Secondary Logo

Sustaining multistakeholder alliances

D’Aunno, Thomas; Hearld, Larry; Alexander, Jeffrey A.

doi: 10.1097/HMR.0000000000000175
Features
Free
SDC

Background: Multistakeholder alliances that bring together diverse organizations to work on community-level health issues are playing an increasingly prominent role in the U.S. health care system. Yet, these alliances by their nature are fragile. In particular, low barriers to exit make alliances particularly vulnerable to disruption if key stakeholders leave. What factors are linked to the sustainability of alliances? One way to approach this question is to examine the perceptions of alliance participants, whose on-going involvement in alliances likely will matter much to their sustainability.

Purpose: This study addresses the question: “Under what conditions do participants in alliances consider that their alliances are well positioned for the future, will perform well over time, and will be able to deal effectively with future challenges?”

Methods: We draw on cross-sectional survey data collected in the summer of 2015 from a total number of 638 participants in 15 alliances that participated in the Robert Wood Johnson Foundation’s Aligning Forces for Quality program.

Results: Results from regression analyses indicate that alliance participants are more likely to view their alliances as sustainable when they (a) share a common vision, goals, and strategies for the alliance and (b) perceive that the alliance has performed effectively in the past.

Practice Implications: Leaders of multistakeholder alliances may need to ensure that alliances are collective efforts that build success one step at a time: to the extent that participants believe they share a vision and strategies and have had some prior success working together, the more likely they are to view the alliance as sustainable.

Thomas D’Aunno, PhD, is Professor, New York University Wagner School of Public Service. E-mail: tda3@nyu.edu.

Larry Hearld, PhD, is Professor, University of Alabama at Birmingham.

Jeffrey A. Alexander, PhD, is Emeritus, University of Michigan, Ann Arbor.

This study was funded by the Robert Wood Johnson Foundation and was approved by the institutional review boards of the authors’ institutions.

The authors have disclosed that they have no significant relationship with, or financial interest in, any commercial companies pertaining to this article.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.hcmrjournal.com).

Multistakeholder alliances in the health care field involve voluntary efforts to bring together diverse organizations and individuals to work on health-related issues in a community (Hasnain-Wynia et al., 2003; Shortell et al., 2002). These alliances, which often include providers, employers, insurance companies, government agencies, and consumer groups, play an increasingly prominent role in the U.S. health care system in addressing community-level issues such as poor quality of care, inadequate measurement and reporting of quality information, and racial and ethnic disparities in care (Alper, 2014; Young, 2012). By leveraging the expertise and resources of their members, alliances can potentially promote improvements in community health more effectively than actors working independently or at cross-purposes (Hearld, Alexander, & Shi, 2015). For example, results from one recent study show that alliances in several U.S. communities were able to increase awareness of public reports on provider performance among racial and ethnic minority groups that typically know little about such reports (Yonek, Mittler, & Hasnain-Wynia, 2014).

Yet, despite their efforts to improve health and health care at the community level, these alliances by their nature are fragile, raising questions about their sustainability (Hearld et al., 2015). Sustainability is important, in large part, because prior research shows that alliances take time—typically years—to develop (Kale & Singh, 2009). Of course, sustainability is not an end in itself—high levels of performance also matter to stakeholders. But, without longevity, it is unlikely that an alliance can mature to make significant contributions to community health.

Among the common challenges to sustaining the activities of alliances is private financial support for activities and organizations that are considered public goods (Beich, Scanlon, & Boyce, 2010). Moreover, alliances face the challenge of maintaining commitment and investment from diverse partners that operate within dynamic and, frequently, competitive health care markets (Simpson, 2012). This challenge, coupled with low barriers to exit, makes alliances particularly vulnerable to disruption if key stakeholders leave.

Given these challenges, an important question is: What factors are linked to the sustainability of alliances? One way to approach this question is to examine the perceptions of alliance participants, whose on-going involvement in alliances likely will matter much to their sustainability. Participants in an alliance are, in effect, synonymous with the alliance given the central role they play in alliance activities and developing and implementing strategy. Indeed, the work of most alliances is carried out by a collection of several groups of volunteers; each group focuses on particular projects to attain alliance goals or to support alliance operations. For example, there may be a group of individuals working on a project to provide the community with useful report cards on the quality of care from local health care organizations, another group that promotes local public health initiatives (e.g., bicycle paths, park, and playground renovation), a third group that helps individuals sign up for insurance plans, and a fourth that manages the alliance’s budget and finances. For this reason and because participation in alliances is voluntary, it is difficult to imagine that an alliance can be sustained when its members are not optimistic about its positioning for the future (Alexander, Hearld, Wolf, & Vanderbrink, 2016).

