Elsevier

Social Science & Medicine

Volume 150, February 2016, Pages 85-94
Social Science & Medicine

Analysing collaboration among HIV agencies through combining network theory and relational coordination

https://doi.org/10.1016/j.socscimed.2015.12.006Get rights and content

Highlights

  • Unites network analysis and relational coordination to study interagency collaboration.

  • Innovative method with theoretical and methodological potential.

  • Gives theoretically-informed insights into reasons that underlie interagency ties.

  • Provides insights for policymakers to improve interagency collaboration.

Abstract

Agencies with different foci (e.g. nutrition, social, medical, housing) serve people living with HIV (PLHIV). Serving needs of PLHIV comprehensively requires a high degree of coordination among agencies which often benefits from more frequent communication. We combined Social Network theory and Relational Coordination theory to study coordination among HIV agencies in Baltimore. Social Network theory implies that actors (e.g., HIV agencies) establish linkages amongst themselves in order to access resources (e.g., information). Relational Coordination theory suggests that high quality coordination among agencies or teams relies on the seven dimensions of frequency, timeliness and accuracy of communication, problem-solving communication, knowledge of agencies' work, mutual respect and shared goals. We collected data on frequency of contact from 57 agencies using a roster method. Response options were ordinal ranging from ‘not at all’ to ‘daily’. We analyzed data using social network measures. Next, we selected agencies with which at least one-third of the sample reported monthly or more frequent interaction. This yielded 11 agencies whom we surveyed on seven relational coordination dimensions with questions scored on a Likert scale of 1–5. Network density, defined as the proportion of existing connections to all possible connections, was 20% when considering monthly or higher interaction. Relational coordination scores from individual agencies to others ranged between 1.17 and 5.00 (maximum possible score 5). The average scores for different dimensions across all agencies ranged between 3.30 and 4.00. Shared goals (4.00) and mutual respect (3.91) scores were highest, while scores such as knowledge of each other's work and problem-solving communication were relatively lower.

Combining theoretically driven analyses in this manner offers an innovative way to provide a comprehensive picture of inter-agency coordination and the quality of exchange that underlies collaborative ties. These methods together can identify areas that could be targeted to promote closer ties.

Introduction

The needs of people living with HIV (PLHIV) are heterogeneous and span both medical and social needs (Abramowitz and Obten, 2000) such as medical care, psychosocial counseling and legal aid, among others. HIV services in the US are delivered through a mix of agencies such as local health department units, state government agencies, not-for-profit agencies, private for-profit agencies, medical services agencies, government agencies and faith-based agencies, among others. Each agency typically offers only a few specialized services (e.g. nutritional, counseling and assistance for PLHIV) and may serve only a few target groups (e.g. men who have sex with men). Agencies therefore need to collaborate in order to serve the multiple needs of PLHIV that cannot be served by agencies working alone.

Problems in coordination, fragmentation of services and people “slipping through the cracks” may occur because of the wide diversity among agencies that serve PLHIV. Problems may arise due to lack of awareness about other organizations, rigid organizational boundaries (Foster-Fishman et al., 2001), fragmentation of funding, duplication of services, differing ideologies and competition for scarce resources (Mor et al., 1994). Poor coordination among service agencies may exacerbate existing problems in service delivery at the client level that arise due to lack of information, perceived stigma associated with HIV, lack of accessibility and lack of insurance coverage.

The nature of the HIV epidemic is different now than in the early years of the epidemic. Groups such as women and African-American men that have sex with men (African-American MSMs) are now among the more severely HIV-affected groups. Newer needs such as employment for PLHIV and managing the HIV infection over a lifetime are now emerging as PLHIV are living longer due to the availability of antiretroviral drugs. These changes require coordination among HIV agencies be examined in order to improve the system of care. We seek to contribute to the limited literature on coordination among HIV agencies by combining the two theories of social network analysis and relational coordination.

Researchers have advanced many theories to explain inter-organizational collaboration. Organizational theory postulates that organizations are motivated to establish linkages with other organizations in order to be more effective at accomplishing their tasks (Parmigiani and Rivera-Santos, 2011). Such linkages yield several advantages for organizations, such as reputation enhancement, access to a more diverse resource base and the acquisition of more powerful allies, (Parmigiani and Rivera-Santos, 2011). The theoretical traditions in this perspective include resource dependency theory, stakeholder theory, institutional theory, and social networks theory (Parmigiani and Rivera-Santos, 2011). Resource dependency theory explains that organizations establish linkages with other organizations to increase their control over scarce resources by establishing strategic partnerships (Pfeffer and Salancik, 1978).

