Inter-organizational relationships of health partnerships: characteristics of the Fulton County SPARC program.
Abstract: Local governments have long used partnerships and collaborations with the nonprofit and private sectors to improve services, support innovation, and save cost. Nowhere is this more evident than in the public health and human services sectors. This case study follows up on an earlier report on the attributes of a collaborative public/private-nonprofit health partnership; Fulton County, Georgia's Sickness Prevention Achieved Through Regional Collaboration program, (SPARC). Fulton County has been able to sustain its SPARC partnership despite the demise of similar efforts within the region. This study examines the characteristics of the Fulton County SPARC partnership that may have contributed to its endurance. A series of interviews and surveys were administered to Fulton County SPARC partners in 2009 and 2011 to capture operational characteristics of the partnership and the degree to which partners interacted among themselves. Results indicate that despite changes thought to strengthen the partnership, the developmental stages of partnerships and collaborations are shown to be non-linear; therefore, movement from one stage to the other is not guaranteed thus warranting leaders' regular attention to both individual and organizational commitment to collaboration dynamics.
Article Type: Report
Subject: Interorganizational relations (Management)
Public-private sector cooperation (Management)
Public health administration (Research)
Author: Sullivan, Brenda A.
Pub Date: 06/22/2012
Publication: Name: Journal of Health and Human Services Administration Publisher: Southern Public Administration Education Foundation, Inc. Audience: Academic Format: Magazine/Journal Subject: Government; Health Copyright: COPYRIGHT 2012 Southern Public Administration Education Foundation, Inc. ISSN: 1079-3739
Issue: Date: Summer, 2012 Source Volume: 35 Source Issue: 1
Topic: Event Code: 970 Government domestic functions; 200 Management dynamics; 310 Science & research Computer Subject: Company business management
Geographic: Geographic Scope: United States Geographic Name: Fulton County, Georgia Geographic Code: 1USA United States
Accession Number: 304051554
Full Text: OVERVIEW OF THE SPARC INITIATIVE

The SPARC health collaborative model was birthed within a strategy to provide preventive health care to senior citizens. In 2006, at the invitation of the Atlanta Regional Commission and national SPARC founder Dr. Douglas Shenson, several aging and health providers within the Atlanta, Georgia area accepted an invitation to a community-wide forum where the SPARC leadership discussed their idea of relieving the burden of primary care on providers by establishing preventive care clinics for senior citizens. The meeting organizers envisioned a regional arrangement in Atlanta, similar to the successful program they had developed in four counties that bordered New York, Massachusetts and Connecticut (Brodeur, 2007).

SPARC's general approach is to enlist collaboration among public, nonprofit and private health providers, local government agencies, community groups, and others to use existing community resources for delivery of preventive care for seniors. However, until the SPARC visitors called the first meeting in Atlanta, very few health care providers in Fulton County had seen a need for special preventive care to senior citizens. In fact, only a few of the organizations that were called to the meeting had previously partnered with the others that met around the table regarding service delivery. Hence, at the time of the introductory meeting there was no group within Fulton County identified as a collaborative leader in the area of preventive healthcare for seniors.

Although many of the invited organizations were attracted to the idea of a coordinated effort to address the health care needs of the region's senior citizens, only a few had the resources (time, personnel, etc.) to become involved in a regional effort. Consequently, only Fulton and Fayette counties agreed to serve as pilots for an expansion of the SPARC model into Georgia. In Fulton County , the County Department of Aging quickly became the 'lead' organization, convening partnership meetings and organizing SPARC health clinics. After a successful pilot study, SPARC became integrated into the regular programming of Fulton County's Department of Aging. Although providing some of the same deliverables as the Fulton program, the Fayette program was not so successful and did not last long beyond the pilot period. Fulton County's SPARC, however, has been able to sustain an active partnership and this study examines the attributes of the Fulton County SPARC partnership from 2009- 2011 that may have contributed to its endurance as a successful collaborative.

LITERATURE

The terms collaboration and partnership are used interchangeably, and defined in many ways. Most definitions contain the following attributes: "a mutually beneficial and well-defined relationship entered into by two or more organizations to achieve common goals. The relationship includes a commitment to mutual relationships and goals; a jointly developed structure and shared responsibility; mutual authority and accountability for success; and sharing of resources and rewards" (Parkinson, 2006). Throughout this paper, the terms collaboration and partnership are used interchangeably since the Fulton SPARC is a partnership, but identifies itself by name as a collaboration.

The literature on health and human service related community partnerships and collaborations contain five main themes:

1) descriptions of how to start and maintain collaborations;

2) historical assessments of successful and unsuccessful collaborative efforts;

3) outcome evaluations of collaborations;

4) explanations for why collaborations are sought; and more recently,

5) the development typologies of collaborations.

