Interagency teams: A vehicle for service delivery reform

A R T I C L E INTERAGENCY TEAMS: A VEHICLE FOR SERVICE DELIVERY REFORM Nicole E. Allen University of Illinois Urbana Champaign Pennie G. Foster-Fishman and Deborah A. Salem Michigan State University 䡲 Human service delivery organizations often have great difficulty implementing new service delivery technologies. This study examines the extent to which interagency service delivery teams facilitate the implementation of a popular reform that significantly challenges the status quo: family-centered service delivery. Survey data from 121 providers representing 25 agencies within one county suggest that interagency teams may promote provider implementation of new service delivery practices. Teams members were more likely than nonteam members to implement practices consistent with family-centered service delivery. Consumer focus group data support these findings, with consumers noting that services received from providers in the team context were more individualized, empowering, and comprehensive than the services they typically received in the county. The implications of these findings for practitioners and scholars interested in facilitating human service delivery reform are discussed. © 2002 Wiley Periodicals, Inc. Human service delivery organizations ~e.g., child welfare organizations, mental health agencies! often have difficulty implementing new service delivery practices, specifically those that require radical change in providers’ approach to service delivery ~Williams, 1995!. Concerns about the feasibility of changing human service delivery The order of the first two authors was alphabetically determined. This research was supported by grants from the State of Michigan Family Independence Agency, Michigan State University Outreach Office, the Institute for Public Policy and Social Research at Michigan State University, the Miller Foundation, and the Binda Foundation. The authors wish to thank the following groups and individuals for their contributions to this research: The County’s Multi-Purposive Collaborative Body and its evaluation subcommittee, Applied Developmental Science at Michigan State University, Mark Schauer, Stacy Curtis, Carrie Preston, David Loveland, Juliette Mackin, and Deborah Bybee. Correspondence to: Nicole E. Allen, Department of Psychology, University of Illinois Urbana Champaign, 603 East Daniel Street, Champaign, IL 61820. JOURNAL OF COMMUNITY PSYCHOLOGY, Vol. 30, No. 5, 475–497 (2002) © 2002 Wiley Periodicals, Inc. Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jcop.10020 476 • Journal of Community Psychology, September 2002 practices are particularly salient today, given the growing recognition that traditional service delivery approaches have failed to address the needs of many high risk children and families ~Knitzer, 1982; Weissbourd, 1990!. Numerous national, state, and local initiatives have called for significant reform in how services are provided to children and families, including altering the financing, structure, and guiding ideological frameworks ~e.g, Foster-Fishman, Salem, Allen, & Fahrbach, 1999; Friedman, 1994; Golembiewski, 1985; Stroul & Friedman, 1986!. One popular reform is the current emphasis on providing family-centered service delivery ~e.g., Nelson & Allen, 1995!. Family-centered service delivery is an individualized empowering approach that emphasizes consumers’ strengths and encourages their active involvement in the design, implementation, and evaluation of services ~Schorr, 1988; VanDenBerg & Grealish, 1996; Yoe, Santarcangelo, Atkins, & Burchard, 1996!. This orientation differs in important ways from the traditional service delivery model which is typically driven by a medical model approach that is often more deficits oriented, and professionally and programmatically driven ~Rappaport, 1977; Tyler, Pargament, & Gatz, 1983!. In contrast to this traditional approach, family-centered service delivery involves the implementation of four distinct service delivery practices. First, providers are expected to focus on consumer strengths and capacities rather than consumer deficits ~e.g., poor parenting skills; Dunst, Johanson, Trivette, & Hamby, 1991!. Second, there is an expectation that providers will extend their focus beyond the traditional assessment of treatment needs ~e.g., counseling! and also focus on families’ broad-based needs ~i.e., a broad range of needs! including, for example, basic living needs ~e.g., food, clothing! and social support needs ~e.g., making new friends, attending a support group; Dunst et al., 1991!. Third, providers are expected to focus on the needs and strengths of the entire family, rather than only a target client ~Dunst et al., 1991; Garbarino, 1988; Weissbourd, 1990!. Finally, given that families are viewed as experts on themselves, providers are required to include them in goal identification ~Dunst et al., 1991; Rounds, 1991!. Given these expectations, the family-centered service delivery approach requires providers to deliver services in new and sometimes unfamiliar ways ~Dunst, 1985; Williams, 1995!. The emphasis on family-centered service delivery can be found in reforms targeting a variety of service delivery domains, including child welfare ~Yoe et al., 1996!, developmental disabilities ~Dunst & Trivette, 1987!, mental health ~Morrissey, Johnsen, & Calloway, 1997! and special education ~Duchnowski, 1994!. Across these domains, the positive implications for consumers served from a family-centered service delivery model have begun to be established ~Bradley, 1983; Marcenko & Smith, 1992; Scannapieco, 1994; Trivette, Dunst, & Hamby, 1996; Weiss & Jacobs, 1988; Weissbourd & Kagan, 1989!. For example, Trivette et al. ~1996! found that consumers who were receiving services from a family-centered organization compared to a traditional professional-centered organization were more likely to feel empowered by the service delivery process, indicating greater control over accessing needed resources. In the field of child abuse and neglect, Scannapieco ~1994! found that home-based familycentered services had a positive impact on family functioning, and reduced out-ofhome placement of children with both low- and high-risk families. Given the promise of family-centered service delivery to result in positive outcomes for consumers, it has been adopted by policy makers, state-level administrators, and organizational leaders in many communities and service delivery contexts ~Flint, 1993!. Although leaders have adopted this model and made it a requirement of the Interagency Teams and Reform • 477 service delivery process, it is still unclear whether or not the adoption of such reform by human service system leaders actually shifts how services are delivered to consumers by providers. In fact, there is often a significant gap between the adoption of change by leaders and the actual implementation of change by employees ~i.e., front line service providers engaging in new behaviors; e.g., Cameron & Vanderwoerd, 1997; Foster-Fishman & Keys, 1997; Mahoney & O’Sullivan, 1990; McBride, Brotherson, Joanning, Whiddon, & Demmitt, 1993; Nelson & Allen, 1995!. Williams ~1995! observes that, “articulating an interagency mission and objectives @regarding service delivery reform# is difficult, but implementing them is fraught with obstacles and challenges” ~p. 416!. Organizational scientists argue that while the adoption of new technologies by key leaders is a necessary step towards successful change, it is not a sufficient means for guaranteeing its successful implementation ~e.g., Klein & Sorra, 1996!. Many innovations fail because employees do not, or cannot, implement the required changes ~e.g., Klein & Sorra, 1996; Mahoney & O’Sullivan, 1990!. Further, the implementation of family-centered service delivery practices requires secondorder change ~Watzlawick, Weakland, & Fisch,1974!. That is, service delivery providers are required to make fundamental shifts in their behavior that are inconsistent with the governing norms of professionals as experts and clients as the passive recipients of their expertise. To facilitate the diffusion of family-centered service practices in human service delivery systems, we must identify interventions that foster employee implementation of this reform. One popular mechanism for facilitating the implementation of family-centered service delivery has been the development of interagency teams. Interagency teams consist of service providers from diverse agencies ~e.g., mental health, substance abuse, judicial, and educational organizations! within the community who meet regularly to plan for and provide services to targeted consumers ~e.g., Adams & Krauth, 1995; Lewis, 1995!. These teams have become a popular vehicle for reform implementation across a variety of service delivery domains. Although they have been used extensively to help implement family-centered service delivery, we know little about their effectiveness in facilitating the implementation of this service delivery model. Thus, the purpose of this study is to examine whether or not involvement in interagency teams facilitates service provider implementation of the family-centered service delivery model. INTERAGENCY TEAMS Interagency teams may be an effective intervention for promoting employee implementation of new service delivery practices for several important reasons. First, interagency teams serve as new social settings ~Kelly, Ryan, Altman, & Stelzer, 2000! within their communities, and as such, may be well positioned to help employees “unfreeze” old attitudes and behaviors and adopt the significant changes these reforms require ~Bartunek & Moch, 1987; Lewin, 1951; Schein, 1985!. Unfreezing requires members to interpret a system’s capacity to change and perceive the proposed change as desirable and necessary ~Armenakis, Harris, & Mossholder, 1993!. Because