Similarly, the views of alliance participants about its future can easily have self-reinforcing effects. These individuals have ties to other important organizations and groups in a community; thus, “bad news” about alliance prospects can easily spread to other stakeholders. Finally, individuals who participate in alliances are insiders to these organizations who have first-hand experience that informs their assessments of alliance sustainability. Thus, this study addresses the question: “Under what conditions do participants in alliances consider that their alliances are well positioned for the future, will perform well over time, and will be able to deal effectively with future challenges?”

Back to Top | Article Outline

Conceptual Approach and Hypotheses

Alliance sustainability. Following Hearld, Bleser, Alexander, and Wolf (2016), we define alliance sustainability as the ability of an alliance to engage in activity over an extended time period while remaining responsive to emergent issues and external changes. This article examines the central conditions that position an alliance for sustainability. We focus on positioning for sustainability because sustainability cannot be measured directly: By definition, it is a future state that cannot be predicted reliably (Alexander et al., 2016). Specifically, we argue that there are four central conditions that position an alliance for sustainability (Hearld et al., 2015; Wiltsey Stirman et al., 2012).

First is the extent to which an alliance has goals, strategies, programs, and leadership that position it well for the future. In other words, participants may perceive sustainability as more likely when they view the alliance as having made appropriate decisions about its aims and how to achieve them into the future (Gomez, Greenberg, & Feinberg, 2005; Greenberg, Feinberg, Gomez, & Osgood, 2005; Pluye, Potvin, Denis, & Pelletier, 2004).

A second condition for sustainability is the extent to which an alliance is well positioned with respect to its economic and political environment. This dimension means that alliance participants have a favorable view of the alliance’s ability to perform well in the external context that it is likely to face in the future.

Third is the extent to which participants believe that the alliance will handle adversity well in the future. Challenges to alliances can stem from several sources, including competitors and loss of financial and human resources (e.g., staff, leaders, board members) needed to perform well. Indeed, Hearld et al.’s (2015) recent review of literature on the sustainability of multisector alliances emphasizes the importance of maintaining adequate resources (Rog et al., 2004).

Finally, a fourth important condition for alliance sustainability is the extent to which alliance members believe that their alliance will perform well overall in the future. By performance, we refer not only to alliance reputation but also to its ability to make measurable improvements in community health and quality of care (e.g., Edwards et al., 2007; Polacsek et al., 2014).

Predictors of alliance sustainability. To begin, it is important to distinguish what we term “conditions for sustainability” from “predictors of sustainability.” A key distinction is temporal: the conditions for sustainability are future-oriented; they focus on participants’ views of how their alliances will fare in the next several years. In contrast, predictors of sustainability are past-oriented; they focus on participants’ views of what has occurred in their alliance since its inception, including their commitment of time and effort to alliance development.

We argue that three past-oriented factors can account for the extent to which participants in alliances are likely to perceive that they are well positioned for sustainability as indicated by the conditions discussed above. First, we argue for the importance of a path dependence perspective (Arthur, 1994; Vergne & Durand, 2010). In this view, history matters a great deal. If we want to understand where an organization is headed, we need to understand where it has been. Gulati’s (1995a, 1995b) research on alliances shows the importance of a history of prior ties among organizations for their continued partnerships.

In this case, participant views of alliance effectiveness to date also play an important role: To what extent do participants believe that their alliances have met goals and create value for their organizations? From the perspective of their participants, alliances that have been able to achieve objectives, such as making measurable improvements in health outcomes or securing financial or human resources, are more likely to be sustainable.

Thus, we hypothesize:

H1: Alliance participants are more likely to view their alliances as sustainable to the extent that they have positive views of past alliance effectiveness in key performance areas.

Second, we argue that participants are more likely to view an alliance as sustainable to the extent that they share a vision for the alliance and share strategies to attain their vision. In alliances characterized by shared views on such central issues, participants are more likely to have a sense of ownership that prompts them to participate in activities that promote sustainability. In other words, developing shared views on alliance vision, goals, and strategies likely means that individuals have been involved in relatively high levels of interaction that bind them to the alliance and build their commitment to its future success. In turn, prior research shows that such organizational commitment motivates individuals so that they are willing to go “above and beyond” in their performance. Thus, we hypothesize:

H2: Alliance participants are more likely to view their alliances as sustainable to the extent that they report agreement on alliance vision and strategies, as well as feel a sense of ownership for the alliance.

