental goals, all concepts aligned with the social work focus upon consumer strengths (Chamberlin, 1990; Hodges & Segal, 2002; Segal, Silverman, & Ternkin, 1995a, 1995b, 1997; Zimmerman, 1990). The current study explores factors associated with hope among long-term members of four consumer-run mental health SHAs.

BACKGROUND ON HOPE AND MENTAL HEALTH

A growing body of literature has begun to address hope in populations with psychiatric disabilities (Littrell, Herth, Kaye, & Hinte, 1996; Holdcraft & Williamson, 1991; Miller & Powers, 1988; Landeen et al., 2000). A common theme is the subjective, highly intrapersonal nature of hope (Holdcraft & Williamson, 1991; Landeen et al., 2000). While presenting significant challenges for researchers, progress is being made in the conceptualization and measurement of hope and its role in psychiatric recovery/rehabilitation (Miller & Powers, 1988; Landeen et al., 2000).

Snyder (1991) suggests that hope can be operationalized as two separate components—the consumer’s perceived ability to generate strategies toward desired outcomes (termed “pathways”) and the perceived ability that these strategies can be enacted successfully (termed “agency”). Our measure of hope in this study (see Methods) incorporates both of these concepts, as the Beck scale from which it was adapted reflects two dimensions: pessimism about the future and resignation to the futility of changing the future (Steer, Beck, & Brown, 1997).

Hope has been conceptualized as separate from other mental health constructs such as depression, self-esteem, self-efficacy, optimism, and problem solving (Snyder et al., 2002). Particularly important is the distinction between hope and depression. While no doubt related, hope is not merely the absence of depression. In fact, the absence of hope is a hallmark of depression, and hopelessness has been shown to be a stronger predictor of suicide than depression itself (Beck, Steer, Kovacs, & Garrison, 1985; Holdcraft & Williamson, 1991). These findings lend evidence that hope and depression are separate constructs. Accordingly, this study controls for the effects of depression as they relate to hope.

The presence of hope has been identified as a factor in the successful recovery and community adjustment of the severely mentally ill (Farran, Herth, & Popovich, 1995; Landeen et al., 2000; Holdcraft & Williamson, 1991). Hope is also positively related to problem-solving skills, psychological health, coping with stress, and recovery from depression and other psychiatric illnesses (Kashdan et al., 2002).

Recent findings suggest that hope among those with severe schizophrenia is not related solely to the severity of psychiatric symptoms (Landeen et al., 2000). This research team wondered if the supportive program environments of self-help agencies might prove to be more related to hope than a simple measure of psychiatric symptoms. Using a Hope scale that draws upon the work of Zimmerman (1990) and Beck (1974), this study seeks to understand factors related to the presence of hope among members of mental health self-help agencies. These factors considered to be important in creating and maintaining hope include both individual factors (i.e., demographics, psychological and psychiatric characteristics) and organizational characteristics (i.e., the program environments at the SHAs).

FACTORS RELATED TO HOPE AMONG OUR SAMPLE

A key factor in creating and maintaining hope is social support (Farran & McCann, 1989; Foote et al., 1990), also the predominant service modality at SHAs. The self-help philosophy can be described as one of mutual support (Rapport et al., 1985). Therefore, the central and distinctive characteristic of SHAs is reliance upon peer, rather than professional help. Self-help users believe that a peer who has endured similar struggles to one’s own can understand and help in ways thought impossible for professionals without first-hand understanding of mental illness (Long & Van Tosh, 1988; Zinman, Harp, & Budd, 1987). SHAs should thus be viewed as service delivery mechanisms capable of fostering hope through mutual peer support.

Faith in the helping process within the psychotherapeutic context has also been shown to be a powerful predictor of positive outcomes (O’Connell, 1983; Senger, 1987). This suggests that SHA members with faith in the peer-help model are likely to exhibit more hope than those members who do not subscribe to this philosophy.

