Are All Consumer-Operated Programs Empowering Self-Help Agencies?

Abstract

The literature on consumer-operated-service programs (CQSPs) distinguishes two organizational types based on their leadership styles: the self-help agency (SHA)-participant democracy and the board-staff-run COSP. This study considers whether the characteristics of these two organizational leadership styles are recognized by members and whether these characteristics are associated with membership degree of empowerment. Two-hundred and fifty new entrants to five COSP drop-in centers rated the programs’ leadership style using the COPES System Maintenance Scale and assessed their own empowerment on four empowerment measures. ANOVA with Bonferroni post-hoc tests were used to evaluate differences between settings; MANCOVA to assess differences in member empowerment. COSP system maintenance differences distinguished the two organizational types (p < .000). SHA-participant democracy members scored significantly better than board-staff-run program members on three of the four empowerment measures. SHA-participant democracies, with a lower focus on system maintenance, and an emphasis on power sharing between staff and non-staff members, appeared to more effectively use organizational decision-making processes to empower their members.

Proponents hold that consumer-operated service programs (COSPs) for persons with serious mental illness effectively empower their members in a fashion that promotes recovery (President’s New Freedom Commission on Mental Health, 2003). Found in numerous countries, COSPs are recognized as a major component of the mental health system (President’s New Freedom Commission on Mental Health, 2003). COSPs thus are likely to play an increasing role in the treatment plans of all individuals seeking service in public mental health systems (Mental Health: A Report of the Surgeon General, 1999). Mental health consumers/survivors developed these programs as alternatives to disempowering professionally run services that limited participant self-determination. COSP simply means consumer-run—as an organizational descriptor, it includes but may not be limited to self-help agencies (SHAs). While both the COSP and the SHA focus on the role of consumers in helping each other, SHAs add the principle that participants can help themselves and each other through peer support, power sharing, and client control of services (Zinman, 1987). This study addresses the need to better understand the relationship between the program characteristics of the SHA and its ability to empower its members, its putative raison d’être (Zinman, 1987). It further asks whether all COSPs are empowering self-help agencies, or whether the need to share power with all members and provide peer support has a special role in empowerment.

“Empowerment” has become part of the mental health lexicon but remains an inconsistently defined concept. In general, it connotes a process by which individuals with lesser power gain control over their lives and influence organizational and societal structures within which they live. In the context of community services, the exercise of power implies the “ability to get what one wants, and the ability to influence others to feel, act, and/or behave in ways that further one’s own interests” (Dodd & Gutierrez, 1990). It is “the capacity to influence the forces which affect one’s life space for one’s own benefit” (Pinderhughes, 1983, p. 332). Empowerment, then, connotes both a process and an outcome; as consumers gain power to obtain resources on multiple levels, they are enabled to gain greater control over their environment (Hasenfeld, 1987). For persons with serious mental illness, such a process may include gaining through their own efforts new resources or competencies such as the capacity to help others, group leadership skills, organizational leadership abilities, and influence in the civic and political spheres (Rappaport, Reischl, & Zimmerman, 1992; Zimmerman & Rappaport, 1988).

Mental health service providers have generally adopted “empowerment” as a program principle in recovery-focused services geared toward meeting the needs of people with mental illness. These services may include programs designed to foster increased social skills, greater client decision making in program operations, and supportive peer interactions (Berman-Rossi & Cohen, 1988; Cohen, 1989; Mowbray, 1990; Susser, Goldfinger, & White, 1990). However, consumers and others argue that empowerment in any context cannot be bestowed by those with greater power on those with less; it must be initiated’ from the bottom up, by those who seek self-determination (Gruber & Trickett, 1987; Pinderhughes, 1983; Rappaport, 1985; Simon, 1990; Yeich & Levine, 1992).