Finally, individuals’ past commitment to an alliance, that is, the duration of their membership and amount of time they have devoted to alliance activities, is likely to contribute positively to their views of alliance sustainability. Individuals who have committed time and effort to an alliance are likely to have done so, at least in part, based on their views that the alliance has the ability to succeed (Herscovitch & Meyer, 2002; Meyer, Srinivas, Lal, & Topolnytsky, 2007). That is, individuals highly value their time and likely would not participate in alliance activities without some expectation of alliance sustainability (Staw, 1981). Moreover, individuals with high commitment levels promote perceptions of alliance sustainability among other alliance members (and community members more generally) who have observed this commitment in action. Thus, we hypothesize:

H3: Alliance participants are more likely to view their alliances as sustainable the greater their past commitment to the alliance (duration of membership and percentage of time devoted to alliance activities).

Back to Top | Article Outline

Methods

Study Context

This study was part of a larger investigation of Aligning Forces for Quality (AF4Q), a multisite, 10-year initiative (2005–2015) funded by the Robert Wood Johnson Foundation (RWJF; Scanlon et al., 2016). The premise of AF4Q is that the greatest improvements in the quality of care for the chronically ill can be achieved when aligning the efforts of key actors, including health care providers (physicians, physician groups, nurses, and hospitals), health care purchasers (employers and insurers), and health care consumers (patients), through multistakeholder alliances. Table 1 shows important characteristics of the 15 participating alliances, including their organizational form (type and status as nonprofit), mission statement, location, and size of market.

Table 1

Table 1

We emphasize three important points about this empirical context that make it useful for testing the proposed hypotheses. First, these alliances had very similar founding conditions: They received funds to launch the AF4Q program at the same point in time (2006–2007). Second, the alliances faced uniform pressure to produce results by the time that funding from the RWJF ended (in mid-2015). In particular, these alliances operated under the same general vision of improving the quality of care in their respective communities and were expected to use the same general strategies to pursue that vision. These strategies focused on initiatives for public reporting, quality improvement, consumer engagement, and equity associated with race and ethnicity. The operational aspects of these strategies (e.g., how the initiatives were developed and implemented), however, were determined by the alliances and thus differed across the sites. As Table 1 shows, the alliances also used different organizational structures (e.g., partnership among local organizations, single organization coordinating efforts of other organizations) to implement these strategies. Finally and most important, the alliances faced the common challenge of sustainability when the RWJF funding terminated in 2015.

Sampling frame and survey data. This study draws on data from an Internet-based survey of alliance participants in 15 of the 16 AF4Q alliances administered at the end of the AF4Q program (one alliance was unable to participate because it was in the process of closing operations at the time of survey implementation). Hearld et al.’s (2015) systematic, critical review of the literature on health care alliance sustainability informed the survey content.

We developed the survey sampling frame from a comprehensive list of alliance participants provided by each alliance (i.e., staff and consultants, board and leadership team members, committee and work group members, advisory group members, and members-at-large). We conducted the survey from June 2015 to September 2015. A total number of 638 individuals (38.6%) completed the whole survey (range across alliances 21.8%–92.9%). In addition, 77 (4.7%) provided responses for a portion of the survey. Survey respondents included representatives from a broad range of stakeholder groups including insurers (9.9%), providers such as physicians and hospitals (26.1%), government agencies (11.2%), employers (5.1%), consumers (4.5%), alliance staff (17.5%), and unaffiliated participants (2.1%).

Dependent variables. We assessed perceptions of sustainability with survey items that correspond to the dimensions of alliance sustainability discussed above. Supplemental Digital Content 1 (see Supplemental Digital Content 1, http://links.lww.com/HCMR/A32) shows the variables and survey items that measure each variable. Respondents answered the items about alliance positioning on a 5-point scale, ranging from 1 (strongly disagree) to 5 (strongly agree). Respondents answered the items about challenges to sustainability and overall future performance on a 5-point scale, ranging from 1 = not at all likely to 5 = completely likely.

We began by conducting a principal component analysis of these survey items with a varimax rotation to assess the extent to which the items reflected the proposed dimensions of sustainability and, if so, to create indices for multivariable analyses. Using the Kaiser criterion of eigenvalues greater than 1 (Kaiser, 1958) and scree tests, the analysis indicated a four-factor solution. Supplemental Digital Content 2 (see Supplemental Digital Content 2, http://links.lww.com/HCMR/A33) shows the results of the principal component analysis.

