Latin American Community-based Mental Health Initiative: Stakeholder matrix tool LATIN AMERICAN COMMUNITY-BASED MENTAL HEALTH INITIATIVE: STAKEHOLDER MATRIX TOOL Javier Mignone University of Manitoba � This article presents a matrix tool that facilitates a systemic view, both for planning and evaluation purposes, of the roles and interactions between key stakeholders in community-based mental health programs. The tool emerged from a community mental health initiative centered on natural caregivers, that commenced in 1995 in Edmonton, Alberta, Canada, and involved the Latin American community, social agencies, and health institutions. The article describes the Latin American Community-based Mental Health Initiative and exemplifies the use of the stakeholder matrix tool. The purpose of the tool is to assist community-based initiatives to explicitly examine, on an ongoing basis, roles, interactions, and areas of tension between its stakeholders. The lack of this examination undermines the possibility for community-based initiatives to overcome the many challenges they face. The stakeholder matrix tool offers a relatively straightforward structure from where to initiate this process, be it from a planning and0or evaluation perspective. © 2002 Wiley Periodicals, Inc. Community mental health programs involve, by their very nature, a series of distinct players, both individual and institutional. Often a key component, the interaction and areas of tension between stakeholders, are overlooked when planning and evaluating these programs. This article presents a matrix tool that facilitates a systemic view, both The author wishes to recognize the participation in discussions of the matrix tool the following individuals involved in the project: Laurel Borisenko ~The Mennonite Centre for Newcomers!, Michael Cairns ~The Boyle Street Coop!, Yvonne Chiu, Luis D’Elia, Denise Spitzer, Adrienne Wiebe ~Royal Alexandra Hospital!, and Joan Wright ~The Support Network!. The contents, however, are the responsibility of the author. Partial funding for this study was provided by the Muttart Foundation. The author is currently a research associate with the Centre for Aboriginal Health Research, Community Health Sciences, University of Manitoba. Correspondence to: Javier Mignone, Suite 715, 7th Floor, Buhler Research Centre, The University of Manitoba, 715 McDermot Avenue, Winnipeg, MB, R3E 3P4, Canada. B R I E F R E P O R T JOURNAL OF COMMUNITY PSYCHOLOGY, Vol. 30, No. 3, 235–245 (2002) © 2002 Wiley Periodicals, Inc. Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jcop.10006 for planning and evaluation purposes, of the roles and interactions between key stakeholders. This tool emerged from a community mental health initiative centered on natural caregivers, that commenced in 1995 in Edmonton, Alberta, Canada, and involved the Latin American community, social agencies, and health institutions. The article describes the Latin American Community-based Mental Health Initiative, and exemplifies the use of the stakeholder matrix tool. DESCRIPTION OF THE LATIN AMERICAN COMMUNITY-BASED MENTAL HEALTH INITIATIVE A series of suicides and suicide attempts by members of the Salvadoran community in Edmonton during 1995 prompted the start of a Latin American Community-based Mental Health Initiative ~LACMHI!. These incidents had the effect of mobilizing interest in suicide prevention and on ways of dealing with the issue in this immigrant community. Two Edmonton agencies, The Mennonite Centre for Newcomers and The Support Network, became involved and worked with members of the entire Latin American community to strategize options for effectively responding to these con- cerns. Together with a series of immediate interventions, these agencies ~with funding from Health Canada and The Muttart Foundation, and staff support from the Royal Alexandra Hospital! facilitated a process “for building more in-depth community participation and dialogue.” ~Mennonite Centre for Newcomers, 1996! A first step culminated in the organization of a 10-hour Suicide Prevention Workshop for 26 individuals of the Latin American community in Edmonton. A series of recommen- dations came out of this workshop, which outlined future steps. A second workshop of similar characteristics was held 3 months later. Before this second workshop a broader consultation and information gathering process in the form of a survey was carried ahead. Eighty-one Salvadorans living in Edmonton were sampled with the purpose of identifying community and agency “recommendations regarding a multi-cultural model that would enable an immigrant community to adequately access formalized mental health crisis services . . . ~and to! identify ways of strengthening natural supports including natural caregivers, existing within the com- munity so that the community is better able to self-support around mental health issues” ~Mennonite Centre for Newcomers & Support Network, 1996!. A Latin Amer- ican Community-based Consulting Committee was formed as a way for the Latin American community to take ownership of the survey process and to establish a body that would continue educating and serving the community in issues related to mental health in the future. From this phase of the project the following basic recommendations originated: take steps to increase community awareness of existing services; natural caregivers should be encouraged, supported and trained; efforts should be