Specifically, the health collaboration literature includes studies of outcomes for service delivery (Brown, 2006; Parker-Oliver, 2005; Hicks, 2008); and, a few articles which focus on the application of general partnership characteristics to health collaborations (Alexander, 2001). The literature on health collaborations echoes the themes found in the general collaboration literature and there is agreement that clear goals, trust and communication are factors necessary for successful collaborations.

A recent trend in the research on collaborations, which is the focus of this paper, is to develop typologies of collaborations to explain their development. One of the forerunners of this theme is Kernaghan (1993). He describes partnerships as ranging from collaborative partnerships (real) to phony partnerships as summarized in Table 1 below. The Kenaghan model focuses on the interactions of the partners rather than the structure and leadership of the collaboration. Although writing from a specific historical experience, the work is important in that it includes both public and public/private sector examples, demonstrating that the characteristics of successful partnerships and collaborations extend beyond economic sectors.

Drawing upon Kenaghan and others, Frey (2006) and Thompson (2009) provide tools to measure the degree to which organizations have reached a certain level of collaboration. Their studies are useful and important in that community partnerships can use this type of information in their self-assessments and future planning. These works are of particular relevance for this discussion in that our goal is to understand why and how the Fulton County SPARC has continued.

In Houge's Levels of Community Linkage Model, as described by Frey (2006), partnerships start as loosely formed networks and advance toward the final level of collaboration. They begin as independent Networks that develop into Cooperations, which share information, begin to define their roles and formalize communications. The third stage of the model is Coordination where both information and resources are shared, roles are defined, communication is open, and independence fades into some shared decision making. The fourth stage, Coalition, consists of increased sharing of ideas and resources. In Coalitions, partners have an equal vote in decision-making. The final stage, Collaboration, is characterized by a unified system of trust and consensus in decision-making. Organizations do not necessarily fit neatly into each stage, carrying remnants of the previous stage as they move forward and reverting to previous stages as they develop. Frey builds upon and expands this model by designing a tool to qualitatively measure the existence of these stages and includes options for being within a collaboration without interacting with all of the partners.

The Thompson Measure of Collaboration Survey, (2009), explores several organizational dimensions of partnerships: organizational autonomy; mutuality of interest; administration of operations, existence of group norms; and organizational governance. This model states that in successful partnerships:

a) collaborative partners must understand how to jointly take partners from governance to action;

c) partners maintain distinct identities and organizational authority separate from a collaborative identity;

d) partners must experience mutually beneficial interdependencies based on differing or shared interest, and;

e) partners must trust each other.

This current study examines the degree to which the organizational dimensions and operational characteristics, independently identified by Frey and Thompson, exist within the Fulton County SPARC collaboration. Knowledge about these aspects of partnerships are useful in strengthening and growing community collaborations. The study also builds upon previous work by jointly utilizing stage theory and organizational interaction (Frey, 2006); and examining the organizational characteristics of collaborative partners (Thompson, 2009) to explore the various aspects of a local government health partnership/collaboration. The study will also discuss some of the changes made within the Fulton County SPARC since 2009, and whether those changes have inhibited or enhanced the collaboration.

METHODOLOGY

Using a time-series design, two features of the SPARC partnership are explored in this study:

* operational characteristics, and

* developmental stages based on a network-collaborative continuum.

These attributes of the Fulton County collaboration are explored through:

1) expert interviews with the 'principal leadership' of SPARC;

2) a survey to determine the intensity of SPARC's collaborative efforts, Levels of Collaboration Scale; and

3) a second survey to ascertain where Fulton County's SPARC lies on the developmental stage continuum, Measures of Collaboration Survey.

The two surveys differ in that the Levels of Collaboration Scale measures organizational interaction according to a collaborative continuum, whereas the Measures of Collaboration Survey examines the strength of the partnership based on the nature of its organizational structure and activities. Questions for the expert interviews were designed from a combination of subject items from the Levels and Measures surveys.

Expert Interviews

Expert interviews were conducted with the leadership of the collaborative, in 2009 and 2011, before distributing the self-administered surveys to the remaining partner organizations. In 2009, all four of the then current members of the Fulton County 'SPARC Leadership' team participated in the interviews, but only two Fulton County community outreach staff members participated in the 2011 interviews.