On the basis of these results, we created four composite indices by averaging responses to items that loaded on the respective factors, which we labeled (a) “alliance positioning with respect to goals, strategies, and leadership” (α = 0.85), (b) “alliance positioning with respect to the economic and political environment” (α = 0.78), (c) “challenges to sustainability” (α = 0.84), and (d) “overall future performance” (α = 0.90).

Predictor variables. We assessed participants’ perceptions of alliance effectiveness with a survey question that asked respondents to indicate how effective the alliance has been in nine functional areas (e.g., recruiting staff, establishing strategic relationships with community leaders). All responses were recorded on 5-point scale (1 = not at all effective; 5 = extremely effective). Factor analysis indicated a single-factor solution, so we created a single composite index by averaging responses across these nine items (α = 0.91).

We assessed alliance as a collective with a survey question that asked respondents to indicate how much they agree that participants in their alliance have a clear shared vision, agree on strategies to achieve priorities, lack a sense of ownership (reverse scored), and lack information or access to alliance plans (reverse scored). All responses were recorded on a 5-point scale (1 = strongly disagree; 5 = strongly agree). Factor analysis indicated a single-factor solution. Accordingly, we created a single composite index by averaging responses across these four items (α = 0.72).

We assessed individual commitment with two sets of dummy variables that pertained to an individual’s participation in the alliance. Duration of participation was assessed with a single question that asked respondents to indicate how long they have participated in the alliance, which we used to create three dummy variables: (a) less than 1 year (referent), (b) 1–5 years, and (c) 5 years or more. Intensity of participation was assessed with a single question that asked respondents to indicate what percentage of their time they devoted to alliance activities over the past 6 months, which we used to create three dummy variables: (a) less than 5% (referent), (b) 5%–24%, and (c) 25% or greater.

Control variables. Using data from survey items, we controlled for the effects of key characteristics of alliances and their members that could influence the results. First, we created seven dummy variables to account for differences in the types of organizations that individual members represented: (a) staff of the alliance (referent), (b) insurer, (c) provider, (d) government agency, (e) employer, (f) consumer, and (g) other.

We also controlled for alliance size (number of individual members), diversity (alliance diversity is measured by a Gini coefficient, which indicates the dispersion of individual members across the organizational types listed above; higher Gini scores indicate more diversity of organizational members in the alliance), and organizational structure (partnership—participating organizations formed an alliance as a partnership, subsidiary—an alliance was incorporated as a subsidiary of one of the member organizations, and single organization—an alliance was formed as a free-standing entity). Much prior research in organizational theory shows that organizational size, diversity, and structure often have important influences on individual behavior as well as organizational performance (Daft, 2015).

Data analyses. Individual respondents were the unit of analysis. We used univariate statistics to describe the study sample. We used two multivariate regression model approaches to test the study hypotheses. First, we used random effects regression models to account for clustering of individuals’ survey responses within alliances. Second, as a check on these models and results, we used regression models with alliance-level fixed effects to account for unobserved factors.

Importantly, given that the data to measure both the predictor and dependent variables come from responses to the same survey, we diagnosed the extent to which common method variance (CMV) was a problem with two tests (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003; Richardson, Simmering, & Sturman, 2009). First, we conducted Harman’s single factor test whereby all variables are loaded onto a single factor via factor analysis and constrained so that there is no rotation (Eichhorn, 2014). CMV may be present if the single factor explains more than 50% of the variance. In our case, the single factor accounted for 35.3% of the variance. The second test (common latent factor) is similar to the Harman’s single factor test in that all observed variables are loaded onto a single factor; however, this test retains the latent factors and their relationships that were included in the main analysis. CMV may be present if the common shared variance of this new latent factor exceeds 50%. In our case, the shared variance was 29.9%. The results of these analyses indicate that CMV was not a significant issue for the study.

Back to Top | Article Outline

Results

Sample characteristics. Data in Table 2 show that survey respondents were predominantly health care providers (31.9%), followed by other stakeholders (18.9%) and alliance staff members (17.5%). Most of the alliances were operating as independently incorporated (stand-alone) organizations (60%), whereas the others were structured either as partnerships (20%) or subsidiaries of a parent organization (20%). There were 110 members, on average, per alliance.

Table 2

Table 2

On average, survey respondents believed the alliance had been relatively effective to date (M = 3.8) and believed that participants had a shared vision for health care in the community and the means of achieving that vision (M = 3.8). Over one half (50.3%) of the respondents reported participating in the alliance between 1 and 5 years, followed by nearly 43% of the respondents who indicated participating more than 5 years. Slightly less than one half of the respondents reported devoting 5% or less of their time to alliance activities, whereas 18.1% of all respondents reported devoting 25% or more of their time to alliance activities.