made to recruit bilingual ~Spanish-English! volunteers for the distress line; additional suicide preven- tion workshops for community members. These recommendations led to the obtain- ment of partial funding from Health Canada for a 3-month follow-up project to recruit and train 11 individuals from the Latin American community. The training, of 60 hours, was on suicide prevention, crisis intervention, bereavement0loss support skills, posttraumatic debriefing, and listening skills. The intention was that this group of 11 continue in a role of natural caregivers. As well, of this group of 11, five individuals were selected to receive further training to form a Spanish speaking “Liaison 236 • Journal of Community Psychology, May 2002 Team” to enable the Mobile Mental Health Crisis Team to provide on-call para- professional support in Spanish. This model was later replicated with other immigrant communities in Edmonton. In the years that followed, further funding from Health Canada was secured for what became known as the “Community-based Immigrant Mental Health Project,” involv- ing five immigrant0refugee communities. The goal was to address the lack of readily accessible and culturally relevant services in mental health, with the rationale that immigrants and refugees experience difficulties in accessing and using the formal mental health system because of linguistic and cultural barriers. The intent was to provide accessible and appropriate mental health services through a network of sup- port and services available to immigrant and refugee families. A further goal was to strengthen community capacities to undertake mental health promotion in the five immigrant0refugee communities involved, through the training of natural caregivers and supporting collective initiatives to respond to specific mental health issues. As stated by Health Canada ~ Joubert, 2000!, “ultimately, the project will demonstrate an authentic process of community–institution partnership, which will have long-term implications in the development of inter and multi-sectoral collaboration in health.” The stakeholder matrix tool introduced in this article is a product of the initial stages of this process of the community–institution partnership, and is thus exemplified here through the LACMHI, and not through its more recent and broader development. The purpose, to enable this type of partnership to assess the initiative from a systemic viewpoint, is nonetheless applicable, with adjustments, to the broader project. RATIONALE FOR THE LATIN AMERICAN COMMUNITY-BASED MENTAL HEALTH INITIATIVE An initiative of the type described above assumes that social support and community participation are determinants of health ~Patrick & Wickizer, 1995; Wallston, Alagna, DeVellis, & DeVellis, 1983; Wilkinson, 1996!. This appears to be particularly pertinent to mental health ~Corin, 1995!. Isolation is a key determinant in mental illness and a particular risk that immigrants face. Although the initiative under study appeared to have been mostly triggered by suicides of community members ~acute episodes!, it opened the door to explore broader mental health concerns. Consequently, the role of natural caregivers within the community did not relate only to suicide. The expec- tation was that these caregivers were a “natural” source of contact for Latin Americans in Edmonton at whatever level of emotional or mental health distress. These caregiv- ers would, if possible, help community members at a primary prevention level. If other interventions were required, they would facilitate the accessibility to secondary and tertiary mental health care of community members. Their role was conceived along the lines of what some authors have called “natural helpers . . . lay people to whom others naturally turn for advice, support, and tangible aid” ~Eng, 1993; Israel, 1982, 1985!, with elements of what Jane Jacobs ~199301961! has called “brokers in public community life.” The creation of the Spanish-speaking liaison team to provide on-call para- professional support to the Mobile Mental Health Crisis Team, addressed the “crisis phase” of mental health. It integrated Latin American mental health workers with a mainstream Canadian crisis intervention agency. This was an important and concrete initiative that had been recommended by the community. Therapeutically speaking, these mental health workers had the cultural and linguistic capability of observing Community Mental Health: Stakeholder Matrix Tool • 237 what is “unobservable” to mainstream mental health workers. Several studies have stated similar notions ~Budman, Lipson, & Meleis, 1992; Minas, Stuart, & Klimidis, 1994; Musser-Granski & Carrillo, 1997!.1 They, however, became one of the first visible sources of tension. In health in general, and in mental health in particular, crisis interventions, because of their urgency and tangible nature, have a tendency to “steal the show” in health initiatives. Institutionalized care and emergency interventions are the “f lashy” sides of mental health ~and health care in general!. Despite their appro- priateness, there seems to be a natural tendency both from the community members’ and institution’s perspectives, to perceive them as meeting broader mental health needs. This became an initial risk and source of tension for the project. If institution- alized care took over, the mental health initiative could have become alienated from broader community