The 2009 leadership team consisted of the following persons:

1) the lead community outreach worker from the Fulton County Division of Aging, viewed as the SPARC coordinator;

2) a Fulton County Division of Aging staff person who assists in specifically organizing outreach for SPARC and manages the technical aspects of the meetings and clinics;

3) a manager from the Fulton County Division of Aging who was involved with the original plans for SPARC; and,

4) a staff person from the Fulton County Department of Health and Wellness who has been instrumental in assuring communication with partners. None of the interviewees held formal titles as SPARC leaders; however, they had evolved as point persons either because of the County's commitment to SPARC or their personal early involvement in the program. (1)

There was not much change in the interview schedule between the 2009 and 2011 administrations. In both periods, SPARC leaders were asked 24 open-ended questions. The questions, divided into 6 sections, were derived from the dimensions of the Measures of Collaboration Survey and included measures of : a) durability of the partnership; b) administration of the collaboration, c) mutuality within the partnership, d) governance of the collaboration, e) authority within the partnership and f) norms within the collaboration that are specific to the Fulton County SPARC. Embedded within each of the dimensions are questions developed from the Levels of Collaboration Scale seeking clarification on, 1) the roles of partner organizations, 2) how information and resources are shared within the collaborative, and 3) the vehicles used for communication among partners. In this way, the leadership of the collaborative could speak qualitatively to the same issues explored in the surveys.

Only one section of the interview schedule was changed from 2009 to 2011. In 2009, the questions relating to the durability of the collaborative relationship centered on the continued need for SPARC and why organizations belonged to the collaborative. In 2011, drawing upon Norris-Tirrell and Clay's (2011) work on strategic collaboration, the durability questions centered on the impetus for bringing the Fulton County SPARC partners together. Another change from 2009 to 2011 is that the 2009 interviews were conducted via face to face; however due to changes in the author's access to personnel in 2011 the interviews were conducted via telephone.

There were a number of partnership changes within the Fulton collaboration between the 2009 and 2011 administrations. In 2009, organizations that served as physical sites for the clinics were not considered SPARC partner members and did not appear on the sampling frame. However, by 2011, Fulton's SPARC began to include the managers of the site locations as part of the partnership planning meetings and current clinic site partners were included in the sampling frame and received one of the two surveys. Thirteen (13) partners from 2009 were still part of the collaboration in 2011. However, fifteen (15) partners from 2009 had left the collaboration by 2011. These partners were replaced by ten (10) new partners in the 2011 administration of the surveys. Forty percent (40%) of the 2011 members were not part of the collaboration in 2009.

The Levels of Collaboration Scale and Measures of Collaboration Survey

In both 2009 and 2011, Fulton SPARC partners were informed about the surveys at one of their regular SPARC meetings. This was followed by an e-mail from the researcher providing the partners access to the surveys. Two additional e-mails were sent as reminders. The 2011 administration took place in the summer, which may explain, in part, the lower return rates, and will be discussed later in the paper. Each of the instruments used in the study had undergone rigorous review and were within the public domain. No major changes were made to the instruments therefore no pre-testing was conducted. The authors and developers of the two instruments were contacted and informed that their surveys would be used in the study.

Using a random selection method, Fulton County SPARC partners were sent one of two surveys, Measures of Collaboration (Measures) and Levels of Collaboration Scale (Levels). Each survey measured a different aspect of the partnership. Levels examined the interaction among partners to determine where the collaboration lies on a continuum. Measures sought information on the strength of the collaboration by exploring the degree to which the collaboration had certain characteristics. Current lists of partners served as the sampling frames for 2009 and 2011. Using a table of random numbers the SPARC partners were divided into two groups, Group A and Group B. In 2009 and 2011, Group A received the Levels survey via e-mail and Group B received an e-mail with a link to the Measure of Collaboration Survey via Survey Monkey. Due to accessibility issues, all surveys were sent via e-mail in 2011.

The Surveys

The Levels of Collaboration Scale, sent to Group A, asked respondents to rate their level of interaction with each of the 33 organizations that were SPARC partners between 2007 and early 2009; and the 28 organizations that were part of the partnership in 2011. The actual scale included the following categories and weights: No Interaction (0), Network (1), Cooperative (2), Coordination (3), Coalition (4), and Collaboration (5). As one moves up the Scale from 0-5, partnerships are assumed to increase in structure, leading to true collaborations. Knowledge of this progression may have influenced the responses; therefore for this study, only the descriptions of the levels were provided on the survey form and respondents were not given the labels or weights for each description.

Respondents were instructed to indicate if their organization interacted with another identified partner organization with varied degrees of sharing information, having lines of communication, and decision-making. Respondents were asked not to rate their own organizations, and to choose only one level that described their interaction with each of the other partners.

For both administration periods, the Measures of Collaboration Survey (Group B), adopted from Thompson's model for conceptualizing and measuring collaboration, consisted of 17, seven scale, Likert statements that measured 1) degrees of organizational autonomy; 2) mutuality of interest; 3) administration of operations, 4) existence of group norms; 5) durability of the collaboration, and 5) organizational governance. The dimensions of governance, administration, mutuality, and norms were expressed in positive statements about relationships within collaborative partnerships. The Likert scores increase so respondents giving higher scores indication by their response that they perceive the dimension exists to a greater degree. Only the autonomy dimension was expressed in negative terms; therefore, a lower score on autonomy should be interpreted as a positive aspect of the partnership.