Consumers maintain their programs truly empower people because consumer control and delivery of services facilitates this grassroots process (Chamberlin, 1990; Clay, Corrigan P, & Schell, 2005; Segal, Silverman, & Temkin, 1993; Zinman, 1987). They regard empowerment as the principle underlying consumer program goals, processes, and outcomes. Zinman (1987) defines the essential characteristics of a consumer-operated organization as client control of all program aspects with autonomy from the mental health system; voluntariness of all services; emphasis on addressing the economic, cultural and social needs of members; and—central to the focus of this study—sharing of power within a structure that seeks to minimize hierarchal relationships. Empowerment through the organizational characteristics of a self-help organization is seen as enabling members to regain hope, self-esteem, and self-confidence lost through stigmatization as persons defined as “mentally ill” (Chamberlin, 1978; Kaufmann, Freund, & Wilson, 1989; Leete, 1988)—these goals are reflected in the principles and practice of the recovery model (Ralph, Kidder, & The Recovery Advisory Group, 2000). Thus, although consumer-operated services are an essential component of self-help organizations, they foster empowerment not merely by providing peer-based services but by allowing members to participate in organizational decision making and governance (Segal, Silverman, & Temkin, 1995a).

COSPs, as they have developed in the mental health services system, are usually incorporated as non-profits, have a director who is/was a consumer, require at least 50% consumer representation on their boards of directors, and have a consumer leadership that controls the budget and makes personnel decisions (Clay, Corrigan P, & Schell, 2005; Mowbray et al., 2006). All such organizations claim to allow their membership participation in decision making regarding organizational operations, both informally in the interactions of staff and members and formally through participation in community meetings. The consumer movement, however, in its founding texts expresses concerns about the ability of all consumer-operated organizations to be empowering (Budd, 1987; Chamberlin, 1994).

The COSP literature (Budd, 1987; Chamberlin, 1994) distinguishes two types of consumer-operated organizations based on their governance structures: the SHA-participant democracy and the board-staff-run program. SHA-participant democracies allow members a direct voice in major organizational decisions such as program planning, hiring, and firing, and budgeting through participatory processes such as committees and community meetings. In the board-staff-run COSP, this authority resides in hierarchically structured consumer leadership rather than collective decision-making processes. Leaders tend to be selected for their vision and ability to get things done (Budd, 1987, p. 126). The challenge of the leadership in the board-staff-run COSP is to be accountable to the membership (Goldstrom et al., 2006; Harp & Zinman, 1994; Zinman, 1987). This study considers whether the characteristics of these two organizational leadership styles are perceptible to member participants and whether they are associated with member reports of their degree of empowerment.

METHOD

Setting

Five consumer-operated programs in the Greater San Francisco Bay Area participated in the study. Common service elements of the COSPs included peer support groups, drop-in spaces for socializing, and direct services such as help securing food, clothing, and shelter; peer counseling; money management; payeeship services; advocacy; and information and referral.

A COSP was defined by the status of its leadership as noted above and its claim to follow an ideology of empowering its members. The COSPs did not specifically define themselves as SHA-participant democracies or board-staff-operated agencies. However, pre-study qualitative observations of the operations of the five study COSPs indicated that the characterization appeared to depend on the extent to which major organizational decision making devolved to community meeting participants (Segal, 1998a, and 1998b).

Based on investigator and research staff observations of these settings, including their community meetings, the five COSPs could be classified as follows. One was a clearly top-down organization with a dominant leader where all major decision making was confined to the board and staff, with community meeting decisions restricted to activity planning. Two COSPs had mixed organizational characteristics: they had strong leadership and organization, yet made significant attempts to involve the membership in the community meeting in major organization decision making (e.g., budget allocation and hiring decisions). Two others practiced power-sharing by delegating major decision making to members at community meetings (Segal, 1998a, and 1998b).

Sample

Two-hundred and fifty (85% of 294) new entrants to five COSP drop-in programs between 1996 and 2001 agreed to participate in the study and were interviewed at 1 month and 8 months following their enrollment in the study. A “new entrant” was an individual who had not received services in such an organization for at least the 6 months prior to agency entry. No significant differences were found when study participants were compared with the refusal group in terms of gender, ethnicity, and housing status.

Assessment

Interviews were conducted by former mental health clients and professionals trained by the Center for Self Help Research, Berkeley, CA. Informed consent for human investigation was obtained from all study participants. All members responded to an extensive interview schedule that included an adaptation of the System Maintenance Scale of the Community Oriented Programs Environment Scale (COPES) (Moos, 1974) and four empowerment measures (Segal, Silverman, & Temkin, 1995a, 1995b).

