Provider and Coach Perspectives on Implementing Shadow Coaching to Improve Provider-Patient Interactions

Abstract

BACKGROUND.

Healthcare organizations want to improve patient care experiences. Some use “shadow coaching” to improve interactions between providers and patients. A Federally Qualified Health Center (FQHC) implemented a half-day observation of individual primary-care providers by a “shadow coach” during real-time patient visits, including an in-person verbal debrief afterwards and a written report with specific recommendations. Shadow coaching identified areas for improvement. We aimed to characterize lessons and barriers to implementing shadow coaching as a mechanism to improve interactions with patients and change organizational culture.

METHODS.

We examined provider and coach perceptions of shadow coaching through interviewing coached providers, stratified by provider type and Consumer Assessment of Healthcare Providers and Systems (CAHPS®) performance, and the coaches who coached the most providers. We interviewed 19 coached providers and 2 coaches in a large, urban FQHC. Content analysis identified implementation barriers, facilitators and themes.

RESULTS.

Coaches reported needing “buy-in” throughout the organization and the need to be credible and empathize with the providers being coached. Most providers reported behavior changes based on recommendations. Almost all providers recalled at least one coaching recommendation that was actionable. Providers and coaches highlighted patient-level and practice-level barriers that impeded their ability to implement recommended improvements. CAHPS data was reported as an effective performance management metric for measuring change, counseling providers, and evaluating provider-level efforts but was not always specific enough to yield tangible recommendations.

CONCLUSIONS.

Regular messaging by leadership about the priority and purpose of shadow coaching was essential for both physician engagement and its mature implementation across the organization. Coaching could be embedded into a long-term strategy of professional development with periodic re-coaching. Re-coaching sessions could target issues raised by providers, such as dealing with difficult patients or specific populations. Research on the timing and content of re-coaching is needed.

1. INTRODUCTION

Healthcare organizations want to improve patient care and experiences.^1-3 Their efforts have included building data monitoring systems using data such as the Consumer Assessment of Healthcare Providers and Systems (CAHPS®) patient experience survey data to provide insights on quality of care and to identify areas for improvement across site locations and providers within an organization.⁴ One method is for organizations to take data insights and communicate potential improvements to providers as a method of one-on-one provider counseling or coaching.⁵

Coaching can range from providing recommendations to providers based on quality scores, or using quality scores to identify providers in need of improvement and provide targeted interventions.^6-8 One such targeted intervention is “shadow coaching” which specifically involves selecting providers for coaching based on patient experience scores and commissioning coaches to observe health practitioners in real-time encounters at a point of care for various lengths of time.^5-12 Unlike traditional coaching, shadow coaching uses real patient encounters over an extended period of time to identify areas of improvement, and often includes follow up coaching to observe providers putting into practice earlier learnings. The coaches then provide structured, feedback to encourage target behaviors among providers.¹² The recommendations are specific to how an individual provider could improve their interactions with patients and emphasize the provider’s current strengths (i.e., encourage current beneficial behaviors) and amplify the provider’s capacity (i.e., recommend new behaviors for improvement).⁹ After observation, coaches draft recommendations for improvement and observe providers on implementation of recommendations.¹⁰

Previous literature examines shadow coaching in various forms,^13,14 including compliance training,^15,16 simulated patient encounters¹⁷ with both physicians and nurse practitioners,¹⁸ and individual provider coaching. Evidence focuses on the motivations¹⁹ for and impacts of shadow coaching on the competencies^20,21 and providers’ behavior.²² It is less clear on how to implement a shadow coaching program in general or specifically to monitor care quality and provider-patient interactions. Evidence is lacking on what facilitates or hinders shadow coaching in transforming the organization’s culture (and all providers) to focus on continual improvement and quality.

We examined the perspectives of coaches and coached providers on the implementation of shadow coaching as a quality improvement program to identify implementation barriers and insights on how to use shadow coaching to change organizational culture.

2. METHODS

2.1. Setting.

The study was conducted in a large, urban Federally Qualified Health Center (FQHC) with 44 primary care practices employing more than 300 providers and receiving nearly 1 million patient visits annually. The FHQC’s practices have fewer resources and their providers receive lower salaries than in other primary care practices in the local area, evidenced by the FQHC’s 15-20% annual turnover rate for providers. The FQHC implemented a company-wide quality monitoring system based on the Clinician and Group CAHPS (CG-CAHPS) patient experience survey, using the 0-10 overall rating of the provider and the provider communication composite.²³ Three years later, the organization introduced shadow coaching as part of its quality monitoring and to improve patient experiences. Every 6 months, they identified “medium-performers” as eligible for coaching based on the 6-month rolling average of an individual provider’s overall provider rating (transformed to 0-100 possible range) being below 90. Those eligible received coaching from March 2015 to August 2018.

2.2. Intervention.

To implement shadow coaching, the organization selected providers who were consistent high performers on patient experience scores and had positive performance input from superiors to act as shadow coaches. There were eight full-time, high-performing, coaches in total. Coaches were provided training via a one-day coaching seminar hosted by the SullivanLuallin group.^10,24-26 Coaches observed providers for 4 or more patient encounters during a half-to-full day and provided verbal feedback after the observation time on provider strengths and areas for improvement with patients. Coaches also drafted a written report to the provider delineating feedback and summarizing comments and recommendations. Improvement was monitored in subsequent coaching sessions.^10-12

Shadow coaching focused on provider-patient interactions that a provider could improve when interacting and caring for patients, with a focus on communication, as it is the strongest driver of overall provider ratings on the CG-CAHPS survey. Coaches focused on the specific behaviors included in the provider communication composite as measured by the CG-CAHPS, including explains things in a way that is easy to understand, listens carefully to the patient, shows respect for what the patient has to say, and spends enough time with patient (using a four-point Never/Sometimes/Usually/Always response scale).²⁷

2.3. Sample and Data Collection.

In 2017 and 2018, 89 providers were eligible for and received shadow coaching. From these, 61 were active providers at the FQHC at the time of interviewing. We randomly sampled 35 providers (35/61, 57%), stratified by provider type and CAHPS overall provider rating. We informed those sampled about the study, made first contact, and recruited them via email.

We conducted hour-long semi-structured interviews (by DQ, NQ, AP) in July/August 2019 with 19 providers (19/35; 54%) about experiences with shadow coaching, including perceptions of its usefulness and barriers to implementation. We (DQ) also interviewed the two coaches (of eight) who coached the most providers across the two years in early September 2019. We asked the coaches how shadow coaching was set up, how they were selected, what training they received, and their perspectives on the implementation and success of coaching. Interviews were audio-recorded and transcribed for analysis.

2.4. Analytic Approach.

We entered transcripts into Dedoose,^28,29 a web application for analyzing qualitative data. We established codes that mapped to key research questions³⁰ and developed a code structure using systematic, inductive procedures to generate insights from responses.^31,32 Three researchers conducted such coding to identify topics,³³ coding early transcripts independently and refining the codebook.³⁰ We used regular team meetings to reach consensus on topics, identify discrepancies, refine concepts, codebook changes, define codes and dialogue about concepts and themes.³⁴ We employed interrater reliability exercises among the coding team to refine codes and descriptions. After code training, we compared differences among the coders in applying codes to interview text and obtained a pooled kappa coefficient of 0.80-0.85, indicating “very good” coder agreement.^35,36 We employed ongoing training for coding among the coding team on emerging sub-codes using the Dedoose training module. Subsequently, we conducted thematic analysis to identify main themes, such as key barriers and facilitators to implementing shadow coaching.

Study protocols were approved by RAND’s Human Subjects Protection Committee (IRB Assurance Number: FWA00003425; IRB Number: IRB00000051; Project ID:2018-0191).

3. RESULTS

Among the 19 interviewed, 12 were medical doctors (i.e. Doctor of Medicine (MD) or Doctor of Osteopathic Medicine (DO)), and seven were either nurse practitioners (NP) or physician assistants (PA). Interviewed providers saw an average of 22 patients per day and had an average 6-month mean on the CAHPS overall provider rating of 76 (on a 100-point scale) four to five months prior to coaching, indicating they were medium performers; high performers were those providers with a 90 or above. Providers’ tenure at their clinic averaged 6.5 years. We conducted a non-response bias analysis using chi-square tests or t-tests, as appropriate, to test for differences (p-value<0.05) between the 19 interviewed and those not interviewed by gender, provider type, specialty, primary language, or mean overall provider rating and found no differences.