The original instrument items developed by Thompson asked respondents to rank the statements based on their organizational viewpoint. For this study, respondents were asked to comment from the perspective of the partners. For example a statement such as: My organization always receives fair treatment was changed to Partners in the collaboration always receive fair treatment. In this way, individual organizations would not have to feel uncomfortable about making comments concerning their own relationships within the collaborative, but would make general statements on how the collaborative operates.

RESULTS

Return Rates for 2009 and 2011: Expert Interviews and Surveys

All return rates were lower in 2011 than in 2009, including interviews with the experts. It is important to note that the interview method changed from face to face in 2009 to telephone in 2011, but there is no information on how this change may have impacted the leadership's responses in 2011. In 2011, only half of the four (4) leaders participated in the interviews. One of the 2009 leaders had diminished their involvement with the Fulton County SPARC over time. In fact, at the 2009 interview this interviewee noted that they had been assigned to more non-SPARC duties than SPARC duties although they were still viewed as part of the 'leadership' team due to their position in the lead organization. A second leader that participated in the 2009 interview opted to complete a survey instead of an interview in 2011.

Fourteen (14) Levels of Collaboration and thirteen (13) Measures of Collaboration surveys were distributed in 2009 and 2011. However, as indicated in Table 2 below survey return rates were lower for both surveys in 2011. This may be explained by the timing of the 2011 administration. In 2009, the surveys were distributed during the Spring, when there were less personal distractions such as summer vacations. Due to circumstances, the 2011 surveys were distributed during the summer when fewer respondents may have been available over a short period of time to participate in the study.

Other factors may have influenced return rates. Although thirteen organizations were identified as partners in both 2009 and 2011, fifteen organizations that were identified as members of the collaboration in 2009 were not re-identified in 2011; they were replaced by fifteen new partners in 2011. It is possible that since many of the partners were new, they did not feel they had enough information to participate in the 2011 study. (2)

Results of the In-Depth Interviews

The Fulton County SPARC is part of a larger effort operating in other parts of the United States. Overall, the mission of SPARC is to provide preventive health clinics to senior citizens, especially those that cannot afford the services of private providers. An early objective in Fulton County, and one of the measures used to determine its success in the past, was to increase the number seniors that participated in the clinics. Since 2009, Fulton County has concentrated more on providing better quality service at the clinics. Instead of just providing information, more testing is done and there is a better attempt at follow-up.

In 2009 the Fulton County leadership reported that SPARC did not have MOUs, organizational charts, or other formal agreements among the local partners. However, there has always been an agreement with the parent organization, the original SPARC, to conduct the clinics and post-service surveys in a prescribed manner. Prior to 2011 the closest Fulton County SPARC came to formal agreements were planning meeting minutes, which recorded the services or resources partners agreed to provide. By 2011, when asked about the idea of formal agreements, leadership was less willing to assume there were no formal guidelines. However, Fulton SPARC partners were still verbally agreeing to their level of participation at the clinics. Today, amidst the verbal agreements, there is a formal agreement to allow some of their partner organizations to operate 'mini-SPARCS, but only during the time period when the county led SPARC is not operating its own quarterly clinics. These mini-SPARCS follow the same protocol as the county-led SPARC, but the clinics are managed by one of the SPARC partners. The SPARC partners leading the mini-SPARC can establish their own partnerships and collaborate with new organizations to deliver the service. There is now a guidebook on how to set up a SPARC. Although during the interviews, there was no mention of sanctions, partners that establish 'mini-SPARCs" are expected to follow the SPARC protocol. With the advent of mini-SPARCs, partners can now conduct their own fundraising if they choose. Even with this change, resource exchange rarely includes monetary support and is usually limited to personnel, and testing supplies, bandages, syringes, etc. In previous years, formalized operations existed only in terms of the overall clinic model presented by the developers of SPARC.

The perspective of Fulton's SPARC leadership shifted somewhat between 2009 and 2011 as indicated in the summary table below (see Table 3). Short-term planning still takes place by group consensus and is usually limited to the clinic events. Partners continue to self-identify their roles at the clinic based on what they are able to supply for each event. There continues to be a concern that partners are less willing to make long-term, formal commitments beyond setting up a mini-SPARC. Communication still takes place primarily through the SPARC event planning meetings and the timely distribution of the minutes from those meetings. There is still little interaction among the partners outside of the SPARC meetings and events.

