Study findings are presented in terms of key issues that emerged relative to the evaluative questions. In reading the findings, the following context is of note:
• Implementation was being driven by a central agency within the VA health care system external to practice sites and was focused on creation of standard evidence-based practice (EBP) throughout the system.
• The PARIHS model was familiar to many of the investigators but was not used prescriptively; it also was not the only potential definition of facilitation available to the participants or QUERI teams.
• The concept of facilitation had to be operationalized by each QUERI project, as none were given a detailed, operational set of guidelines for the emerging role.
• Most projects were placed under a tight timeline for demonstrating rapid, targeted improvements in current practice.
• Findings relate to an "external" facilitator role as opposed to an "internal" facilitator role. [See Table .] Local individuals directly involved in or assigned to an active local role in a QUERI implementation project, and interacting with the external facilitator, are henceforth referred to as "internal change agents." Internal change agents within implementation sites were variously termed by QUERI project teams as a clinical champion, opinion leader, site coordinator, site leader, or site team leader – but not as a facilitator. As one interviewee noted, these internal players were expected to implement the new practice and "figure out how to ensure that patients received the recommended care." In some projects, the person filling this internal site role was a unit manager or held another formal leadership role; in many projects it was a physician; and in others, the external facilitator worked with "emergent groups" or "different individuals for different interventions," as each required a particular skill or role.
The nature of facilitation
As experienced by these VHA implementation researchers, facilitation is a valuable and critical process of interactive problem-solving and support, which occurs in the context of a recognized need for improvement and a supportive interpersonal relationship. The recognized need is one derived through research of best practice and diagnostic analysis of a site's performance gaps [15
In general, interviewees viewed the end goal of facilitation as helping people in health care settings modify their work to incorporate a specific evidence-based clinical practice. This "helping" effort was seen to occur in general through the support of locals' use of an evidence-based implementation strategy – originally developed by the research team (additional file 1
: Study Projects
) – and through provision of other assistive and encouraging activities.
More specifically, interviewees identified the objectives of external facilitation as follows:
1. To help internal change agent/s at implementation sites understand what needs to change and how change can occur. This involved identifying organizational or provider factors that could make it easier or harder to achieve the adoption of a best practice; providing information regarding the evidence, related changes and/or useful problem solving linkages; and assisting in the reduction of identified barriers to progress.
2. To provide support to internal change agent/s in the form of encouragement, mentoring, and positive feedback, as appropriate.
The following individual participant comments highlight these objectives ['you' and 'them' herein refer to the internal change agents as applicable]:
• "We can help you with problems and can help explain the implementation interventions to you; these are the tools that will help you, and we can help you use them."
• "Facilitation leads to enabling staff at the sites put the interventions in place, and to maintain and/or modify them over time."
• "The essence of facilitation is trying to make things easier or easiest to make changes. To help them understand how they need to change, give them tools, monitor, and keep providing support as necessary."
Facilitation and other roles
A key question about facilitation is whether it is different from or part of other implementation interventions, processes or roles. Based on the experiences of these QUERI interviewees, the following observations were made about facilitation:
1. In its most concrete form, i.e., as an implementation change agent or facilitator role, facilitation can be viewed as a distinct intervention. However, in general interviewees suggested that facilitators will likely use or integrate other implementation interventions while performing this problem-solving/supportive function. For example, they might provide education to enhance an internal change agent's knowledge about the targeted evidence and its credibility. The difference between education and facilitation is, one suggested, related to "whether you are using predetermined materials, identified at the beginning of the project or whether you discover a need and develop or modify preexisting information for the use of those with the problem. You might even develop the education to meet requests of the local personnel." Another said, "On the site visit, I came in with a PowerPoint presentation. That is education. When they called me for help ... that was different. It was facilitation."
Facilitators also might help internal change agent/s understand and utilize audit and feedback data provided by an implementation research team, or use persuasive advocacy/championing to reinforce the need for a targeted evidence-based change in relation to local practice gaps. In terms of the latter, another participant suggested, "Facilitation is not equivalent to marketing ... or networking. But, facilitation can encompass those particular roles and be part of making these roles easier." Yet another said, "It really has a lot to do with creating buy-in/activating the sites/championing the recommendations."