Among the sustainability indicators, respondents were most positive about the alliance’s positioning with respect to its goals, strategies, and leadership over the next 5 years, with an average rating of 3.9 (SD = 0.9). Respondents were less positive about whether the alliance had sufficient financial resources and participation, with an average rating of 3.0 (SD = 1.0). Respondents were moderately optimistic about the likelihood of the alliance’s future work leading to measurable improvements in the quality of health care and the health of the people in its service area (M = 3.4, SD = 0.9). Similarly, respondents believed that the alliance was slightly to moderately likely, on average, to face challenges attracting the right people to the alliance and securing resources over the next 5 years (M = 2.5, SD = 0.8). Supplemental Digital Content 3 (see Supplemental Digital Content 3, http://links.lww.com/HCMR/A34) shows bivariate correlations among the continuous variables in the study.

Back to Top | Article Outline

Multivariate Results

We organize the discussion below around the dependent variables, that is, the conditions indicative of sustainability. For each of the dependent variables, we focus on the three groups of predictors that are of primary interest in the analysis. We conclude with a summary of the relationships and discussion of the extent which these relationships provide support for the study hypotheses. Because the results from fixed effects models did not differ from the results from random effects models, we report only the latter below.

Positioning with respect to its goals, strategies, and leadership over the next 5 years. Controlling for other factors, respondents who believed the alliance had been more effective were more likely to agree that, on average, the alliance was better positioned with respect to its goals, strategies, and appropriate leaders in the coming years (b = 0.69, p < .001; Table 3). Similarly, respondents who believed more strongly that alliance participants had a shared vision for health and health care in the community and who were in agreement about the means to achieve that vision reported that the alliance was better positioned with respect to its goals, strategies, and appropriate leaders over the next 5 years (b = 0.23, p < .001). None of the measures of commitment were significantly associated with participants’ perceptions of alliance positioning with respect to goals, strategies, and leadership.

Table 3

Table 3

Positioning with respect to the economic and political environment over the next 5 years. Respondents who believed more strongly that alliance participants had a shared vision for health and health care in the community and who were in agreement about the means to achieve that vision were more likely to report that the alliance was better positioned with respect to its economic and political environment (b = 0.25, p < .01).

Challenges facing alliance over next 5 years. Controlling for other factors, respondents who felt the alliance had been more effective were less likely to believe that the alliance would face adverse challenges in the coming years (b = −0.35, p < .001; Table 3). Likewise, respondents who believed more strongly that alliance participants had a shared vision for health and health care in the community and who were in agreement about the means to achieve that vision reported that the alliance was less likely to face adverse challenges over the next 5 years (b = −0.27, p < .001). Once again, none of the commitment variables were significantly associated with perceptions of future adverse challenges.

Likelihood that alliance’s future performance will lead to positive outcomes. Respondents who believed the alliance had been more effective were more likely to report that, on average, the alliance’s work would lead to measurable improvement in the quality of health care and health of people in its service area (b = 0.64, p < .001). Likewise, respondents who believed more strongly that alliance participants had a shared vision for health and health care in the community and who were in agreement about the means to achieve that vision, on average, reported that the alliance’s work would be more likely to lead to measurable improvement in the quality of health care and health of people in its service area (b = 0.26, p < .001).

Finally, respondents were more likely to report that the alliance’s work would lead to measurable improvements in the quality of health care and health of people in its service area when they devoted more time to alliance activities. Specifically, respondents were more likely to report that the alliance’s work would lead to measurable improvements in the quality of health care and health of people in its service area when they devoted 6%–25% of their time to alliance activities (b = 0.26, p < .001) and 26%–100% of their time to alliance activities (b = 0.53, p < .001), relative to respondents who devoted 0–5% of their time to alliance activities.

In summary, our results provide the strongest support for Hypothesis 2—that alliance participants are more likely to view their alliances as sustainable when they share a common vision, goals, and strategies for the alliance—with all four potential relationships statistically significant in the direction hypothesized. The analysis also provides support for Hypothesis 1, with three of the four relationships between perceived effectiveness and conditions indicative of sustainability statistically significant in the direction hypothesized. In contrast, the hypothesized relationship between commitment and conditions indicative of sustainability received no support.

Back to Top | Article Outline

Discussion

The results from multivariate analyses show that participant perceptions of alliance effectiveness and their alliance as a collective effort with shared vision, strategies, and sense of ownership are robust correlates of perceptions of sustainability. Indeed, these two predictors are significantly correlated with at least three of the four measures of sustainability in both random effects and fixed effects models.

