‘In the Moment’: An Analysis of Facilitator Impact During a Quality Improvement Process

Abstract

Facilitators frequently act ‘in the moment’ – deciding if, when and how to intervene into group process discussions. This paper offers a unique look at how facilitators impacted eleven primary care teams engaged in a 12-week quality improvement (QI) process. Participating in a federally funded QI trial, primary care practices in New Jersey and Pennsylvania formed practice-based teams comprised of physicians, nurses, administrative staff, and patients. External facilitators met with each team to help them identify and implement changes aimed at improving the organization, work relationships, office functions, and patient care. Audio-recordings of the meetings and descriptive field notes were collected. These qualitative data provided information on how facilitators acted ‘in the moment’ and how their interventions impacted group processes over time. Our findings reveal that facilitators impacted groups in multiple ways throughout the QI process, rather than through a linear progression of stages or events. We present five case examples that show what acting ‘in the moment’ looked like during the QI meetings and how these facilitator actions/interventions impacted the primary care teams. These accounts provide practical lessons learned and insights into effective facilitation that may encourage others in their own facilitation work and offer beneficial strategies to facilitators in other contexts.

Introduction

Facilitation has grown rapidly and an expansive literature has articulated definitions, processes, and techniques that are popular in facilitator practice (see Hunter et al., 2007; Jenkins & Jenkins, 2006; Justice & Jamieson, 2006; Bens, 2005; Ghais, 2005; Wilkinson, 2004; Schwarz, 2002). Medical translational research has also increasingly incorporated external facilitators into quality improvement (QI) interventions (Gordon and Flocke, 2005; Stroebel et al., 2005; Thomas et al., 2005; Baskervilie et al., 2001; McCowan et al., 1997).

As scholars and practicing facilitators have emphasized, facilitators work intuitively, and often need to act in the moment. “It is a matter of choosing, in a particular moment, what to do, whether or not to intervene, and how to intervene” (Hunter et al., 2007, p. 38). Significant questions remain about how facilitators act ‘in the moment’ and the impact such decisions can have on groups over time. Recent work has pointed out that facilitator effectiveness is often directed at immediate ends such as self-reported perceptions during and after an intervention (Wardale, 2008).

With the understanding that facilitators are bombarded with stimuli during a meeting and must constantly make decisions in the moment, we posed the following question: what is a facilitator’s impact on a QI team and their change process over time? We provide a unique look into the actual facilitation processes of 11 Primary Care Practices (PCPs) during a QI trial. We present two overarching findings: 1) the frequency in which facilitators intervened over time (based on our “facilitator impact codes”), and 2) case examples showing what facilitators’ ‘in the moment’ decisions looked like and how these impacted the QI teams.

Background

With an extensive body of literature on group development (Hill & Gruner, 1973), Smith (2001) divides a wide number of group development models into three categories: linear-progressive, cyclical and pendular models, and non-sequential or hybrid models. The central characteristic of linear-progressive models is that groups develop in a linear fashion, moving from one phase or stage to another (sec, for example, Tuckman & Jensen’s forming, storming, norming, performing and adjourning, 1977). In contrast, cyclical and pendular models propose (hat groups revisit stages or swing back and forth among issues over the course of (heir work together (see, for example, Drexler et al.’s Team Performance model, 1991). Non-sequential models posit that groups do not follow a prescribed pattern of developmental events, but are shaped by various contingencies such as the task of decision making (Poole et al, 1985), time and deadlines (Gersick, 1988, 1991), and Die group’s external environment, internal group relations, and temporal boundaries (McCollom, 1990). While these models may be helpful for understanding how groups form, bond, and develop, many studies were based on groups of students or patients in highly-con trolled environments (Smith, 2001). Such models do not take into account how a facilitator may impact actual workplace group processes and development

In a short definition, facilitators focus on “helping groups do better” (Schuman, 2005, p. xi). In doing so, they typically perform a unique service that requires that they understand group processes and that they be able to assist the group as it diagnoses and solves process problems. As organizations increasingly use facilitators to assist with group dynamics and process issues, existing theories of group development must be reexamined in order to understand how facilitators impact groups.

Many groups involve an intentional, external facilitator that is, an organizational outsider who takes on a special role to intervene in the ongoing group process. To define what we mean by “facilitator”, we use the label ‘intentional’ to distinguish a purposeful (or formal or professional) facilitator from one who facilitates a group meeting with only a tacit awareness of the reasons or motives behind actions or with limited knowledge of facilitation tools and techniques (Thomas, 2004; Brockbank. & McGill, 2007). We also use the adjective “external” purposefully. While many organizations ascribe facilitation activities to a group member because s/he is recognized as a leader in some way (Keltner, 1989), our focus is on the “facilitator specialist”. As Keltner (1989) writes, “…the specialist does not deal with group content, but functions either as a non-group member or as a member with a very special and restricted role” (p. 37). Rather than being “concerned about his own learning in the group, his own gratification, and his standing with other members, the intentional, external facilitator is not a stakeholder in the tasks or interpersonal struggles that may affect group members” (Kingsbury, 1972, p. 110). This docs not mean the facilitator presents him/herself as an indifferent or unfeeling change agent, but as someone who has the advantage of distance from the issues at hand (Keltner, 1989, p. 38) and can balance the power issues at play within hierarchical organizations and teams.

Methods

For an understanding of how we collected data for this analysis and came to our findings, this methods section details the practices we studied/worked with, the organizational change process that was facilitated within the 11 practices, the facilitators (e.g., their background and training), our data collection methods, and our analysis procedures.

Settings

PCPs (comparable to ‘general practices’) play a significant role in the U.S. health care system, providing numerous benefits to the general population (Green & Fryer, 2002; Starfield & Simpson, 1993; Rosenblatt, 1992). Primary care clinicians typically treat acute and chronic illnesses and provide preventive care for all ages and both sexes. Sixty PCPs from New Jersey and Pennsylvania, USA were recruited into a federally funded QI trial called ULTRA (Using Learning Teams for Reflective Adaptation) from 2003–2008. PCPs included a range of ownership arrangements (including private community-based practices and university-owned practices), were located in rural, suburban, and urban areas, and ranged in size from 2 to 65 practice members. All names in this paper have been changed to protect confidentiality.

The ULTRA study was designed to improve adherence to multiple chronic disease guidelines through the implementation of a QI process. ULTRA uses a change process based on complexity science that views PCPs (and many organizations) as complex adaptive systems (Stroebel ct al„ 2005). This change process is called Multi-method Assessment Process (MAP)/Reflective Adaptive Process (RAP). MAP consists of observations (e.g., of various office functions, physician-patient encounters), in-depth interviews, and collection of various practice documents (e.g., policies, mission statement). MAP provides opportunities for the facilitator to better understand the practice and its members, and to develop rapport with participants. MAP data is analyzed and summarized in a report which is delivered to and discussed with practice members. Following this, RAP begins, which consists of weekly, facilitator-guided QI meetings with a cross-functional team in each PCP. QI team members were initially identified by the facilitator whose primary goal was to have a diverse team (based on functional work areas and length of employment with the practice). Table 1 details the QI team composition across practices.

Table 1.

QI Team Composition for ULTRA PCPs

	Physicians	Clinical staff	Admin/clerical Staff	Patient	Size of Practice
Hope Practice	1	3	4	1	37
River Practice	1	3	6	1	22
Oak Practice	1	3	4	1	14
Rosewood Practice	2	0	2	0	6
Meadow Practice	1	2	3	1	39
Snowy Practice	2	2	5	1	50
Elm Practice	3	3	1	1	11
Falls Practice	1	1	3	0	5
Care Practice	2	1	2	1	15
Hilltop Practice	1	2	2	0	9
Valley Practice	2	2	5	0	65

Facilitators

A core philosophy of the ULTRA study was built upon two fundamental aims: 1) to create time and space for learning and reflection as teams implemented QI changes, and 2) to draw on the wisdom of the group. Facilitators did not provide consultation or expert advice on the content of a team’s QI plans; rather, facilitators focused on process issues and guided teams through steps to implement their practice improvement plans.

The central elements of the facilitator’s role in the ULTRA study included the following:

1. A trained facilitator from the research project conducted an initial assessment of the practice (MAP – Multimethod Assessment Process) (Crabtree et al., 2001)

2. That same facilitator then facilitated a QI process with a cross-functional team of practice members, as well as a patient member (RAP – Reflective Adaptive Process) (Stroebel et al., 2005)

3. QI teams met at the practice for one hour per week for 12 weeks

4. The facilitator followed loosely-structured steps designed to help the QI team generate ideas, develop an improvement plan (on an issue of the team’s choosing), implement that plan, and monitor its effectiveness

5. The facilitator tailored the change process to the needs and desires of the practice

In this capacity, facilitators were external members of the PCPs, and took on a specialist role as they used facilitation tools and techniques to help teams “do better.”

Facilitators were hired for the ULTRA study at several points over the course of the five-year project. Positions were advertised locally (New Jersey and Pennsylvania) and applicants were expected to be adept at qualitative data collection and group facilitation. As such, most of the ULTRA facilitators had social science backgrounds and experience using qualitative methods and leading various kinds of groups (see Table 2).

Table 2.

Description of ULTRA Facilitators

	Gender	Level of Education	Field of Study	Facilitated PCP:
Facilitator A	F	MSW	Social Work	Hope Practice, River Practice
Facilitator B	M	PhD	Political Science	Hope Practice
Facilitator C	F	MPH	Public Health	Oak Practice
Facilitator D	F	PhD	Psychology	Snowy Practice, Elm Practice
Facilitator E	M	PhD	Sociology	Rosewood Practice, Meadow Practice
Facilitator F	F	PhD	Sociology	Fails Practice, Care Practice
Facilitator G	F	PhD	Anthropology	Hilltop Practice, Valley Practice

ULTRA facilitators received training on qualitative methods and facilitation based on the MAP/RAP model of practice improvement. This model incorporates many common facilitation tools, techniques, and strategies such as agenda-setting, establishing ground rules, brainstorming, collective decision making, learning current processes, and evaluating outcomes. Training sessions were conducted by UMDNJ faculty and staff. The primary facilitation expert was a public health professional with over 19 years of facilitation experience in domestic and international health care, research, and community settings. She had formal training/experience including facilitating total quality management teams with the Group Health Association in Washington, DC, and group facilitation and participatory strategic planning with Technology of Participation (ToP) and the Metropolitan Health Department of Nashville, Tennessee.

Training sessions incorporated various methods. Formal training occurred during a two-week in-house training where trainees read extensively about facilitation methods and participated in didactic sessions conducted by more seasoned facilitators/researchers. They also engaged in experiential and role-playing exercises to simulate the processes and techniques they would be using in practices. Understanding that facilitation is more than using tools and techniques, facilitators also participated in regular, ongoing meetings (2 hours, twice a week) with their colleagues to advance their learning and enhance their facilitator competencies (such as self-awareness). Occasionally, a study team member would observe the facilitator as s/he facilitated a QI meeting. And during regular facilitator meetings, a facilitator would often play portions of a QI meeting audio-recording. These provided opportunities for facilitators to critically reflect on their own work and to receive feedback and support from colleagues. This helped ensure that ULTRA facilitators were calibrating their techniques, reflecting on actual ‘in the moment’ decisions, and learning from their successes and failures.

Data Collection

With prior written consent from each QI team member, the weekly meetings were audio-recorded. In most cases, the quality of the recordings was good; however, in some instances, technical difficulties or environmental noise interfered and no recording was available. After each QI meeting, the facilitator wrote a descriptive field note capturing details of attendance, body language, verbal dialogue, and other impressions of the meeting.

DATA ANALYSIS

Selection Criteria

To reduce potential problems with missing data, we considered only those practices for which we had at least 75% of the audio-taped QI meetings and field notes. Although the trainings and ongoing facilitator meetings were intended to standardize the facilitator role and the methods used to brainstorm ideas, resolve conflict, or reach consensus, we recognized that variations of individual facilitator styles (based on personalities and facilitator attributes such as self-awareness or flexibility) were inevitable. To consider how individual facilitator styles may impact QI teams and their change process over time, we selected two practices for each facilitator. (However, one facilitator worked with only one practice; in another practice, there was a facilitator change midway through the 12 weeks). Our final sample consists of 11 practices and 7 facilitators.

All audio-recordings were transcribed by members of the analysis team. Because multiple people spoke during meetings, it was not possible to capture all discussions verbatim; however, all comments made by the facilitator (including pauses, tone of voice, and other utterances) were transcribed, as well as the contextual dialogue and length of the facilitator’s comments. Taken together, these data afforded rich empirical information on ‘real time’ facilitator interactions with the teams.

Development of the Facilitator Impact Codebook

Through an iterative process, the analysis team developed facilitator impact categories. As a secondary analysis, this process was done after all of the QI meetings were completed for the study. Initially, three members of the analysis team identified seven areas of facilitator impact based on prior knowledge of the literature and experience as facilitators. Listening to the audio-recordings from one meeting prompted the addition of three more codes, for a final set of ten (see Table 3). Many of these codes bear similarities to competencies for facilitators (e.g., see the International Association of Facilitator’s Core Competencies for Facilitators: http://iaf-world.org/i4a/pages/Index.cfm?pageid=3331).

Table 3.

Conceptual Codes and Definitions of Facilitator Impact

Code	Definition
Process	Facilitator’s efforts to move the group as established by the study design (e.g., agenda setting, team rotes, consensus building); everything to do with meeting management and team composition
Reflection	Facilitator’s attempts to have the QI team step back and took at issues by using questions and probing for clarification; learning current processes; asking how participants felt about a topic; everything to do with evaluating a meeting or the larger QI process
Conflict	Facilitator’s attempts to manage disagreement and conflict (conflict can vary in intensity and manifestation)
Safety	Facilitator’s efforts to create a safe place where QI team members can ask questions, challenge the facilitator or the process, disagree with others, and raise unpopular subjects; using warm-up exercises; establishing and working from ground rules
Socialization of New Skills	Facilitator communicates principles of mindfulness of others in the system or impact of decisions; Facilitator teaches, demonstrates or explains new facilitation tools and behaviors
Self-Disclosure	Facilitator shares personal insights, experiences, or stories
Task ID & Completion	Facilitator’s efforts to direct or re-direct the team to stay on task and to take actions aimed at promoting productivity and accomplishment of team goals; establishing accountability; facilitating the delegation of responsibility; setting timelines
Use of Humor	Facilitator’s efforts to ease tensions through playful comments; facilitator can be self-deprecating; laughing and joking
Emergence	Facilitator recognizing and allowing for new ideas to come out; nurturing unplanned suggestions and developments; the ‘a-ha’ moments; legitimizing concerns and observations of team members; establishing legitimacy of complaints and concerns
Encouragement, Support and Appreciation	Facilitator’s expressions of positive reinforcement for QI team members and the practice; noting the value of their work and personal contributions; rewarding enactment of new skills

Subsequent meetings with the analysis team contributed to a collective agreement on definitions and coding techniques. Remaining practice data were then coded independently by pairs of analysts and reviewed by the lead author to catch discrepancies. In cases of disagreement, die analysts reviewed the transcriptions and audio-recordings together to resolve different interpretations. Differences sometimes arose when a particular interaction could be reasonably coded two ways (e.g., a facilitator’s funny, personal story could be simultaneously “use of humor” and “self-disclosure”). Otherwise, analysts attempted to reach agreement on a single predominant category.

Quantifying Facilitator Impact by Code

To understand a “global picture” of how often facilitators intervened based on our 10 impact codes, we quantified the frequency of the impact codes for each meeting. While not intended to be statistical in nature, the frequency scores present some useful indicative results. Scores were based on the number of facilitator interventions and facilitator “air time” (time speaking) per code. Scores ranged from 0 to 4 (with 0 = None, 1 = Few, 2 = Some, 3 = Often, and 4 = Very Often). For example, in the first meeting at Hope Practice, the facilitator made numerous remarks and devoted extensive time to explaining the meeting process; the ‘process’ code for this meeting, therefore, was assigned a ‘4’ or ‘Very Often.’ A score of ‘0’ for a particular code (e.g., “conflict”) does not mean that no conflict occurred during that given meeting; rather, it means that the facilitator did not directly intervene to address conflict during the meeting.

Results

Global Trends of Facilitator Intervention

We first present a global picture of facilitator intervention by averaging the scores from each code for all 11 practices (See Figures 1-2-3). Little is known about how often facilitators intervene in group processes and how the frequency of these interventions may vary over time. We do not present these findings as “ideal” levels of facilitator engagement; rather, these serve as an empirical basis that can help facilitators be aware of their own frequency of intervening and how this may vary over time.

Figures 1-2-3.

Global Scores of Facilitator Intervention

Three patterns of facilitator intervention stand out:

• 6At the beginning of the QI meetings, facilitators intervened very often on ‘process’ (coded ‘4’ on average) and ‘safety’ (3.36 on average) which then declined over time. Since these PCPs tended to have patriarchal hierarchies, establishing an equal playing field with cross-functional teams required that facilitators devote considerable time and energy up front to create psychologically safe environments.
• 7In the areas of ‘reflection,’ ‘conflict’, ‘socialization of new skills’, and ‘task identification/completion’, facilitators increased the number of interventions as the QI process continued, then decreased their interventions towards the end of the QI process. Although facilitators intervened in these areas over the 12 weeks, their involvement was heightened as teams put their QI plans into action. We found considerable variation in how QI teams tackled issues: some identified issues almost immediately, while others struggled to articulate doable improvements. We also saw a small increase in reflection at the end of the 12 weeks when facilitators prompted teams to discuss their accomplishments and plan for next steps after the facilitator left.
• 8Facilitators intervened in relatively low (‘few’ to ‘some’ interventions) but stable levels in the remaining categories of ‘self-disclosure’, humor’, ‘emergence’, and ‘encouragement/support’ throughout the 12 meetings. This pattern suggests that teams needed relatively low but constant levels of involvement from external facilitators to interject personal insights and experiences, humor, and encouragement. Furthermore, although there is some variability of ‘emergence’ over the course of the intervention, we see that facilitators routinely intervened by nurturing new ideas and developments.

In the Moment: Case Examples

Stories from the qualitative data provide a closer look at what these facilitator interventions “looked like” and their impact on the complex and dynamic group processes. While there are numerous stories to draw from, we selected those that typified some common elements from the impact codes.

Case Example: Self-Disclosure

At the start of meeting #6 in Hope Practice, the facilitator began by admitting she was “frazzled” that day because her daughter had just left for the airport. Several team members sympathized with her, and one asked about her daughter’s trip, prompting the facilitator to share additional details about her daughter’s work.

The facilitator then transitioned to a warm-up exercise to engage the QI members in different kinds of conversations in order to help them get to know each other better and feet safe in sharing.

The facilitator asked everyone to “share something about your grandmother.” With hardly a pause, one person said that her grandmother was an “angel” and gave details of what made her such a “wonderful person”. This elicited very positive reactions from other members. With some additional prompting from the facilitator and assurance that this was a voluntary exercise, others continued to share personal stories of their own grandmothers. At one point, the facilitator interjected, “I’m going to have to disagree … because my grandmother is the best!” The whole team laughed. She then described how her grandmother had taught her to cook, although she has always resented having to cook after work. This revelation sparked more laughter,

Even though this exercise took 7–8 minutes, it proved to be a valuable use of time; the facilitator’s authentic self-disclosure fostered an open, safe environment as evidenced by most team members responding with their own personal recollections. By simply admitting to the team that she was ‘frazzled’ the facilitator displayed her willingness to be vulnerable and genuine - qualities that help break down barriers that inhibit communication and relationship-building (see Jenkins & Jenkins, 2006 for additional useful discussion).

Case Example: Emergence

As a small office with a solo physician and four staff members, Falls Practice was struggling to reorganize alter the dissolution of a tense partnership. Business matters were shaky, yet the physician was committed to implementing changes that would increase their capacity to care for patients. In early QI meeting discussions, the team realized that they needed to sort out job responsibilities and realign daily tasks.

At the 5th QI meeting, the facilitator entered the practice with an established agenda in mind and had written the agenda up on a flip chart The first agenda item asked QI team members to report on office processes. In a field note, the facilitator wrote:

I asked the staff if they had time to collect data on the slow points over the past week. They hadn’t. The nurse said ‘I think it’s a little better. But the pressure is on. He [the physician] gets too mad (she looked at him and corrected her word choice) or annoyed if I talk too long with the patients.’

This remark opened up deeply held feelings. The facilitator interjected carefully, asking for more clarification. At one point, the physician explained his approach to scheduling. Throughout this discussion, the facilitator made a number of specific interjections, emphasizing to the team that this should be a safe place to air these feelings.

Opting to let the discussion continue rather than strictly following the agenda, the facilitator attempted to tease out the meaning beneath the team members’ emotional statements. Doing so created an opportunity for the physician to acknowledge the difficulties of their transition and his appreciative understanding of their pressures. He stated:

“My point is, I know where you’re coming from… What can we do, bow can we make it easier? Alert other people so they can help out.… You know what? You’re going to have a lot of stuff coming at you from all sides. I just need ideas from you. Give me solutions. How can we solve this? Maybe we should do this or that … When I come to you and I seem upset, you might feel it’s something you’ve done … (but) I’m not trying to beat up on you.”

During this intense exchange, a number of important issues were addressed including time management, patient wait time and opportunities for patient education. In a field note, the facilitator wrote that this meeting represented a breakthrough in their communications and trust.

We contend that the groundwork for emergence occurs in an additive way throughout the QI cycle, as facilitators encourage team members to trust each other and to speak their minds and share ideas.

Case example: Conflict

The agenda for the 4th QI meeting at Meadow Practice focused on prioritizing their QI action items. Halfway through the meeting, the team reached a tie vote on the two top issues. The facilitator asked how the team wanted to proceed, and several suggestions were offered. The lead physician then said (jokingly): “Anybody who voted for [the second issue] as their priority - they’re all fired!” One staff member laughed uneasily, (The facilitator was already familiar with the physician’s tendency to joke with practice members. He had documented several of the doctor’s jabs from previous meetings.) At this point, the facilitator decided to intervene and said: “I have to ask …(stumbling for words). I know you’re just joking but … (chuckling with some uneasiness).” Other team members appeared anxious. The physician immediately replied (also struggling with his response): “I, I, ah, 1,1, no, I, they realize I’m joking too, you know.” The facilitator interjected again, “Are you sure though? At this, the nurse practitioner chimed in: “Some truths are said in jest.” After a slight pause and chuckles from die group, me physician added (with a genuine tone): “But anyway, no, I’m, I’m obviously just kidding. I have no problem with [the order of the action items], I think [both of] those [items] are very important …” The team then moved into a productive discussion of how to begin work on the first issue.

Though the facilitator struggled and ultimately intervened with minimal words, he drew attention to the potential power of the physician’s comments. Since the lead physician had organizational authority over the practice members, his humor was counterproductive to group cohesion and trust. While the facilitator’s intention was not to stifle humor, his comments made die physician aware of the impact of his joking on practice members.

Case example: Safety

Facilitators are able to help groups establish ground rules or agreements for how they interact with each other. By articulating these guidelines early in the process, group members can become more comfortable speaking up without feeling intimidated by more powerful members of the system. Facilitator field notes from an early QI meeting at Oak Practice demonstrate this point:

Before the close of the meeting, Priscilla (the clerical supervisor) made a statement regarding her concerns for open-minded and honest discussions. She wanted confirmation from the providers that there would be no repercussion regarding comments made during the team meeting. She directed her question to the general group but her eyes were focused on the provider. The provider looked directly at Priscilla as she offered her concerns. The provider’s body was leaning forward, and she was looking directly at Priscilla, acknowledging what she said by nodding. Priscilla reiterated that she wanted to make sure that no one would be fired after sharing comments. The group laughed, but the practice leader acknowledged that no one would be fired from open and honest discussions. She encouraged all members to share whatever they would want to share.

Facilitators may work with newly-formed groups where members know little about one another as well as groups that have members with long-standing relationships. In many of the PCPs under study, practice members knew one another through their working relationships but the QI team was a newly-formed group. Facilitators did not assume that members would be comfortable sharing ideas or raising dissenting opinions. Facilitating a discussion of ground rules early in the process provided the opportunity for team members to air concerns and for practice leaders to address these concerns. In many cases, practice leaders helped lay a foundation for open and safe discussions by directly stating that there wouldn’t be repercussions and that they welcome team members’ opinions and ideas.

Case example: Reflection

The QI team at Valley Practice began their ninth meeting by addressing their major issue of transitioning from paper files to an electronic medical record (EMR) system. Almost in passing, the Information Technology (IT) specialist mentioned that the system would be down for about six days later that month because the hospital’s system was scheduled for an upgrade. “This doesn’t necessarily have to stop the training process,” she added. Team members nodded in agreement and the conversation moved on. But the facilitator stepped in and began asking questions about the scheduled shutdown: Was there a plan in place to guide the practice in its absence? How might this cause unexpected problems?

The IT specialist answered that it wouldn’t be a problem: everything would just have to be done on paper that week. A resident commented, “T don’t think most residents are even aware of this.” The facilitator then asked how the shutdown would affect people’s jobs and began calling on each team member in turn. As each team member began reflecting on how this shut down would affect him/her, it became clear that the impact would be significant and the conversation shifted to figuring out how they should prepare for the shut down. As they brainstormed ideas, the facilitator wrote them on the flipchart and interjected questions to help concretize their plans. When the plan was finalized, the facilitator verbally summarized it, naming who would be doing what and when each task would be completed. The facilitator told the team that at the next meeting, they would revisit each step and reflect on how it was handled.

As promised, QI meeting #10 started with a check-in. Right away, the IT specialist said, “This was a true test of the practice. We had people all over the place. Some of the residents were at the hospital, but even though there were plenty of phone calls, it went pretty smoothly.” A physician added, “There were no major problems and the paper system we developed to use in place of the EMR worked fine.” “So the forms were created and distributed?” the facilitator asked. Several team members called out, “Yes!” One nurse spoke up: “Also, the doctors were excellent and prescription refills were done quickly.” The facilitator asked front office staff for their opinions, and they responded that their plans worked very well. Other team members nodded in confirmation. Then the practice leader asked, “What I want to know is this: will this system work again at the next downturn?” Various team members shouted out: “Yes! Yes! We have the forms now.”

Finally, the facilitator posed a last key question: “What did you learn from this?” One person answered, “There was LOTS of great teamwork between sectors and that really got the frustration level down.” A physician added, “Making a plan was huge! Especially having the whole practice on board for it.” Others agreed, and the team expressed enthusiasm for the successful outcome of their planning. In subsequent meetings, they recollected their success in dealing with this major change, and morale continued to rise as the QI cycle moved towards closure.

In this example, the facilitator reacted in the moment to a comment that was said in passing. Certainly, the facilitator did not anticipate facilitating this particular conversation that day, but realized potential implications for the system shut down. Asking “obvious” questions (e.g., how will the shut down affect their work?) prompted each team member to reflect on this. And returning the next week after the plans had been implemented, the facilitator prompted the team to reflect again on how things went. While facilitators want teams to accomplish agreed-upon QI plans, providing time and space for reflection wherein team members publicly share personal thoughts on the team’s efforts is evidence of success as well.

Discussion

Acting in the moment, facilitators can sometimes look like heroes or fools. Mastering facilitation tools and techniques is not always sufficient preparation for what takes place in a team meeting. Yet sometimes a facilitator’s impact can be effective despite its awkwardness or stumbling articulation. This paper offers some unique empirical analyses of how facilitators intervened in the moment during a QI process.

Our findings suggest that as intentional, external facilitators work with groups to achieve their goals, they do not tend to help groups through a linear progression of stages or events. Because”…groups must constantly address similar issues and problems at multiple time periods and settings” (Smith, 2001, p. 25), facilitators are prompted to act in the moment to respond to group conflict as it surfaces, and to give steady doses of encouragement and supportive feedback at each meeting so that group members can feet good about even small steps of success.

Using our 10 facilitator impact codes derived from the data, we presented a global picture of how often facilitators intervened over the 12-week meetings. We found that certain actions (such as intervening on ‘process’ and ‘safety’ matters) predominated in early meetings which then steadily declined. Facilitators’ frequency of intervention in the areas of ‘reflection’, ‘conflict’, ‘socialization of new skills’, and ‘task identification/completion’ varied over the course of the meetings. Interventions in these areas peaked during the middle of the 12-week timeline suggesting that as teams became more engaged in implementing their QI plans, facilitators became more active in prompting teams to reflect on what was working and what could be improved, managing disagreements and conflict, engaging teams in learning appropriate facilitation and organizational change skills, and helping teams stay on top of their agreed-upon QI tasks. Finally, in the areas of ‘self-disclosure,’ ‘humor’, ‘emergence,’ and ‘encouragement/support’, findings indicate low but constant levels of facilitator intervention during the meetings.

The case examples show what acting ‘in the moment’ looked like during the QI meetings and how these decisions impacted groups. We present these case examples so that the lessons available in the field notes and audio recordings may encourage others in their own facilitation work and that facilitators and researchers may glean insights for other contexts. The subjective aspects of facilitation - intricately tied to style, experience, and training - reveal the complexities of facilitating QI endeavors as facilitators must constantly listen to the content of discussions, gauge group dynamics, and pay attention to larger group processes. Over the course of a single meeting, facilitators must make multiple decisions ‘in the moment’ about when and how to act. Our findings provide insights into the benefits that intentional, external facilitators offer work groups as they are able to ask critical questions, hold people accountable, and even “see” processes or dynamics that the insider may not.

We invite questions and open dialogue about our MAP/RAP model relative to other facilitation work and organizational improvement models. We acknowledge that the specifics of ULTRA create both benefits and limitations for this study. While it provides us with a delimited timeframe to analyze naturally-occurring work groups engaged in QI efforts, this structure may limit how generalizable the findings are to other kinds of facilitation endeavors. ULTRA required that participating practices undergo an intense QI process by meeting with a facilitator for 12 weeks. If other kinds of work groups have different meeting structures and timeframes, our findings may look different from these types of groups. Also, because ULTRA facilitators were not hired as “professional, certified facilitators,” their decisions and actions in the moment while facilitating may likely differ from more seasoned facilitators. Moreover, we can say little about how variations in facilitator style impacted our findings. While we analyzed QI meetings from multiple practices for each facilitator, we were limited in our ability to systematically assess individual facilitator attributes. Future research should continue this line of inquiry to better understand what is the experience for each facilitator in the moments just preceding the ‘action in the moment’ (e.g., what is it that allows them to see what has emerged in the group, what do they draw on, and what assists them to intervene?) Nevertheless, this paper provides important empirical findings for facilitators and researchers on group development and organizational change, and presents a clearer picture of how external facilitators act ‘in the moment’ and impact QI teams over time.

Biographies

Eric K. Shaw, PhD is an Assistant Professor at the University of Medicine and Dentistry of New Jersey – Robert Wood Johnson Medical School. Dr. Shaw completed his PhD in sociology at Rutgers University where he conducted qualitative research on helping behaviors in various contexts. In 2003, he joined the Department of Family Medicine at UMDNJ as a facilitator for the ULTRA study. He is now the project director for a National Cancer Institute-funded grant that uses external facilitators and quality improvement learning collaboratives to increase cancer screening rates in primary care practices. He also works with the University of Colorado Health Sciences Center on a facilitated intervention to improve diabetes care in primary care practices. Contact Eric K. Shaw, Department of Family Medicine, UMDNJ-Robert Wood Johnson Medical School, 1 World’s Fair Drive, Somerset, New Jersey, USA 08873, Phone: 732-743-3274, Fax: 732-743-3395, shawek@umdnj.edu

J. Anna Looney, PhD is an Assistant Professor in the Pre-Doctoral Education division of the Department of Family Medicine, Robert Wood Johnson Medical School. She holds Masters degrees in English Literature and in Sociology, and a doctorate in Sociology from Rutgers University. Dr. Looney is responsible for Community-based Education for medical students which enables her to work collaboratively with organizations and agencies serving the under-served. Dr. Looney has research experience in quality improvement interventions in primary care practices as an ethnographic data collector and facilitator. She mentors medical students in community service projects through the Distinction in Service to the Community program. Recently, Dr. Looney was named Director of the New Jersey Family Medicine Residency Network. She has presented her research at STFM and NAPCRG conferences. Contact J. Anna Looney, Dept of Family Medicine, UMDNJ-RWJMS, 675 Hoes Lane N107, Piscataway, NJ, USA 08854, Ph: 732-235-4200, Fax: 732-235-4202, looneyja@umdnj.edu

Sabrina Chase, PhD is a medical anthropologist and Post-Doctoral Research Fellow in the Research Division of the Department of Family Medicine, Robert Wood Johnson Medical School in New Brunswick, N.J. She received her Ph.D. in medical anthropology at Rutgers University in January 2005. Her dissertation traced the help-seeking pathways of urban HIV+ Puerto Rican women living in the greater Newark, New Jersey metropolitan area. As a Fellow, Sabrina has worked with both the ULTRA (Using Learning Teams for Reflective Adaptation) and SCOPE (Supporting Colorectal Cancer Outcomes through Participatory Enhancements) projects as a facilitator and quality improvement team leader. Contact: Sabrina Chase, Department of Family Medicine, Research Division, Robert Wood Johnson Medical School, 1 World’s Fair Dr., Somerset, New Jersey, USA 08873, Phone: 732-309-5032, Fax: 732-743-3395, chasesm@umdnj.edu

Rohini Navalekar, MPA is an Oncology Clinical research coordinator at the Cancer Institute of New Jersey. She holds a Masters degree in Public Administration from the University of Louisville and a Bachelor of Environmental Engineering degree from L.D. College of Engineering, India. Before joining the Cancer Institute of New Jersey, Rohini worked as a Research Teaching Specialist at the Department of Family Medicine as a member of project SCOPE (Supporting Colorectal Cancer Outcomes through Participatory Enhancements). She consented and recruited more than 600 patients in 24 different family practices. After completing her Masters Degree, she worked as Research Interviewer for the Southern Community Cohort Study Cancer Research Project, a landmark study of racial disparities in cancer incidence and mortality, at Family Health Centers, Inc. Louisville, KY. She consented and interviewed more than 1700 patients about their health and lifestyle at the Community Health Center. Her previous work also includes a public sector internship on Bio-Terrorism Preparedness Planning at Louisville Metro Health Department, Louisville, KY. Contact Rohini Navalekar, Cancer Institute of New Jersey, 195 Little Albany Street, New Brunswick, NJ USA 08901, Ph: 732-235-4926, Fax: 732-235-7690, navaleru@umdnj.edu

Brian Stello, MD, is director of the Eastern Pennsylvania Inquiry Collaborative/Network (EPICnet), a practice-based research network at Lehigh Valley Hospital in Allentown, PA. He is a graduate of Jefferson Medical College. In 1994, Dr. Stello became a founding faculty member of the Family Medicine Residency, and an executive member of the Department of Family Medicine at Lehigh Valley Hospital, In 1999, he moved on to become the Director of Network Development for Family Medicine, with responsibilities for new practice development, new physician recruitment, and existing practice recruitment. His work on behalf of the Department of Family Medicine includes quality-improvement activities, physician education, and evidence-based medicine programming. He began participating as an investigator in practice-based research in 2000, and became director of EPICnet in 2006. Contact Brian Stello, Family Medicine, Lehigh Valley Hospital, PO Box 7017, Allentown, PA USA 18105-7017. Phone 610-969-4950. Fax 610-969-4952. brian.stello@lvh.com

Oliver Lontok, MD is a National Research Service Award (NRSA) Fellow at the UMDNJ Robert Wood Johnson Medical School (RWJMS) Department of Family Medicine. He earned his medical degree from the Universidad Autonoma de Guadalajara and a Masters in Public Health at the UMDNJ School of Public Health. His research interests include Preventive Medicine with a focus on cardiovascular and cancer care, and he has worked with faculty from the Cancer Institute of New Jersey, UMDNJ School of Public Health and RWJMS. As an NRSA Fellow, his education focuses on qualitative and quantitative methodologies. He plans to utilize his clinical skills and expertise in research methodologies to decrease disparities in access and treatment outcomes related to primary care. Contact Oliver Lontok, UMDNJ, Department of Family Medicine – Research Division, 1 World’s Fair Drive, Somerset, New Jersey USA 08873, Phone: (732) 743-3368, Fax: (732) 743-3395, lontokol@umdnj.edu

Benjamin F. Crabtree, PhD is a medical anthropologist and Professor Somerset, NJ 08873and Director of Research in Family Medicine, UMDNJ-Robert Wood Johnson Medical School. He is also Program Co-Leader for Population Science at Cancer Institute of New Jersey. He has contributed to numerous articles and chapters on both qualitative and quantitative methods covering topics ranging from time series analysis and log-linear models to in-depth interviews, case study research, and qualitative analysis strategies. He is co-editor on two books, Doing Qualitative Research, now in its 2nd Edition, and Exploring Collaborative Research in Primary Care. He has received funding from AHRQ, NCI, and NHLBI to support his research on organizational change strategies for enhancing quality of care in primary care practices using a conceptual framework based on complexity science concepts. Contact Benjamin F. Crabtree, UMDNJ Robert Wood Johnson Medical School, Department of Family Medicine – Research Division, 1 World’s Fair Drive, Somerset, New Jersey USA 08873, Phone: (732) 743-3220, Fax: (732) 743-3395, crabtrbf@umdnj.edu