participation, defeating in the end the overall goal of the project. The stakeholder matrix tool was created, in part, as way of preventing this type of situation. From a therapeutic perspective, the basic component of the LACMHI was the natural caregiver. The natural caregiver was defined as an active member of the community who had also received training to enhance their normal capabilities, and who had access to resource information. Natural caregivers would play different roles according to their particular circumstances. Some would remain latent within the community, able to involve themselves at the level they considered appropriate when there was a mental health need within their immediate community. Others would have a more active role because of their jobs or their other volunteer activities. As well, a small group, as already mentioned, took on the task of on-call workers with the Mobile Crisis Team, in addition to their general role as natural caregivers. Finally, some natural caregivers would have at times active participation in the Community Consultation Committee and0or other groups that would occasionally be formed to work on particular issues. The common element, at whatever level of participation was that their role would be mostly of primary prevention, both in its health promotion and illness prevention aspects and as a source of appropriate referral and follow-up ~primary prevention also includes follow-up after secondary or tertiary prevention interventions!. The expectation of this model of intervention was to enable emotional0 mental health to be addressed jointly at societal, familial, and individual levels. An added benefit ~which can be of particular relevance for some immigrant groups! was its potential to minimize the taboo around emotional0mental health. DEVELOPMENT OF THE STAKEHOLDER MATRIX TOOL The need to develop a tool to assist in the planning and evaluation of the initiative became apparent early on. Many issues and questions related to the feasibility of the project were raised, which required discussions among the various stakeholders. The following were some of the main questions that arose. How representative was the Latin American Consulting Committee of the Latin American community? How par- ticipatory was it? Was the commitment of the Consulting Committee members waver- ing, becoming more passive vis-à-vis institutional players? What was the source of conf lict 1 “. . . with bilingual clients in a therapeutic session, interpreters are called upon to do much more than translate words. They must communicate subtle meanings, idiomatic expressions, sayings, implied mean- ings, affect, tone of voice, facial expressions, and other non-verbals” ~Egli, 1991!. 238 • Journal of Community Psychology, May 2002 between the Spanish-speaking liaison team and Mobile Mental Health Crisis Team? Was the agenda progressively being taken over by the social agencies or by some of the institutional players? Why was there some resistance from professional mental health staff ? Three premises were considered essential for dealing with these types of questions: that the answers be sought in a collective manner; that they be framed in a nonaccu- satory way; that a relatively simple tool be used to facilitate the analyses. The matrix presented in this article fits the above criteria. Although at first glance it may appear complicated, it does not involve a complex process. It requires the input of key stakeholders, and it enables the exploration of the questions from a systemic view, consequently minimizing the possibility of “scapegoating” any of the players. The sources of information used to develop the matrix were the following: inter- views and group discussions with project stakeholders, and project background doc- umentation ~Bigras, Chagnon, & Eastace, 1996; Healthy Community Development Network, 1997; Mennonite Centre for Newcomers, 1996; Mennonite Centre for New- comers & The Support Network, 1996; The Support Network & Mennonite Centre for Newcomers, 1997; Walter, 1995!. Five interviews were held with family members of suicide attempters of the Latin American community; one focus group with five natural caregivers, two interviews with the main facilitator of the project, one inter- view with the coordinator of the mobile crisis team, and one interview with director of The Mennonite Centre for Newcomers. Five discussion meetings were held with project stakeholders during the elaboration of the matrix. The first meeting essentially worked on clarifying the purposes of the tool. Following a naturalist inquiry approach ~Lin- coln & Guba, 1985!, the subsequent meetings entailed group discussions that helped to verify information, generate ideas, analyze data, receive feedback and negotiate outcomes with stakeholders at different stages of its development. IMPLEMENTATION OF THE STAKEHOLDER MATRIX TOOL The use of the matrix tool involves several steps. First, stakeholders need to be identified2 and grouped in meaningful categories. Second, the roles of each category of players should be specified. Third, the interactions determined, and, finally, ten- sion areas examined. Table 1 identifies components of the matrix as a general case. Stakeholders are headers of rows and columns. Roles are located at the intersection of the same stakeholders. Interactions are specified as column for a stakeholder in interaction with row of a different stakeholder. Tension areas