These results support arguments that stem from path dependence: Prior organizational success provides a foundation for future success. Participants who believe that their alliance has been effective in the past also are more likely to believe that their alliance is sustainable. This result may imply that stakeholders concerned with sustaining alliances and their contributions should emphasize the importance of timely, “small wins” (Weick, 1984). In other words, setting objectives, achieving at least some of them early in the life of an alliance, and then performing consistently over a period of years likely build perceptions of alliance reliability and effectiveness. In turn, these perceptions build momentum for sustainability. In contrast, alliances that rely on last-minute heroics or “moon-shot” projects to attain goals may not engender the same level of confidence, including confidence about future performance.

These results also support the importance of alliances as collective efforts: To the extent that participants believe they share a vision and strategies as well as feel a sense of ownership for an alliance, the more likely they are to view the alliance as sustainable. The message here for alliance leaders seems to one of inclusion: How can alliances involve participants in developing a vision and strategies (D’Aunno, Alexander, & Jiang, 2017)? The implication is that participants in alliances need to move carefully—and probably slowly—in the early days and months of launching an alliance to ensure that individuals have opportunities to make their concerns and views known and further feel that such exchanges are respectful, inclusive, and not rushed.

Though this advice may seem obvious, it is common for leaders to fall into “efficiency traps” in which they place a high premium on “getting things done” at the expense of effectively listening to individuals and thus building a strong foundation for alliance development that rests on as much member input as possible. In fact, some stakeholders (e.g., from the business community) may be accustomed to and prefer “top-down” leadership approaches that do not involve much discussion. Going slowly and exercising patience are likely to be challenging, but rewarding.

Last, descriptive statistics show that, compared to other dimensions of sustainability, respondents are less optimistic about how well positioned their alliances are to deal with future political and economic conditions. Perhaps alliance participants feel that they have less control over and less certainty about these conditions than they have over key characteristics of their alliances (e.g., goals). As a result, respondents may feel less optimistic about the ability of their alliances to deal well with these external conditions.

Limitations. Despite the study’s strengths, it also has limitations. In particular, we note that the article relies on data from a common source, that is, surveys of participants. Thus, the data may be subject to common method bias (Podsakoff et al., 2003). Though we checked for common method bias in the survey data we draw on and found no evidence of any, future research should draw on multiple sources of data.

Second, the study response rate (39%), though within a widely accepted range, may be a concern. Yet, researchers using data from earlier waves of this survey conducted analyses to identify possible nonresponse bias in the survey data and found no evidence of any (e.g., D’Aunno et al., 2017). These multivariate analyses compared responders to nonresponders in three time periods between 2006 and 2013 and used several important variables as covariates (e.g., measure of perceived benefits from alliance participation). Results show that responders did not differ significantly from nonresponders; this is particularly important because these results indicate that nonresponders were not individuals who were disgruntled about their alliance.

Similarly, there is the possibility of omitted variable bias. But, the results from the random effects and fixed effects models do not differ, suggesting that this is not a substantial threat to the validity of the results. Furthermore, the data are cross-sectional, making it difficult to interpret the direction or possible causal nature, if any, of the relationships we observed. We note, however, that survey respondents typically have been participating in the alliances over many years (43% of respondents have been alliance members for 5 years or more), and as a result, they are quite familiar with their alliances and can likely distinguish past performance and alliance characteristics from their assessments about the future. Moreover, a rigorous literature review informed the survey items.

Last, we want to be clear that, though participants' perceptions about an alliance are important, they form only one element within a set of factors that shapes the sustainability of an arrangement such as the AF4Q organizations. Participants may be convinced, for example, that an alliance has done great work, identify with its goals, and view it as well prepared to meet future challenges—and yet see the alliance deteriorate as environmental changes occur.

Future research. The study results suggest some important directions for research on alliance sustainability. Most important is the need for follow-up research to identify which alliances continue to operate and what factors among those we examined (or others) are related to sustainability. The results we report provide some “good news” in the sense that respondents generally are optimistic about the positioning of their alliances for the next 5 years when considering their goals, strategy, programs, and leadership. Importantly, respondents also are somewhat optimistic about the likelihood that their alliances’ efforts will lead to improvements in quality of health care services and health in their communities. The key question remains: How well-founded is the respondent optimism we observed?

Furthermore, it will be important to understand how alliance leaders and participants were able to meet the challenges of sustainability. Identifying key factors that contribute to alliance sustainability provides only part of the understanding we need to promote the work of alliances. We also need studies of the processes that alliance participants engage in to sustain their work. In other words, alliance leaders need to “know what” matters for sustainability, but they also need to “know how” to promote these outcomes. Such studies typically require qualitative or mixed-methods approaches.