Demographic characteristics that are thought to be related to hope include gender, housing, ethnicity, and income. Hope is shown to be related to factors such as self-efficacy, perceived control, and depression—all issues for which there is evidence of differential gender effects. Additionally, low levels of income may decrease hope by preventing access to necessary resources (Diblasio & Belcher, 1993). Finally, ethnicity and housing are associated with access to societal resources; thus, these demographic factors may either facilitate or block the development of hope, particularly in SHA members.

Organizational factors that are potentially related to hope among this population include program environment characteristics and organizational involvement. These factors may occupy an even larger role in the presence of hope for users of self-help agencies, many of who regularly spend large quantities of time at the SHAs. The inclusive and supportive philosophy found at SHAs is reflected in efforts to empower members through involvement within the organization (Segal, Silverman, & Temkin, 1995b). This empowerment link to hope draws upon the work of Zimmerman (1990), who found that participation in voluntary organizations is directly related to hope.

Organizationally-mediated empowerment is the mechanism by which SHAs include members in significant decision processes, enabling them to exercise control and influence within the organization (Segal, Silverman, & Temkin, 1995b). For this reason, organizationally-mediated empowerment is a good index of the agency’s role in fostering hope—i.e., the notion that a member can indeed achieve independence and resume productive functioning in the community. It is further thought that participation in the co-creation of services and shared organizational maintenance may ultimately lead to increased hope.

Psychological factors such as self-efficacy, personal empowerment, and locus of control have also been associated with hope in the context of severe depression (Moore & Paolillo, 1984). Farran, Herth, and Popovich (1995) suggest that “Further studies (among clinical populations) are needed to explore the nature of the relationship between hope and hopelessness and depression and other mediating variables such as self-esteem, locus of control, and uncertainty” (p. 138). The current study accounts for the distinct role that depression plays in hope as well as incorporating other measures, such as self-efficacy, that are thought to be distinct entities from hope (Magaletta & Oliver, 1999).

Drawing upon previous studies of hope among psychiatrically disabled populations (e.g., Holdcraft & Williamson, 1991; Landeen et al., 2000) as well as the self-help literature, we examine member attitudes toward peer help, program environments of the SHAs, member characteristics, and levels of psychiatric disability to enhance our understanding of the relative importance of these factors in promoting hope. Given that previous research has found a mixed relationship between hope and symptom severity (Landeen et al., 2000), SHAs, drawing on their non-medical approach and reluctance to focus upon clinical pathology (Long and Van Tosh, 1988), may provide a supportive, hope-instilling environment.

METHODS

Study Design

A cross-sectional design was used with an emphasis on multivariate predictive models, using data collected by the Center for Self-Help Research (CSHR) in affiliation with-the Public Health Institute and the University of California, Berkeley School of Social Welfare. A sample of members was drawn from four mental health self-help agencies in the San Francisco Bay Area (N = 310). Each agency is consumer-run, independently incorporated, has a governing board with a majority of consumers, and offers a wide range of services (Segal, Silverman, & Temkin, 1995a).

All measures are scales and questionnaires administered through in-person interviews. Interviewers were trained by the Center for Self-Help Research in Berkeley, California, and included consumers as well as mental health professionals. Consumers also had input on adapting the scales in the interview schedule for use with a self-help population.

Participants and Sampling Procedures

Respondents were long-term users of the four member-run mental health self-help agencies, having attended at least twelve times over the previous three months. Overall, 96% of those asked to participate in the study agreed. Written informed consent to participate in the interview schedule was obtained from each respondent.

Demographically, the sample had a mean age of 38 years (SD = 8.4), 64% were African-American, 72% were male, 46% were homeless, 87% had DSM-III-R diagnoses, and 20% had a diagnosis of drug or alcohol abuse. Twenty-four percent of the sample was employed at time of the interview, working a mean of 23 hours per week (SD = 20), with mean monthly income of $996 (SD = $2339). The diversity of the sample meets one recommended direction for research on hope—assessing hope among “demographically and culturally diverse clinical populations” (Farran, Herth, & Popovich, 1995, p. 76). See Table 1 for a description of the participants across a range of demographic, clinical, and employment variables (largely from Segal, Silverman, & Temkin, 1995a).

TABLE.