individual attitudes and behavior are influenced by the demands, character, values, and norms of the contexts within which people live and work ~Bronfenbrenner, 1979; Kelly et al., 2000!, this unfreezing is often facilitated by the creation of new social settings ~Kelly et al., 2000!. New social settings may be particularly important considering that existing organizational settings contain values and practices incongruent with the desired 478 • Journal of Community Psychology, September 2002 change ~Cameron & Vanderwoerd, 1997; Nelson & Allen, 1995!. Given that traditional human service delivery organizations include numerous policies and practices incompatible with a family-centered approach ~i.e., they typically focus on and treat consumer deficits, ignore consumer strengths, and have services that are providerdirected!, the creation of new social settings that support the attitudes and behaviors required by this reform seems critical. Second, interagency teams may provide a more tightly coupled, structured context for introducing change than the traditional service delivery system. Human service delivery organizations are often considered loosely coupled ~Weick, 1976! because their technology is indeterminate and ambiguous and the administration has a limited ability to observe and control human service providers’ interactions with consumers. This loose-coupling makes it difficult for human service delivery organizations to effectively inform their members of a change endeavor and to monitor its implementation ~Weick, 1976!. The interagency team context, on the other hand, may be a more tightly coupled setting. For example, within an interagency team that has been created to implement a particular service approach, the desired technology ~i.e., family-centered service delivery! is typically well-defined and service delivery guidelines and expectations are often quite explicit. Team members have the opportunity to monitor and provide feedback on each other’s service delivery behavior because they meet to discuss and develop consumer service delivery plans and the progress of their implementation. While traditional human service delivery organizations could be more tightly coupled ~e.g., via case presentation and frequent staff meetings!, teams may be well situated to accomplish this more easily given consistent direct contact among providers. Third, interagency teams involve employees in a manner that may help to promote their commitment to the change initiative and their perception of the desirability of the change. Because service delivery providers have significant autonomy in deciding how services are actually provided to clients ~Kouzes & Mico, 1979; Lipsky, 1980!, change needs to be implemented in a manner that promotes employee support of new service delivery technologies ~Glisson, 1978!. Rogers ~1995! suggests that individuals’ perceptions of the relative advantage of an innovation influences the rate of adoption and that interpersonal channels are an effective way to communicate this advantage. Teams may be adept at promoting employee buy-in because they involve active employee participation and interaction with other service providers, enabling providers to witness and experience, first-hand, the feasibility and desirability of these new practices, an important step towards promoting employee support for change ~Armenakis et al., 1993!. Overall, within the current zeitgeist of service delivery reform, interagency teams may play an important role in helping to mitigate employee resistance to mandated change ~Argyis, 1970! by promoting a venue that fosters “unfreezing” of old attitudes and behaviors and validates providers’ new attitudes and behaviors, increases employee accountability, and enhances employee understanding of and commitment to the desired reforms. While many states are utilizing interagency teams, to date research addressing their effectiveness has been limited. There is some evidence that the development of interagency teams result in changes in the service delivery system. For example, Pandiani & Maynard ~1993! found that the existence of interagency teams improved interagency collaboration. In addition, interagency teams have also been linked to the adoption of positive attitudes towards service reform efforts ~Foster-Fishman et al., Interagency Teams and Reform • 479 1999!. While these studies highlight some of the ways teams may alter the service delivery system and the attitudes of service providers, there is very limited research addressing whether or not team involvement actually impacts the service delivery practices of participating providers. For example, one study found that assessments of family needs developed by teams provided a better match to the needs identified by mothers when compared to the assessments developed by individual providers ~Garshelis & McConnell, 1993!. While comparisons of providers who are team members and those who are not are rare, such comparisons are critical to understanding whether or not teams are a promising intervention for fostering provider implementation of new service delivery practices. This understanding is particularly important given that many communities are viewing interagency teams as an expensive service delivery experiment ~i.e., requiring many provider hours!, and are beginning to question if they are effective in promoting the implementation of new service delivery practices. CURRENT STUDY Within a county where all service providers were mandated to provide family-centered services, we examined whether or not providers involved in interagency teams deliver services in a manner that is more consistent with the family-centered service delivery approach than providers who are not members of these teams. To do this we compared providers on the extent to which they implemented the specific components of a family-centered service delivery approach in the development of their service delivery plans. Specifically, based on the literature describing family-centered service delivery, we would expect a provider involved in these teams to be more likely to implement practices including: ~a! identifying client strengths ~Dunst et al., 1991!; ~b! focusing on the broad-based needs of clients ~e.g., material needs, social support needs! ~Dunst et al., 1991!, rather than only traditional, deficit-oriented needs ~e.g., diagnosis for treatment, parenting skills training!; ~c! attending to the needs and strengths of the entire family, rather than only a single family member ~e.g., Garbarino, 1988; Weissbourd, 1990!; and ~d! involving clients and0or families in the creation of goals ~Dunst et al., 1991; Rounds, 1991!. In addition, we sought to validate our findings by conducting focus groups with consumers who were served by an interagency team. METHOD To insure that the questions asked and the methods employed accurately represented the experiences of service providers within this county, this study employed a joint insider–outside methodology ~Bartunek & Louis, 1996!. A committee of organizational leaders, service providers, and one consumer representative worked collaboratively with the research team on sample selection, measurement development, data collection, data analysis, and feedback. This committee was a subgroup of the county’s interagency coordinating council ~ICC!, a governing body formed approximately 10 years prior to the onset of this study to improve the quality of life of county residents by improving service delivery processes and outcomes. The ICC included representatives from 32 agencies including both human service and criminal justice systems and had recently become the state’s designated collaborative body. In that role, it became responsible for the implementation of service delivery reforms within the county and the distribution of state funds allocated to support the interagency teams. 480 • Journal of Community Psychology, September 2002 This study utilized quantitative survey methods to investigate whether or not providers who are involved in an interagency team are more likely to implement a family-centered service delivery model than those who are not. The survey data were collected from human service providers in the county. To validate our findings, focus group data was also collected from consumers who received services from one of the county’s interagency teams. Setting Creek County is a mid-sized community in a Midwestern state. Prior to data collection, this county became one of a few selected in the state to pilot a new approach to service delivery—family-centered services. While the specific components of this reform effort emerged from a state-sponsored initiative to improve services, they reflected the goals and previous attempts of the ICC to improve services within the county. The ICC engaged in numerous activities to stimulate this reform including developing a shared mission regarding the adoption of family-centered service delivery in the county, gaining support for this approach from the leaders of member and nonmember organizations, making training available to service providers throughout the county on family-centered service delivery, and creating interagency service delivery teams. The ICC was committed to implementing this reform and was working with our evaluation team to examine the degree to which their efforts were successful. This study was part of this larger evaluation and focuses on the role of interagency service delivery teams as a mechanism used to facilitate the implementation of a family-centered service delivery approach. At the time of the study, there were two interagency teams in the county: one focused on early intervention with families with children aged 0–3 who were at risk for child abuse and neglect or struggling with disabilities, and one focused on children with mental health issues and their families. Given the desire of this community to stimulate