Finally, it will be important to assess the sustainability of alliances from the perspective of multiple stakeholders. To be sure, this study draws on data from several types of stakeholders who are participants in alliances. Yet, it is likely that, to understand the sustainability of alliances more fully, we will need to include the perspectives of important actors who are not alliance participants. These could include, for example, foundations or other funders that provide alliances with financial support, community leaders, local elected officials, and community members who use services that an alliance facilitates. In short, the best efforts of alliance leaders and participants may be necessary, but not sufficient, factors in alliance sustainability; support from key actors and stakeholders in local environments may play a fundamental role as well.

Back to Top | Article Outline

Practice Implications

Our interpretation of the results suggests a critical challenge for leaders and participants in alliances: How to balance the need to create early “small wins” that show progress toward goal attainment versus the need to avoid falling into “efficiency traps” in which leaders focus on “getting things done” at the expense of relationship development among alliance partners. In other words, alliance leaders and participants face somewhat conflicting pressures that, if not handled well, threaten their success and sustainability. Alliances may need to simultaneously satisfy participants who are focused on a series of wins early on in the partnership and participants who are more focused on partnership processes and building consensus across members.

We propose that alliance leaders can consider two related practices for managing this challenge. To begin, they should emphasize the need to be mindful of the potential conflict between timely goal attainment and relationship building. Indeed, recent research shows important benefits from mindfulness (Good et al., 2015). Being mindful should include, for example, open discussion among participants at the launch of an alliance about the pitfalls associated with neglecting (or overemphasizing) either early wins or relationship building.

Furthermore, building on such discussions, alliance leaders can ensure that participants have opportunities to experience both some quick “wins” (and celebrate them) and collaborative relationship building. That is, it may be important for alliance leaders to consciously pursue both of these goals simultaneously. This can be done, for example, by creating project teams that vary in the pace of their work and difficulty of their goals. Some teams can focus on quick wins, whereas others focus on longer-term gains that require more consensus, participation, and collaborative work. Individuals then can self-select into project teams that match their preferences. In summary, though we clearly need more research to recommend evidence-based practices to build alliance sustainability and performance, this study suggests important practical considerations for alliance leaders and participants.

