The importance of symbolic and engaged participation in evidence-based quality improvement in a complex integrated healthcare system: response to “The science of stakeholder engagement in research”

Abstract

In this commentary, we respond to the commentary provided by Goodman and Sanders Thompson regarding our paper on multilevel stakeholder engagement in a VA implementation trial of evidence-based quality improvement (EBQI) in women’s health primary care. We clarify our overall approach to engagement (comprised of both symbolic and engaged participation, according to the authors’ classification rubric), highlighting that symbolic participation is of more import and value than the authors suggest, especially in the context of a hierarchical healthcare system. We contend that the issue of power—and how power matters in stakeholder engagement—needs to be considered in this context rather than in global “community” terms. In response to the authors’ call for greater detail, we clarify our planning processes as well as our approach to veteran engagement. We concur with Goodman and Sanders Thompson that the science of stakeholder engagement necessitates a broader understanding of best practices as well as the impact of engagement on implementation outcomes.

We appreciate the thoughtful commentary provided by Goodman and Sanders Thompson [1] regarding our paper on multilevel stakeholder engagement in a VA implementation trial of evidence-based quality improvement (EBQI) in women’s health primary care [2]. The stakeholder engagement classification (categorized as non-participation, symbolic participation, and engaged participation) articulated by Goodman and Sanders Thompson is provocative, as are their comments on implementation and evaluation of stakeholder engagement. They pose key questions in order to push forward the science of stakeholder engagement. Here, we briefly clarify our approach to engaging multilevel stakeholders in the improvement of VA’s patient-centered medical home for women veterans, a numeric minority in the male-dominated VA healthcare system [3].

Goodman and Sanders Thompson categorized our approach to engagement as “falling somewhere between cooperation and collaboration,” thereby straddling symbolic and engaged participation. Stakeholder engagement in the symbolic category, they contend, may not change the “status quo,” with researchers still having “ultimate decision-making power.” Engaged participation, in contrast, includes shared decision-making, with those who “traditionally have limited power” being “given shared decision-making authority with powerful stakeholders.” Collaboration, in the context of engaged participation, means that researchers and community members are involved in design, implementation, and interpretation of findings. It also means that “all stakeholders benefit in some way from working together,” and that “community health stakeholders collaborate in decision-making and resource allocation with an equitable balance of power…” The authors seem highly inclined toward engaged participation as being the preferred or highest form of engagement of the three categories, suggesting that stakeholders in a symbolic participation approach are not “meaningfully engaged.”

We do not dispute the characterization of our approach to engagement as both symbolic/cooperative and engaged/collaborative. Rather, we would like to reconsider symbolic participation as perhaps of more import and value than the authors suggest. We suggest an alternative conceptualization of symbolic participation that considers who has power (traditionally, historically, institutionally) in the implementation context; who needs, warrants, and/or deserves more power; and how is the power of various stakeholders enacted in decision-making processes (and how could this enactment be changed or shifted). The issue of power—and how power matters in stakeholder engagement—needs to be considered in context rather than in global “community” terms, or mainly with reference to researchers’ power. In a federal organization such as the $60 billion VA healthcare system, hierarchy is inherent, with power “at the top” (i.e., at the level of the US President, who nominates the Secretary of Veterans Affairs) affecting all below; this “status quo” [1] is not likely to change in the foreseeable future. At least in the VA healthcare system, stakeholder engagement can shift the needle of change, but engagement processes and activities must be designed with explicit leadership commitment: VA colleagues have found across multiple implementation efforts that those with power (e.g., regional and facility leaders with access to resources) must support frontline clinical QI initiatives in order for them to have a chance of success [4].

Our research team’s deep understanding of the clinical implementation settings (based on over 14 years of VA women’s health research)—and the relative power of individuals working within these settings—motivated us to use EBQI to tailor women’s primary care, rather than other implementation strategies [5], because of its emphasis on top-down/bottom-up processes of evidence- and practice-based decision-making. Our task as researchers was to help put women’s health on leaders’ radars by sharing evidence of quality gaps. We did not undertake this task with the illusion that we had any decision-making power. In fact, we directly conveyed to the sites that we were aware of our lack of authority over their clinical activities. Instead, we offered technical assistance, facilitation, and data, as well as a decision-making infrastructure (EBQI) that brought multilevel organizational stakeholders together to effect meaningful changes. We therefore suggest that symbolic and engaged forms of participation in VA settings (and likely in other settings) need to incorporate as many levels of stakeholders as possible in order to tap into and collaborate across multiple complex spheres of clinical and policy influence [6, 7]. While laudable and aspirational, we contend that Goodman and Sanders Thompson’s vision of an “equitable balance of power” across all stakeholders is not realistic in a hierarchical organization.