Stakeholder theory (Freeman, 1984) and institutional theories (DiMaggio and Powell, 1983, Scott, 1987) place more emphasis on the motivation to gain legitimacy with stakeholders and to conform to professional norms respectively by forming linkages with other organizations. Social network theory emphasizes gains in information and knowledge as motivation for creating a new tie or strengthening a current one. In sum, organizational theory states that organizations may be motivated to establish linkages with other organizations in order to reduce dependence on the environment, to gain legitimacy, status, and power and/or to increase access to information and knowledge (Parmigiani and Rivera-Santos, 2011). Even organizations that sometimes compete with each other may cooperate at other times, or for other purposes, sometimes called co-opetition (Chalhoub, 2007, LeTourneau, 2004).

The social network theoretical tradition, on which we focus in this paper, emphasizes that the main motivation for establishing linkages with other organizations is to expand access to information and knowledge (Kadushin, 2012). This perspective emphasizes the embeddedness of organizations in a larger social structure among other organizations (e.g., an organizational field). Stronger, long-term relationships are likely to result in greater trust and collaboration among organizations. Networks are frequently characterized by such descriptive measures as density and centralization for the network as a whole and centrality for the positions of individual actors within the networks. Table 1 provides definitions of network measures used in this paper. Social network theory predicts that central actors have greater influence than peripheral actors because of their positional access to information and the power created by having the choice of whether or not to share information.

Organizational theory also features a body of literature concerned with coordination among parts of an organization, or across organizations, that represent different parts of a process (such as providing a full range of services to PLHIV) and are therefore interdependent. Coordination requires information sharing via methods that have sufficient bandwidth to meet the information processing demands of the work (Argote, 1982, Daft et al., 1987, Galbraith, 1973, Thompson, 1967). Relational theories of coordination suggest that communication and relationship ties provide the means by which to coordinate work (Crowston and Kammerer, 1998, Faraj and Xiao, 2006, Gittell, 2002). Organizations may better achieve their goals through a high degree of coordination with partnering organizations. Stronger relationships should yield better collaborative outputs.

A measure of relational coordination (RC) can be used to better understand the strength of relationships among collaborating organizations. An RC instrument developed by (Gittell, 2009) measures seven dimensions of coordination (frequency, timeliness, and accuracy of communication, problem-solving, knowledge of others' work, mutual respect and shared goals) on a scale of 1–5, with 5 being highest. Fig. 1 depicts the seven dimensions of relational coordination. Gittell has made some changes to the instrument and provides data collection services using the new version (Gittell, 2011). Table 2 presents the relational coordination questions used in this study.

Our study contributes to the literature on interorganizational collaboration by creatively combining two tools to capture the breadth and depth of collaboration among HIV agencies. We use network analysis to capture the web of relationships that exist among HIV agencies, and relational coordination to provide a meaningful scale to compare the quality of relationship ties among HIV agencies.

Despite calls by the local health department to improve interagency collaboration among HIV agencies in Baltimore, there has been little information on the actual patterns of collaboration at the city level that might aid the health department in fostering greater interaction. We collected data in Baltimore, Maryland. Both the city and the state are severely affected by HIV. The state of Maryland had an estimated 30,558 PLHIV at the end of 2010 (IDEHA, 2013). This was the ninth-highest number of cases for any US state or territory (IDEHA, 2013). The population of Baltimore City in the 2000 US census was 620,961 (US Census, 2010). The Baltimore–Towson Metropolitan area had 18,318 people living with HIV, making this the third-highest rate for any region in the US (IDEHA, 2013). Despite calls by the local health department to improve interagency collaboration among HIV agencies in Baltimore, there has been little information on the actual patterns of collaboration at the city level.

A previous study reported results of a network analysis of relations among thirty organizations that received Ryan White Care Act funding in 1995–96 or were otherwise engaged in HIV care or treatment (Kwait et al., 2001). The authors studied five types of ties as bases for collaboration: client referrals to other organizations, client referrals from other organizations, written linkage agreements, exchange of information about shared clients and joint programs (Kwait et al., 2001). Most organizations were linked either directly or indirectly (Kwait et al., 2001). Providers tended to work directly with others as client needs arose rather than negotiate through “clearinghouse” types of organizations (Kwait et al., 2001). The “central” organizations “tend to be those … created specifically for HIV related services or that specialize in HIV/AIDS care” (Kwait et al., 2001, p. 484). We think that these central agencies handle more PLHIV than other agencies and thus have more connections. The authors recommended designing interventions to promote collaboration that are feasible for agencies given their individual contexts and resources.