Back to Top | Article Outline

References

Alexander J. A., Hearld L. R., Wolf L. J., & Vanderbrink J. M. (2016). Aligning forces for quality multi-stakeholder healthcare alliances: Do they have a sustainable future? The American Journal of Managed Care, 22(12 Suppl.), s423–s436.
Alper J. (2014). Population health implications of the Affordable Care Act: Workshop summary. Washington, DC: National Academies Press.
Arthur W. B. (1994). Increasing returns and path dependence in the economy. Ann Arbor, MI: University of Michigan Press.
Beich J., Scanlon D. P., & Boyce P. S. (2010). A community-level effort to motivate physician participation in the national committee for quality assurance diabetes physician recognition program. Population Health Management, 13(3), 131–138.
D’Aunno T., Alexander J. A., & Jiang L. (2017). Creating value for participants in multistakeholder alliances: The shifting importance of leadership and collaborative decision-making over time. Health Care Management Review, 42(2), 100–111.
Daft R. L. (2015). Organization theory and design. Boston, MA: Cengage Learning.
Edwards J. C., Feldman P. H., Sangl J., Polakoff D., Stern G., & Casey D. (2007). Sustainability of partnership projects: A conceptual framework and checklist. The Joint Commission Journal on Quality and Patient Safety, 33(12 Suppl.), 37–47.
Eichhorn B. R. (2014). Common method variance techniques. Cleveland State University, Department of Operations & Supply Chain Management. Cleveland, OH: SAS Institute, Inc.
Gomez B. J., Greenberg M. T., & Feinberg M. E. (2005). Sustainability of community coalitions: An evaluation of communities that care. Prevention Science, 6(3), 199–202.
Good D. J., Lyddy C. J., Glomb T. M., Bono J. E., Brown K. W., Duffy M. K., … Lazar S. W. (2015). Contemplating mindfulness at work: An integrative review. Journal of Management, 42(1), 114–142.
Greenberg M. T., Feinberg M. E., Gomez B. J., & Osgood D. W. (2005). 3.5 Testing a community prevention focused model of coalition functioning and sustainability: A comprehensive study of communities that care in Pennsylvania. Preventing Harmful Substance Use: The Evidence Base for Policy and Practice, 129.
Gulati R. (1995a). Does familiarity breed trust? The implications of repeated ties for contractual choice in alliances. Academy of Management Journal, 38(1), 85–112.
Gulati R. (1995b). Social structure and alliance formation patterns: A longitudinal analysis. Administrative Science Quarterly, 619–652.
Hasnain-Wynia R., Sofaer S., Bazzoli G. J., Alexander J. A., Shortell S. M., Conrad D. A., … Sweney J. (2003). Members’ perceptions of community care network partnerships’ effectiveness. Medical Care Research and Review, 60(4 Suppl.), 40S–62S.
Hearld L. R., Alexander J. A., & Shi Y. (2015). Leadership transitions in multisectoral health care alliances: Implications for member perceptions of participation value. Health Care Management Review, 40(4), 274–285.
Hearld L. R., Bleser W. K., Alexander J. A., & Wolf L. J. (2016). A systematic review of the literature on the sustainability of community health collaboratives. Medical Care Research and Review, 73(2), 127–181.
Herscovitch L., & Meyer J. P. (2002). Commitment to organizational change: Extension of a three-component model. Journal of Applied Psychology, 87(3), 474–487.
Kaiser H. F. (1958). The varimax criterion for analytic rotation in factor analysis. Psychometrika, 23(3), 187–200.
Kale P., & Singh H. (2009). Managing strategic alliances: What do we know now, and where do we go from here. Academy of Management Perspectives, 23(3), 45–62.
Meyer J. P., Srinivas E., Lal J. B., & Topolnytsky L. (2007). Employee commitment and support for an organizational change: Test of the three-component model in two cultures. Journal of Occupational and Organizational Psychology, 80(2), 185–211.
Pluye P., Potvin L., Denis J. L., & Pelletier J. (2004). Program sustainability: Focus on organizational routines. Health Promotion International, 19(4), 489–500.
Podsakoff P. M., MacKenzie S. B., Lee J. Y., & Podsakoff N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879–903.
Polacsek M., Orr J., O'Brien L. M., Rogers V. W., Fanburg J., & Gortmaker S. L. (2014). Sustainability of key Maine Youth Overweight Collaborative improvements: A follow-up study. Childhood Obesity, 10(4), 326–333.
Richardson H. A., Simmering M. J., & Sturman M. C. (2009). A tale of three perspectives: Examining post hoc statistical techniques for detection and correction of common method variance. Organizational Research Methods, 12(4), 762–800.
Rog D., Boback N., Barton-Villagrana H., Marrone-Bennett P., Cardwell J., Hawdon J., … Reischl T. (2004). Sustaining collaboratives: A cross-site analysis of the National Funding Collaborative on Violence Prevention. Evaluation and Program Planning, 27(3), 249–261.
Scanlon D. P., Beich J., Leitzell B., Shaw B. W., Alexander J. A., Christianson J. B., … McHugh M. (2016). The aligning forces for quality initiative: Background and evolution from 2005 to 2015. American Journal of Managed Care, 22(12 Suppl.), S346–S359.
Shortell S. M., Zukoski A. P., Alexander J. A., Bazzoli G. J., Conrad D. A., Hasnain-Wynia R., … Margolin F. S. (2002). Evaluating partnerships for community health improvement: Tracking the footprints. Journal of Health Politics, Policy and Law, 27(1), 49–92.
Simpson L. A. (2012). That was then, this is now. American Journal of Managed Care, 18(6), S109.
Staw B. M. (1981). The escalation of commitment to a course of action. Academy of Management Review, 6(4), 577–587.
Vergne J. P., & Durand R. (2010). The missing link between the theory and empirics of path dependence: Conceptual clarification, testability issue, and methodological implications. Journal of Management Studies, 47(4), 736–759.
Weick K. E. (1984). Small wins: Redefining the scale of social problems. American Psychologist, 39(1), 40.
Wiltsey Stirman S., Kimberly J., Cook N., Calloway A., Castro F., & Charns M. (2012). The sustainability of new programs and innovations: A review of the empirical literature and recommendations for future research. Implementation Science, 7, 17.
Yonek J. C., Mittler J. N., & Hasnain-Wynia R. (2014). Why and how six aligning forces for quality communities have focused on reducing disparities. Medical Care Research and Review, 71(5), 435–449. 1077558714533826.
Young G. J. (2012). Multistakeholder regional collaboratives have been key drivers of public reporting, but now face challenges. Health Affairs, 31(3), 578–584.
Keywords:

alliance sustainability; conditions for alliance sustainability; multistakeholder alliances

Supplemental Digital Content

Back to Top | Article Outline
Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved