|Home | About | Journals | Submit | Contact Us | Français|
In 2005, New Mexico implemented a unique reform in managed behavioral health services that seeks to ensure delivery of consumer-driven, recovery-oriented care to low-income individuals. Distinguishing features of the reform are the Local Collaboratives (LCs), regionally based community organizations designed by state government to represent behavioral health concerns of New Mexico's diverse cultural populations. We examine community response to the LCs, focusing on two broad sets of themes derived from 18 months of ethnographic fieldwork. The first set—structure and function—encompasses several issues: predominance of provider versus consumer voice; insufficient resources to support internal operations; imposition of state administrative demands; and perceived lack of state response to LC efforts. The second set—participation and collaboration—reveals how problems of information flow and other logistical factors impact involvement in LCs and how the construction of “community” introduced under this initiative exacerbates tensions across localities with varied histories and populations.
In July 2005, the State of New Mexico undertook a unique reform in managed care, entrusting a single for-profit corporation, referred to locally as the Statewide Entity (SE),1 to administer all publicly financed behavioral health care to its vulnerable populations, including individuals from predominantly poor and rural areas and those with serious mental illness or substance-abuse problems. The reform's architects emphasized that behavioral health care—typically encompassing a range of preventative and treatment services for mental and emotional disorders or chemical addictions—would be based on principles of cost effectiveness and performance. Services would also be recovery oriented, consumer driven, and responsive to the state's diverse communities (Hyde 2004).
In keeping with national trends to include “consumers”2 in the development and evaluation of health services, leaders of state agencies central to the reform, grouped together as the New Mexico Interagency Behavioral Health Purchasing Collaborative, or the Purchasing Collaborative (PC),3 called for the creation of Local Collaboratives (LCs). The PC consists of cabinet secretaries and agency directors from the state's largest purchasers of behavioral health care—the Departments of Human Services, Health, and Children, Youth and Families, along with representatives from agencies specializing in labor, vocational rehabilitation, education, housing, and corrections. The PC is designed to advance partnerships at the highest levels of state government to streamline service delivery, reduce administrative costs, and improve quality oversight within the public behavioral health system.
The charge of the LCs is to advise the PC and the SE on strategies to enhance care and quality of life for consumers and families, based on local needs and realities. The PC and SE, in turn, are to provide LCs with technical assistance to meet their goals. In the state-issued recruitment flyer, consumers are encouraged to “claim” their “voice” by joining LCs. These consumers include persons with debilitating illnesses, such as schizophrenia, major depression, and bipolar disorder, who commonly experience social and economic marginalization. Traditionally, these persons have not performed active roles in state policy reform and have had limited opportunities to garner skills to maximize their contributions in this area. Nevertheless, key state officials envision the LCs as a national precedent for greater consumer and community input into public behavioral health systems.
In this article, we examine community responses to the LC model and how the model influences the expression of “voice” by consumers, family members, and providers. We consider how, despite an unprecedented recognition of consumer voice, power asymmetries have fundamentally shaped the model's implementation, replicating social inequalities already existent between providers and consumers, and within historically entrenched, culturally distinct communities. The following questions guide this discussion: (1) Whose voices are represented in the LC initiative? (2)What factors affect participation in LCs? (3) To what degree do LC participants feel engaged in direct dialogue with state officials about policy reform? (4) Does the LC model facilitate effective community input into public policy?
Community-based participatory approaches have gained momentum within the past decade and are increasingly accepted as essential components in addressing social problems and health disparities (Butterfoss 2007; Israel et al. 2005; Minkler and Wallerstein 2008). Proponents contend that community input enhances public behavioral health care, by helping to ensure that services are responsive to local needs (Gonzales and Moten 2006).
Anthropologists have long argued for the importance of involving community members in policies and programs that impact local concerns, based on notions of justice, acknowledgment of local expertise, and recognition that engaged communities are more likely to support programs they help shape (Greenwood and Levin 1998; Schensul and Schensul 1992). Anthropologists have also critiqued the potential that exists for cooptation of community needs and values to promote external agendas, if “participation” is not truly collaborative and democratic (Cooke and Kothari 2001).
Interest in collaborative approaches has grown in public health research, notably as community-based participatory research or CBPR, an umbrella term for an assortment of strategies aimed at building “community capacity” to tackle health issues at local levels (Israel et al. 2005; Minkler and Wallerstein 2008). CBPR principles emphasize a cooperative process whereby contributions of community members and researchers have “equal” weight. Research questions, data collection, analysis, and subsequent action are guided by community members. One fundamental purpose of community-based approaches is to overturn the traditional power imbalance between researchers and communities by advancing local ownership over research and democratizing knowledge. A second is to mobilize this knowledge to effect social change and improve conditions within communities (Minkler and Wallerstein 2008; Stringer 2007).
Because CBPR represents a paradigmatic shift in how researchers interact with community members—the traditional subjects and objects of their studies—the entire CBPR enterprise is about power. Research has examined why and how issues of race, ethnicity, privilege, and historical trauma shape relationships between “outside researchers” and marginalized communities (Chávez et al. 2008). Other work explores strategies to engage diverse stakeholders (Minkler and Wallerstein 2008). Increasingly, explicit attention is given to power vis-à-vis participation throughout the CBPR experience, from planning to implementation (Gaventa and Cornwall 2008; Hampshire et al. 2005; Wallerstein 1999), and various approaches have been proposed to address power imbalances to create more equitable relationships (Wallerstein and Duran 2006).
Community partnerships are also being viewed as vital to health promotion and disease prevention. In recent years, collaborative groups and coalitions of concerned citizens and institutional partners have addressed a host of public health issues in community settings (Butterfoss 2007), and there is also interest in such partnerships within the behavioral health arena (Garland et al. 2006; Institute of Medicine 2006; Wells et al. 2004). The President's New Freedom Commission on Mental Health (2003)—a group of policymakers, practitioners, and administrators who conducted a comprehensive study of behavioral health service delivery in the United States—have challenged states to develop such partnerships, as have other government agencies. The U.S. Substance Abuse and Mental Health Services Administration (SAMHSA) has spearheaded the nationwide charge for “meaningful participation” from consumers and families in all aspects of behavioral health research, legislation, and service delivery.4 The National Institute for Mental Health (2008) also underscores the importance of community engagement in its strategic plan.
The literature on CBPR and community partnerships emphasizes factors that affect collaboration, including legitimacy accorded to partnerships in the broader community; political and social support; funding and resource access; pursuing goals at a reasonable pace; openness to varied decision-making modes; recognition of local expertise; awareness of cultural differences and one's own biases; understanding the historical context of community and research problems; incorporating critical reflection into the participatory process; and fostering respect, understanding, and compromise (Butterfoss 2007; Israel et al. 2005; Minkler and Wallerstein 2008). Despite attention to such factors, the actual process by which power shifts within community partnerships still remains “elusive” in much of the literature (Minkler and Baden 2008).
Participation is a celebrated ideal in research and policy (Cleaver 2001; Hampshire et al. 2005). However, as Bill Cooke and Uma Kothari (2001:13) warn, participation can “conceal and reinforce oppressions and injustices.” Participatory approaches, particularly those implemented at the behest of a dominant agent, can bolster power differentials in subtle ways that may never appear in public transcripts—meeting minutes, reports to funding agencies, and academic articles. Analysis of power asymmetries that shape relationships framed as participatory is therefore central to the study of collaborative processes (Gavanta and Cornwall 2008).
Likewise, normative constructions of “community” warrant further examination. Raymond Williams (1976) observes that community, which entered into Western vocabularies in the 14th century, is a “warmly persuasive word” that can describe either existing or new sets of social relationships, but unlike all other terms of social organization, such as state, nation, or society, it tends to have positive connotations. Recent theorists, such as Gerald Creed (2007), argue that invocations of community in Western economic and political discourse, particularly in development projects, often rest on problematic assumptions of consensus, conformity, and solidarity within a particular set of social relationships. The very term serves as a “heuristic designation” steeped in emotionally evocative, “taken-for-granted understandings … that are so internalized or routinized as to escape comment” (Creed 2007:4). In government-led policy reform, notions of community can simplify and homogenize local contexts, making them “legible” and “subordinate” to the state, while also facilitating administration (Creed 2007:8). Such imposed definitions of community, if they do not reflect local understandings and realities, can create barriers for participatory efforts.
New Mexico is culturally diverse, with the highest proportion of the total population identifying as Hispanic or Latino, 42.1 percent as compared to 12 percent of the U.S. population (Guzman 2001). It is second highest with respect to Native Americans, at nearly 10 percent as compared to under 1 percent of the U.S. population (Ogunwole 2002). New Mexico leads the nation in death rates due to alcohol and drug use, suicide, and homicide (Morgan and Morgan 2008). Behavioral health disparities affect ethnic minorities disproportionately, especially those of low income who face substantial structural inequalities (Institute of Medicine 2003).
Several other social and demographic factors distinguish New Mexico from the rest of the country. It recently ranked 47th in personal income per capita ($24,291), third in persons below the poverty level (18.4%), and second in lack of health insurance (22.1%). About 21 percent of the population of 1.8 million is covered by Medicaid. New Mexico is largely rural, averaging 15 persons per square mile, and all but one of its counties are designated as Health Professional Shortage Areas or Medically Underserved Areas (Waitzkin et al. 2008).
Our research on earlier reform suggested that introducing Medicaid managed care into this context adversely impacted behavioral health services for many low-income populations (Waitzkin et al. 2008). Our research clarified how consumers, advocates, and providers inserted themselves into state and federal policy discussions concerning inappropriate care provision and statewide reductions in services (Willging et al. 2003). These stakeholders had been excluded from such discussions until media attention began focusing on serious system deficiencies under the reform. Such stakeholders, many of whom had not previously collaborated in any structured capacity, developed informal coalitions throughout the state and with national organizations, such as the Bazelon Center for Mental Health Law, to lobby successfully for federal intervention by the Centers for Medicare and Medicaid Services (CMS).
Stakeholders in this unprecedented community organizing effort presented firsthand account testimony to CMS and the state legislature illustrating how state officials had not exercised adequate oversight of the Medicaid program and had permitted managed care organizations to place mentally ill persons in inappropriate levels of care. They also argued that managed care had weakened the state's mental health safety net as a minimum of 60 facilities had closed during the reform.
Despite public outcry that had once galvanized diverse stakeholder groups to effective action, New Mexico lacked a cohesive network of interested parties, independent of the state, when the most recent reform was announced in October 2003. From the outset, state leaders sought to rectify this situation while responding to calls for local input issued by the President's New Freedom Commission on Mental Health (2003) and federal agencies. They laid the groundwork for LCs to emerge as venues for community participation in New Mexico's revamped system. Given the dismissal of community perspectives by state officials under the previous reform, public receptivity toward the LC concept was generally positive; consumers, in particular, were pleased that the government would be taking their concerns seriously.
To its credit, the state government made strides in establishing the LC network, creating internal staff positions to support the burgeoning initiative. State officials deployed new web-based strategies and used public PC meetings to disseminate information about the reform to LCs and other interested parties. By March 2006, 13 LCs were established, based pragmatically on New Mexico's existing judicial districts (each incorporating from one to four counties). Their demographics, provided in Table 1, demonstrate similarities and differences in terms of ethnic group distribution, population density, and poverty level. Two additional LCs, not defined geographically (thus, not listed in Table 1), representing the state's sovereign tribes and pueblos, were created—one for the Navajo Nation and a second for the 19 Pueblo, Jicarilla, and Mescalero Apache Nations. It is important to note that each LC is distinct because of the varied nature of New Mexico's urban, rural, and frontier communities; this has created challenges for the initiative.
The ethnographic study on which this article is based takes a critical multilevel and social systems approach (Baer et al. 2004) that strives to accomplish what Hugh Gusterson (1997) has called “polymorphous engagement.” In our attempt at studying up and studying down (Nader 1969), our research team has interacted with “informants across a number of dispersed sites, not just in local communities … collecting data eclectically from a disparate array of sources in many different ways” using techniques ranging from participant observation to formal interviews to critical readings of newspapers and official documents (Gusterson 1997:116).We focus our attention not only on the levels of individual, family, and community—those most likely to experience consequences of reform changes—but also take what Laura Nader terms a vertical slice of the behavioral health system, focusing upward on those with power. This includes providers responsible for reform implementation and state officials and SE administrators actively engaged in creating behavioral health policy.
This article is informed by baseline data collected during the first 18 months (November 2005 to May 2007) of a five-year multi-method assessment of the New Mexico reform. In addition to survey research and document reviews, this assessment includes an ethnographic study of how the reform affects provider organizations and consumers in three urban counties and three rural counties that were selected for study because of their unique social and demographic characteristics. Our team of eight ethnographers conducted participant observation in 14 behavioral health agencies and semi-structured interviews with various stakeholders: 189 with administrators, providers, and staff; 325 with adults with a serious mental illness or substance-use disorder; and 217 with family and friends of consumers.
We implemented reputational case selection to recruit participants in each group (Schensul et al. 1999). Within agencies, we first interviewed a lead administrator who referred us to providers and staff. Similarly, consumers were identified with help from agency personnel. They, in turn, recommended family, friends, and others instrumental in their recovery process. Recommendations from community experts (county-based health council members, consumers, and advocates) helped identify key leaders, policymakers, and SE administrators involved in the new reform or the LC initiative.
Four complementary interview guides were created for each stakeholder group. Pertinent to this article, we solicited knowledge and perceptions of the LC initiative from each participant. We first asked each participant if he or she were aware of the LCs, and we posed additional questions to those who answered affirmatively. These questions concerned the extent of the participant's involvement in the LCs, the role and composition of the LCs in the broader behavioral health system, and the relative effectiveness of the LCs in representing diverse stakeholder interests.
Our ethnographic team undertook participant-observation of monthly meetings of LCs located in the study areas. We also observed monthly PC meetings and other public forums pertaining to the reform. For each event, we documented the setting, the individuals involved and the nature of their interactions, and all discussion pertaining to the LCs. We also recorded direct quotes whenever possible and collected and inventoried written materials that were distributed.
All data collection protocols and informed consent procedures were approved by the Pacific Institute for Research and Evaluation Internal Review Board. Verbal consent to conduct observations was obtained at LC meetings (but not at public forums, such as PC meetings). Participants were compensated $30 per interview (except for respondents in the policymaker interviews).
Observation and interview notes on LCs were handwritten, transcribed, and uploaded into an electronic database. Interviews were digitally recorded, transcribed, and coded, using an iterative process of reviewing the textual data, which were analyzed by open coding to discover themes and issues, followed by focused coding to determine which of these were repeated or represented particular concerns (Emerson et al. 1996). We strengthened our qualitative inquiry through triangulation of interview and observation data (Patton 2002), which contributed to a coherent account of cross-cutting themes relevant to the LC initiative. This multilayered analytic process generated a number of themes relevant to this initiative, which fell into two broad categories: (1) structure and function; and (2) participation and collaboration.
The first category encompasses several concerns: predominance of provider versus consumer voice; insufficient resources and training; state administrative demands; and perceived lack of state response to LC efforts. The second category reveals how problems of information flow, transportation, and stigma (esp. concerning illness disclosure) impacted involvement in the LC initiative and how the construction of “community” introduced under this initiative exacerbated tensions across localities with diverse histories and populations. Case examples emblematic of commonly shared experiences and quotations that underscore the perspectives of study participants are presented below to illuminate these themes. Through this ethnographic lens, we first provide an overview of the LC recognition process, followed by snapshots of LC meetings. We then illustrate how bureaucratic considerations, power, and status differentials among LC members and disparities across geographic localities affected LC implementation. Finally, we consider the views of state officials and SE administrators, and we discuss implications for state-mandated community input processes.
LCs are local only in relation to the larger unit of New Mexico. State officials rationalized that there was “no simple way to define what is the appropriate local geographic area for planning or local input.” If state agencies or the SE were expected to work with and help support LCs, “common boundaries had to be identified and maintained” (New Mexico Interagency Behavioral Health Purchasing Collaborative 2004:5).
In this new system, consumers, families, and even providers who had not historically collaborated were encouraged by state-issued documents “to come together to decide how to put forth one application” to obtain LC status, which would then be reviewed by the PC. Approval would confer formal recognition to these individuals as officially comprising the LC for their specific region, thus rendering local contexts “legible” for state planning and administration.
The applications required much accountability. Aspiring members had to demonstrate that their proposed LC:
Even though state officials provided assistance, community members expressed amazement over the magnitude of work required for recognition. This included spending thousands of “voluntary” hours traveling great distances to attend meetings within regions or in Santa Fe (where the PC convened), filling out paperwork, and undertaking preliminary capacity building and needs assessment.
The following example demonstrates the elaborate efforts that some groups felt were necessary to become recognized. When Native American members of one aspiring LC met with the PC in Santa Fe for the recognition process, they gave an elaborate PowerPoint presentation punctuated by statements from consumers, providers, politicians, and spiritual leaders. Presenters stood before the dais where the PC members sat to address the room.
After a brief introduction, a provider–member of the petitioning LC highlighted the need for extensive community outreach because the region served was rural and poor. She noted, “Tribal members come from a long history of historical oppression and learned helplessness. We have long-standing issues with trust. People here don't like it when outsiders say, ‘We'll tell you what is best for you, and what is good for you.’” PC members listened intently, nodding encouragement. The woman added, “That is why this is a historical event for many of our people. Never before have they been asked what they want.” The presenters proceeded to describe resources and potential LC partners, their proposed organizational structure, and culturally appropriate ways of discussing behavioral health issues. To close the presentation, a Medicine Man offered a blessing and then stated, “We appreciate the respect you have shown us, and we acknowledge the government-to-government relationship between [our community] and the State of New Mexico. … [We] are crossing historical prejudicial boundaries.” State officials praised the presentation and proclaimed those who stood before them to be an LC.
During PC meetings, state officials characterized LCs as “revolutionary” and fundamental to the success of the reform. Promotional literature declared that “the Local Collaboratives will be the voice of local communities.” Nevertheless, among our study participants, many consumers (79%), their family and friends (99%), and providers (63%) had not yet heard of LCs. Among those aware of the LC initiative, some expressed optimism over “giving voice to the mentally ill,” whereas others who reportedly participated in earlier attempts at behavioral health reform in the state were more cautious, with one advocate going as far as to lament participation in another “mandated governor-appointed body without real power.”
Each LC consisted of volunteers with multiple, often contradictory, opinions, levels of knowledge, and competing interests related to behavioral health care. By the state's design, consumers and family members were to make up at least 51 percent of each LC's membership, while the remainder could include providers, advocates, and other interested parties. Based on our observations of LC meetings and reports from LC members, attendance fluctuated, ranging from eight to 50 participants. Most meetings had a variety of attendees, but in some cases, only providers were present. In some LCs, we documented strong, vocal consumer participation: in others, consumers tended to listen to providers and advocates, rarely engaging in group conversation. Meetings were generally held in accessible, neutral public venues such as libraries and community centers. However, in rural settings, where conference rooms were unavailable, meetings were sometimes held at hospitals and criminal justice facilities, spaces that inadvertently favored participation of one stakeholder group over another.
Various features of the LC structure and process made it difficult to achieve effective collaboration. For example, the meeting format tended to be bureaucratic and, as reported by research participants, often intimidating for those inexperienced with such venues. Providers were most likely to possess knowledge of technical behavioral health terminology and abstract concepts regarding systems change, as well as experience that facilitated active involvement in structured meetings.
LC engagement favored persons with strong listening and oral abilities, proficiency in English, long-term reasoning skills, the ability to quickly put “an idea into words,” and to integrate the idea into group discussion in a clearly relevant way. Those able to forge alliances with other LC members were more effective in having their voices heard. This kind of engagement was difficult for persons struggling with behavioral health issues. Also, provider networks and advocacy groups already in existence tended to benefit from institutional support, making it easier for them to attend meetings regularly and impact agenda setting. Not surprisingly, at the onset of this initiative, LC leadership largely fell to providers and advocates, rather than consumers and family members.
Even when consumers did play a leading role, they found administrative demands quite onerous. One consumer, who had served as co-chair of an LC, told our ethnographer that the workload was “tremendous. … If I couldn't keep up with what the state wanted, how can other consumers?” When her friend assumed this same co-chair position and became overwhelmed, this consumer expressed frustration that she could not offer assistance because the stress associated with LC leadership “was still too much to deal with.”
Problems also arose within the LC initiative that reflected urban–rural differences and varying levels of expertise. One urban site, Las Cruces, is a large, university town with a sizable population of people with serious mental illness. This is where Local Collaborative 3 (LC3) typically meets. Although underserved according to national standards, Las Cruces offers greater access to care than other areas of the state and has an established network of providers and an active advocacy group. With the creation of LC3, these existing groups now were expected to work with consumers and families, who had not previously participated in public policy discussions.
At one LC3 meeting we observed, many consumers were present, but the agenda was largely controlled by providers and advocates and focused on items reflecting their interests, such as creating a new facility for the mentally ill. Some consumers, however, had more basic needs. One individual with serious mental illness and physical limitations, who received care from a local agency, asked, “Why can't I get transportation? And what about someone to help me clean my house?” Two other women representing a more rural, Hispanic community within LC3 expressed concern that “everything will just go as far as Las Cruces” and asserted that they needed “resources and help for people in our community.”
Rural residents elsewhere in the state shared similar concerns—making it clear that geographical distance, isolation, and poverty not only impact service utilization but comprise barriers to knowledge of and participation in LC events. Such individuals had other pressing priorities occupying their attention, as we documented at a forum involving state officials and LCs, where one rural resident—the sole voice of her LC—explained: “People in my community don't have phones [or] the Internet. I don't even have running water. They are not diagnosed, but that doesn't mean they don't have mental health needs.” She then added, “Oh well. That's okay. You can just ignore us. We're used to it.”
During the first 18 months of the LC initiative, the goal of “giving voice to the mentally ill” drifted slowly toward realization, as illustrated by a PC meeting attended by members from various LCs. Many presented their concerns to high-ranking state officials, and an attending reporter published some of the comments. One woman declared, “People with mental illness don't have a way to get to their doctor.” Another stated that most services are “in Albuquerque or along the Rio Grande corridor;” in her rural community, she “feels ignored.” Dozens came forward to ask state officials directly for help and support.
The presiding official admitted that the state lacked many resources for people with behavioral health issues but responded to the pleas by throwing a “challenge back in the consumers' laps.” The official stated, “It's only by speaking out and getting local officials' support that you'll see action. We can make all the proposals we want, but we're just government bureaucrats. Your voices are ten times as powerful as ours.” Later, when the reporter spoke with one advocate, she was told, “People feel helpless because they don't know what is available or where to go for help” (Jadrnak 2006: B1). At one level, such frustration reflected tension emanating from a government system in which consumer voice was strongly encouraged yet did not carry the full weight of authority needed to compel action, as suggested by the above-mentioned official. At another level, this deflection of frank consumer requests for services and assistance demonstrated how state officials could inadvertently undermine active consumer involvement in policy discussions on pressing issues even through well-intentioned comments.
State requirements that LCs establish governing structures, identify and assess local needs, and create legislative priorities in rapid succession meant that hectic agendas and formal parliamentary procedures at LC meetings often trumped open, friendly discussion that could otherwise advance group cohesiveness. One advocate complained that the state had delegated LCs numerous tasks with quick turnaround times but had given “no roadmap for how they want [the tasks] to be accomplished.” Citing a lack of “staff support,” the advocate added that “the government has not given the collaboratives their own resources to come into power.” Despite state attempts to allocate modest financial and personnel support to each LC, such concerns were commonly articulated in LC meetings and in our interviews with LC participants. Similar concerns also surfaced independently in two evaluations conducted for the state (Kinney 2007; MacArthur Foundation Network on Mental Health Policy Research 2007).
A prime example illustrates these failures and their implications. In 2006, state officials encouraged LCs to develop two top funding priorities that reflected behavioral health care needs of their community. An LC member would then participate in a state-level committee that would rank legislative priorities for all of New Mexico and select those that the PC would present before the state legislature for the 2007 fiscal year.
In response, each LC identified and agreed on priorities. After forwarding the results to state officials within the specified timeframe, a master document of overlapping priorities was then produced, but not in time for presentation during the legislative session. An advocate remarked, “They made us go through this sham process, as if they cared what we think.” When state officials distributed the priorities one month later, many were disappointed with the product. The advocate noted, “[They released a] bullshit document that is wrong.” Such attitudes on the part of LC members were reinforced when this process was replicated in 2007. Once again, state officials failed to distribute the priorities to the legislature in time to influence key behavioral health–funding directives for the 2008 fiscal year.
A closer look at the specific case of Local Collaborative 1 (LC1) illustrates the complexities of arranging diverse stakeholder groups in solidarity and having all voices represented. It also underscores the artificial nature of “community” because LC1 represents three considerably different counties in Northern New Mexico that are unique in terms of history, geography, and culture: Santa Fe, Los Alamos, and Rio Arriba.5 These structural differences led to inequalities in power magnified by local imbalances in ethnic residency, education and income levels, and access to behavioral health resources, which contributed to problems with communication, collaboration, decision making, and adequate consumer and family representation.
Santa Fe County includes the oldest inhabited capital city in North America, best known for its adobe architecture, indigenous artists, and spectacular vistas. Its population is equally divided among Hispanic and White residents with only 3 percent Native American. Residency is largely determined by those who can afford homes in a high-priced market. Because of its affluence, Santa Fe has many behavioral health services compared to other regions of New Mexico, and, thus, Santa Fe providers are highly invested in the reform.
Los Alamos County has been called an “epicenter of U.S. nuclear weapons research” (Zeman and Amundson 2004). Originally founded by Hispanic settlers in the 1880s, it was taken over by the Manhattan District of the U.S. Atomic Energy Commission in 1943, displacing Hispanics and becoming a model of 1950s suburban culture. Its population is 79 percent White and 13 percent Hispanic, with the highest per capita income in the state, allowing many residents to access behavioral health care in the private sector.
These two counties differ greatly from the more impoverished and rural Rio Arriba County, where 78 percent of residents were born in the region. Ethnically, it is 72 percent Hispanic, 16 percent Native American, and 11 percent White. In terms of education, only 15 percent of those in Rio Arriba have a bachelor's degree, as compared to 40 percent of Santa Fe and 61 percent of Los Alamos residents.
Rio Arriba County is one of the first Spanish settlement sites in the United States. Thirty miles north of Santa Fe, it is located in a naturally stunning area, yet beauty is rarely mentioned as attention is usually focused on the region's intense poverty, inordinately high drunk-driving arrests, and drug-related crises. These historically and culturally situated phenomena, when combined with excessive disability and low median income levels, reinforce socioeconomic patterns that contribute to despair for community members. Behavioral health services are nearly nonexistent despite substantial need. This situation has fostered distrust of government among many residents, as demonstrated by one Rio Arriba consumer's comment, “I think the governor is just turning his back and waiting for us to kill each other.”
Under the LC initiative, members of these disparate counties were expected to work together to identify and help meet the needs of their newly created community, and to establish social and advocacy networks for consumers and families. To further this goal, they met as separate “branches.” After more than one year of meetings, the chair of the Rio Arriba branch called the January 2007 tri-county meeting to order, commenting, “Right now, if I was to classify us, I would call us a dis-organization. … It has been incredibly difficult to pull this region together.” Members then conversed about who was responsible for taking notes at meetings, compiling minutes, disseminating information, and making decisions for the LC. The frustration felt by many was evident in the comments of one provider: “We are not getting anywhere. We come to these meetings. …We talk about the same things over and over again, and nothing gets decided. …We come up with things we want to happen, yet no one takes minutes, and we have to have the same discussions over and over.”
Despite much dialogue, the meeting adjourned with no decisions reached. Such frustrations intensified tendencies for some members to view others with suspicion, believing them less committed to the LC process. One Rio Arriba member blamed Santa Fe people, who “don't always come. And then we have to go over everything again. We are spinning our wheels.” Similar statements were made by a Santa Fe member: “Some [Rio Arriba] leadership doesn't want this process to succeed.” Such comments led one LC chair to note that “dysfunctional patterns” emerged from the inability of the counties “to get over their differences.”
This finger pointing underscores both the artificial nature of this diverse “community” and the fact that all counties within the state have historically been forced to vie for scarce resources in the drastically underfunded behavioral health system. This competition, coupled with the shared perceptions of LC members that their cultural differences are simply too great, has thwarted collaboration among those who attend these meetings.
Another pressing issue for LC1 was lack of consumer representation at meetings. Of the 40 people who attended the above-mentioned meeting, we noted only one family member from Santa Fe, one consumer who worked for the SE, and another who worked for state government; the remaining attendees identified as providers. Similarly, at a meeting of the LC1 Santa Fe branch, there were 23 providers but only one consumer and two family members present.
Lack of consumer participation was owing to several factors. One Los Alamos provider explained that potential members from her area would not identify openly as behavioral health consumers because of their work at the national labs. A Santa Fe representative, in turn, said consumers from her county did not want to be part of a process so dominated by providers. Even more serious was the fact that many Rio Arriba residents were not even aware of LC1 and, thus, did not truly have a choice regarding participation. When asked specifically about the LCs, 65 of the 72 consumer and family member respondents answered with blank looks, confused expressions, or statements like, “I have no idea what you're talking about.” LCs, in turn, lacked support structures (e.g., staff, volunteers, and technical assistance) for facilitating community outreach to recruit consumers and families, as well as capacity-building activities that might empower new members, particularly those grappling with serious behavioral health concerns, to participate regularly in LC processes.
State officials recognized the need to recruit consumers and families into LCs but were slow in accomplishing this. Obstacles to community participation were more complex than reform supporters had anticipated. Having no prior experience implementing large-scale community organizing mechanisms, state officials underestimated the financial resources, incentives, and range of creative strategies needed to draw in significant numbers of consumers and family members. Representation in LCs was even lower in areas where isolation, poverty, and lack of behavioral health resources shaped peoples' daily experiences.
Failure by reformers to address these two levels of marginalization—expert versus non-expert and urban versus rural—perpetuated a discourse of equality that masked, albeit superficially, alterity based on class, culture, and behavioral health disparity, and ensured continued lack of representation in traditionally underserved communities. Such exclusion of consumer, family, and rural voices reified existent social divides in these counties, incapacitating this LC and others within the state.
In an attempt to promote the appearance of solidarity and consensual self-governance in the LC initiative, state officials glossed over important participant differences, thus hindering their own attempts at democratizing the policy decision-making process. This may be a common occurrence: “Many participatory projects rest on the dubious assumption that simply identifying different ‘stakeholders’ and getting them around the table will result in a consensus being reached that is ‘fair’ to all” (Hildyard et al. 2001:69).
Such misconceptions were publicly articulated since the initiative's inception: state-issued brochures claimed “an equal voice” for consumers and families in LC decisions. There was a fundamental disjuncture between this idealized view of community participation, promulgated by state officials, and the lived realities of people marginalized by extant social structures. Although such rhetoric captured public interest by promising a dialogue with state officials that would impact them positively, even when this occurred, the promises benefited some and not others.
In critically assessing such processes, Gianpaolo Baiocchi (2001:50) notes that “participatory democracy … reproduces class hierarchies [and] hierarchies of political competence of ‘experts’ against non-experts within the field of politics.” We suggest that participation in the LC process likewise masked reproduction of inequalities that already existed in New Mexican society by privileging the perspectives of provider–experts over consumers, and the voices of urban upper- and middle-class consumers and family members over those more removed from the discussion because of distance, poverty, lack of resources, and general mistrust of government.
Some state officials acknowledged shortcomings in the LC initiative, noting that engaging those left out would require “a systematic and consistent approach, with appropriate resources … on issues meaningful to consumers and families” (Kinney 2007:21). Nevertheless, the effects of culture and economic stratification in the lives of consumers remained unaddressed. The state's own evaluation of the LC process treated consumers and families as a homogenous entity, free of race, ethnicity, class, and geography. Rather than offering new outreach strategies, proposed solutions focused on administration, including added state controls and enhanced “planning, measurement and management” (Kinney 2007:21).
Other state officials interviewed for this study responded to criticisms by referring to the complexity of this “revolutionary” process, shuffling blame onto fledgling LCs, and impugning local politics. Most argued that this process “allowed people to be at the table who have never been at the table before,” echoing the sentiment that participation is an “intrinsically good thing” (Cleaver 2001:36). Some were more critical. One official conceded state control, observing that the state was “trying to balance how much power to give the LCs” and that he was “afraid the collaboratives might shape themselves in a way that we don't want.” He went a step further: “It is meaningful to give them this power, but it is a bone.”
Equally telling was a comment made by an SE administrator: “I give the state a lot of credit for supporting the Local Collaborative system. … It is a struggle for folks because it is all volunteer. Local Collaboratives will succeed if they see themselves as advisors, not policymakers.” In other words, despite the reform's rhetoric concerning empowerment, the reality was merely a performance of power by “advisors,” rather than creation of real consumer power.
As Uma Kothari points out, “The very act of inclusion, of being drawn in as a participant, can symbolize an exercise of power and control over an individual” (2001:142). One could argue that bringing community stakeholders into the LC process incorporated them into a “framework in which the micro is set against the macro … the powerless against the powerful” (Kothari 2001:140).On the one hand, this process valorized consumers willing to take part in this time- and labor-intensive system, much like the “deserving poor” of the 1980s and 1990s who “embrace[d] the spirit of entrepreneurship, voluntarism, consumerism, and self-help” (Goode and Maskovsky 2001:8). On the other hand, the comments of those who did not participate could be more easily dismissed by policymakers and SE administrators. Even those who did actively participate in the LCs felt that they were not guaranteed a voice, a view reinforced by their experience of having legislative priorities that they set as part of a collective process ostensibly ignored.
Toward the close of our 18-month fieldwork period, it appeared that this situation was changing. State officials began questioning the workability of the LC model and sought input from members of the research community to identify problem areas and develop strategies to improve the model. Three research teams, including ours, met twice to discuss key issues, such as barriers to implementation. Both times state officials took seriously the concerns raised by the teams. One official expressed her desire to learn from earlier errors, noting that “Most mistakes are made because we have never done this before and our actions often have unintended consequences [for LC members].” She and her colleagues offered to work more closely with the research teams to identify better consumer engagement models, concluding that state officials and LC members were entering into a co-learning process to “build a strong cadre of engaged consumers” to emerge as an “active voice” in the behavioral health system. Clearly, changes in the LC model would need to focus on addressing barriers to greater consumer involvement.
Regardless of social location, those involved in the LC process—state officials as well as community members—generally concur that the current model requires new forms of participation that address the needs of underserved populations and increase their role in the decision-making process. This demands a critical focus on social and demographic disparities and how issues of class and culture affect participation in this and similar community-input processes, regardless of whether they fall under the rubric of CBPR or state-mandated reform.6 We have called attention to the power-laden context in which the LC model has evolved. Consideration of this context can help policymakers, behavioral health service administrators, researchers, providers, and concerned citizens alike to better understand the historically based systems of inequality that shape implementation of such initiatives, and the unintended duplication of such systems via adoption of well-intentioned community input mechanisms.
In many respects, the emphasis on creating a “consumer and family-driven” system is emblematic of neoliberal reforms that facilitate the devolvement of responsibility for service provision to the personal domain of individual and family, away from the state and its corporate partners (Clarke 2004; Goode and Maskovsky 2001). The justification for privatizing publicly financed behavioral health care lies in the assumption, largely unproven (Geyman 2004), that private enterprises operate more efficiently than government entities. And yet, cost effectiveness is only one part of a health care equation that must also include quality, access, and compassion of care that meets public needs. For these reasons, we recognize that this model of consumer engagement is problematic.
However, if better adapted to real-life situations, this model could begin to disperse the totalizing power of state institutions, making engagement, input into policy, and evaluation of services a reality for historically disenfranchised parties (e.g., consumers and families), thus providing new opportunities to genuinely affect the course of system change. At the same time, this model can provide a way in which community stakeholders might develop new “competencies for social integration” that build their confidence (Ware et al. 2007:472). If those in need of behavioral health services have become, as some scholars suggest, “second-class citizens” because of restrictions of “rights and privileges” enjoyed by others in a democratic society based on “grounds of mental incompetence,” then full participation in this initiative is significant (Ware et al. 2007:471).
Overall functioning of the LC initiative could be improved by state coverage of basic operating expenses and by offering training and greater technical assistance to members. Equally critical is the need to address power imbalances. We offer the following recommendations regarding the LC model: (1) Providing training to LCs (by state government or the SE) in communication strategies for engaging consumers and families, thus promoting what Kim Hopper (2007:875) calls the “social recovery” of individuals marginalized as a result of behavioral health issues; (2) aggressive outreach efforts (e.g., presentations and brochures) in both clinics and public venues, including churches, community events, and libraries, to help connect with those who have yet to hear about LCs; (3) appropriate (non-stereotyping) cultural competence training that emphasizes openness and self-reflexivity for state and SE personnel, in tandem with leadership training for all LC members; (4) allocation of state resources to address structural barriers, such as distance and transportation, so consumers can attend LC meetings; (5) altering meeting structure to allow flexibility, giving all participants equal chance to express opinions and concerns; and (6) finally, state government needs to realistically assess the demands of volunteerism, on which the LC initiative has relied since its inception. Given the workload, paying LC members whose involvement is not supported through employment, as is often the case for consumers and families, would encourage participation and reaffirm the value of their labors.
In addition to identifying barriers to real community participation, our study highlights the key point that voice is not a monolithic construct. If the real-world messiness of community participation is not addressed, such processes are likely to result in more image than substance. In the LC case, the reform's emphasis on consensual decision making, implying a single unified voice, has often glossed over difference and dissent and has not always inspired trust among all partners (Cooke and Kothari 2001), thus reinforcing power inequalities.
Although this ethnographic work provides an in-depth description of the “on the ground” dynamics that affect how community voices are incorporated into state policy concerning behavioral health care for underserved populations, we recognize certain limitations. Our qualitative investigation of specific LCs is limited to six counties where we are conducting a broader investigation of adult consumer service delivery and utilization issues under the reform; we did not undertake a comprehensive assessment of LCs outside these settings. However, findings reported in state-supported evaluations (Kinney 2007; MacArthur Foundation Network on Mental Health Policy Research 2007) suggest that the experiences documented here were common throughout New Mexico. State officials currently characterize the New Mexico reform as a ten-year process; in this article, we report only on the first 18 months of LC implementation. Additional research, which we are undertaking, is examining the ongoing implementation of the initiative and the extent to which community input ultimately contributes to changes in local and statewide service-delivery systems.
As problems of cost, quality, and disparities in care and access continue to plague the U.S. health care system, states are stepping in to fill the growing void created by the devolution of federal responsibility. In instituting various kinds of reforms, states need to provide structures for public input and participation, so as to ensure that these reforms truly address public health and access issues and not just reify existing economic and power imbalances. Such structures must also include room for a sustained analysis of how deployment of taken-for-granted concepts, such as participation and community, can promote social exclusion, rather than inclusion.
It is conceivable that the New Mexico experiment may be a model that other states follow. As this study demonstrates, such sweeping attempts at participation-based reform can only succeed in an open process of mutual learning and discovery that ensures all voices an opportunity to be heard regardless of ethnic, class, cultural, and geographic differences.
This study was funded by the National Institute of Mental Health (R01 MH076084) and the Substance Abuse and Mental Health Services Administration. The methods, observations, and interpretations put forth in this study do not necessarily represent those of the funding agencies. We thank our ethnographic team—Caroline Bonham, Elizabeth EnglandKennedy, Louise Lamphere, Jill Reichman, Gwen Saul, Paula Seanez, and Margaret Watson—for their contributions to this study. We also wish to express our appreciation to our research participants for their generous assistance and cooperation.
1Through a competitive bidding process, ValueOptions, Inc., a subsidiary of FHC Health Systems, was selected as the Statewide Entity. It is the largest privately owned managed behavioral health care company in the United States, serving over 23 million people through publicly funded, federal, and commercial contracts.
2As employed in this article, the term consumer represents an emic category used by state officials, behavioral health system administrators, providers, and laypersons who participated in our study, to identify individuals who utilize behavioral health services. Historically, this term evolved with the “consumer movement” in the 1960s, focused on increasing public power and autonomy in various social arenas, including education, product safety, law, and social welfare. In the health care context, the term emerged in reaction against physician hegemony and toward more patient input and autonomy (Reeder 1972). More recently, it has been used to frame health care services in economic and marketing discourses (Pellegrino 1999).
3For more information, see http://www.bhd.state.nm.us/.
4See the Substance Abuse and Mental Health Administration website, at http://mentalhealth.samhsa.gov/consumersurvivor/default.asp.
5Although race and ethnicity are touched on in this section, we consider these issues more comprehensively during the next phase of research.
6The problem of garnering participation is not unique to the LC initiative. Admittedly, the community advisory board for our study—a group of local leaders, consumers and families, behavioral health professionals, and policymakers—confronts many of the same issues discussed in this article.