Goodman and Sanders Thompson suggest that our planning processes were “perhaps insufficient for the variety of teams formed,” because we did not describe a phase of partners getting to know each other and understand each other’s unique perspectives. It is true that our multilevel stakeholder planning meetings, described in our paper, were not protracted gatherings designed to foster bonds amongst stakeholders. Rather, as we describe in our paper [2], they were targeted efforts to convene relevant stakeholders (many of whom already knew or were familiar with each other) in order to educate them about the goals of EBQI, review evidence about regional and facility-level gaps in quality of care, and map out strategic plans (“QI roadmaps”) using a modified Delphi panel process specifically designed to level the playing field [8]. With busy stakeholders such as regional directors, physicians, and staff, convening even these relatively brief meetings was challenging but extremely valuable for strengthening buy-in across organizational levels. In recognizing multiple levels of power and influence, we had national VA policy leaders in primary care and women’s health call in to the stakeholder planning meetings and provide documentation of their commitment, which served important symbolic purposes. Our facilitated discussions tackled strategic and resource issues, staffing problems, and other challenges within the context of consensus-derived priorities. Thus, we argue that planning processes must fit the implementation context, including the availability of and access to stakeholders at the highest levels necessary for project goals.

We would also like to respond to the authors’ encouragement to “take this work further to include meaningful engagement of patients (or the community/population impacted by the proposed work).” We neglected to point out in our paper that upwards of one third of the VA workforce is comprised of veterans, many of whom are also VA patients. When veteran stakeholders were “at the table” for our planning meetings or on the phone during our interviews, their vantage point as veterans (and for some, as VA patients) was inextricably informing their engagement in our efforts. As Murtagh and colleagues recently noted, all contributions to engagement efforts “are produced from specific subject positions which may change in different settings or be deployed for different effects” [9]; we believe this holds true for stakeholders who have multiple, sometimes tacit stakes in implementation. In this implementation trial, we chose to engage women veterans specifically at a strategic policy level via our Women Veterans Council, which weighs in on patient-level implications and dissemination of our initiatives. These nuances that we did not describe in the original paper suggest that it is important to clearly articulate the rationale behind the selection of which stakeholders to involve when and in what ways, and for what purposes [10].

In conclusion, we concur with Goodman and Sanders Thompson that the “science of stakeholder engagement” needs to be advanced by moving “beyond individual projects to a broader understanding of what works and why (best practices).” We endeavor to contribute to this broader understanding by expanding the notion (and recognized value) of symbolic participation and embracing a community-engaged research [11] approach in our women’s health research initiatives within a complex integrated healthcare delivery system.

Compliance with ethical standards

Conflict of interest

The views expressed within this study are solely those of the authors and do not necessarily represent the views of the Department of Veterans Affairs or the US Government. None of the authors has any actual or potential conflicts of interest.

Reporting and control of data

The reported findings have not been previously published and the manuscript is not being simultaneously submitted elsewhere.

The data has not been previously reported.

The authors have full control of all primary data and agree to allow the journal to review their data if requested.

Ethics approval and consent to participate

This study protocol has been approved by the VA Greater Los Angeles Healthcare System IRB-B Subcommittee (IRB Chair, Paul Shekelle, MD, PhD) (Approval Number 2013-040589). Individual study components, including key stakeholder interviews, were reviewed and approved by IRB-B as serial project modifications to an umbrella IRB application under the study title. For each study component, we obtained IRB approval for waivers of documentation of consent for all data collections. Consent was provided verbally and recorded in the case of interviews.

The VA Central IRB (Washington, DC) also determined that the participating VA medical centers (n = 12) were not engaged in research and that the intervention (or implementation strategy, evidence-based quality improvement [EBQI]) was QI. Documentation of IRB approval (and in VA settings, Research and Development Committee approval as well) and Central IRB determination of non-research for the implementation strategy are available on request.

Statement on welfare of animals

No animals are or will be involved in this study protocol. Similarly, use of animal or human data or tissue is not applicable to this study protocol.

HELSINKI statement

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.