Contrasting Perspectives of Practice Leaders and Practice Facilitators May Be Common in Quality Improvement Initiatives

Abstract

Practice facilitation is an approach for advancing quality improvement (QI), but its success depends on effective relationships and communication among participants. Our goal was to identify patterns of contrasting perspectives on implementation issues between practice leaders and their practice facilitators, and factors that may contribute to them. We conducted individual interviews with practice leaders and the practice facilitators assigned to them as part of a program focused on preventive cardiology within primary care practices. We used summative content analysis to quantify the incidence of contrasting perspectives, and bivariate and qualitative analyses to explore relationships between contrasting perspectives and contextual factors. Among the 16 dyads, contrasting perspectives commonly related to the easiest or hardest QI interventions to implement (8 of 16 dyads) and the practice’s success implementing interventions (5 of 16 dyads). There was a nonsignificant, inverse correlation (r = −0.19, p = .47) between the number of in-person QI visits from the facilitator and the total number of contrasting perspectives. Turnover of staff was frequently reported in dyads with contrasting perspectives. Although the impact of contrasting perspectives warrants additional study, planners of QI initiatives using practice facilitation should consider taking steps to minimize contrasting perspectives, or the potential adverse consequences of them, by addressing turnover challenges and encouraging opportunities to share perspectives.

Introduction

Practice facilitation, sometimes referred to as practice coaching, is a scientifically supported approach for implementing quality improvement (QI), particularly for small, independent physician practices with limited resources.^1–5 Practice facilitators are specially trained individuals who help practices engage in QI projects and develop capacity for continuous QI.^6,7 They offer practices, a variety of supportive services, for example, using practice-level data to drive change, implementing best practices in QI structures and methods, and capacity building in the use of health information technology to support improved clinical care and office efficiency.⁸ Practice facilitator-supported QI initiatives have been shown to be positively associated with evidence-based guideline adoption within primary care,^4,9 improved patient outcomes for chronic conditions,¹⁰ and may be cost-neutral if they result in two fewer hospital admissions per practice per year.²

Given the promise of practice facilitator-supported QI, a number of prominent stakeholders have encouraged the use of practice facilitation and made resources available to encourage implementation.^11–13 These resources emphasize that practice leaders and practice facilitators play complimentary and vital roles in QI initiatives, and that effective facilitation is based on mutual trust and good communication.^11,14 However, little is known about the relationships between practice leaders and practice facilitators, including areas of agreement and disagreement. Indeed, qualitative assessments of practice facilitator-supported QI initiatives typically include interviews with either practice leaders or practice facilitators—not both.^15–17

Our goal was to identify and describe contrasting perspectives between practice leaders and practice facilitators regarding implementation of a practice facilitator-supported preventive cardiology QI initiative. By examining our qualitative data in pairs, we aimed to identify patterns of contrasting perspectives between practice leaders and their practice facilitator partners, and the contextual factors that may contribute to contrasting perspectives. Incidence of contrast may signal a breakdown in communication, and information about contrasting perspectives may be used to inform future efforts and trainings.

Methods

Intervention

Healthy Hearts in the Heartland (H3) was a research program that worked with small- and medium-sized primary care practices to implement and evaluate QI strategies for cardiovascular disease preventive care using practice facilitation. Practices were eligible to participate if they had fewer than 20 primary care clinicians and were located in Indiana, Illinois, or Wisconsin. All participating practices were assigned a practice facilitator for 12 months who met with practices as often as requested, ideally at least once a month. Practice facilitators received structured training on clinical topics and QI strategies related to four common quality measures (additional details in Supplemental Digital Content 1, Appendix 1, https://links.lww.com/JHQ/A98): antiplatelet therapy for ischemic vascular disease, blood pressure management, cholesterol management, and smoking assessment and cessation (ABCS). Quality measures in these domains are used in national quality incentive programs, such as Meaningful Use, the Physician Quality Reporting Program for the Centers for Medicare and Medicaid Services, and Accountable Care Organizations Shared Savings Programs.^18,19 The practice facilitators’ training included formal, weekly, web-based sessions delivered by subject matter experts for 12 months, and three 2-day interactive in-person sessions over 15 months. Practice facilitator-led discussion sessions, both virtual and in-person, were held weekly over 24 months and then became biweekly.