The System Maintenance Scale of the COPES was selected for the assessment of the COSP leadership style because of its specific focus on the characteristics believed to distinguish top-down/board-staff-run settings from the less hierarchical participant democracy settings. The Scale also had descriptive relevance to the environment of COSP drop-in centers, established reliability, and was widely used in evaluations of mental health service and rehabilitation settings (Moos, 1972, 1974).

The COPES System Maintenance Scale includes 21 forced-choice yes/no items reworded in a previous study of long-term clients of self-help agencies to be relevant to the COSP environment (Segal, Silverman, Temkin, 1995b). System Maintenance Scale ratings in the later study obtained from 310 long-term self-help agency drop-in center clients had an internal consistency of Alpha = .76 (Moos, 1972; Segal, Silverman, Temkin, 1995b).

The System Maintenance Scale has three subscales: order and organization, program clarity, and staff control. The order and organization subscale includes five items asking whether clients believed the Center “was very well organized,” “looked messy at times,” and “strongly encouraged members to be neat and orderly,” among other items. The program clarity subscale’s seven items ask, for example, whether clients believed that “… everyone knows who is in charge,” “… rules are clearly understood by clients,” and “… clients who break the rules know the consequences.” The staff control subscale includes nine items addressing whether “… staff make and enforce all the rules,” “… staff order clients around,” or “… clients are suspended from the Center if they don’t obey the rules.” Cronbach’s Alpha reliabilities of the three subscales, based on the response of a long-term client sample were, respectively, .59, .66, and .60 (Moos, 1972; Segal, Silverman, Temkin, 1995b).

The four measures of individual empowerment, describe various behaviors associated with the empowerment construct, have established construct validity and appear to measure the extent to which a person could be considered empowered (Segal, Silverman, & Temkin, 199Sa).

The Self Efficacy Scale (SES) can be considered a measure of self-confidence in one’s ability to effect certain actions, and constitutes a bridging concept between two dimensions of the empowerment construct—the control the individual has over their own personal life and material situation, and their experiences in exercising control and influence over others within and outside of the self-help organization (Segal, Silverman, & Temkin, 1995a). This bridging characteristic of the SES links the individual’s confidence in his or her ability to be efficacious in common life activities with his or her sociopolitical or group actions. The SES has a reliability of Alpha between .89 and .92, and a stability coefficient of .62 (Segal, Silverman, & Temkin, 1995a).

The Personal Empowerment Scale (PES) measures the amount of control individuals have over their own common life domains, including shelter, income, and service provisions, as well as their ability to minimize the chance of unwanted occurrences such as personal danger or homelessness. Using a Likert scale format, the 20-item PES poses questions such as “How much choice do you have about how to spend any money you might have?” and “How much choice do you have about how you will spend your free time?” The PES has a reliability of Alpha between .84 and .85, and a stability coefficient of .49 (Segal, Silverman, & Temkin, 1995a).

The Organizationally Mediated Empowerment Scale (OMES) assesses the extent to which COSP members were empowered by organizational participation. According to The Levi-Strauss Company (n.d.), organizations give power to their members by increasing the exercise of authority and responsibility of those in the organization. As Perrow (1967) notes, the task structure of an organization revolves around issues of control and coordination. The former addresses the discretion an individual possesses in carrying out tasks within the organization, and the power of the individual to mobilize scarce resources within the organization. Coordination, on the other hand, involves the exercise of responsibilities. The 17-item OMES presents yes/no questions to the respondent, such as “[At your Center] have you taken part in deciding what rules people need to follow?”, “… in deciding whether to hire someone?”, “… in deciding how much money should be spent on a service or program?”, and “… [have you] helped set up a meeting?” The OMES has a reliability of Alpha between .87 and .90, and a stability coefficient of .62 (Segal, Silverman, & Temkin, 1995a).

The Extra-organizational Empowerment Scale (EES) assesses participation in community efforts. The 15 scale items look at the respondents’ involvement in political and other community activities outside their service agency, such as “[Have you] spoken on a panel or given a speech at a local, state, or national conference?” “Worked on a political campaign?” and “Attended a meeting or hearing of a government board or commission?” The EES has a reliability of Alpha between .72 and .73, and a stability coefficient of .61 (Segal, Silverman, & Temkin, 1995a).