Results from the Surveys

Through a random selection process, fourteen (14) of the current partners received the Levels survey and thirteen (13) received the Measures survey. Six of the organizations that received Levels in 2011 also received a copy of Levels in 2009. No organization that received a copy of the Measures survey in 2009 received a copy in 2011. Therefore, the Measures survey was new to most organizations. It has not been determined how the familiarity with the instruments may have impacted the return rates for the current study as there is no way to determine which of the 2009 respondents also responded in 2011.

Partner Perspectives from the Levels of Collaboration Scale

The Levels of Collaboration Scale examines the types of interactions among partners within a collaborative arrangement along a continuum of No Interaction to a 'true Collaboration. As modified for this study, a description is provided for each type of interaction i.e., share information, share clients, etc. Respondents are provided a list of all organizations within the collaborative and descriptions of the interactions and asked to identify their interaction with that organization. If there is no interaction with an organization because of lack of knowledge of the organization, respondents are asked to place a U in the No Interaction column.

In 2009, thirty-three organizations were identified as Fulton County SPARC partners. By 2011, this number had decreased to 27 partners. The table below represents the total percent of time partners were identified as having a particular interaction. There were fewer surveys returned in 2011 than 2009, therefore caution should be taken when drawing definitive conclusions. There are however, a few interesting results.

Frey advises that, "Data collected with the Level of Collaboration Scale can be reported in a variety of ways depending on the interests of stakeholders. Collaboration can be reported as the mean level of perceived collaboration across all respondents for all partners, summarized in other meaningful ways." For the purposes of this study we have provided mean scores across all partners and percentages of aggregate scores for all stages.

In both 2009 and 2011, over 50% of the responses indicated there is some interaction among the SPARC members, although some organizations interact with only a limited number of partners. The fact that the "no interaction" percentages remain consistent is surprising in light of the differences in response rates over the administration time period. Participants in the expert interviews did confirm that the partners often only interacted within the confines of SPARC activities. This appears to be true as well for the levels of cooperation among the Fulton County partners. Several Divisions from the Fulton County Health Department partner with the Aging Division of the Fulton Human Services Department to carry out SPARC activities. Beyond that, there is little interaction aside from Vax and Vote vaccination campaigns that take place during government elections.

The mean collaboration score for the Fulton SPARC in 2009 was 1.66, and .55 in 2011, suggesting that in 2009, there may have been a higher level of interaction. Some of that may have been due to the length of time organizations had been members of SPARC.

Survey results from 2009 indicated most interactions among partners identified the Fulton County SPARC as a Network, a group of organizations that had knowledge of each other but loosely defined roles and limited communication. By 2011, and maybe due to the development of the mini-SPARCS and the mini-SPARC guidebook, twenty-eight percent (28%) of the respondents felt SPARC operated more like a Cooperative; where partners share information and resources, have defined roles, communicate frequently and have some shared decision-making. In 2009, only 5 percent of the interactions were described as a Cooperative. There is not enough information at this point to determine if the more active members responded in 2011. However, the respondents to the Levels survey in 2011 did note that their affiliation with SPARC ranged from 4-6 years; half of the respondents had been members of the collaboration for 6 years and may have had more information about the internal operations of the collaboration, and felt more comfortable responding. (3)

Partner Perceptions from Measures of Collaboration Survey

The Measures of Collaboration Survey gives a diagnosis the health of a collaborative and measures the degree to which 'healthy' habits exist among partners in five key areas: governance, administration, norms, autonomy, and mutuality. As noted in the literature section, these areas have been identified by many scholars as necessary for a successful collaboration.

As noted in the table below, in 2009 the Fulton County SPARC scored high on governance, administrative functions, mutuality among partners, and organizational norms in both administrations of the Measures survey. The mean scores were lower in 2011; but SPARC still scored high on governance and their administrative processes; but only average on mutuality and norms. Organizational autonomy went from 2.26 in 2009 to .84 in 2011, suggesting organizations felt a lot of independence.

All of the respondents to the 2009 Measures survey indicated that they had only joined the partnership within the last three years. In 2011, four of the respondents had been involved with SPARC for 2 years or more, only one partner reported being a partner for less than one year. This may have impacted the scores. The scores may also have been influenced by the fact that the Fulton County SPARC had instituted the mini-SPARCs option, giving partners nearly complete autonomy to operate their own SPARCS. However, with this autonomy has come a loss of local level administrative control. There may also be less certainty among the partners that shared organizational norms are still present. With the addition of new partners at the table, it may not yet be clear that all partners are still working toward similar goals.