Practice facilitators had a menu of QI strategies that practices could choose from related to the four clinical ABCS topics. The QI strategies (detailed in Supplemental Digital Content 2, Appendix 2, https://links.lww.com/JHQ/A99) included audit and feedback, clinical decision support within the electronic health record, standing orders, workflow improvements, and patient education and outreach. H3 is funded by the Agency for Healthcare Research and Quality as part of the EvidenceNOW initiative, which is dedicated to improving the heart health of Americans.²⁰

Study Participants and Recruitment

The 226 H3 practices were grouped into four waves based on the date they were recruited for H3, and each wave began their 12-month interventions approximately 3 months apart. Wave 2 practices were the focus of this qualitative study. The wave included 40 practices, although only 33 were actively participating in H3 after 9 months when we began recruitment for the qualitative interviews. Practice leaders from the 33 practices received up to six contact attempts through phone, email, or fax asking them to complete a one-time, 30-minute telephone interview. Once we received a commitment from the practice leader, we invited the corresponding practice facilitator to complete a separate telephone interview. We conducted individual interviews, rather than joint interviews, because it enables each participant to offer their own perspective, without having to consider the reaction of the other when discussing sensitive topics or providing criticism.^21,22 The interviews were conducted between March and May 2017, which corresponded to the final 3 months of the practices’ 12-month practice facilitation period. Interviews were digitally recorded and transcribed. This study was approved by our university’s Institutional Review Board (STU00202126).

Data Collection and Analysis

We used semistructured interview protocols constructed based on the Consolidated Framework for Implementation Research (CFIR),²³ a conceptual framework developed to guide assessment of implementation contexts to identify factors that might influence intervention implementation.²⁴ The framework considers how the design of the intervention, the QI process, the internal and external environments, and the characteristics of individuals influenced program implementation. The interview questions were worded similarly for both practice leaders and practice facilitators, and the protocols are accessible in Supplemental Digital Content 3 (see Appendix 3, https://links.lww.com/JHQ/A100). Two experienced qualitative researchers (M.M. and T.B.) led the interviews.

We began our analysis by creating and applying deductive codes, corresponding to the CFIR, in Atlas.ti, a qualitative software program.^25,26 We analyzed the data to identify general themes about the enablers and barriers to implementation of H3, which we reported on elsewhere.²⁷ Next, for the dyadic analysis, M.M. and T.B. independently read the transcribed responses from each dyad (a practice leader and his or her assigned practice facilitator) arranged side-by-side in Atlas.ti. Based on that reading, the two authors developed a coding scheme to identify examples of contrasting perspectives within a dyad.²¹ Specifically, contrasting perspectives were tagged as “major” or “minor.” Text was tagged as a “major” contrasting perspective when responses from the practice facilitator and practice leader clearly contradicted each other (Table 1). Text was tagged as a “minor” contrasting perspective when there was a clear difference in perspective, but not necessarily a contradictory one. All dyads were coded independently by the two authors, every instance of a contrasting perspective was discussed, and disagreements were resolved through discussion.

Table 1.

Examples of Major and Minor Contrasting Perspectives

Major Contrasting Perspectives	Question: How well prepared do you think the [your] practice was to adopt H3 changes and tools?
	Practice Facilitator: “I don’t think the practice was prepared.”
	Practice Leader: “We were very ready.”
Minor Contrasting Perspectives	Question: To what extent were you able to develop a plan at the outset for implementing H3 changes or tools at the [your] practice?
	Practice Facilitator: “My plan was pretty loose. I got the reports quarterly and talked through them with the providers. I don’t think I did a good job or pushed the practice to aim for clear goals over a certain timeline.”
	Practice Leader: “We did have a plan. [Our practice facilitator] is so good s/he doesn’t do anything that isn’t going to work, I can tell you that.”

We used summative content analysis to quantify the incidence of contrasting perspectives, and to identify patterns and the underlying context of their occurrence.²⁸ Using Stata, we calculated Pearson correlation coefficients to investigate whether the number of contrasting perspectives was related to the number of QI visits, the practice facilitator made to the practice, and t-tests to examine the relationship whether there was a difference in number of contrasting perspectives, based on whether the practice leader and practice facilitator had worked together previously. We also categorized practices into three equal-sized groups (high, medium, and low) based on their average improvement across all four ABCS measures after 12 months of participation in H3. We used one-way analysis of variance to determine whether there was a difference in the average number of contrasting perspectives across improvement groups. Finally, we report on the themes from our qualitative results that may have contributed to contrasting perspectives between practice facilitators and practice leaders.