Analysis

All analyses were completed using SPSS 16.0 (2009). COSPs were constituted as agencies to provide mutual assistance (Segal, Silverman, Temkin, 1995b). These agencies call participants members, implying a participatory responsibility within the organization exercised in their community meetings. COSPs also emphasize the importance of social network building in their helping role. COSPs that host drop-in centers, however, serve individuals who attend with varying degrees of commitment—some individuals simply “drop-in” for a cup of coffee or to get out of the rain; others are engaged in a variety of services and become through their engagement members in the true sense of the term. Early work on such programs reported on the positive experience of “long-term users,” those involved with the organization for an extended period of time (Segal, Silverman, Temkin, 1995b) This analytic approach, based on consultation with COSP membership, focused on the use of duration of participation as an indicator of true “membership” and the responsibility and benefits derived from the mutual assistance offered by the organization. Herein, in order to assess the experience of those fully participating in the COSPs, using SPSS’s GLM weighting option, responses of individuals in the sample who reached 8 months of service-participation were weighted by the inverse of the probability of reaching this level of service involvement in the member’s agency (i.e., the number of people enrolled in the condition divided by the number reaching 8 months of service). Those failing to reach eight months of service were given a zero weight. This process recognizes the importance of the membership experience for those completing the eight months and discounts the casual service use of those failing to do so.

System maintenance differences between the board-staff-run COSP and the SHAs were evaluated using analysis of variance (ANOVA) with Bonferroni post-hoc tests for differences between programs. The relationship between program differences and member empowerment was evaluated with a multivariate analysis of covariance (MANCOVA) that adjusted for preexisting demographic and diagnostic differences in the populations served by the different organizations (i.e., age, gender, race [Caucasian, African American, Other], and diagnosis [schizophrenia/schizoaffective disorder, major depression, other]).

The project was reviewed and received Institutional Review Board (IRB) approval for procedures insuring the protection of human subjects. The authors have no known conflicts of interest and certify authorship.

RESULTS

Consumers visited the programs during their first month 8.6 (SD 7.8) times on average. There were no statistical differences in the average number or the variance in the number of visits between programs. No duration of visit information was recorded and multiple entries on a given day were reported as one visit.

Program Differences

Differences on the average ratings each COSP received from its members on the COPES System Maintenance dimension are reported in Table 1. The ANOVA shows what appear to be two program clusters based on the results of the Bonferroni post-hoc tests. One program cluster, a “unique” cluster comprised of a single program referred to as Program B, differed significantly (between p = .015 and p < .000) from the other programs in the sample in its emphasis on strong system maintenance—it evidenced the highest scale score on this organizational characteristic. Investigators’ observations of many other COSP drop-in centers would indicate that while Program B (noted in Table 1) may be “unique” to this COSP sample it is not unique to the universe of COSP drop-in centers. We thus, treat it as potentially representative of a cluster of such organizations.

TABLE 1.

One Way ANOVAs Testing for Consumer-Operated Program Differences In Leadership Style Measured by the COPES System Maintenance Scale

Statistics	Descriptive statistics1			Bonferroni post hoc tests for program differences
Consumer-operated service	n	Mean	Std. deviation	Mean difference (I-J)2	Std. error	Sig.
Clearly defined participant democracy: Program A.	88	15.26	3.20	4.24*	.50	.000
Clearly defined top-down board-staff-run: Program B.	49	19.50	1.49	—	—	—
Clearly defined participant democracy: Program C.	22	16.23	3.72	3.27*	.72	.000
Strong leader—Participant democracy: Program D.	24	16.53	2.00	2.97*	.70	.000
Strong leader—Participant democracy: Program E.	44	17.62	2.89	1.88*	.58	.015
Total	227	16.86	3.22

ANOVA

¹F = 19.24; df 221, 4; p < .000.

²(I) = Program B.

(J) = Programs A, C, D, and E.

^*Significant at p < .05.

Table 2 displays the two observed organization clusters as separate organizational types distinguished primarily by high versus lower system maintenance scores. Analysis via ANOVA (Table 2) reveals that the system maintenance differences between the clusters derive from significant differences (p < .000) on all three system maintenance subscales (i.e., staff control, order and organization, and program clarity).