3.1. Coach Perspectives on Implementation

Both coaches acknowledged several themes: Buy-in, prioritizing quality communication, and emphasizing all providers understanding the rationale for shadow coaching. Specifically, the need for buy-in among all providers, not just buy-in by those who received coaching, underlining the needed scope of messaging surrounding the rationale and purpose of coaching so that it supports a culture of improvement, rather than a culture of criticism. They indicated that the organization had purposefully put a premium on quality communication that includes quickly building rapport and trust between providers and patients. They mentioned that all providers needed to understand the rationale for shadow coaching and the priority that shadow coaching gave to improving patient experiences, specifically provider-patient interactions. Both coaches indicated that buy-in was needed both for successful implementation and for improving provider-patient interactions. One coach explained:

“The first critical element of shadow coaching is to communicate the vision and purpose very effectively to the entire provider group so that, number one, all providers recognize that this is something that is prioritized…and that it is an organizational strategy being offered to providers throughout the organization” –Coach-A, MD

Both coaches mentioned two key implementation factors related to the type of coach that is needed: empathy and credibility. Both indicated the characteristics used to select them as coaches, and what characteristics the FQHC sought. The chief characteristics were “having empathy and credibility with the providers being coached.” Successful coaches were able to have empathy with individual providers, knew how their clinics were run, understood the obstacles that providers faced, and knew first-hand the type of patient needs the providers were dealing with on a daily basis. As one coach said, the coaches needed to be “a peer who understands the unique challenges of the provider’s work situation.”

Both coaches also noted they needed credibility with those being coached. Coaches gained this credibility not only by having empathy and first-hand knowledge of providers working environment, but also having consistently high patient experience scores in communication as well as “evidently excellent interpersonal skills.” Coaches needed credibility to be able to offer recommendations on how to improve provider-patient interactions and coaches needed to be able to show their credibility while coaching. Both coaches said they and the providers wanted their recommendations to be clear and useful and relevant to the provider work environment, whether regarding new or existing behaviors.

“When giving constructive feedback, it should never be presented as a judgement of the provider’s personality or character traits, but presented as specific observations on what the provider wanted to demonstrate and what the provider actually…projected” –Coach-A, MD

Coaches identified several main barriers. Coaches mentioned cultural and language barriers as well as the number of patients seen in a given day as major impediments to improving provider-patient communication and to the effectiveness of shadow coaching. They added that available resources at the FQHC were less than they would be elsewhere. They lauded providers for their ability to perform as well as they do at their large, urban FQHC serving vulnerable populations.

Coaches had mixed perspectives about CAHPS scores. Coaches explained that CAHPS scores as a metric for the quality improvement program and coaching may lag behind actual behavior change in providers because of delays in collecting survey data, calculating scores, and providing results to providers. Both coaches noted that expectations of improved CAHPS scores should be made clear to all providers to specifically ensure their continued acceptance of shadow coaching and more broadly of the organization-wide assessment and monitoring quality framework. Using CAHPS scores as a basis for coaching was considered a key part of the organizational focus; integrating performance on CAHPS into the performance management approach by medical directors and organizational leadership of all individual providers was seen as critical. Coaches highlighted several metrics for assessing whether shadow coaching was effective, including the CAHPS overall rating, CAHPS provider communication score, and direct patient feedback. Coaches also utilized direct provider feedback to determine whether coaching was effective.

3.2. Provider Perspectives on Coaching and its Aim of Improving Provider-Patient Interactions

Seventeen (89%) of the 19 providers remembered being coached and 14 (82%) found the recommendations to be actionable. Recommendations had three main themes: engaging and spending time with patients, listening to patients, and ensuring information presented was easily understood (see Table 1).

Table 1:

Perceptions of Coaching by Provider

Provider ID/ Provider Type	Provider Remembered Coaching	Area(s) of Actionable Feedback from Coach	Behavior Change Made by Provider	How Provider Indicated They Identify Any Improvement	Suggestions on How to Improve Coaching from Provider
1111/SMD	Yes	Engaging and spending time	No change made	Unclear - Not sure changes can be reflected in CAHPS scores	None offered
1218/NP	Yes	Engaging and spending time	Sit with patient	Coach feedback	Improved coaching, More coaching
1114/MD	Yes	Engaging and spending time	Sit with patient - though this may not be realistic; Position self between computer and patient - though this may not be realistic	Increased CAHPS scores	Additional training
102/MD	Yes	Engaging and spending time	Sit with patient	Unclear - Not sure changes can be reflected in CAHPS scores	Improved coaching
1112/SMD	Yes	Engaging and spending time	Sit with patient at start of encounter	Unclear - Not sure changes can be reflected in CAHPS scores	Improved coaching, More coaching
106/MD	Yes	Engaging and spending time	Sit with patient, make eye contact, move computer	Unclear - Not sure changes can be reflected in CAHPS scores	Don't know
208/NP	Yes	Engaging and spending time; listening	Give patient time to speak at beginning of visit - makes patient feel like visit is longer	Patient feedback	Additional training
104/MD	Yes	Engaging and spending time; listening	Let patient speak for a minute uninterrupted	Increased CAHPS scores	More coaching
105/MD	Yes	Engaging and spending time; listening	Make eye contact, don't stand by the door	Patient feedback	More coaching
1217/NP	Yes	Information is understood	Repeat and make sure patient understands medication instructions	Patient feedback	Additional training
1116/MD	Yes	Information is understood	Repeat and make sure patient understands medication instructions; Sit with patient at start of encounter	Patient feedback	Additional training
207/NP	Yes	Listening; Information is understood	Ask patient open ended questions to clarify at end of visit	Unclear - Not sure changes can be reflected in CAHPS scores	Additional training, Patient follow up
1115/MD	Yes	Listening; Information is understood	Ask for primary concern	Increased CAHPS scores; Patient feedback	Additional training, Improved coaching
1113/MD	Yes	Listening; Information is understood	Validate complaints and indicate will take care of them in future appointment - Difficult to incorporate this recommendation	Unclear - Not sure changes can be reflected in CAHPS scores	Translation services
1209/PA	Yes	No tangible area of feedback	Not actionable; No change made	N/A	Additional training, More coaching
101/MD	Yes	No tangible area of feedback	Not actionable; No change made	N/A	Improved coaching
1219/NP	Yes	No tangible area of feedback	Not actionable; No change made	N/A	None offered
103/MD	No	N/A	N/A	N/A	None offered
210/PA	No	N/A	N/A	If I made a change, I would see it in improved CAHPS scores; Provider intuition	Additional training

NOTE: SMD = Site Medical Director; MD = Medical Doctor; NP = Nurse Practitioner; PA = Physician Assistant. N/A = Not applicable.

The most common theme was engaging and spending time patients during patient visits. Recommendations for better engagement with patients included making more eye contact, not standing by the doorway towards the end of a clinical encounter, sitting with the patient, and positioning yourself between the patient and the computer. One physician mentioned:

“The coach gave me a good tip, saying if the parents seem like they have something more serious to talk about, then don’t hold the doorknob on your way out. Just make sure you give them some time. Sit down at eye-level and listen. I learned that [from coaching], and that’s been very effective.” –MD, female, speaks Asian language, ID0105

The second-most common theme was listening to the patient. Example recommendations include asking open-ended questions about their needs and giving patients periods of uninterrupted time to speak. One nurse practitioner noted:

“The coach told me the first thing to do when I sit down with the patient is to not talk for a few minutes. Let them talk. Let the patient talk. I didn’t think that was going to make a big difference, but it does!… especially in an initial visit this is so important, because you can either connect with a patient or lose them. They can either perceive you are going to make time for them, or that you’re just trying to get out the door, be done with the visit and take care of the next patient. So, allowing time and space is crucial.” –NP, female, speaks English, ID0208

The third-most common theme was ensuring that patients understand information given by the provider. Example recommendations include having the patient repeat instructions given by the provider, the provider asking whether the instructions were clear, and asking patients who needed more time to come back for another appointment, thereby validating their additional needs.

“After coaching, I’ve made sure that definitely before I leave the room, I repeat any instructions and information. One of the things that I have been told by the coach to do is to ask the patient to repeat what I told them, which I think I’m lacking most of the time and am working on. The coach’s feedback to me was to tell patients again what any medication is for and what my instructions are. The repeating of information and instructions is one thing that I continually work on.” –MD, male, speaks English, ID1115

All providers thought the recommendations were actionable and tangible, but one provider reported not making the recommended behavior change (eye contact). This provider felt that the space in the exam room was not conducive to making eye contact with the patient given the need to attend to electronic medical records.

Three-fourths of providers who remembered being coached identified how they would know if their communication improved. Several indicated they could track the impact of their behavior changes through direct patient feedback. Others said they would know about changes through “their own intuition and understanding of how well they are communicating with their patients,” or from additional coaching feedback or follow up from the coach.