Description of the Participants in the Sample (N = 310)

CHARACTERISTIC		PERCENT OF SAMPLE
Gender:	Female	28%
Ethnicity:	African-American	64%
	Caucasian	17%
	Other/Did not answer	19%
Primary diagnosis:	None	13%
	Drug or alcohol abuse	20%
	Antisocial personality disorder	12%
	Anxiety disorders	24%
	Affective disorder	19%
	Schizophrenia	13%
Homeless:		46%
Never married:		49%
Age:	18 to 24	5%
	25 to 44	76%
	45 to 64	19%
	Over 65	<1%
Education:	Less than high school	27%
	High school	30%
	Some college	31%
	Bachelor’s degree or higher	8%
Employed in paying position:		24%
Mean hours worked per week:		23 hours
Mean monthly income:		$996
Source of income:	SSI or SSD	36%
	General assistance (GA)	36%
	AFDC	5%
	Employment	24%
	Food stamps	33%
	Money from family or friends	16%

Study Sites (Agencies)

All four SHAs were located in urban settings of the San Francisco Bay area. Three of the agencies are located in densely populated, low-income, ethnically diverse neighborhoods, and the fourth in a residential area of single-family dwellings, though this last agency was within walking distance of an urban downtown from which it drew much of its clientele. Three of the agencies targeted services to homeless persons with mental disabilities, and one targeted services to all individuals with mental disabilities (Segal, Gomory, & Silverman, 1998).

Each SHA offers multiple services, including basic resources such as food, drop-in center services, laundry, as well as advocacy, peer-counseling services, and assistance in meeting goals. Other services offered include: financial advocacy/payee assistance, housing assistance, vocational counseling, substance abuse counseling and groups, case management, social interaction opportunities, and resource referrals. Each agency also focuses upon community activism and social change-oriented activities such as lobbying for the rights of the mentally ill (Segal, Silverman, & Temkin, 1995a).

Data Collection and Measures

The CSHR Interview Schedule (Segal, Silverman, & Temkin, 1995a), developed jointly by researchers and consumers, obtained information on hope, demographics, psychiatric disability, psychological factors, program environments at the SHAs, and attitudes toward peer helping.

Hope was measured using a scale created by Segal and colleagues, based upon the work of Zimmerman (1990) and Beck (1974) (Segal, Silverman, & Temkin, 1995a). Beck’s scale is one of the most frequently used measures in the mentally ill population (Farran, Herth, & Popovich, 1995), making Segal ‘s adaptation appropriate for this study’s sample. This scale comprises 10 items with high internal consistency (α = .81). Higher scores indicate more hopefulness. Demographics: Gender, ethnicity, income, and homelessness were each measured via single questions in the interview schedule. Income was measured using a monthly amount, and homelessness was defined in terms of a current lack of housing at the time of interview. Psychiatric disability: The Brief Psychiatric Rating Scale (BPRS) (Overall & Gorham, 1962) was used to measure severity of psychiatric disability. The BPRS is a 24-item scale measuring symptoms such as guilt, hostility, depression, anxiety, hallucinations, and conceptual disorganization. The scale is a symptom-based index that has been frequently employed in drug trials (Rhoades & Overall, 1988) and used by Segal and colleagues (Segal & Kotler, 1993) in their studies of former psychiatric patients in residential board and care settings. Staff members, both consumers and mental health professionals, were trained with the aid of the Clinical Research Center for Schizophrenia and Psychiatric Rehabilitation at UCLA. Inter-rater reliabilities during training were in the .9 range. In the current sample, the Scale’s internal consistency was α = .79. Depression was measured by the Center for Epidemiologic Studies Depressed Mood Scale (CES-D) (Radloff, 1977). Its internal consistency for this sample is α = .85. The CES-D is a self-report measure that asks respondents to agree or disagree with statements regarding the presence of a depressed mood. Personal empowerment was measured using a 20-item scale developed by Segal and colleagues, with internal consistency of α = .85 (Segal, Silverman, & Temkin, 1995b), to assess the amount of personal choice and control that respondents have in various areas of their lives. Self-efficacy was measured via Segal and colleagues’ (1995b) self-efficacy scale, based upon the work of Alfred Bandura (1977, 1982) and developed with Bandura’ s consultation. This is a 15-item scale (α = .89 in our sample) measuring perceived confidence in completing a number of everyday tasks. In this context, self-efficacy is perhaps best described as a measure of self-confidence. Because of considerable overlap between the self-efficacy and locus of control constructs, we elected to include only one of the two factors in our model. Satisfaction with the SHA was measured via a 7-item scale, with each item measuring satisfaction with aspects of services at the SHA on a 5-point Likert scale (from “very dissatisfied” to “very satisfied”). Organizational empowerment was measured via a scale created by CSHR staff in conjunction with SHA members, α = .87 in this sample (Segal, Silverman, & Temkin, 1995b). This scale measures members’ participation and involvement in daily agency tasks and ongoing organizational maintenance. Program Environment at SHAs was measured using the Community-Oriented Program Environment Scale (COPES) developed by Moos (1988). The COPES is a l00-item true/false scale that measures the character of social interactions in service settings (Segal, Silverman, & Temkin, 1997). The COPES consists of ten subscales measuring the structures, processes, and interpersonal re1ationships within the service setting, with average α = .66 for each subscale. The Mutual Assistance Program Environment Scale (MAPES) is an adaptation of the COPES scale, with 74 items added to COPES to factor in the concerns of SHA program environments. Taken together, the COPES and the MAPES effectively address program environments in the SHA setting. MAPES reflects user perceptions of the following organizational dimensions of SHAs: dependence on external resources, insularity, service value preferences, staff/user egalitarianism, and the agency’s internal structure. Internal consistency of the MAPES scale is α = .91. Attitudes toward peer/professional helping were measured via agreement with the following statement: “The best help comes from a professional, not a peer.”