Our analysis provides updated network analysis results from a larger sample and enhances the findings by including results from relational coordination. Network analysis tells us the pattern of the relationships but not the quality of the ties or what aspects of the ties might be strong or weak. Relational coordination in turn allows us more in-depth information about the nature of coordination/collaboration among these agencies. The methodological innovation demonstrated in this study can provide ideas for other organization behavior researchers to similarly combine these methods in their studies. Furthermore, a third party (e.g., government) might seek to increase collaboration, and understanding the strong or weak points of interagency coordination could provide important insight as to how to bring that about.

Section snippets

Materials and methods

We collected and analyzed data in two phases.

Phase I

We received 51 responses to Phase I data collection (large network analysis of interagency ties). The density was 39% when considering all interactions and 20% when considering interactions of only monthly or higher frequency. We focused subsequent analyses on interactions of monthly or higher frequency. Not-for-profits and university-affiliated agencies had higher centrality than government agencies with no subgroups emerging. Degree centralization was 28% and betweenness centralization 6.72%.

Discussion

We discuss below the implications of the Phase I network analysis, followed by those of the Phase II relational coordination analysis. The network analysis revealed a well-connected network (except for one isolate which was a small agency that closed) comprising agencies that offer a wide range of HIV-related services. The pattern of connections allows resources to potentially flow from one agency to another through the web of interagency linkages, unlike networks that contain several isolated

Conclusion

The limitations in capturing the details of collaborative ties among agencies can be reduced by combining methods that we demonstrated in this study. Health care especially for chronic conditions such as HIV infection requires collaboration among different agencies that offer different services. Social network analysis can provide an important context to HIV research and practice by showing the nature and pattern of relationships among agencies. Relational coordination analysis can pinpoint

Acknowledgment

The research was supported through the Distinguished Doctoral Research award received by Nidhi Khosla, from the Department of Health, Behavior and Society of the Johns Hopkins Bloomberg School of Public Health. No other funding was received for this research. We thank the respondents for sharing their time and insights with us. We would like to thank Dr. Jane Banaszak-Holl and three anonymous reviewers for their helpful comments and suggestions for improving the manuscript.

References (31)

  • S. Abramowitz et al.

    Ryan White CARE act title IV: building networks to improve healthcare delivery to the HIV infected

    AIDS Public Policy J.

    (2000)
  • L. Argote

    Input uncertainty and organizational coordination in hospital emergency units

    Adm. Sci. Q.

    (1982)
  • J. Banaszak-Holl et al.

    Organizational characteristics associated with agency position in community care networks

    J. Health Soc. Behav.

    (1998)
  • C.T. Butts

    sna: Tools for Social Network Analysis

    (2010)
  • M.S. Chalhoub

    A framework in strategy and competition using alliances: application to the automotive industry

    Int. J. Organ. Theory & Behav.

    (2007)
  • K. Crowston et al.

    Coordination and collective mind in software requirements development

    IBM Syst. J.

    (1998)
  • R.L. Daft et al.

    Message equivocality, media selection, and manager performance: implications for information systems

    MIS Q.

    (1987)
  • W. de Nooy et al.
    (2011)
  • P.J. DiMaggio et al.

    The iron cage revisited: institutional isomorphism and collective rationality in organizational fields

    Am. Sociol. Rev.

    (1983)
  • S. Faraj et al.

    Coordination in fast response organizations

    Manag. Sci.

    (2006)
  • P.G. Foster-Fishman et al.

    Facilitating interorganizational collaboration: the contributions of interorganizational alliances

    Am. J. Community Psychol.

    (2001)
  • E.R. Freeman

    Strategic Management: A Stakeholder Approach

    (1984)
  • J.R. Galbraith

    Organization design: an information processing view

  • J.H. Gittell

    Coordinating mechanisms in care provider groups: relational coordination as a mediator and input uncertainty as a moderator of performance effects

    Manag. Sci.

    (2002)
  • J.H. Gittell

    Relational Coordination: Guidelines for Theory, Measurement and Analysis

    (2009)
  • Cited by (0)

    View full text