DISCUSSION

We have explored the interactions among the Fulton County SPARC partners on two levels, first applying stage theory to discuss the interactions among partners along a collaboration continuum, and secondly, examining organizational characteristics of the partnership that can help determine the strength of the partnership. At the conclusion of the 2009 report, SPARC was a young partnership, still early in its development. This may explain why at that time the Levels of Collaboration Survey and Expert Interviews supported the identification of SPARC as a Network. The Levels of Collaboration Scale defines partnership interactions as a Network when partners are aware of other member organizations although they do not interact much. The fact that over seventy percent (70%) of the interaction responses in 2009 indicated that the SPARC partners had knowledge of each other supported this contention. In addition, at that time partners scored SPARC as providing independence for member organizations suggesting that organizations acted independently in their relations with other partners. The low scores on the 2009 mutuality measure of sharing information suggested that the partners did not share information outside of preparations for SPARC events, which could explain the lower levels of interaction and communication.

In this most recent study (2011), more respondents to the Levels Scale defined their interactions with other partners as a Cooperative (28%). According to the Levels Scale in 2011, roles within the Fulton County SPARC are somewhat defined and there is some shared decision-making; but there is still little communication among the partners. Evidence of this lies in the fact that the percent of respondents indicating there is no interaction among the partners remained stable from 2009 to 2011 at 45%. However, the Fulton County SPARC developed a Resource Toolkit that provides guidelines for partner organizations to follow as they develop their own 'mini' SPARC clinics; providing some support for the contention of some defined roles. In addition, all organizations involved in SPARC clinics, including the operators of the SPARC sites now sit at the planning table, increasing shared decision-making.

In 2011, perceptions of both the leadership and partners had shifted somewhat. The leadership has begun to embrace, ever so gently, the concept of formalizing some of their activities. This seems to coincide with the creation of the mini-SPARCs. With this new development, the Fulton County Department of Aging moved from being a project manager to a new role of technical advisor. Prior to this, the Fulton County Division of Aging was considered the project manager for all of the local SPARCs.

According to the Measures of Collaboration Survey the strength of the organizational characteristics of partnerships determine a certain level of operational success. These characteristics include norms, governance and administration, autonomy and mutuality. Scores for the Fulton County SPARC differed on most of these measures from 2009 to 2011, due in part to changes that had taken place within the partnership.

Norms

Norms refer to acceptable behavior among partners, trust, reliability and reciprocity. The Fulton County SPARC scored high on the Measures dimension of Partnership Norms in 2009 (6.73 out of 7.00). However, norms assume a level of trust that may only come from long-term interaction, something that most of the respondents in 2011 indicated they did not have within their current partnerships, as evidenced by the 3.06 score in 2011.

Governance and Administration

Despite the fact that the Fulton SPARC did not have formal lines of authority or written rules, respondents in 2009 gave high scores for governance and administration. Governance is defined as the mechanism(s) through which the group makes joint decisions. Administration is closely related to governance, but differs in that administration focuses more on implementation and management. In 2009, the scores for administration and governance were nearly equal 6.30 and 6.20 respectively; and at that time, it was concluded that the Fulton SPARC possessed some characteristics associated with a Cooperative, namely defining roles in a limited way through verbal agreements at meetings. By 2011, SPARC's decentralized administrative structure had morphed whereby members of the Fulton County Human Services Department were still seen as the principal coordinators of the program, but others, that lead mini-SPARCs could now become project managers for their own efforts. Fulton County was still viewed as a governing body, but now they provided technical assistance and were not required to manage the details of every SPARC event. Where the principals from Fulton County had taken on the responsibility for management and implementation of SPARC clinics, that role is now shared with whatever organization decides to sponsor a mini-SPARC. This may explain why the scores for Goverance and Administration decline respectively to 5.34 and 4.65 in 2011.

Autonomy

Organizational autonomy refers to the ability of organizations to maintain their distinct identities and organizational authority separate from the collective identity. Partners scored SPARC as providing independence for member organizations suggesting that organizations act independently in their relations with other partners. The Fulton SPARC scored low on having barriers to organizational autonomy in both 2009 (2.26) and 2011 (.84), possibly influenced by the creation of Mini-SPARCS.

Mutuality

The final dimension, "mutuality," is the existence of mutual benefits, and interdependence. The decrease in these scores from 6.03 to 3.68, could possibly be explained by the changes that have taken place in the partnership since 2009. In 2009, all SPARC partners provided health services to seniors as part of their regular operations, and had the health of citizens, including seniors as part of their organizational missions. The lower scores on the 2009 mutuality measure of sharing information suggested that the partners did not share information outside of preparations for SPARC events, which could explain the lower levels of interaction and communication. By 2011, organizations that just provided space for the clinics were invited to the decision-table and served as part of the partnership. The site partners may not be involved in the daily business of delivering health services to seniors, or even that involved in the overall goals of SPARC aside from providing space for the clinics, causing some disconnects in partners working toward the same goals. Moreover, understanding levels of mutuality also comes from knowledge of other partners, something that a large proportion of the members do not have because they are new to the organization. It is possible that with limited inter-communication partners are not able to perceive their joint interests in broader terms.