are identified within the same cell, but in italics. Table 2 illustrates the first step where the LACMHI identified five categories of stakeholders, specifying who the players were in each category. First, the immigrant community itself; second, the community consultation group formed by community members; third the natural caregivers, who given the nature of the initiative had a central role; fourth, social agencies collaborating with the project; and last, institu- tional players, for the most part members of the formal health care system. This categorization came about by understanding the distinct nature of each group. Most of the stakeholder categories are self-explanatory. However, the Institutional Player 2 In a process similar to that proposed by Patton ~1997! for utilization-focused evaluations. Community Mental Health: Stakeholder Matrix Tool • 239 merits some clarification. Under this term the matrix table includes health care institutions and health professionals that deliver direct service ~hospitals, health cen- ters, general practitioners, psychiatrists, psychologists, etc.! as well as governmental or para-governmental entities like Health Canada, Alberta Health, Alberta Mental Health, and Regional Health Authorities. Under Institutional Players the matrix also includes governmental and nongovernmental agencies that are not exclusively involved in health issues, but that may provide funding or have other inf luences on health initiatives. Table 2 spells out specific information of what took place in interactions between stakeholders. Linked to this categorization was the clarification of distinct roles played by each stakeholder. The use of the matrix imposed the need to make explicit these roles from the very beginning ~with the understanding that roles may evolve and change over time!. In some cases the role may not be initially clear, as is evident with the category Community. The analysis of the interactions imposed a dynamic understanding of these roles, by focusing specific observations of how one stakeholder interacts with another. The in- teractions can have both an ideal component, i.e., how the stakeholders as a whole con- ceive the interactions should be, and an empirical component , where evidence is interpreted about how these interactions actually occurred. Table 3 illustrates this pro- cess of interpreting evidence, which lead to the final step, i.e., the identification of ten- sion areas. Tension areas are essentially indicants of role conf licts between players, of deviance from original roles, or of overly passivity or aggressiveness by some stakeholders. DISCUSSION The LACMHI can be identified with ideas put forward by researchers studying social networks and social support in community mental health. Gottlieb ~1981a, p. 17! states, among several themes foreshadowing the role of informal helping networks in health maintenance, the “importance of a diversity of informal helping resources that are ubiquitous in the community and that ought to be drawn into any comprehensive plan to meet the mental health needs of citizens.” He proposes the mobilization of informal resources focusing “on improving the supportive quality of network con- Table 1. Stakeholder Matrix: General Case Stakeholder 1 Stakeholder 2 Stakeholder 3 Stakeholder 4 Stakeholder 5 S1 ROLE Interactions0 tension areas Interactions0 tension areas Interactions0 tension areas Interactions0 tension areas S2 Interactions0 tension areas ROLE Interactions0 tension areas Interactions0 tension areas Interactions0 tension areas S3 Interactions0 tension areas Interactions0 tension areas ROLE Interactions0 tension areas Interactions0 tension areas S4 Interactions0 tension areas Interactions0 tension areas Interactions0 tension areas ROLE Interactions0 tension areas S5 Interactions0 tension areas Interactions0 tension areas Interactions0 tension areas Interactions0 tension areas ROLE Matrix should be read as follows: columns in interaction with rows ~for example, stakeholder 1 in interaction with stakeholder 2!. 240 • Journal of Community Psychology, May 2002 T a bl e 2. L A C M H I St a ke h ol d er M a tr ix : St a ke h ol d er Id en ti fi ca ti on a nd A ct io ns P la ye rs C om m u n it y C om m u n it y C on su lt at io n G ro u p N at u ra l C ar eg iv er s So ci al A ge n ci es In st it u ti on al P la ye rs C L at in A m er ic an co m m u n it y in E d m o n to n R ep re se n te d C o m m u n it y co n ce rn s o n m en ta l h ea lt h is su es . A ct iv e m em b er s o f L at in A m er ic an C o m m u n it y; h av e d o n e p ri m ar y p re ve n ti o n w o rk w it h in co m m u n it y; h av e w o rk ed as li ai so n w o rk er s at M o b il e C ri si s T ea m . M en n o n it e C en tr e an d Su p p o rt N et w o rk st ra te gi ze d w it h L at in A m er ic an co m m u n it y o n h o w to d ea l w it h su ic id e is su e. H ea lt h C an ad a an d M u tt ar t F o u n d at io n p ro vi d ed fu n d in g; R o ya l A le xa n d ra H o sp it al p ro vi d ed st af f su p p o rt ; M o b il e C ri si s T ea m — Su p p o rt N et w o rk in cl u d ed L ia is o n W o rk er s in th ei r te am . C C G W as re p re se n te d to so m e d eg re e th ro u gh th e L at in A m er ic an C o n su lt in g C o m m it te e. L at in A m er ic an co m m u n it y- b as ed co n su lt in g co m m it te e So m e N at u ra l C ar eg iv er s p ar ti ci p at ed in th e L at in A m er ic an C o n su lt in g C o m - m it te e; N