Results

Characteristics of Respondents

We completed interviews with practice leaders from 17 of the 33 eligible practices, and all 10 practice facilitators assigned to those practices (Five of the practice facilitators were assigned to multiple practices). We eliminated one dyad from the analysis because responses from one of the dyad members were incomplete. Ultimately, our analysis included 16 dyads. Seven of the practice leaders were physicians, six were nurses, and three were practice managers. Most were from small practices, with a median of three clinicians, and on average, the practices received approximately six QI visits from their practice facilitator before our interviews. Practices that chose to participate were representative of the overall group of practices in Wave 2 (Table 2).

Table 2.

Characteristics of Participating Practices and All Wave 2 Practices

	Characteristics of practice leader interviewees and their practices (n = 16)	All Wave 2 practices (n = 33)
No. of providers in the practice	Median: 2.5 (range 1–10)	Median: 2.0 (range 1–10)
Part of larger health system, % yes	5 (31%)	7 (21%)
State
IN	6 (38%)	12 (36%)
IL	7 (44%)	17 (51%)
WI	3 (19%)	4 (12%)
No. of H3 QI encounters	Mean: 6.8 (SD: 2.4); median: 6.5 (range 4–13)	Mean: 6.4 (SD: 3.0); median: 6.0 (range 0–14)
Practice leader had worked with the practice before H3, % yes	4 (25%)	Not available

QI, quality improvement.

Among the 10 practice facilitators, three had 0–2 years of experience working as a practice facilitator prior to H3, 5 had 3–5 years of experience, and 2 had 6 or more years of experience. Five practice facilitators had backgrounds in QI, four in health information technology (e.g., a meaningful use certification specialist), and three had clinical backgrounds (e.g., a nurse).

Incidence of Disconnect

There was a mean of 2.3 instances of contrasting perspectives per dyad (Table 3). Minor contrasting perspectives were more common than major contrasting perspectives. Contrasting perspectives were most commonly related to the easiest or hardest H3 interventions to implement (8 of 16 dyads) and the practice’s success (or lack thereof) implementing H3 interventions (5 of 16 dyads) (Table 4). There was a nonsignificant, inverse correlation between the number of in-person QI visits made by the practice facilitator and the number of major (r = 0–.35, p = .19), and total (r = −0.19, p = .47) contrasting perspectives, and there was essentially no correlation with the number of minor (r = 0.02, p = .94) contrasting perspectives. There was no difference in the number of major (t = −1.42; p = .18), minor (z = −.83; p = .42), or total (t = −1.53; p = .14) contrasting perspectives based on whether the practice leader and practice facilitator worked together before H3. There was also no difference by improvement group in the dyads’ number of major (F = 0.90; p = .47), minor (F = 0.18; p = .94), or total (F = 1.56; p = .26) contrasting perspectives.

Table 3.

Incidence of Contrasting Perspectives per Dyad

Type of contrasting perspective	Incidence of contrasting perspectives (N = 16)
	Mean	Median	Range
Major	0.7	1	0–3
Minor	1.6	2	0–4
Total	2.3	2	0–5

Table 4.

Topics With the Highest Incidence of Contrasting Perspectives

Topic	No. of dyads (and percent) where there was a major or minor contrasting perspective (N = 16)
Easiest/Hardest H3 interventions to implement	8 (50%)
Success of the practice in implementing H3 interventions	5 (31%)
Characteristics of the practice that helped with H3 implementation	4 (25%)
Whether H3 was a good fit for the practice	4 (25%)
Whether H3 was a priority within the practice	4 (25%)
Whether there was an implementation plan at the outset of H3	4 (25%)

Contextual Information Surrounding Contrasting Perspectives

Qualitative findings suggest that two contextual factors—turnover of staff and a general misunderstanding of the nature of H3—may have contributed to contrasting perspectives. Within dyads that had a contrasting perspective, some practice facilitators remarked about a key individual leaving the practice, for example, “The person who was most engaged [in H3] left the practice,” and “I did have a plan with the chief medical officer, but she left right after the kick-off.” Similarly, among the dyads that had a contrasting perspective, a few of the practice leaders noted that H3 did not meet their expectations, for example, “H3 wasn’t what we envisioned,” and “We thought H3 was going to be run by cardiovascular researchers or educators who could come in to talk to us about guidelines.”