widespread reform in their service delivery system and their use of teams to achieve their reforms goals, this county was an excellent setting for this study. The interagency teams in this county were purposely created to include all agencies relevant to a given population. For example, for the Early Intervention Community Team, agencies who could provide potential resources to a family with a child 0–3 were invited and most chose to participate ~approximately 20 agencies in total!. These included agencies that provided: mental health ~e.g., individual or family counseling!, physical health ~e.g., medical clinics!, educational ~e.g., parenting classes, special education, Head Start!, child protective, supportive ~e.g., child care!, and specialized ~e.g., services for children with developmental disabilities, the local domestic violence shelter! services as well as material resources ~e.g., food, clothing!. When an organization agreed to participate they were required to ~a! send a representative to team meetings, and ~b! make organizational resources available to support families’ service plans. Leaders of participating agencies would typically assign service delivery providers to attend team meetings or ask for volunteers. The team coordinators indicated that the majority of providers were assigned by organizational leaders and required to attend team meetings. Occasionally, more than one provider would represent an organization or multiple providers would rotate attending team meetings. A larger community team would meet biweekly, with child and family teams forming for particular families and meeting as necessary. Interagency Teams and Reform • 481 Families were referred to teams by their individual providers. When providers thought that a family could benefit from receiving team-based services ~i.e., the range and complexity of their needs were not well met by the traditional service delivery system!, they would present the family with the option of working with a team. If the family agreed, the provider would present the family with a list of all community team members, provide the family with information regarding each participating organization and ask the family to decide which organizations they wanted represented at their meeting. Families would then come to a regularly scheduled community team meeting ~with representation from only the agencies selected by the family! to begin the process of building a service plan. The primary goals of this meeting were: ~1! to allow families to tell their story in their own voice and to discuss a plan of action with the family, not in their absence which is typical of traditional service delivery models, ~2! to develop a service delivery plan that would build on a family’s strengths and meet their diverse needs, and ~3! to form a smaller team of providers ~i.e., a child and family team! who would work with the family to further develop and meet their service delivery goals. After an initial plan was developed and a child and family team was formed, tasks would be delegated to different team members including family members. Typically, the service provider who brought the family to the community team would act as the coordinator of the child and family team, but often family members would be encouraged to lead their child and family team meetings. The team’s flexibility in meeting family needs was increased by access to noncategorical funds ~i.e., funds that can be used for assistance not typically provided in the traditional service delivery system!. This provided a pool of flexible funds to meet family needs that could not be met by existing community resources ~e.g., to repair a car motor or to fund home health care that would not be covered by a family’s insurance!. A subcommittee of the community team would review requests from child and family teams for noncategorical funds. Survey Data Collection Organizational Sample. Considering the purpose of this study, we purposively created a sample that included the organizations most central to the family-centered service delivery reform effort. Given the role of the ICC in directing these reforms, the initial sample included all 20 agencies that were members of the ICC who employed direct service providers and who were implementing these reforms. To ensure that our sample was representative of the core service providing agencies in the county targeted in these reforms, we included an additional 13 organizations identified by our evaluation subcommittee as important agencies in the County and as involved in the reform implementation. One organization deemed appropriate declined participation. Thus, while service delivery organizations were purposely rather than randomly sampled, the organizations targeted in this study represent a wide array of agencies in the county ~e.g., domestic violence shelters, Head Start, substance abuse programs, Community Mental Health, Public Health! with regard to service domains ~e.g., mental health, physical health, education!, organization types ~not-for-profit and profit organizations, government and community-based agencies!, and populations served ~e.g., families with small children, adults with mental illness!. Provider Sample. For a larger study, surveys were distributed to a sample of 530 providers across the 32 targeted organizations. Three hundred twenty-eight surveys were 482 • Journal of Community Psychology, September 2002 Table 1. Demographic Characteristics Demographic Characteristics Age 20 to 29 30 to 39 40 to 49 50 to 59 60 and over Gender Female Race0ethnicity White African American Latino Asian Other Highest educational level High school Technical school Some college College degree Some graduate Graduate degree % (N ⫽ 121! 17.4 25.6 40.5 14.9 1.7 81.0 82.5 14.2 1.7 .8 .8 6.6 .8 16.5 23.1 17.4 35.5 returned, yielding a response rate of 62%. Because of the nature of the larger study, three survey versions were randomly distributed to the targeted providers—two of these included a family-centered practices outcome measure. Of the 203 surveys returned including the family-centered practices outcome measure, data from 121 service providers ~60%! across 25 organizations was included in subsequent analyses. Surveys from providers were excluded when they did not provide adequate data or did not complete the outcome measure portion of the survey ~10%! or they did not provide direct services that involved developing service delivery plans 1 ~e.g., administrators, administrative assistants, day care providers, nurses providing medical care to the elderly, nurses aides who provided home visits to clean patients, and teachers; 30%!. The majority of providers were between 30 and 49 years old ~66.1%! and had completed at least a college degree ~76%!. Participants were predominantly female ~81%! and White ~82.5%!. Twenty-three percent of providers were interagency team members ~N ⫽ 28!. Teams primarily included service providers who were mandated to attend team meetings. See Table 1 for the demographic characteristics. Data Collection Procedures. A presentation describing the purpose of the study and the planned methodology was made in person to interagency council leaders and via 1 The 30% of participants who were not included in subsequent analyses either indicated that the measure was irrelevant to their work or this determination was made based on examination of the data they provided. For example, one provider completed the form based on a 15-minute telephone contact with a client. Another completed the form for an entire class of Head Start students. Finally, service providers serving elderly clients in a medical capacity ~mostly public health and visiting nurses services! were excluded from these analyses. Interagency Teams and Reform • 483 phone to noncouncil leaders. Interested leaders provided a list of all eligible directcare workers and managers. Given the large number of organizations involved in our sample it was critical to adapt data collection procedures to gain entry into such a diverse array of settings. Thus, dependent upon the leaders’ desires and the size and structure of the organizations, surveys were distributed either to the entire population of providers or to a random sample.2 A standardized protocol was presented to the targeted staff in either written or oral form. Surveys were either distributed to staff during a group presentation, left in staff mail boxes, or mailed. To encourage participation, staff were invited to enter a lottery for one of five gift certificates for a local mall. Staff either completed and returned the survey immediately after the presentation, returned it in a sealed drop box left in their organization, or returned it by mail.3 Extensive follow-up procedures were pursued until at least a 60% return rate across the county was achieved. Survey Instrument Demographic Information. Categorical data was collected on providers’ age, gender, ethnic background, education, organizational level, and years employed by their organization. Team Membership. Providers responded to two items assessing the extent of their involvement in each of two interagency teams. To indicate whether or not a provider was a member of a team a dichotomously scored variable was created ~0 ⫽ not at all involved, 1 ⫽ involved!. Indicators of Family-Centered Service Delivery To assess the implementation of family-centered service delivery, we asked providers to report information from their most recent service delivery plans. We selected service delivery plans because all direct service providers were required to develop these plans for all clients. In addition, service delivery plans reflect the types of services to be provided and how the client was involved in the planning process providing excellent indicators of the services actually put into place. We anticipated that the nature of the plans would be impacted by the degree to which a provider was implementing a family-centered service delivery model. A family-centered practice measure was developed to assess the degree to which providers’ self-reported service delivery plans reflected a family-centered service delivery approach. While we ideally would have liked to examine and code providers’ actual plans, given the confidentiality issues involved in accessing client treatment plans and the resource constraints of the study this was not possible. Instead, providers were asked to reflect on the most recent client for whom they completed a plan for services and detail information about the service plan that reflected key elements of family-centered service delivery ~e.g., assessment of strengths, assessment of needs vs. deficits, and involvement of clients and families in the creation of goals!. Provider reports were then coded to 2 In some cases random sampling was stratified by organizational role ~e.g., social workers, nurses!. While using a variety of data collection methods is not optimal, participants were always provided with standardized information via oral or written form. Response rates did not vary considerably across organizations. 