at u ra l C ar eg iv er s w as in fo rm ed o f L at in A m er ic an C o n su lt in g C o m - m it te e’ s re co m m en d at io n s. H el p ed o rg an iz e Sa lv ad o ra n C o n su lt in g C o m m it te e an d L at in A m er ic an C o n su lt in g C o m m it te e. N o d ir ec t in te ra ct io n . N C G U se d N at u ra l C ar eg iv er s w it h in C o m m u n it y. P ro vi d ed in fo rm at io n to N at u ra l C ar eg iv er s o n C o m m u n it y n ee d s an d p er ce p ti o n s. W o rk sh o p p ar ti ci p an ts ; gr o u p o f 11 ; li ai so n w o rk er s P ro vi d ed o rg an iz at io n al su p p o rt fo r W o rk sh o p s; re so u rc e su p p o rt fo r G ro u p o f 11 ; tr ai n in g fo r L ia is o n W o rk er s. P ro vi d ed su p p o rt to w o rk sh o p s; T ra in ed L ia is o n W o rk er s; In co rp o ra te d L ia is o n w o rk er s in M o b il e C ri si s T ea m . SA C o m m u n it y st ra te gi ze d w it h M en n o n it e C en tr e an d Su p p o rt N et w o rk o n h o w to d ea l w it h su ic id e an d b ro ad er m en ta l h ea lt h is su es . W o rk ed w it h M en n o n it e C en tr e an d Su p p o rt N et w o rk p ro vi d in g in fo rm at io n o n C o m m u n it y n ee d s an d re co m m en d at io n s. R ec ei ve d tr ai n in g fr o m M en n o n it e C en tr e an d Su p p o rt N et w o rk ~t h ro u gh W o rk sh o p s! ; re ce iv ed su p p o rt fr o m M en n o n it e C en tr e an d su p p o rt n et w o rk . M en n o n it e C en tr e fo r N ew co m er s; S u p p o rt N et w o rk W o rk ed w it h so ci al ag en ci es fo r w o rk sh o p s. P ro vi d ed fu n d in g fo r p ro je ct . IP U se d M o b il e C ri si s T ea m ~3 5 ca ll s si n ce h av in g L ia is o n W o rk er s! ; So u gh t co u n se li n g. P ro vi d ed re co m m en d at io n s th at le d to in cl u si o n o f L ia is o n W o rk er s in M o b il e C ri si s T ea m . R ec ei ve d tr ai n in g fr o m M o b il e C ri si s T ea m ~S u p p o rt N et w o rk !; re ce iv e tr ai n in g su p p o rt fr o m R o ya l A le xa n d ra H o sp it al ; so m e N at u ra l C ar eg iv er s w o rk w it h M o b il e C ri si s T ea m as L ia is o n W o rk er s. M en n o n it e C en tr e an d Su p p o rt N et w o rk w o rk ed w it h M o b il e C ri si s T ea m an d R o ya l A le xa n d ra H o sp it al to o rg an iz e w o rk sh o p s an d tr ai n in g o f L ia is o n W o rk er s. M o b il e C ri si s T ea m ; M u tt ar t F o u n d at io n ; H ea lt h C an ad a; R o ya l A le x an d ra H o sp it al M at ri x sh o u ld b e re ad as fo ll o w s: co lu m n s in in te ra ct io n w it h ro w s ~f o r ex am p le , co m m u n it y in in te ra ct io n w it h co m m u n it y co n su lt at io n gr o u p o r so ci al ag en ci es in in te ra ct io n w it h n at u ra l ca re gi ve rs !. C o d es : C � C o m m u n it y; C C G � C o m m u n it y co n su lt at io n gr o u p ; N C G � N at u ra l ca re gi ve rs ; SA � So ci al ag en ci es ; IP � In st it u ti o n al p la ye rs . It al ic s � T en si o n ar ea s. Community Mental Health: Stakeholder Matrix Tool • 241 T a bl e 3. L A C M H I St a ke h ol d er M a tr ix : R ol es , In te ra ct io ns , T en si on A re a s P la ye rs C om m u n it y C om m u n it y C on su lt at io n G ro u p N at u ra l C ar eg iv er s So ci al A ge n ci es In st it u ti on al P la ye rs C Im m ig ra n t C o m m u n it y R ep re se n ts C o m m u n it y m em b er s; w o rk s w it h C o m m u n it y is su es an d co n ce rn s; ad vo ca te s fo r C o m m u n it y. Is ol at es it se lf fr om C om m u n it y, be - co m es a sm al l al ie n at ed bo dy . D ie s du e to la ck of co m m u n it y pa rt ic ip an ts . W o rk w it h in C o m m u n it y; li st en s to C o m m u n it y co n ce rn s. R es p o n d s to C o m m u n it y n ee d s. L in k C o m m u n it y m em b er s w it h o th er re so u rc es . N C G ’s pr of es si on al iz e th em se lv es ; co m - m u n it y ha s di ff ic u lt y ac ce ss in g N C G . St ra te gi ze w it h C o m m u n it y o n h o w to d ea l w it h m en ta l h ea lt h is su es . So ci al A ge n ci es im po se th e ag en da ; n ot hi n g is do n e w it ho u t So ci al A ge n ci es in vo lv em en t. P ro vi d e fu n d in g fo r C o m m u n it y in it ia ti ve s; li st en to co m m u n it y co n ce rn s an d su gg es ti o n s an d ch an ge s ac co rd in gl y. D is re ga rd C om m u n it y is su es an d re co m m en da ti on s: Im po se ow n m od el as ex cl u si ve m en ta l he al th m od el . C C G R ai se s is su es to C o m m u n it y C o n su lt at io n G ro u p ; p ro - m o te s p ar ti ci p at io n in C o m - m u n it y C o n su lt at io n G ro u p . Ig n or es th e ex is te n ce of C om - m u n it y C on su lt at io n G ro u p; do es n ’t ra is e is su es to C om - m u n it y C on su lt at io n G ro u p. R ep re se n ts an d A d vo ca te s Im m ig ra n t C o m m u n it y A re in vo lv ed in th e C o m m u n it y C o n su lt at io n G ro u p ; li st en to C o m m u n it y C o n su lt at io n G ro u p re co m m en d at io n s an d ta ke id ea s to C o m m u n it y C o n su lt at io n G ro u p . D o n ot ac kn ow le dg e ro le of C om m u n it y C on su lt at io n G ro u p; is ol at e fr om C om m u n it y G ro u p is su es . P ro vi d e in it ia l o rg an iz at io n al su p p o rt to C o m m u n it y C o n - su lt at io n G ro u p an d so m e m in im al o n go in g re so u rc e su p p o rt . D ri ve C om m u n it y C on su lt at io n G ro u p; C om m u n it y C on su lt at io n G ro u p be co m es de pe n de