Two interviewees described their view of the distinctness of facilitation as follows:
• "Facilitation is more general than other change roles – more flexible. I think facilitation is a concept that addresses change and the individuals who will create the change. Its precise character and activities will depend on the purpose of facilitation, the structure in which people work, and the people with whom the facilitator will work."
• "Facilitation is more two-way than other implementation strategies, not as prescriptive, and is more adaptive and respectful of what is in place."
2. Facilitation can occur at multiple levels within an implementation project. Multiple individuals can contribute to such facilitative support in formal or informal ways. Clinical leaders, for example, might help to remove a barrier to change or a study team member (i.e., other than the facilitator) can assist with a needed linkage to an outside expert or a problem solver.
3. Facilitation may be considered a mediating
implementation intervention or process [4
] because it often enables and supports actualization of other implementation interventions, such as electronic clinical reminders, audit and feedback, or operational system changes. In reviewing the final manuscript, one participant described it as the "engine that drives the other
... [an] ingredient that makes them work
4. Within the context of action-oriented implementation research such as QUERI [15
], formal facilitation was seen to begin when the external facilitator starts to establish a working relationship with an internal change agent. Prior to that point, as a member of the research team, facilitators may participate in pre-implementation tool development or other groundwork. As one participant explained, "I did a lot of the developing of the implementation interventions and related educational materials... this was not a part of my facilitation role."
This facilitator also described how "my work changed from education to facilitation over time
." Another interviewee talked about the difference between facilitation and "getting them ready
" for implementation, and included educational outreach activities as part of the latter groundwork.
5. The study team project management role within QUERI was often separate from the facilitator role, or at the least was seen as a separate set of functions. In addition to pre-implementation groundwork such as decision-making about interventions, diagnostic analysis of targeted sites, development of toolkits, or engagement of leadership, QUERI project management often focused on data collection, report development, scheduling of team meetings and record maintenance.
Key components of external facilitation
As noted above, facilitators appear to help internal change agents actualize an implementation plan and, thereby, EBP through two key components of the facilitation process, i.e., interactive problem solving and support. A summary regarding distinct behaviors and activities related to these two components of the external facilitator role are described below, with further detail provided in Tables , , and .
The external facilitator role in problem identification/resolution: potential activities/behaviors per QUERI experiences
The external facilitator role in communication and formative use of data: potential activities/behaviors per QUERI experiences
The external facilitator role in a supportive relationship: potential activities/behaviors per QUERI experiences
The QUERI facilitator was described by participant accounts as engaging in interactive, contingent problem solving. QUERI facilitators were commonly depicted as a problem solving resource for internal change agents, with the intent to enable these internal agents to solve their own problems. Facilitator – internal change agent problem solving was seen as interactive because of the need for ongoing dialogue and collaboration, and as contingent in terms of the pre-determined evidence-based goal of the QUERI project, the complexity of involved changes, and the potentially differing needs of different sites.
Analysis of participant accounts resulted in identification of three enabling and inter-connected aspects of problem-solving, as illustrated below:
1. Problem identification and resolution (Table ): This usually involved helping the internal change agent/s in each site identify local problems and viable solutions, which often varied from site to site. As three of the participants described it: "Sometimes people were having trouble and we helped them name the problem so we, or they, could seek help"; "It involves helping others to get whatever they are trying to do done ... to help with the process... helping them through barriers, talking them out, and giving advice"; and "We facilitators engaged these players to identify problems and then go solve it themselves."
Occasionally the external facilitator would take direct action (i.e., the "doer/task function" in the PARIHS model [7
]) but only in particular circumstances. For example, the facilitator might take such actions in the case of a problem requiring special expertise, or for role-modeling purposes. As one participant explained, "Facilitation focused on the idea of doing for and then they would do."
2. Use of formative data (Table ): An inherent component of problem identification and, at times, resolution is the acquisition, review and use of data. QUERI external facilitators obtained and used multiple types of formative data, which they may or may not have collected themselves. They used such information to enhance their ability to make needed changes at the site, provide real-time feedback to internal change agent/s, and understand and fulfill their own role. Examples of such formative data are as follows:
• Status of pre-requisite implementation factors in the local context, i.e., clinician perceptions regarding credibility of available clinical practice guidelines;
• How implementation was progressing from a fidelity/integrity of innovation [11
] and activity point of view, i.e., the degree to which the internal change agent/s and others actually did institute identified components of the implementation strategy;
• How implementation was progressing in relation to outcomes, e.g., the degree to which clinicians had adopted recommended guidelines; and
• The nature of local factors that appeared to be essential to the spread of implementation and progress, e.g., the degree of visible leadership support or cooperation of needed departments.
As individual participants said, "The facilitator worked to gather data from them about barriers and problems"; "I asked questions about the local context and sought to understand it"; and [At the start of implementation] "I made site visits to learn about current clinic practices including, for example, what they were doing about [X] ...by whom, how often, when.... And [I] did give them a lot of feedback on their practice once they started the intervention."
(Table ): As the link between the study team, implementation sites, and other relevant stakeholders, external facilitators supported the exchange of information through multiple communication channels. Thus, facilitators tried to establish a means of regular, goal-focused contact and became a resource and boundary spanner [3
] for ongoing information exchange or networking. As individual participants said, "If the opinion leaders needed input from
[another research team member], they went through the facilitator"; "I could tell them how other sites did it – give them information about solving problems";
and "Sometimes I could put them in contact with other clinics that were doing well
... [and] I collected protocol books from all the clinics, put them together, and shared it across clinics so they could decide what to do."
Such communication activities appeared to indirectly assist sites to solve problems and to provide mechanisms for moral support among internal change agent peers.
One interviewee described the essence of facilitation as "support and encouragement – more of a relationship where you work with the team or identified person at the site, rather than an outsider coming in with educational materials." Other interviewees described encouraging and helping internal change agents [ICAs] feel that they – the ICAs – could actualize implementation and, thus, enhance adoption of the targeted evidence. For example, they talked about "allowing and encouraging them to do their best to achieve these goals"; and "You have to share that sense of goal achievement ... attainment in order to get there, in helping them. It means cheerleading."
Overall, external facilitators tried to focus on enhancing the ability of ICAs to succeed, strengthening their sense of accomplishment, and reinforcing the fact that assistance was at hand. Table lists in more detail the essence of establishing and maintaining a reciprocal, supportive, problem-oriented interpersonal relationship, as experienced by these interviewees. This support function also appeared to involve helping ICAs understand what is required to facilitate, which may involve clarifying expectations for both the external change agency facilitator and the ICA, and keeping these expectations realistic.
Factors related to the perceived degree of success of external facilitation
There are barriers to and enablers of external facilitation, at times reflecting either the absence or presence of the same critical factor. Table depicts this relationship in terms of common themes across interviewee experiences and sample quotes illustrating contrasting circumstances.
Barriers and facilitators of external facilitation per QUERI experiences
Four key factors and related observations were described as follows:
1. Motivation/leadership at the sites: These interrelated elements came in the following forms:
• An identified or assigned individual at the site, i.e., the internal change agent, who needed both commitment and time to put into the change process.
• Buy-in, or conversely, lack of support from formal administrative and clinical leaders for the change initiative and internal change agent role/s. Active leadership buy-in and support was observed, e.g., through provision of needed resources, verbal reinforcement of the importance of the initiative, and integration of changes into routine QI structures. Conversely, lack of support was noted in, e.g., a lack of responsiveness to needed assistance from key departments.
2. Research team understanding and support of the external facilitator role:
• Facilitation was a new concept to many of the projects and not uniformly understood. However, some teams were reported as having supportive members, such as those who recognized facilitation as a distinctive role that was critical to the team's work. An example of this support was the perceived protection of the facilitator's time to "facilitate."
• Study teams that did not make facilitation a distinct or inherent part of the project were perceived as having made support and communication with sites more difficult, thus impeding optimization of implementation. As one participant further explained, "The problem was with how the implementation project was organized and the emphasis on the outcome rates. If anything, I think team members thought that telling staff to 'do something' would suffice to 'implement' the interventions."