TABLE 2.

Descriptive Statistics and ANOVA Differences in System Maintenance Subscale Scores for Program B versus Other Programs

Descriptive statistics		n	Mean	Std. deviation	ANOVA: Contrasting system maintenance subscale means in Program B: Board-Staff-Run vs. SHA-participant democracies
System maintenance subscales	Program
Order and organization	SHA-Participant Democracies	190	3.83	1.44	F = 22.66; df = 1, 247; p = .000
	Top-down: Program B: Board-Staff Run	60	4.75	.66
	Total	250	4.05	1.36
Program clarity	SHA-Participant Democracies	180	6.94	1.84	F = 44.05; df = 1, 230; p = .000
	Top-down: Program B: Board-Staff Run	52	8.66	.68
	Total	232	7.33	1.80
Staff control	SHA-Participant Democracies	185	5.43	1.07	F = 19.42; df = 1, 232; p = .000
	Top-down: Program B: Board-Staff Run	49	6.16	.85
	Total	234	5.58	1.07

Member Differences in Empowerment

The 250 members who rated the COSPs were more likely to be male (60.5%) and to never have been married (56.8%). Their average age was 41.5 (SD 9.4). Their ethnicity was: 44.3% Caucasian, 40.2% African American, and 15.5% other ethnic groups. Thirty-two percent were literally homeless at the time of the interview; 34.9% had failed to complete high school, 19.8% completed high school, and 45.3% had more than a high school education. Axis I DSM IV diagnoses, as assessed with the Diagnostic Interview Schedule (2009), included 20% with schizophrenia/schizoaffective disorder, 54% with major depression, 7% with bipolar disorder, and 18% with “other” conditions.

Overall, the MANCOVA results for the four multivariate tests conducted with the procedure (Pillai’s Trace = .130; Wilks Lambda = .870; Hotelling’s Trace = .149; and Roy’s Largest Root = .149) indicated that board-staff-run high system maintenance program membership was significantly associated with lower empowerment scores (F = 5.89; df = 4, 158; p < .000; Partial Eta Squared = .13) even after controlling for preexisting demographic and diagnostic differences in the groups served in the different programs. Table 3 indicates that these results were attributable to significant differences favoring members in the SHA-participant democracy programs on three of the four empowerment measures: Self Efficacy Scale scores (p = .019), Organizationally Mediated Empowerment Scale scores (p = .001), and Extra-organizational Empowerment Scale scores (p = .011).

TABLE 3.

Descriptive Statistics and MANCOVA Evaluating Membership Empowerment Differences Between SHA-Participant Democracies and Board-Staff-Run Programs

Dependent variables	Independent variable	Mean	Std. deviation	n	MANCOVA: Tests of between subjects effects
Self-efficacy	SHA-Participant democracies	53.4	11.12	114	F = 5.59; df = 1; p = .019; Eta2 = .03
	Board-staff-run program	45.0	17.52	50
	Total	50.8	13.89	164
Personal empowerment	SHA-Participant democracies	63.0	10.84	114	F = .02; df = 1; p = .895; Eta2 = .000
	Board-staff-run program	64.4	10.21	50
	Total	63.4	10.64	164
Organizationally mediated empowerment	SHA-Participant democracies	3.1	2.79	114	F = 11.99; df = 1; p = .001; Eta2 = .07
	Board-staff-run program	1.6	1.07	50
	Total	2.5	2.46	164
Extra-organizational empowerment	SHA-Participant democracies	.2	.51	114	F = 6.59; df = 1; p =.011; Eta2 = .04
	Board-staff-run program	.0	.00	50
	Total	.1341	.43644	164

DISCUSSION

The two observed COSP program clusters correspond to the two program types reported on in the consumer self-help literature: board-staff-run organizations and SHA-participant democracies (Budd, 1987; Chamberlin, 1994). While both organization clusters were consumer-operated, the board-staff-run cluster was distinguished by a leadership style focused on system maintenance activities. SHA-participant democracies, with a lower focus on system maintenance, and an emphasis on power sharing between staff and non-staff members, appeared to more effectively use organizational decision-making processes to empower their members.