n t on So ci al A ge n ci es . A w ar e o f th e ex is te n ce o f C o m m u n it y C o n su lt at io n G ro u p ; li st en to C o m m u n it y C o n su lt at io n G ro u p re co m m en d at io n s. D is re ga rd le gi ti m ac y of C om m u n it y C on su lt at io n G ro u p; im po se ow n he ge m on ic m od el ov er C om m u n it y C on su lt at io n G ro u p’ s in it ia ti ve s. N C G M ak es go o d u se o f N at u ra l C ar eg iv er s; fo st er s th e d ev el o p - m en t o f m o re N at u ra l C ar eg iv er s. D oe sn ’t u se N at u ra l C ar eg iv er s; do es n ’t co n si de r N at u ra l C ar eg iv er s as a C om m u n it y re so u rc e; is ol at es N at u ra l C ar eg iv er s fr om C om m u n it y. P ro m o te s N at u ra l C ar eg iv er s an d p ro vi d es th em w it h su p p o rt , id ea s. Se ek s fu n d in g fo r tr ai n in g. L ea ve s m aj or de ci si on s in ha n ds of N at u ra l C ar eg iv er s; do es n ot in vo lv e N at u ra l C ar eg iv er s or is ol at es th em . T h er ap eu ti c ro le w it h in C o m m u n it y. B ro ke r ro le . P ro vi d e tr ai n in g su p p o rt ; re - so u rc e su p p o rt ; co n ta ct s an d in fo rm at io n . T ak e ov er N at u ra l C ar eg iv er s’ in i- ti at iv es ; pr of es si on al iz e N at u ra l C ar eg iv er s; de n ie s th em n ec es sa ry su pp or t. P ro vi d e te ch n ic al tr ai n in g an d su p p o rt ; p ro vi d e re so u rc e in fo r- m at io n . D o n ot re co gn iz e u n iq u e kn ow le dg e of N at u ra l C ar eg iv er s; pr of es si on al iz e N at u ra l C ar eg iv er s. SA C o m m u n it y ap p ro ac h es So ci al A ge n ci es to w o rk to ge th er o n m en ta l h ea lt h is su es . C om m u n it y le av es is su es in ha n ds of So ci al A ge n ci es an d re m ai n s pa ss iv e. W o rk s in p ar tn er sh ip w it h So ci al A ge n ci es to fo st er co m m u n it y- b as ed m en ta l h ea lt h ap p ro ac h . L ea ve s le ad er sh ip ex cl u si ve ly to So ci al A ge n ci es ; do es n ot in fl u en ce So ci al A ge n ci es . W o rk w it h So ci al A ge n ci es fo r o n go - in g tr ai n in g, co n ta ct s, in fo rm at io n . D ep en d ex cl u si ve ly on So ci al A ge n ci es ; do n ot in fl u en ce So ci al A ge n ci es . In it ia l o rg an iz at io n al fa ci li ta to r ro le . P ro vi d e so m e re so u rc e su p p o rt . H el p w it h in st it u ti o n al li n ka ge s. W o rk as p ar tn er s w it h So ci al A ge n ci es to su p p o rt co m m u n it y m en ta l h ea lt h in it ia ti ve s. O ve rr id e So ci al A ge n ci es ; m ak e So ci al A ge n ci es fi n an ci al ly de pe n de n t on In st it u ti on al P la ye rs . IP C o m m u n it y m em b er s ac ce ss ap p ro p ri at e ca re w it h In st it u - ti o n al P la ye rs ; co m m u n it y in fl u - en ce s In st it u ti o n al P la ye rs . C om m u n it y re m ai n s pa ss iv e in re la - ti on to In st it u ti on al P la ye rs ’ po li ci es ; ty pe of ca re an d ac ce ss ib il it y. L o b b ie s In st it u ti o n al P la ye rs fo r cu lt u ra ll y ap p ro p ri at e ca re , fo r ac ce ss ib il it y, et c. A cc ep ts st at u s qu o of In st it u ti on al P la ye rs ; in it ia ti ve s ar e ab so rb ed by In st it u ti on al P la ye rs ’ he ge m on ic ca re m od el . K n o w le d ge o f In st it u ti o n al P la ye rs ’ ty p es o f ca re an d re fe r co m m u n it y m em b er s to ap p ro p ri at e ca re ; so m e al so w o rk w it h in h ea lt h ca re in st it u ti o n s. R el y on In st it u ti on al P la ye rs ca re m od el n ot va lu in g ow n po te n ti al ; ar e ab so rb ed by In st it u ti on al P la ye rs . W o rk w it h In st it u ti o n al P la ye rs to o rg an iz e tr ai n in g: In fl u en ce In st it u ti o n al P la ye rs p o li ci es ; lo b b ie s th em fo r fu n d in g. R em ai n s pa ss iv e to In st it u ti on al P la ye rs he ge m on ic ca re m od el ; to ta l de pe n de n cy on In st it u ti on al P la ye rs fu n di n g. P ro vi d es cr is is in te rv en ti o n . P ro vi d es se co n d ar y an d te rt ia ry p re ve n ti o n in te rv en ti o n s. M at ri x sh o u ld b e re ad as fo ll o w s: co lu m n s in in te ra ct io n w it h ro w s ~f o r ex am p le , co m m u n it y in in te ra ct io n w it h co m m u n it y co n su lt at io n gr o u p o r so ci al ag en ci es in in te ra ct io n w it h n at u ra l ca re gi ve rs !. C o d es : C � C o m m u n it y; C C G � C o m m u n it y co n su lt at io n gr o u p ; N C G � N at u ra l ca re gi ve rs ; SA � So ci al ag en ci es ; IP � In st it u ti o n al p la ye rs . It al ic s � T en si o n ar ea s. 242 • Journal of Community Psychology, May 2002 tacts.” The LACMHI parallels his suggestion of “focused strategies of teaching helping skills . . . to categories of people who are known to be approached frequently as informal helping agents . . .” ~Gottlieb, 1981b, p. 211! It also goes further, along the lines of what other authors ~Froland, Pancoast, Chapman, & Kimboko, 1981! identify as fostering a linkage between formal and informal support systems. Ultimately, the LACMHI links social support and social networks, consistent with what Israel ~1985! categorizes as “programs enhancing entire networks through natu- ral helpers.” This social network model is favorably argued by D’Augelli ~1983! as having the potential to provide “a framework to help the isolated and to enrich the development of those embedded in social networks.” For this to be so, the monitoring of a program of these characteristics