Analyses

The demographic and descriptive characteristics of the sample are first described. Then, a two-stage least-squares regression model was utilized to predict hope scores based upon the hypothesized factors, with the second-stage including site of empowerment. The agencies were included as dummy variables to control for clustering effects by site.

RESULTS

Our data indicate that the sample was relatively hopeful. Self-help members had a mean hope score of 8.3 (SD = 2.37) on a ten-point scale, higher values indicating more hope. Normative data are lacking on this scale, either with a clinical or the general population, as it is from several previous scales, taking into account mental health consumer input in its development.

Multivariate Analysis

The second stage of the two-stage least squares regression model (agency site) was not significant and added no predictive ability to the model, thus it was dropped. The final multiple linear regression model is significant (adjusted R² = .382, df = 13, F = 12.498, p < .0000). One demographic factor was significant in the multivariate model, homelessness (p = .015). Being currently homeless (when accounting for each of the other variables in the model) was significantly associated with higher scores on the Hope scale. Psychological factors significant in the model were depression (p < .0000), severity of psychiatric symptoms (p = .024), and self-efficacy (p = .000). Controlling for all factors in the model, less depression, fewer psychiatric symptoms, and higher self-efficacy were significantly associated with the presence of more hope. Finally, among program environment factors, only the MAPES factor was found to be significant (p = .005), indicating that program environments stressing mutual peer support were associated with higher levels of hope (see Table 2).

TABLE 2.

Linear Regression Model: Factors Associated with Hope (N = 310)

Independent Variables	Standardized Regression Coefficient/b	T	p
Demographics Gender	−.107	−2.055	.052
Ethnicity	.046	.828	.409
Income	.000	−.002	.998
*Homelessness	−.141	−2.457	.015
Psychiatric Disability Indicators *Psychiatric severity (BPRS)	−.146	−2.271	.024
*Depression (CESD)	−.310	−4.569	.000
Psychological Characteristics *Self-efficacy	.241	3.919	.000
Personal empowerment	.109	1.603	.110
Proqram Environment of SHAs Member satisfaction with help from agency	−.085	−1.396	.164
Orqanizationally-mediated empowerment	−.082	−1.435	.153
COPES scale	.073	1.163	.246
*MAPES scale	.165	2.841	.005
Professional vs. Peer Helpinq The best help comes from a professional, not a peer	−.041	−.747	.456

^*Variable is significant at p is less than or equal to .05

Model summary: R² = .415, df = 13, f = 12.498, p = < .000

Note: The second stage of the two-stage regression model (SHA site) was dropped as it did not add to the predictive ability of the regression model.