CONCLUSION AND FUTURE RESEARCH

Stage theory can often be used to explain the development of collaborative efforts. However, in all stage theory activity is dynamic and there are no linear paths. Norris-Tirrell and Clay (2011) refer to the final stages of a collaboration as having two options: dissolving the collaboration or choosing renewal. The Fulton County SPARC has chosen renewal. It was clear in 2009 that there was no interest in formalizing agreements among the Fulton SPARC partners, but the realities of personnel and organizational changes indicated that if SPARC was to survive and experience growth, change was inevitable. Instead of dissolving, the Fulton County SPARC took on a new role that was conducive to the personnel changes that were taking place within the parent department. The Fulton SPARC has become more like the original SPARC created by Dr. Douglas Shenson. Fulton County shares more of its governance and administration with partners, allowing them to set up their own SPARC clinics within certain guidelines. In this way, SPARC expands, gaining more partners and reaching more of the county. At the same time, the SPARC's informal leadership is released from the day-to-day management of the clinics to provide technical support. While the overall mission of the Fulton County SPARC has not changed, the strategies used to achieve the goals of the collaborative have changed.

This study has explored only a limited aspect of the Fulton County SPARC, and has attempted to answer the question, why has the Fulton County SPARC survived its local counterparts? It appears that the answer lies in part in its willingness to change. Although clearly not interested in formal agreements, the Fulton County SPARC Resource Toolkit clearly outlines the duties of partners from the national, regional and local levels. Although nothing is signed formally, agreeing to do a SPARC is an open commitment to follow the guidelines. Despite early resistance, the partnership is beginning to formalize, still moving from a Network to a Cooperative. Future follow-ups may provide information concerning the partnership's continued ride along the continuum.

Additional research is needed in order to determine the 'true' stage of the Fulton County SPARC partnership. The low return rates in 2011 may have given a misleading impression of the stage of the collaborative. Future studies should include at least 50% of the partners. However, there is no guarantee that an organization's newly formed relationship with the collaborative will make the partners any more willing to participate in this type of study. As long as new members continue to join the collaborative, and the current assessment methodology is used to build longitudinal data, researchers may have to contend with low return rates.

The structure of the Fulton County SPARC and its intra-and inter-organizational relationships beckons for additional study. This work did not explore the degree to which organizations interacted with certain organizations. It is possible that while overall partner interaction is low, interaction may be extremely high with a selected few partners. Does the length of time an organization has been a member of the collaborative determine the level of interactions among partners? Moreover, does the presence of other Fulton County departments increase or decrease the chance of interactions among all SPARC partners? In other words, is there a higher level of interaction among all of the Fulton County partners, than with the non-county organizations?

Other methodological considerations should be taken into account in future studies. Frey (2006) discusses how individuals completing the Levels survey may actually be responding from a personal rather than an organizational viewpoint. To offset this tendency, partners should also receive personal interviews similar to the expert interviews for the leadership. The open-ended questions could then be compared to those from the survey to get an overall view of partner perceptions.

A comparative analysis of the Fulton SPARC with its regional counterparts may now be in order. Case studies of the attributes of these types of collaborations and partnerships may uncover the best ways to maintain local level health partnerships for senior citizens.

REFERENCES

Alexander, J., Comfort, M., Weiner, B., & Bouge, R. (2001). Leadership in collaborative health partnerships. Nonprofit Management and Leadership, 12 (2), 159-175.

Brodeur, P. (2006). SPARC Sickness Prevention Achieved Through Regional Collaboration, To Improve Health and Health Care IX, ed. Stewart, I. and Knickman, J. Robert Wood Johnson Anthology, Hoboken, NJ: Jossey-Bass.

Brown, D., White, J., Leibbrendt, L. (2006). Collaborative partnerships for nursing faculties and health service providers- what can nursing learn from business literature. Journal of Nursing Management, 14 (3), 170-179.

Frey, B; Lohmeier, J., Lee, S., & Tlooefson, N. (2006) Measuring collaboration among grant partners. American Journal of Evaluation, 7 (3), 383-392.

Hicks, D., Larson, C. Nelson, C., Olds, D., & Johston, E. (2008). Influence of collaboration on program outcomes: the colorado nurse-family partnership. Evaluation Review, 32 (5), 453-477.

Kernaghan, K. (1993). Partnership and public administration: Conceptual and practical considerations, Canadian Public Administration, 36 (1), 57-76.