requires that it be assessed from a systemic point of view. Questions about the impact of the project on natural helpers in terms of both their skills and the way they relate to their networks, or about changes in the profes- sional service system ~Mitchell & Hurley, 1981, p. 292!, for example, can only be answered from that vantage point. Further, monitoring of these programs need to be consistent with the main phi- losophy, that is of the involvement of stakeholders. For this to be feasible, a relatively simple process of planning and evaluation should be established to realistically foster this involvement. The stakeholder matrix tool presented here has the potential to help in this endeavour because of its relative simplicity and several of its characteris- tics: It is a relatively clear-cut tool, involving straight steps and a plain structure for analysis. It overtly addresses tensions and potential sources of conf lict in community programs that by their very nature involve various stakeholders of very different char- acteristics. It can help to normalize the existence or potential for existence, of con- f lict. It enables a simplified systemic view. It provides a systemic structure for stakeholder discussion, consequently minimizing the possibility for scapegoating of any particular group. It enables a dynamic examination of the process. Table 3 illustrates the use of the matrix for the LACMHI. It shows five stakeholder categories: Latin American community in Edmonton, community consultation group, natural caregivers0liaison workers, social agencies, institutional players, that resulted from the following process. A specific community ~Salvadoran! that was part of a larger related cultural community ~Latin American! suffered acute mental illness incidents. Community members together with some social agencies agreed on the need to act. Through workshops and a broader consultation process, needs and recommendations were identified. The recommendation that took the most concrete form was the one related to crisis response and to accessibility to formal care ~health care institutions, institutional players!. There was a group ~in a sense intermediating between community and agencies0institutions! put in place for natural caregivers to interact, and a broader community group ~Latin American Community-based Consulting Committee! to advice “its community and the relevant agencies on mental health issues concerning the Latinos” ~Support Network & Mennonite Centre for Newcomers, 1997!. Because there is a tendency to avoid the explicit examination of conf lict or of potential conf lict in planning and evaluating community programs, a tool of this type enabled its observation within a framework where it was normalized. The notion is that interactions between players will eventually create areas of tension, which consti- tute ongoing challenges to the initiative.3 Tension areas simultaneously provide infor- 3 Hoping that tensions will not occur and0or trying to ignore them, ultimately makes things worse. Community Mental Health: Stakeholder Matrix Tool • 243 mation about the vitality of the process, and about possible breakdown points. To exemplify, the following were a series of possible tension areas faced by the LACMHI: tension between formal care and informal care; tension between acute care and prevention0promotion; tension between institutionalization and community participa- tion; tension between agency leadership and community leadership ~or “lack of ” from either player!; tension between “specialized” professionals and “natural” caregivers; tension between natural caregivers with institutional roles and those without institu- tional roles. Although tension areas may seem to be the negative side of the interactions, they are actually a key source of information of the initiative’s dynamics because they signal areas of conf lict, of lack of vitality, etc. They provide light on possible necessary adjustments to the program. Ultimately, the examination and discussion of these tension areas by representatives of the different stakeholders holds the most promise for ensuring the initiative’s ongoing renewal and success. An interesting use of this matrix tool is the possibility of anticipation. If this type of analysis is done during the planning stages, there can be stakeholder speculation concerning future potential areas of tension, enabling useful foresight. Using this matrix tool, the examination of the LACMHI brought to the forefront several challenges that the initiative of this type faced. One has been mentioned previously: the “narrowing” of mental health issues to the crisis phase ~and, therefore, mostly dealing with acute mental health problems!. This also relates to the tendency for institutionalized care to take over as the central player. Added to a tendency of some community-based activities to lose participants and the dynamism of the earlier mobilized period, there was a risk of losing the essential nature of this type of initia- tive ~i.e., that it be community driven!. The versatility of stakeholder matrix derives from the fact that it can be imple- mented at any stage of an initiative or program. As well, it can be used both for planning purposes and evaluation purposes ~mainly, although not exclusively, forma- tive evaluation!. Community-based programs need to explicitly examine, on an ongo- ing basis, the roles and interactions of its different players and identify