3 484 • Journal of Community Psychology, September 2002 create a multidimensional assessment of the extent to which they reflected a familycentered service delivery approach. Coding and Interrater Reliability. A coding scheme developed specifically for this study was used to code providers’ service delivery plans. This coding framework was developed based on the primary tenets of family-centered service delivery as described in the literature on this approach and by the key informants with whom we collaborated. In addition, to assess whether this coding framework was a valid measure of indicators of family-centered service delivery, an expert in the implementation of this model independently reviewed and verified the coding framework ~face validity!. A single rater coded all of the data for this study. To determine the reliability of this coding, a second rater coded a subset of this data ~i.e., for a random selection of 25% of all cases interrater agreement ⫽ 93%!. Raters were not aware of which families were seen by teams and which were not. Each component of this measure and its operation are discussed below. Assessing Strengths. Providers were asked to provide a written description of the client’s strengths and were given the space to record up to six strengths.4 Strengths refer to the characteristics, skills, or knowledge clients or families have which reflect their capacities ~e.g., good communication skills, supportive natural-support network!. To measure the extent to which providers considered consumer strengths, the number of legitimate strengths ~i.e., those responses describing consumer weaknesses were excluded! listed was summed. Assessing Needs. Providers were also asked to describe up to six clients’ needs. To identify needs that were consistent with a family-centered approach, the needs listed by providers were coded into five categories that were emergent from the data and grounded in descriptions of family-centered service delivery ~e.g., Dunst et al., 1991!. Three of these categories reflect broad-based needs and are considered congruent with the family-centered model: basic living needs ~e.g., housing, food!, community resource0natural support network needs ~e.g., support from family0friends, legal services!, and promotional skill building needs ~e.g., education!. The remaining two categories reflect deficit-oriented needs and are considered incongruent with the targeted reform: compensatory skill building needs ~e.g., parenting skills, anger management skills!, and treatment needs ~e.g., counseling, medical assessment!. Those needs that were consistent with the family-centered service delivery approach ~i.e., basic living, community resource and promotional skill building needs! were summed to create a single score representing the total number of broad-based needs identified by providers. Transcending the Individual Level. Another goal of family-centered services is that the client’s natural support system be considered in the service delivery process. To assess 4 In three cases providers listed more than six strengths and0or six needs. Given the open-ended nature of this measure, these strengths and needs were included in the final sum for the number of strengths identified, the number of family-centered needs identified, and the number of strengths and needs identified at the family-level. In two cases these providers were not team members ~therefore, not favoring the hypothesized relationship in this study!. Further, the team member who exceeded the space allotment was excluded as an outlier, further reducing the risk of biasing subsequent analyses in favor of the hypothesized relationship. Interagency Teams and Reform • 485 the extent to which providers considered the client’s context, the strengths and needs identified were assigned one of two categories: individual client level ~e.g., motivated, creative! or family0natural support network level ~e.g., caring aunt, supportive family!. To provide an estimation of the degree to which a provider looked beyond the individual client in the identification of strengths and needs, a single score was computed by totaling the number of times a provider identified strengths or needs on the family0natural support network level. Involving Clients and Families in Goal Setting. To assess the degree to which providers involved clients and families in goal identification, providers detailed the goals that were developed for the client and indicated who was involved in developing each goal ~the provider, client, and0or family!. To estimate the degree to which providers believed that they were including clients and0or families in the creation of goals a ratio was created by dividing the number of goals which involved consumer input ~client and0or family! by the total number of goals ~a score of “1” indicating that consumers were involved 100% of the time!. Focus Group Data Collection Because our survey data was cross-sectional, potentially influenced by a self-selection and self-report bias, and based upon providers’ perceptions of their own service delivery practices, we felt it was important to validate our findings by examining consumers’ service delivery experience. Because we did not have access to client records, we conducted a focus group with a small number of consumers currently involved with the largest interagency team in the community. These consumers had previous involvement with the traditional service system and were thus well positioned to compare service delivery experiences. The purpose of this focus group was to gather data that could serve to validate or invalidate the providers’ self-reports of their service delivery practices. Sample. All families who had received services from the largest interagency team in the county in the last year were invited to participate in a focus group ~32 families!. To protect consumer confidentiality, each family was sent a letter by team leaders notifying them of the focus groups and the opportunity to provide feedback regarding the services they had received. Families were provided with a self-addressed return envelope to indicate interest. After the recruitment letter was sent, providers who had previously had contact with the families made telephone calls or home visits ~if they could not reach the family by phone! to be sure families had received the information and to ask if they had any questions. Providers did not recruit families or ask them to participate, but they did respond to questions about the groups. Of the 32 families who received services, eight families could not be contacted ~e.g., moved to another area!. Of the remaining families ~N ⫽ 24!, 42% ~N ⫽ 10! chose to participate. Transportation and child care were provided for families to facilitate participation. Each consumer who participated received a $25 reimbursement for their time. While a control focus group ~i.e., a focus group including families who did not receive teambased services! would have been ideal, issues of confidentiality in this community and limited funding did not allow for such an inquiry. The sample consisted of primarily low-income consumers ~average income of $1370 per month!. One family member per family participated in the focus group. 486 • Journal of Community Psychology, September 2002 Two participants were African American and seven were White. Two of the consumers had a high school degree and five had some trade school or college, and one had a college degree. The majority of the sample was unemployed ~n ⫽ 7!. Consumers reported working with the interagency team for an average of 11.5 months. All but one family had accessed noncategorical dollars ~flexible funds! to augment their service delivery plan and all had prior experience with the traditional service delivery system in the county. Thus, these families were well positioned to report on their service delivery experiences in traditional and team settings and compare their experiences with the team to previous experiences with the human service delivery system. Focus Group Protocol. Families were asked a series of questions during the focus group that explored their service delivery experiences with the interagency team. These questions addressed ~a! the degree to which families perceived the services provided by the team to be family-centered ~i.e., consumers being actively involved in the service delivery process; e.g., What role did you play in planning services with the team?!, ~b! the degree to which families perceived the services provided by the traditional human service delivery system to be family-centered ~e.g., What role did you play in planning services with the service delivery provider who you worked with?!, ~c! how the service delivery provided by the