Parker-Oliver, D., Bronstein, L. and Kurzejeski, L. (2005). Examining variables related to successful collaboration on the hospice team. Health and Social Work, 30, 4. Retrieved from http://www.thefreelibrary.com/Examining+variable s+related+to+successful+collaboration+on+the...-a0139521226

Parkinson, C., (2006) Building Successful Collaborations: A Guide to Collaboration Among Nonprofit Agencies and Business; Cambridge and North Dumfries Community Foundation, SeptemberRetrieved from: http://www.cfcfcc.ca/linkdocs/collaborationReport.pdf

Norris-Tirrell, D. and Clay, J. (2011). Strategic collaboration in public and nonprofit administration: A practice-based approach to solving shared problems. Boca Raton, FL: CRC Press.

Shenson, D.; Benson, W., & Harris, A. (2008). Expanding the Delivery of Clinical Preventive Services Through Community Collaboration: The SPARC Model; Preventing Chronic Disease: Public Health Research, Practice and Policy. Retrieved from: www.cdc.gov/pcd/issues/2008/jan/07 0139.htm

Thompson, A. M.; Perry, J., & Miller, T. (2009). Conceptualizing and measuring collaboration. Journal of Public Administration Research and Theory, 19 (1), 23-56.

This Symposium is dedicated to the professional work of and devotion to public service ofBrenda Sullivan, Ph.D.

BRENDA A. SULLIVAN

Fulton County Department of Public Health

Brenda Sullivan, Ph.D. most recently served as the Gender Equality Coordinator for Fulton County Government, a position she held since May 2011. Prior to that, she had served as a researcher with the Fulton County Human Services Department. Dr. Sullivan earned her doctorate at Clark Atlanta University in 1988, researching Black women at the 1977 Women's Convention as well as racial disparities in metropolitan transportation. Dr. Sullivan brought 20 years of public management experience as a practitioner and educator. She considered herself a "pracademic"--applying the principles of public administration that she studied and taught at a number of universities in addition to conducting research for and collaboration with a number of organizations. Dr. Sullivan was an active member of the National Conference of Black Political Scientists (NCOBPS) and the National Conference of Minority Public Administrators (COMPA). She passed away in January 2012.

(1) Although an additional staffer from these organizations was encouraged to complete surveys, the individuals that participated in the face-to-face interviews were asked not to participate in the surveys.

(2) In fact, several organizations new to the collaboration stated that they did not know how to respond to the surveys or were only involved for a short period and did not feel prepared to make evaluative statements.

(3) The length of time an organization was a member of SPARC was not asked in 2009.
Table 1

Kernaghan Model

Partnership           Characteristics

Collaborative         pool resources; each partner exercises
Partnerships          power in decision-making; bring equal
(Real Partnerships)   resources; sense of power sharing;
                      surrender autonomy

Operational           work-sharing rather than decisionmaking
Partnerships          power; share resources; power by one
                      partner

Contributory          does not require active participation of
Partnerships          all partners in decision-making;
                      organization provides funding with
                      little operational; involvement

Consultative          advisory committee or council
Partnerships

Phoney Partnerships   established for the co-optation of
                      stakeholders

Table 2

Participation and Overall Return Rates

                 Distribution  Return Rate

Name of Survey   2009   2011   2009      2011

Levels of        14     14     9 (64%)   5 (35%)
Collaboration

Measures of      13     13     5 (38%)   4 (30%)
Collaboration

Table 3

Leadership Perceptions of SPARC Collaboration

Areas               2009              2011

Partner Roles       Loosely defined   Some definition for
                                      Mini-SPARCS

Formal Agreements   Non-existent      Guidebook on how to
                                      conduct a SPARC
                                      clinic

Communication       Through SPARC     Through SPARC
                    leadership;       leadership; limited
                    limited to        to planning
                    planning

Decision-making     Independent       Independent about
                    about             participation;
                    participation     limited decision-
                                      making for
                                      Mini-SPARCS

Table 4

Fulton County SPARC Results From the Levels of
Collaboration Scale

        Type of           2009     2009   2011     2011
Stage   Interaction       Result   Rank   Result   Rank

0       No Interaction    45%      1      45%      1

1       Network           21%      2      14%      3

2       Cooperative       5%       6      28%      2

3       Coordination      8%       4      4%       5

4       Coalition         7%       5      0%       6

5       Collaboration     11%      3      6%       4

        Unknown partner   4%              4%

Table 5

Results from SPARC Measures of Collaboration Scale

                    Administrative   Barriers to
Year   Governance   Process          Organizational
                                     Autonomy

2009   6.20         6.30             2.26

2011   5.3          4.65             .84

       Mutuality   Norms
       Among       Within
Year   Partners    Partnership

2009   6.08        6.73

2011   3.68        3.06

The results are based on a scale of 1-7 where 1 (low) represents
Not at All and 7(high) represents To A Great Extent. A score of
four (4) is considered a mid-point
Gale Copyright: Copyright 2012 Gale, Cengage Learning. All rights reserved.