Limitations

Our findings highlight the importance of collecting perspectives of multiple participants in a QI effort to gain a more holistic picture of implementation issues. An important limitation is that our analysis relied on the recollection and perceptions of only practice leaders and facilitators. Although we believe these individuals are best positioned to report on implementation of H3, other staff within the practices may have offered different perspectives. Also, our study was the first to investigate differences in perspectives between practice leaders and practice facilitators, and, as such, we created definitions of major and minor contrasting perspectives. Alternative definitions may have yielded a different count of contrasting perspectives. We do not have information on all possible causes of contrasting perspectives (e.g., personality conflicts) or on the incidence of contrasting perspectives among dyads that were not interviewed for this study. Finally, the small number of dyads in this study limited our ability to detect significant associations. Larger sample sizes are needed to better explore these relationships. In particular, we caution readers against concluding that contrasting perspectives have little effect on outcomes.

Discussion

Practice facilitation holds great promise for advancing care quality in primary care practices cost effectively,^2,4 but like most QI interventions, its success depends on effective relationships and communication among participants.²⁹ Although there is widespread agreement about the importance of good relationships among key participants, they are often unexamined in QI evaluations, and rarely included in frameworks or theories of change.³⁰ As a first step toward addressing this gap, we examined differences in perspectives among practice leaders and practice facilitators participating in H3.

Our findings indicate that contrasting perspectives were common among practice leaders and practice facilitators participating in the H3 QI program. Although some differences in perspectives are to be expected, contrasting perspectives on key issues, like whether the practice was successful in implementing H3 or whether the practice was a good fit for H3, may signal a breakdown in communication. This breakdown could potentially hinder the implementation and success of practice facilitator-led QI interventions. For example, if practice facilitators do not have a clear understanding of the interventions that practice leaders find most difficult to implement, they may not make sufficient efforts to identify strategies for practice leaders to overcome their challenges. A better understanding of practice leaders’ perspectives may help practice facilitators conduct more effective implementation planning.

Planners of practice facilitator-led QI interventions should consider strategies for addressing differences in perspectives between practice leaders and practice facilitators. For example, practice facilitators could receive additional training in motivational interviewing to support improved communication with practice leaders and to gain a better understanding of their needs, goals, and concerns. Similarly, practice facilitators could be encouraged to share their reflections on the practices’ progress toward implementation and request similar reflections from practice leaders. In addition, planners should try to address some of the issues that may drive contrasting perspectives, such as having practice facilitators establish relationships with multiple individuals within a practice, so that turnover is less likely to disrupt their work.

Although the small number of dyads studied makes our estimations imprecise, our results suggest that there may be an inverse relationship between the incidence of total contrasting perspectives and the number of in-person QI visits made by the practice facilitator to the practice. Although encouraging more in-person visits is an obvious recommendation, findings from H3 and other practice facilitator-led QI interventions indicate that it is often difficult for practice facilitators to gain regular entry into busy practices.^16,27,31 In addition, the analyses exploring the presence of a previous relationship between the practice facilitator and practice leader and incidence of contrasting perspectives were also nonsignificant, but the direction suggested that there may be a benefit to longer-term relationships between practice leaders and practice facilitators. These relationships are in need of further study.

Further study is needed to investigate the link between contrasting perspectives and improvement on QI outcomes, and how other factors known to influence outcomes (e.g., resources and leadership) might moderate that relationship. The linkage between contrasting perspectives and outcomes was preliminarily explored in this study; however, additional analyses with larger sample sizes are needed. If contrasting perspectives are associated with reduced efficacy of this facilitator-led QI initiative, then it will be important for future program planners and participants to have a better understanding of the factors driving contrasting perspectives.

Conclusions

We found that contrasting perspectives about the implementation process were common among practice leaders and their assigned practice facilitators working together in this QI initiative. In particular, practice leaders and facilitators most frequently held contrasting perspectives related to the easiest and most difficult H3 interventions to implement and the practice’s success (or lack thereof) implementing H3 interventions. These contrasting perceptions may be driven by changes in staffing and a general misunderstanding of the nature of H3.

Implications

Planners of practice facilitator-led QI initiatives may consider taking steps to minimize the incidence of contrasting perspectives, or the potential adverse consequences of them by addressing turnover challenges and encouraging opportunities to share perspectives. Future work should explore the implications of contrasting perspectives on QI outcomes.

Biographies

Megan McHugh, PhD, is an associate professor of Emergency Medicine at Northwestern University.

Tiffany Brown, MPH, is a research project manager at Northwestern University.

Theresa L. Walunas, PhD, is an assistant professor of Medicine and associate director for the Center for Health Information Partnerships.

David T. Liss, PhD, is a research assistant professor of Medicine at Northwestern University.

Stephen D. Persell, MD, MPH, is an associate professor of Medicine at Northwestern University.