team differed, if at all, from the services families had received in the past ~e.g., How were the services you received working with the team similar to services you have received in the past? How were the services you received working with the team different from services you have received in the past?!, ~d! how the team impacted families’ lives ~e.g., What impact did working with the team have on your family?!, and ~e! what aspects of team involvement were most beneficial ~e.g., What do you think was most beneficial about working with the team?!. The focus group data was transcribed and then initially content analyzed by the first author. Using QSR-Nudist, the data was organized into the relevant interview question and then an iterative content analysis approach was used. For each interview question, first-order themes that described the character of team service provision and its comparison with nonteam services in Creek County were initially identified. A second-order content analysis of these emergent themes was then conducted to identify the substantive or overarching themes that emerged across the interview questions ~Strauss & Corbin, 1990!. Both first- and second-order themes were independently discussed and confirmed with the other authors. SURVEY RESULTS Providers’ Family-Centered Practices Overall, our survey data suggested that providers were implementing the various components of family-centered service delivery in a limited manner. Providers appeared to focus somewhat on client strengths, with the average provider identifying four client strengths. However, approximately 40% of the providers identified two or fewer individual level strengths. Identifying clients’ basic living needs was also rare, with only half of the providers in the sample identifying at least one basic living need ~e.g., housing, food, clothing!. In addition, few providers targeted the family level in their plan development with only half of the providers identifying at least one strength or need at the family level. However, most providers ~83%! indicated they included clients and0or families in the creation of at least half of the goals identified in the Interagency Teams and Reform • 487 Table 2. Means and Standard Deviations of Outcome Variables by Team Membership Team Members Dependent Variables Family-centered practices Identifying family-centered needs Identifying family–level strengths and needs Identifying individual–level strengths Involving consumers0families in goal setting Nonteam Members M SD n M SD n Univariate F -test ~1,114! 2.67 1.74 4.00 .79 1.90 1.99 1.52 .30 27 27 27 27 1.60 .84 3.56 .71 1.55 1.14 1.75 .38 89 89 89 89 8.87* 7.58* 1.37 1.18 *p ⬍ .01 service plan and approximately half of the providers indicated they included families in the creation of all of the goals identified ~54%!. Team Membership and Family-Centered Service Delivery Practices To examine the extent to which team members and nonteam members differed in their implementation of family-centered service delivery practices a one-way MANOVA was performed including the following dependent variables: ~a! the number of strengths identified, ~b! the number of family-centered needs identified ~e.g., basic living needs, community0natural support network needs!, ~c! the number of strengths and needs which were identified on the family level,5 and ~d! the percent of goals families were involved in creating.6 The overall F -test indicated that team members were more likely to implement family-centered practices than nonteam members ~F ~4,111! ⫽ 3.50, p ⬍ .05!. Univariate F -tests ~Table 2! indicated that interagency team members were significantly more likely than nonteam members to: ~a! identify needs consistent with strengths-based, family-centered service delivery ~F ~1,114! ⫽ 8.87, p ⬍ .01; Team Mean ⫽ 2.67; Nonteam Mean ⫽ 1.60! and ~b! identify strengths and needs of the family rather than only of the target client ~F ~1, 114! ⫽7.58, p ⬍ .01; Team Mean ⫽ 1.74; Nonteam Mean ⫽ .84!. Team members and nonteam members were not different regarding ~a! the number of strengths identified ~F ~1,114! ⫽1.37, p ⬎ .05; Team Mean ⫽ 4.00; Nonteam Mean ⫽ 3.56!, and ~b! the percent of goals in which the provider involved the client and0or family ~F ~1,114! ⫽1.18, p ⬎ .05; Team Mean ⫽ 79%; Nonteam Mean ⫽ 71%!. To determine whether the difference between team and nonteam members identification of family-centered needs was confounded by the possibility that team members served more needy families, we conducted some additional analyses. We found that team members and nonteam members did not differ with regard to the number of treatment needs they identified ~t ⫽ ⫺.24, d.f. ⫽ 114, p ⬎ .5! or the number of compensatory skill building needs they identified ~t ⫽ ⫺.19, d.f. ⫽ 114, p ⬎ .5!. This 5 To adjust for skewness in the number of family level strengths and needs identified, a log transformation was preformed on this variable for subsequent analyses. 6 One outlier ~a team member! was removed from the analyses because her scores on strengths-based, family-centered indicators were considerably higher than those of other providers. The results of the analyses were the same whether or not she was included. 488 • Journal of Community Psychology, September 2002 suggests that the difference in the number of family-centered needs identified by team versus nonteam member was not simply a result of the families served by team members having more needs, but by team members actually being more likely to focus on the broad-based needs of families.7 FOCUS GROUP RESULTS Overall, families who participated in our focus group described the services provided by the interagency team and its members as different from traditional service delivery in a variety of ways. These included team services being more ~a! comprehensive, ~b! individualized, and ~c! inclusive and nonjudgmental of consumers than those provided by the traditional human service delivery system. In other words, families reported that the services they received from interagency teams included more of the service delivery components included in the family-centered services delivery model than those offered by the traditional service system. Comprehensive Services Families described the services they received from the interagency team as more comprehensive then those provided by the traditional service delivery system. Consumers explained that the team met all of the needs of their family, including those of family members other than the target client. Families noted that this holistic focus was an unusual experience for them. In their interactions with traditional human services, they noted that it was more typical for services to focus only on their child. As two consumers stated: It’s not just the children that they’re there for. They’re helping the whole family whether it’s a two-parent, two-kid household or single parent with one child or more children. I feel the team is just that, they are a team. Even though you have these children, my husband has @a serious illness# and they were there for all of our needs, all of our needs, the team. Whereas @with traditional# human services you’re just there. You’re there for the services that they offer, if you can get them to help. Individualized Services Families also described the team as providing services that met their unique and specific needs. This individualized, person-centered approach varied significantly from the service or program-centered nature of most other services in the county. As two consumers described: They’re a totally different type of team. They more in tune with what our real needs are and not as to what their service is. 7 To perform the two group comparisons in this study, we had approximately 80% statistical power to detect a large effect size ~d ⫽ .75! and approximately 65% statistical power to detect a medium effect size ~d ⫽ .5!. Interagency Teams and Reform • 489 I don’t think there’s any similarity to it @traditional services# because when you go out . . . into the community and you stop at a lot of different places . . . They don’t care about the details or what the needs is and @the team# wanted to know details. I mean because they wanted to know what all the reasons were for wanting @the service#, why did we really need this service. And we were able to explain all the experiences that we’ve had, what it’s done to us . . . and that this type of illness put us financially back . . . emotionally back, physically brought us down. And they cared about those things and other places that you go to they don’t care . . . They’re there for this type of service and that type of service only. One consumer noted that such individualized treatment is needed because consumer’s vary in their needs and desires: There’s variance, though. Everybody’s had different experiences . . . you know some people need help with rent or housing or getting food . . . other people . . . have sick children or they have children that have been abandoned by other family members. I myself have no family support. So they’re a basic, they’re a basic support system for me . . . my story isn’t similar to a lot of these. The only thing that’s in common is I have a child and I’m working with the same team. As part of this individualized service provision, consumers reported that the team addressed a broad array of family needs ~i.e., material, physical health, mental health, etc.!. Overall, this focus on the unique and multiple needs of families significantly varied from how services were typically provided in Creek County. Inclusive/Nonjudgmental Services Families described the team as including them fully in the service delivery process. Overall, they described being more involved in goal development, service planning, and service implementation. One family described the initial planning meeting: You know, she sat down with me before we went to the team