Board-staff-run COSPs thus appear to be programs where there is greater concern for order and organization, staff control and program clarity in the organization and where most major organizational decisions—including hiring, firing, program development, and fundraising strategies—are made by consumer staff. Such programs, like professionally run programs, try to enable people in the recovery process. While both organizational types promote the value of peer counseling, SHA-participant democracies are organizations where major organizational decisions are made through extensive member involvement in participatory processes such as community meetings. Such organizations may have strong leadership, but leadership that is committed to sharing power within the organization as a means to empowerment consistent with the recovery principles of self-determination and full participation.

An orientation toward high system maintenance in a hierarchical structure may occur at the expense of member organizational empowerment, as indicated by the significantly lower OMES scores in Program B, the high system maintenance board-staff-run COSP. Its mean OMES score was 1.6, while that of the other programs was 3.1. As a guideline we would suggest that OMES scores below 2 indicate a program environment that does not promote power sharing between staff and non-staff participants.

In the SAMHSA multisite COSP study, OMES scores are reported among the eight sites as generally “quite low and positively skewed”; the baseline-mean was just over two and the modal score was zero (Rogers et al., 2007). The SAMHSA multisite results for the OMES showed “an overall negative effect on [OMES scores] over time (F_{1, 4075} = 47.17, p < 0.001); on average, scores declined over time” (Rogers et al., 2007, p. 792). Possibly the modest effects reported by the SAMSHA study in its empowerment measures of participants were driven by the presence of COSP sites run more in conformity with the Program B- board-staff-run model than as SHA-participant democracies such as the other programs included in the present study. The SAMHSA multisite report actually noted considerable site variation but tended to attribute this to program content differences rather than the absence of an essential ingredient—a true participant democracy self-help orientation allowing for sharing of power in agency decision making—in some of its sites. Such site variation in participant democracy decision making may account for the absence of a significant time × group × site interaction (F_{7, 4075} = 1.87, p = .07) in the SAMSHA multisite OMES scores. It may imply that some sites did not use their organizational context to empower their membership (Rogers et al., 2007).

A lesser emphasis on system maintenance does not in and of itself imply higher OMES scores. Such scores derive from member participation in organizational decision making. That such participation is empowering is perhaps validated by the observed difference in extra-organizational activities in the participant democracies and by the increased self-confidence indicated by the higher self-efficacy scores among members in these organizations. While it does not appear that organizationally mediated and extra-organizationally mediated empowerment facilitated by the SHA-participant democracies was accompanied by increased scores in personal empowerment, the significant difference between the clusters in self-efficacy—the bridge indicator linking changes in the realms of the exercise of social influence to more extensive control of one’s life activities (Segal, Silverman, & Temkin, 1995a)—suggests that such gains should be forthcoming.

The study has limitations; causation cannot be proven with the cross-sectional design and methods employed. We have assumed, not proved; that the board-run program in this sample is representative of other board-run programs. There also is the issue of selection at work; perhaps consumers are referred or choose the different agencies based on perceptions of how well the consumers’ needs or interpersonal styles match the agency culture. Thus, disempowered people go to agencies providing more extensive direction and structure for their clients. Such selection, from the perspective of promoting recovery, while offering more support to the disempowered, may be denying such individuals the opportunity to empower their lives and may lead to dumping of those in need of most support into disempowering COSPs—COSPs exhibiting the same characteristics as organizations that inspired the creation of SHA-participant democracies as an alternative to professionally directed services. It is the case, however, that there was no instance in this study where a COSP and SHA were within reasonable proximity to each other; thus individuals and those who referred them did not have a choice between the two models.

Both board-staff-run consumer-operated services and SHA-participant democracies offer alternative decision-making structures whose relative potential for achieving recovery outcomes is an open issue. This study adds empirical evidence that substantiates theoretical concerns central to the founding of such COSPs. It indicates that a critical distinction should be made between SHA-participant democracies and board-staff-run COSPs: the former require both consumer control and a set of procedures that facilitate member empowerment, while the latter appears to require consumer control with an emphasis on standard social service agency programming and administration more focused on sound system maintenance, perhaps at the expense of efforts to promote member empowerment through organizational structures.