tension areas. The lack of this examination undermines the possibility for community-based initia- tives to overcome the many challenges they face. Central to this examination is that it be done with the involvement of the different stakeholders as part of a common exercise. The stakeholder matrix tool offers a relatively straightforward structure from where to initiate this process, be it from a planning and0or evaluation perspective. REFERENCES Bigras, R., Chagnon, F., & Eustace, J. ~1996!. A comparative needs analysis of mental health crisis response services in Edmonton: An integrative approach. Edmonton, Alberta: The Support Network. Budman, C.L., Lipson, J.G. & Meleis, A.L. ~1992!. The cultural consultant in mental health care: the case of an Arab adolescent. American Journal of Orthopsychiatry, 62~3!, 359 –370. Corin, E. ~1995!. The cultural frame: Context and meaning in the construction of health. In B.C. Amick, S. Levine, A.R. Tarlov, & D. Chapman Walsh ~Eds.!, Society and health ~pp. 272– 304!. New York: Oxford University Press. D’Augelli, A. ~1983!. Social support networks in mental health. In J.K. Whittaker & J. Garbarino ~Eds.!, Social support networks ~pp. 1–106!. Hawthorne, N Y: Aldine. 244 • Journal of Community Psychology, May 2002 Egli, E.A. ~1991!. Bilingual workers. In J. Westermeyer, C.L. Williams, & A.N. Nguyen ~Eds.!, Mental health services for refugees ~DHHS Publication No. ADM 91–1824! ~pp. 90 –110!. Washington, DC: U.S. Government Printing Office. Eng, E. ~1993!. The save our sisters project: A social network strategy for reaching rural black women. Cancer, 72, 1071–1077. Froland, C., Pancoast, D.L., Chapman, N.J., & Kimboko, P.J. ~1981!. Linking formal and infor- mal support systems. In B.H. Gottlieb ~Ed.!, Social networks and social support ~pp. 259 – 275!. Beverly Hills, CA: Sage. Gottlieb, B.H. ~1981a!. Social networks and social support in community mental health. In B.H. Gottlieb ~Ed.!, Social networks and social support ~pp. 11– 42!. Beverly Hills, CA: Sage. Gottlieb, B.H. ~1981b!. Preventive interventions involving social support networks and social support. In B.H.Gottlieb ~Ed.!, Social networks and social support ~pp. 201– 232!. Beverly Hills, CA: Sage. Healthy Community Development Network. ~1997!. Report on the Vietnamese mental health needs assessment survey. Edmonton, Alberta: MCN. Israel, B.A. ~1982!. Social networks and health status: Linking theory, research and practice. Patient counselling and health education, 4, 65–79. Israel, B.A. ~1985!. Social networks and social support: Implications for natural helper and community level interventions. Health Education Quarterly, 12, 65–80. Jacobs, J. ~199301961!. The death and life of great American cities. New York: Modern Library. Joubert, N. ~2000!. Promoting the best ourselves: Mental health promotion in Canada. Keynote paper. National Conference on Mental Health Promotion: Perspectives and Practices. Gal- way: National University of Ireland. Lincoln, Y.S., & Guba, E.G. ~1985!. Naturalistic inquiry. Newbury Park, CA: Sage Publications. Mennonite Centre for Newcomers, & The Support Network. ~1996!. Healing ourselves—Healing each other. Relevance and accessibility of mental health crisis services: A consultation with the Salvadorian community in Edmonton. Edmonton, Alberta: MCN & The Support Network. Mennonite Centre for Newcomers. ~1996!. Suicide prevention initiative with Latin American community members: Final report. Edmonton, Alberta: MCN. Mitchell, R.E., & Hurley, D.J. ~1981!. Collaboration with natural helping networks: Lessons from studying paraprofessionals. In B.H. Gottlieb ~Ed.!, Social networks and social support ~pp. 277– 298!. Beverly Hills, CA: Sage. Minas, I.H., Stuart, G.W., & Klimidis, S. ~1994!. Language, culture and psychiatric services: A survey of Victorian clinical staff. Australian and New Zealand Journal of Psychiatry, 28~2!, 250 – 258. Musser-Granski, J., & Carrillo, D.F. ~1997!. The use of bilingual, bicultural paraprofessionals in mental health services: issues for hiring, training, and supervision. Community Mental Health Journal, 33~1!, 51– 60. Patrick, D.L., & Wickizer, T.M. ~1995!. Community and health. In B.C. Amick, S. Levine, A.R. Tarlov, & D. Chapman Walsh ~Eds.!, Society and health. New York: Oxford University Press. Patton, M.P. ~1997!. Utilization focused evaluation: The new century text. Thousand Oaks, CA: Sage Publications. Support Network, The & Mennonite Centre for Newcomers. ~1997!. Report on the Latin Amer- ican community crisis support project. Edmonton, Alberta: The Support Network & MCN. Wallston, B.S., Alagna, S.W., De Vellis, B.M., & De Vellis, R.F. ~1983!. Social support and physical health. Health Psychology, 2~4!, 367–391. Walter, L. ~1995!. Newcomer suicide prevention project: Immigration, culture and suicide. Edmonton, Alberta: MCN. Wilkinson, R.G. ~1996!. Unhealthy societies: The aff lictions of inequality. New York: Routledge. Community Mental Health: Stakeholder Matrix Tool • 245 INTRODUCTION DESCRIPTION OF THE LATIN AMERICAN COMMUNITY-BASED MENTAL HEALTH INITIATIVE RATIONALE FOR THE LATIN AMERICAN COMMUNITY-BASED MENTAL HEALTH INITIATIVE DEVELOPMENT OF THE STAKEHOLDER MATRIX TOOL IMPLEMENTATION OF THE STAKEHOLDER MATRIX TOOL Table 1. Stakeholder Matrix: General Case DISCUSSION Table 2. LACMHI Stakeholder Matrix: Stakeholder Identification and Actions Table 3. LACMHI Stakeholder Matrix: Roles, Interactions, Tension Areas REFERENCES