leader. @She asked,# what do you need? What would help you in the future? What would keep you on your feet? . . . Sit down and write me a plan. They didn’t come and say, okay, you need this and this and this . . . and we’ll get you this and this . . . Other families shared this experience, one said: I planned, the team and I sat down and made a plan. Because see you’ve got . . . children, you need so many things. We started out with medical. The children needed medical help, dental help, psychological help. You know . . . there was no beds, there was no clothing, there was nothing. Families described the planning process they experienced in traditional human services as very different from this process. They noted that in non-team service delivery settings, providers typically did not take the time to inquire about their needs or, if they did, their needs were ignored in the service delivery plan. 490 • Journal of Community Psychology, September 2002 It’s frustrating. You try to tell’em what you need or what the problem is and it’s like a smack in the face. They don’t care. It’s well, I don’t care. I have my job to do and that’s what I’m gonna do . . . They’re just like well, that’s not my problem. It’s kind of like you go to a store and you ask where something is and they’re like well, that’s not my department. Families also noted that they felt more respected and valued by interagency team members than by other, nonteam providers in the county. Families stated that in their interactions with the traditional service delivery system they often felt judged or looked down upon. However, this was not their experience with interagency team members. One consumer said, “The person ~team member! that comes to your house isn’t just somebody that’s coming to look at your house or your family. They’re more like a friend.” Another consumer echoed this saying, “If something doesn’t go right or I don’t do something right, they’re not there to knock me down. They’re there to provide another avenue or give me something else.” Overall, this experience of inclusion and of being respected helped families feel that the team gave them the tools to make and sustain positive changes in their lives. One participant shared her feelings about how the team had prepared her for the future, “they’re making it able for me to get back up on my feet and when they do pull out and they do leave, I’m not going to fall flat on my face.” Another consumer also expressed her feelings that the team prepared her to make a better life for herself and her children. She said, They’re not just like in your life, then out of your life . . . They make sure that you have the needs you know to take care of your kids . . . to find out your support and things like that and if you need someone to talk to they’re there for you . . . not just to give you things, but to help you focus and make a better life for you and your children or your grand kids . . . Of course, while the participants in our focus group described the services they received from interagency teams and their members as more family centered than those offered by the traditional service system, it is important not to assume that this experience is reflective of all consumer’s interactions with these teams and their members. It is possible that consumers that refused to participate in this focus group were the least satisfied with the services they received, the most critical of the interagency teams, or those most in need. Nevertheless, this focus group data does provide some validation for the survey data results, suggesting that interagency team members are more likely to provide family-centered services than providers within the traditional service delivery context. DISCUSSION The findings from this study suggest that interagency teams may be a promising venue for fostering the successful implementation of some types of human service delivery reform. In Creek County, the service delivery providers involved in interagency teams were more likely to engage in service delivery practices that reflect a family-centered approach to care. Team members were more likely than nonteam members to identify the strengths and needs of the entire family and to meet broad-based family needs. Consumers working with interagency teams described the services they received as Interagency Teams and Reform • 491 more comprehensive, inclusive, and individualized than the services they had previously received in the county. The potential role that interagency teams can play in promoting the successful implementation of family-centered service delivery is highlighted when we consider the extent to which these reforms were not being implemented in other contexts within Creek County. Despite the county’s widespread efforts at diffusing these reforms, and the fact that their implementation was mandated by several funding sources, at the time of this study few service delivery providers were actually implementing these changes. For example, nearly three quarters of the providers in our sample identified no ~47%! or only one ~27%! family level strength or need with their last client. While most providers in the county were limited in their implementation of these new reforms, providers involved in interagency teams were more likely to practice in ways that were consistent with a family-centered approach. While it does appear that providers who were involved in teams were more likely to implement these innovations, methodological limitations of our study require us to be cautious in our interpretation of the data. There are three issues related to selfselection that could potentially bias our findings. First, there is the possibility of a response bias in the sample of participants who completed the survey. Examination of differences across organizations with different response rates suggests that our sample was representative of providers in the county. While our overall response rate was 62%, for 10 organizations we had a response rate of 100% and in another three we had a response rate of over 80%. If our overall sample was biased, we would expect different patterns of findings for organizations with low response rates. While the small number of interagency team members made it impossible to test this in the current analysis, in a previous study using the total sample, we found a consistent pattern of relationships between providers’ perceptions of contextual support for reform and their attitudes towards these reforms across organizations, regardless of response rate ~Foster-Fishman et al., 1999!. There is no reason to think that providers would be any more likely to self-select on the basis of their service delivery practices than on their beliefs. Second, given the cross-sectional nature of this data any causal attributions made are limited. It is possible that providers who were more family-centered in their practice were more likely to join teams. Our experiences observing county planning meetings, interagency team meetings, and feedback sessions suggest that this was not the case, however. Most providers did not volunteer to be team members, but were assigned to the team by their organizational leaders. Many providers expressed skepticism about the team approach at their initial meetings. Over a 12-month period we observed changes in providers’ behavior during interagency team meetings that demonstrated greater alignment with family-centered service delivery practices ~e.g., changes in the language used to describe families, greater knowledge of available resources!. In addition, one service provider testified during our feedback meetings that she was initially resistant to the family-centered approach to service delivery, but had changed her attitude and practice with all of the families she served as a result of her participation on the team. Finally, our sample of families who took part in the focus group was small and self-selected. It is certainly possible that those families who participated in the focus group had more satisfying experiences with the services they received from the teams and0or that they were coping better at the time the focus group was conducted. While these families clearly experienced the services that they received from the team as 492 • Journal of Community Psychology, September 2002 different from those they had received in the past, we cannot generalize their experiences to those of all families who receive team-based services. While we must be cautious in our causal interpretations and in how we generalize our findings, extensive knowledge of the county’s service delivery system, authentification by members of the setting, and validation of our findings with the focus group data give us confidence in the validity of our conclusions. Providers who were team members demonstrated service delivery practices that were more consistent with a family-centered approach and families who had received team-based services reported that their experiences were consistent with this approach. Given that many previous attempts to transform human service delivery practices have been unsuccessful ~e.g., Bargal & Schmid, 1992; Foster-Fishman & Keys, 1997; Glisson, 1978!, the fact that interagency team membership was associated with provider practices that were consistent with the desired changes is particularly notable. They point to the need for future longitudinal research that can more adequately address the questions of causality and generalizability. Our findings also raise the question of how interagency teams might facilitate the adoption of innovation. Klein and Sorra ~1996! provide one framework for understanding how settings can promote a climate that encourages the implementation of innovation. The ways in which interagency teams might promote such a climate is discussed below. Promoting a Climate for Implementation The fundamental challenge faced in innovation implementation is promoting employee support for and use of new practices. Klein & Sorra ~1996! argue that employees are far more likely to implement an innovation if they work in a setting that does the following: ~a! promotes needed skills, ~b! provides incentives for the adoption of innovation and disincentives for failure to adopt innovation, ~c! eliminates obstacles to implementation, and ~d! develops employee attitudes and values that are congruent with the innovation. While our study did not examine how teams impact the implementation of innovative service delivery practices, given what we know about an interagency team’s structure and process ~e.g., Pandiani & Maynard, 1993; VanDenBerg & Grealish, 1996! and what we observed in Creek County, interagency teams appear to have many of the qualities Klein and Sorra ~1996! highlight when describing settings that have the capacity to facilitate the adoption of innovation. One critical component of innovation implementation is the promotion of relevant employee skills. The switch to family-centered care requires providers to deliver services in a manner that is very different from traditional approaches to care and requires skills often not found among service delivery providers ~Dunst, 1985; VanDenBerg & Grealish, 1996!. While Creek County had sponsored several workshops on these reforms, providers outside of the interagency team context were offered no additional implementation support, were not provided with the time needed to practice this behavior, and often faced incompatible policy and procedures within their home organization ~Foster-Fishman et al., 1999!. Teams provide a context where new service approaches can be modeled ~Bandura, 1977!, less experienced service providers can learn from those more experienced in implementing the reforms ~Bailey et al., 1991!, and members can practice new skills and behaviors. In addition to promoting skill development, Klein and Sorra ~1996! suggest that settings must provide incentives for implementing a desired innovation and disincentives for avoiding implementation. The loosely coupled structure of traditional human Interagency Teams and Reform • 493 service delivery organizations significantly inhibits an agency’s capacity to monitor the implementation of reforms and to reward and sanction behavior ~Weick, 1976!. In contrast, in teams where providers practice in a group, behavior that is often invisible to other providers becomes visible. The team may therefore be better equipped to support, reward, and monitor providers’ implementation of reforms. A third strategy for creating a climate for innovation involves eliminating obstacles to successful implementation. In the traditional service delivery system, organizational structures and norms often fail to support the implementation of service delivery innovations ~Cameron & Vanderwoerd, 1997; Cohen & Lavach, 1995; Foster-Fishman et al., 1999!. Implementing innovative services to consumers is difficult considering the fragmentation of a traditional service delivery system that can limit providers’ access to and knowledge of other service delivery organizations. Interagency teams may assist providers in circumventing barriers to addressing broad-based consumer needs because they create opportunities to interact with providers from other organizations. This might increase their knowledge of what other organizations have to offer and how these services can be accessed. Finally, in some communities teams may increase providers’ access to noncategorical funds that can be employed to support a greater implementation of family-centered service delivery. Finally, the implementation of innovations is also more likely to succeed if providers have positive attitudes towards and value the targeted innovation. Positive attitudes towards service delivery changes has been found to be a predictor of providers’ adoption and implementation of a more integrated approach to service delivery ~Glisson & Hemmelgarn, 1998! and more family-centered practices ~Allen, 1997!. Interagency teams may help providers adopt positive attitudes towards these reforms by providing a forum where the benefits of reforms are demonstrated and the barriers to their implementation are minimized ~Foster-Fishman et al., 1999!. The Limitations of Interagency Teams It is important to note that we found that team members were not distinguishable from nonteam members on all indices of family-centered practice. While team members were more likely than nonteam members to address the broad-based needs of clients and to address the needs and strengths of all family members, they were not more likely to identify the strengths of the “target client” or to include clients and0or families in creating goals. These findings could be related to differences in the ease of implementing these separate elements of the reforms. It is possible that the identification of consumer strengths and the inclusion of families in goal setting require less behavior change on the part of providers than the other elements of these reforms ~e.g., attending to the specific needs of family members in addition to the target client! and that the training offered to all staff in Creek County on the familycentered model may have been a sufficient means for facilitating these changes. These findings may also be attributed to differences in the efficacy of the team context in promoting different behavioral shifts. Future research should further explore the impact of interagency team membership on the different behavioral changes required by these reforms. Finally, the lack of differences with regard to including families in the identification of goals may reflect a limitation of our measure. In particular, providers were asked to indicate whether or not they included clients and0or families in goal setting, but were not asked to detail how they included them. Thus, a provider who simply asked a client if a particular goal was acceptable might appear equivalent 494 • Journal of Community Psychology, September 2002 to a provider who asked a family to develop the goal on their own. Future research should attempt to capture providers’ service delivery practices with more sensitive indices. In addition, it is critical to note that while participation on interagency teams is associated with the implementation of new service delivery practices by individual providers, these settings did not appear to impact the overall service delivery system. Within Creek County, significant contextual barriers to this form of service delivery still existed ~Foster-Fishman et al., 1999!. These barriers included rigid financial reimbursement procedures that funded categorical services and restricted flexible funding pools, strict rules regarding when and where providers could deliver services to consumers, and inconsistent leader support of family-centered service delivery. While interagency teams may be a promising avenue for helping individual providers to change their approach to service delivery, their influence may be limited to those who participate in them. IMPLICATIONS FOR PRACTICE AND RESEARCH Overall, the findings from this study have important implications for those interested in promoting and understanding the successful implementation of family-centered service delivery. Given that teams appear to provide a venue for the adoption of positive attitudes towards ~Foster-Fishman et al., 1999! and implementation of practices consistent with this reform, it seems important to maximize the number of providers involved in these settings. However, it is common practice to include only a selected group of providers in these teams. For example, in Creek County, only 21% of all providers surveyed reported at least some experiences working in these interagency teams. Increasing the number of teams and0or rotating team membership among several service providers would expose more providers to this setting. Rotating membership may also avoid the perception that the team setting is a new service delivery program ~Williams, 1995! as opposed to a setting employed to facilitate the implementation of a new service delivery philosophy. While interagency teams appear to be an effective intervention for promoting service delivery reform, it is important to recognize that they are resource intensive, requiring significant amounts of financial support ~e.g., noncategorical dollars! and service provider time ~which is sometimes not reimbursable!. Still, if providers who are team members are better at meeting consumers’ broad-based needs and facilitating their independence, teams may actually result in a reduction in service delivery costs in the long run. Thus, one direction for future research is an examination of the short- and long-term costs associated with team interventions. In addition, in examining the financial costs associated with interagency teams, it is vital to build a better understanding of those factors that contribute to effective team functioning. Given the small number of teams in our study, we were not able to systematically examine what internal group factors contributed to an effective team. Certainly, previous research would suggest that group characteristics such as group cohesiveness, leadership style, and internal communication patterns might significantly affect the interagency team’s capacity to influence group member behaviors and accomplish its goals ~e.g., Hackman, 1990; Larson, Foster-Fishman, & Franz, 1998!. In conclusion, interagency teams appear to be a promising intervention for encouraging providers’ implementation of family-centered service delivery practices. Given the difficulties typically associated with implementing new service delivery technolo- Interagency Teams and Reform • 495 gies, interagency teams may be a promising context for fostering the implementation of service delivery reforms. Future research aimed at establishing the causal role of teams in shifting providers’ service delivery practices, exploring how teams influence their participants, and identifying what types of reforms are facilitated through a team approach and what types of families are best served will help us to better understand the potential role of interagency teams in human service delivery system reform. REFERENCES Adams, P., & Krauth, K. ~1995!. Working with families and communities: The patch approach. In P. Adams & K. Nelson ~Eds.!, Reinventing human services: community and familycentered practice ~pp. 87–108!. New York: Aldine De Gruyter. Allen, N.E. ~1997!. The process of adoption of the family-centered model among service providers. Unpublished master’s thesis, Michigan State University, East Lansing. Argyis, C. ~1970!. Intervention theory and method. Reading, MA: Addison Wesley. Armenakis, A.A., Harris, S.G., & Mossholder, K.W. ~1993!. Creating readiness for organizational change. Human Relations, 46~6!, 681–703. Bailey, D.B., Palsha, S.A., & Simeonsson, R.J. ~1991!. Professional skills, concerns, and perceived importance of work with families in early intervention. Exceptional Children, 58~2!, 156 –165. Bandura, A. ~1977!. Social learning theory. Englewood Cliffs, NJ: Prentice-Hall. Bargal, D., & Schmid, H. ~1992!. Organizational change and development in human service organizations: A prefatory essay. In D. Bargal & H. Schmid ~Eds.!, Organizational change and development in human service organizations ~pp. 1–13!. New York: Haworth Press. Bartunek, J.M., & Louis, M.R. ~1996!. Insider0outsider team research. Thousand Oaks, CA: Sage. Bartunek, J.M., & Moch, M.K. ~1987!. First-order, second-order, and third-order change and organizational development interventions: A cognitive approach. Journal of Applied Behavioral Science, 23, 483– 500. Bradley, C.F. ~1983!. An evaluation of family-centered maternity care. Women and Health, 8~1!, 35– 47. Bronfenbrenner, U. ~1979!. The ecology of human development: Experiments by nature and design. Cambridge: MA: Harvard University Press. Cameron, G., & Vanderwoerd, J. ~1997!. Protecting children and supporting families: Promising programs and organizational realities. New York: Aldine De Gruyter. Cohen, R., & Lavach, C. ~1995!. Strengthening partnerships between families and service providers. In P. Adams & K. Nelson ~Eds.!, Reinventing human services: Community and family-centered practice ~pp. 261–277!. New York: Aldine De Gruyter. Duchnowski, A.J. ~1994!. Innovative service models: Education. Journal of Clinical Child Psychology, 23 ~Suppl.!, 13–18. Dunst, C.J. ~1985!. Rethinking early intervention. Analysis and Intervention in Developmental Disabilities, 5, 165–201. Dunst, C.J., & Trivette, C. ~1987!. Enabling and empowering families: Conceptual and intervention issues. School Psychology Review, 16~4!, 443– 456. Dunst, C.J., Johanson, C., Trivette, C., & Hamby, D. ~1991!. Family-oriented early intervention policies and practices: Family-centered or not. Exceptional Children, 58~2!, 115–126. Flint, S. ~1993!. Family-centered child welfare services: A national overview of jurisdictions that have adopted a systemic family centered approach. ~Available from the Vermont Department of Social and Rehabilitation Services, New England Association of Child Welfare Commissioners and Directors, Judge Baker Center, 3 Black Fan Circle, Boston, MA 02115; http: www.jbcc.harvard.edu0index.htm.! 496 • Journal of Community Psychology, September 2002 Foster-Fishman, P., & Keys, C.B. ~1997!. The person0environment dynamics of employee empowerment: An organizational culture analysis. American Journal of Community Psychology, 25~3!, 345–370. Foster-Fishman, P., Salem, D., Allen, N.E., & Fahrbach, K. ~1999!. Ecological factors impacting provider attitudes towards human service delivery reform. American Journal of Community Psychology, 27~6!, 785–816. Friedman, R.M. ~1994!. Restructuring of systems to emphasize prevention and family support. Journal of Clinical Child Psychology, 23, 40– 47. Garbarino, J. ~1988!. The human ecology of early risk. In S.J. Meisels & J.P. Shonknoff, Handbook of early childhood intervention ~pp. 78–96!. Cambridge: Cambridge University Press. Garshelis, J.A., & McConnell, S.R. ~1993!. Comparison of family needs assessed by mothers, individual professionals and interdisciplinary teams. Journal of Early Intervention, 17~1!, 36 – 49. Glisson, C. ~1978!. Dependence of technological routinization on structural variables in human service organizations. Administrative Science Quarterly, 23, 383–395. Glisson, C., & Hemmelgarn, A. ~1998!. The effect of organizational climate and interorganizational coordination on the quality and outcomes of children’s service systems. Child Abuse & Neglect, 22~5!, 401– 421. Golembiewski, R.T. ~1985!. Humanizing public organizations. Mt. Airy, MD: Lomond. Hackman, J.R. ~1990!. Groups that work and those that don’t. San Francisco, CA: Jossey-Bass. Kelly, J.G., Ryan, A.M., Altman, B.E., & Stelzner, S.P. ~2000!. Understanding and changing social systems. In J. Rappaport & E. Seidman ~Eds.!, Handbook of community psychology. New York: Plenum. Klein, K.J., & Sorra, J.S. ~1996!. The challenge of innovation implementation. Academy of Management Review, 4, 1055–1080. Knitzer, J. ~1982!. Unclaimed children: The failure of public responsibility to children and adolescents in need of mental health services. Washington, DC: The Children’s Defense Fund. Kouzes, J., & Mico, P. ~1979!. Domain theory: An introduction to organizational behavior in human service organizations. Journal of Applied Behavior Science, 15, 449– 467. Larson, J.R., Foster-Fishman, P.G., & Franz, T.M. ~1998!. Leadership style and the discussion of shared and unshared information in decision-making groups. Personality and Social Psychology Bulletin, 24, 482– 495. Lewin, K. ~1951!. Field theory in social science. New York: Harper & Row. Lewis, M.R. ~1995!. Linking schools with family- and community-centered services. In P. Adams & K. Nelson ~Eds.!, Reinventing human services: Community and family-centered practice ~pp. 163–173!. New York: Aldine De Gruyter. Lipsky, M. ~1980!. Street-level bureaucracy. New York: Russell Sage. Mahoney, G., & O’Sullivan, P. ~1990!. Early intervention practices with families of children with handicaps. Mental Retardation, 28, 169–176. Marcenko, M.O., & Smith, L.K. ~1992!. The impact of a family-centered case management approach. Social Work in Health Care, 17~4!, 87–100. McBride, S.L., Brotherson, M.J., Joanning, H., Whiddon, D., & Demmitt, A. ~1993!. Implementation of family-centered service: Perceptions of families and professionals. Journal of Early Intervention, 17~4!, 414 – 430. Morrissey, J.P., Johnsen, M.C., & Calloway, M.O. ~1997!. Evaluating performance and change in mental health systems serving children and youth: An interorganizational network approach. Journal of Mental Health Administration, 24~1!, 4 –22. Interagency Teams and Reform • 497 Nelson, K. & Allen, M. ~1995!. Family-centered social services: Moving toward system change. In P. Adams & K. Nelson ~Eds.!, Reinventing human services: Community and family-centered practice ~pp. 109–125!. New York: Aldine De Gruyter. Pandiani, J.A., & Maynard, A.G. ~1993!. Vermont’s local interagency teams: An evaluation of service coordination and system change. Community Alternatives: International Journal of Family Care, 5~1!, 85–97. Rappaport, J. ~1977!. Community psychology: Values, research, and action. New York: Holt, Rhinehart and Winston. Rogers, E. ~1995!. Diffusion of innovation ~4th ed.!. New York: Free Press. Rounds, K.A. ~1991!. Early intervention services for very young children and their families under P.L. 99– 457. Child and Adolescent Social Work, 8~6!, 489– 499. Scannapieco, M. ~1994!. Home-based service program: Effectiveness with at risk families. Children and Youth Services, 16~5– 6!, 363–377. Schein, E.H. ~1985!. Organizational culture and leadership. San Francisco: Jossey-Bass. Schorr, L.B. ~1988!. Within our reach: Breaking the cycle of disadvantage. New York: Doubleday. Strauss, A., & Corbin, J. ~1990!. Basics of grounded theory methods. Beverly Hills, CA: Sage. Stroul, B.A., & Friedman, R.M. ~1986!. A system of care for children and youth with severe emotional disturbances. Washington, DC: CASSP Technical Assistance Center, Georgetown University. Trivette, C., Dunst, C., & Hamby, D. ~1996! Characteristics and consequences of help-giving practices in contrasting human service programs. American Journal of Community Psychology, 24~2!, 273–293. Tyler, F.B., Pargament, K.I., & Gatz, M. ~1983!. The resource collaborator role. American Psychologist, 38~4!, 388–398. VanDenBerg, J.E., & Grealish, M.E. ~1996!. Individualized services and supports through the wraparound process: Philosophy and procedures. Journal of Child and Family Studies, 5~1!, 7–21. Watzlawick, P., Weakland, J.H., & Fisch, R. ~1974!. Change: Principles of problem formation and problem resolution. New York: W. W. Norton & Co., Inc. Weick, K.E. ~1976!. Educational organizations as loosely-coupled systems. Administrative Science Quarterly, 21, 1–19. Weiss, H., & Jacobs, F. ~Eds.!. ~1988!. Evaluating family support programs. Hawthorne, NJ: Aldine de Gruyter. Weissbourd, B. ~1990!. Family resource and support programs: Changes and challenges in human services. Prevention in Human Services, 9~1!, 69–85. Weissbourd, B., & Kagan, S.L. ~1989!. Family support programs: Catalysts for change. American Journal of Orthopsychiatry, 59, 20–31. Williams, S.D. ~1995!. Integrating the family service system for the inside out: A view from the bureaucratic trenches. Journal of Family and Economic Issues, 16~4!, 413– 424. Yoe, J.T., Santarcangelo, S., Atkins, M., & Burchard, J.D. ~1996!. Wraparound care in Vermont: Program development, implementation, and evaluation of a statewide system of individualized services. Journal of Child and Family Studies, 5~1!, 23–39.