• It was agreed that both the implementation team and external facilitator need the latter's role to be explicitly defined, with a core of facilitation responsibilities and behaviors.
3. Physical aspects of the facilitator role: Overlapping with factors cited above is the element of maintaining contact between the external QUERI facilitator and internal change agent/s.
• In order to fulfill the interactive problem solving and support function, interviewees felt they needed regular communication, face-to-face contacts, and, at times, onsite presence to attend meetings or directly observe local discussions.
• In contrast, various interviewees reported not only inaccessibility problems with internal change agents, but also physical barriers of geographical distance, a large number of sites and thus a large number of internal change agent/s, and the prohibitive cost of travel. For example, external facilitators reported difficulty teaching interactive skills at a distance or evaluating related achievements.
4. Selection and assignment of an individual to the external facilitator role: Interviewees reported that individuals could have an easier or harder facilitator role depending upon their skills, experiences, and/or personal attributes. In general, they felt that an external facilitator had to be able to develop positive relationships with individuals and teams, have skills in problem solving, and have credible, relevant experience. Such experience might relate to time within the VHA system, a previous facilitator role or perhaps, in complex clinical situations, an applicable clinical background.
A multi-faceted interview question was used to assess participant views regarding the degree of "success" of facilitation efforts. Data are from six interviewees, as one felt s/he couldn't judge success. Interviewee perceptions regarding the success of individually identified and critical facilitation behaviors ranged on a scale from 1 (not at all effective) to 5 (very effective/successful). In terms of the success of the overall facilitative effort, on the same scale, two individuals rated facilitation as not effective ("1" and "2" respectively), while remaining summary ratings averaged 3.25. Perceptions of the overall success of individual QUERI implementation projects were similar, with an average rating of 3.4.
The above responses usually were qualified, in that success was said to vary across sites within a project. For example, one participant with an overall negative rating said that it varied from "...site to site because of the intervention and the context. It differed by how much trust the team members at a facility gave me. They didn't all see me as a problem solver."
Only one interviewee reported a significant degree of formal facilitation measurement. That analysis suggested "More facilitation yielded better results – except for one site," which may have had frequent contact due to resistance.
In those facilitation relationships perceived by interviewees to be more successful, one or more of the following factors, consistent with the above barrier/enhancer discussion, were likely to be cited:
1. External facilitators were perceived by internal change agents as having certain attributes, i.e., they were:
• Credible to the internal change agent/s, e.g., seen as understanding the evidence;
• Good communicators, i.e., open to being contacted, friendly, and outgoing, and having established good rapport; and
• Flexible and responsive to the needs of the internal change agent/s, i.e., either having answers to questions or being able to find them.
2. The external facilitator had a history of relevant experience, needed skills, and a commitment to facilitation.
3. The external facilitator was based in a QUERI team that from its inception recognized the value of facilitation and supported the facilitator's ability to be successful.
Some illustrative comments from three interviewees included the following: "They said I was easy to communicate with, readily available. I returned calls. I either had answers or would find them." Another was said to "...have had a great rapport," and a third felt that a facilitator had to be "...someone who is well-liked." In terms of a clinical background, one said, "With no clinical background, it is difficult to talk to MDs. You need to understand what it is like being in the trenches." While another (a clinician by background) indicated "It isn't as important as personal skills...I could always refer them to literature or an MD."
In contrast, those facilitation relationships or encounters perceived as not very successful were more likely characterized by interviewees and interviewers with one or more of the overlapping factors on the following list.
1. The individual who might have played an active facilitator role was more focused, per research team requirements, on project study management activities such as data collection and other evaluation duties.
2. The person who was in contact with the internal change agent/s was not introduced as, nor supported to work as a "problem solver." One said, "My role got watered down. We didn't have answers to barriers so the participants got worn down, discouraged." Another was felt to have had an inappropriate "level of problem solving skills."
3. Responsibility for success of the project was delegated to the sites in the form of internal change agent/s, cited as "local champions" or "site leaders," but without the support of structured QUERI facilitation, internal support, or appropriate selection.
4. Particularly in the early phases of implementation in the VHA, facilitation was not consistently addressed nor supported in an explicit, structured manner at a project team level.