DISCUSSION

A primary goal of this study was to describe the levels of reported hope among long-term users of mental health SHAs. Overwhelmingly, this sample has hope: The mean hope score was 8.3 (SD = 2.37), of 10 possible. This finding alone is of particular interest in that it characterizes mental health consumers who utilize mutual support services as a resoundingly hopeful population. Our findings suggest that the SHA provides an example of how the psychiatric recovery paradigm can be effectively utilized to reframe long-standing notions of illness as well as to foster hope among members. Rather than subscribing to a view of mental disability as chronic and debilitating, self-help organizations offer participation in supportive peer-focused environments that may ultimately foster hope for the future.

It should be noted that there have been criticisms of very high hope levels (termed “false hope”) in the literature, as such hope may be based upon unrealistic expectations, inappropriate goals, or maladaptive strategies used to achieve these goals. However, a key researcher on hope has found no evidence that false hope is indeed a real threat (Snyder et al., 2002). In keeping with the social work and mental health self-help movement’s focus upon strengths and empowerment, the high levels of hope here are interpreted in this light.

There were significant associations between hope and homelessness, severity’ of psychiatric symptoms, self-efficacy, and SHA program environments. There is much support in the literature for some of these findings (e.g., that increasing severity of symptoms leads to decreased hope, that greater self-efficacy is associated with greater hope). Other findings (such as those on homelessness) suggest that further research is needed on the link between hope and these factors. It was surprising to find that those who are currently homeless demonstrate higher levels of hope than those who are housed. It is possible that this is an effect of “having more to hope for,” but this relationship should be further investigated in future research.

Of the organizational findings, the role of program environments at the SHA in the development of hope is perhaps the most important. This was supported by our MAPES finding and suggests that agency environment characteristics can be linked with individual outcomes such as the presence of hope. This again confirms the assertion that SHAs are hope-inducing, empowering agencies. Future research might address specific components of the SHA program environments and their potential use in non-self-help settings to foster hope among this population.

Significant methodological challenges exist in the measurement of hope, a variable of subjective and individualized nature. Despite concerns with validity, these challenges dictate the utilization of self-report data to ensure that we are truly assessing factors considered important to respondents. The variables included were chosen because they were considered relevant in the self-help literature, the hope literature, or the general mental health literature. Other variables not included in the analysis here might also be important to understanding hope among this population. Future research should investigate the roles of alternative variables, such as self-esteem, length of time at agency, and agency setting.

Over a decade after Kurtz’s (1990) predictions about its rise in mental health settings, self-help is now a recognized and highly prevalent service delivery mechanism for those with severe mental illnesses (USDHHS, 1999). Recognizing the relevance of hope as a key construct for those seeking recovery, it is critical to understand the ways in which SHAs support their members’ sense of hope and their resulting movement toward independence and successful community functioning.

Jacobson and Curtis (2000; 335) suggest that “The emotional essence of recovery is hope, a promise that things can and do change, that today is not the way things will always be.” The current study extends the process of connecting hope and psychiatric recovery, particularly in the finding that psychological characteristics such as symptom severity are not solely responsible for hope. Rather, they should be considered in conjunction with organizational and environmental factors. Our findings support the social work emphasis upon consumer strengths, autonomy, dignity, and self-determination. Yet social workers and other professionals seeking to enhance hope (and ultimately recovery) among clients need to develop new skills and competencies around the instillation of hope (Russinova, 1999). We further suggest that such social workers may do well to consider referrals to consumer-run mental health organizations such as the SHAs studied here.

Future research should investigate whether the supportive, peer-focused nature of SHA program environments might prove more effective than the traditional CMHAs in promoting hope, growth, and recovery. Future studies of those using professionally delivered mental health services rather than peer support might also address the nature of specific factors related to hope. Finally, the organizational findings herein have implications for service structure and delivery at the latter type of organization. Professional mental health agencies may benefit from exploring ways to incorporate those program environment aspects associated with increased hope for SHA members.