Six (32%) of the 19 providers said they were not sure whether their behavior changes with patients would be detected in patient experience scores. One physician noted:

“I'm not sure if there is a real way to see the change in my behaviors in patient experience scores. If we look at CAHPS scores before and after coaching, that shows some potential correlation, but it's hard to say if there's any causation. So, I don't know.” –MD, male, speaks Spanish, ID1113

3.3. Perspectives on Facilitators for Provider-patient Interactions

Seventeen (89%) providers and both coaches mentioned that either their practice or they themselves prioritized provider-patient communication. Four coached providers said they personally prioritized provider-patient communication more than their practices, while the others said that provider-patient communication was not a priority for themselves or for their practice.

“I don’t think everybody is 100% on the same level regarding the importance of provider-patient communication. Some people may actually put more emphasis on it than I do, and some may put less. We have discussions on its importance regularly in meetings, and I do think everyone’s heart is in the right place. Certainly, when there’s room for improvement, we all utilize things like coaching and whatever we feel would be helpful to improve our interactions with patients.” –MD, male, speaks Spanish, ID1112

Providers cited several possible drivers of overall patient ratings. The two most common were 1) being approachable and welcoming and 2) being considerate and caring; these were cited by seven providers (37%). Other themes mentioned included being attentive to the patient (21%), developing trust (21%), and making sure all patient needs are met (16%).

Providers mentioned several institutional factors that facilitate high-quality provider-patient interactions and noted that time constraints during patient visits were particularly difficult to navigate. Providers recommended either extended appointments or allowing greater time to see patients during their first encounters. Providers mentioned that having a site-based culture that expects the best communications during all visits facilitated a focus on both high-quality real-time interactions and a focus on making continual improvements in patient communications.

“Leadership at our site and our organization has the expectation that we should be providing good quality care.… We just went through a provider retreat. All providers were expected to come. So, I saw a lot of providers there and one thing that a leader said was ‘we expect the best from you’… I think it’s good to hear from leadership that our main aim is serving our patients and that is expected from our leadership.” –NP, female, speaks English, ID0208

3.4. Perspectives on Provider-Patient Interactions

While acknowledging that coaching can improve patient communication and interactions, all providers interviewed reported significant barriers to improving patient interactions. Commonly reported barriers were culture and language (10/19, 53%), transportation (8/19, 42%), and low health literacy among patients (3/19, 16%). Transportation concerns made it difficult for providers to recommend follow-up visits, leading providers to attempt to address as many patient needs as possible in one visit. Low health literacy and cultural barriers impeded providers’ ability to ensure instructions were clear or to communicate directly with patients. One physician noted:

“Language can definitely be an issue. I think even though we have translator services, a lot still can get lost in translation. It's much easier to do counseling without a third-party present.” –MD, male, speaks English, ID0106

Thirteen (68%) providers noted institutional-level barriers to improving patient communication. All 13 mentioned the time constraint placed on providers for each visit. Providers said they needed more time during visits to attend to all patient needs. One nurse practitioner said:

“We just feel overwhelmed with the amount of patients we see. It's hard to go in, diagnose and plan the care and prescribe or treat with little time. Time is a barrier that won’t let you communicate with patients that well or in a timeframe that we want or in the time that we need to communicate well, to explain and listen. If the patient has complaints, it takes time to address them. Sometimes we're limited in doing that.” –NP, female, speaks Asian language, ID1218

Other barriers identified by providers included having set or fixed attitudes towards patient communication (4/19, 21%) or patient expectations about a visit (4/19, 21%). Providers said these two barriers require different approaches than coaching, such as provider or patient education.

3.5. Provider Perspectives on Improving Shadow Coaching

Providers suggested several ways to improve shadow coaching. Five (26%) wanted more coaching. Both coaches agreed that additional coaching would be valuable to remind providers about recommended changes.

Eight providers (42%) wanted more training (not necessarily coaching) on topics that indirectly impact provider communication. Specific topics recommended were: how to treat patients with multiple chronic conditions, or other complex issues; how to treat patients who often need additional clinic time to address all of conditions; and how to treat older patients with more co-morbidities and socioeconomic needs.

Providers noted several ways to improve coaching. These included: having more targeted, specific recommendations (5/19, 26%); having providers shadow the coaches to view how high-performers conduct patient visits (4/19, 21%); and adding coaching refresher sessions with reminders on best practices for provider communication and other targeted issues (4/19, 21%).

4. DISCUSSION

Our study offered an opportunity to examine coaches and providers experiences in a large shadow coaching program seeking to improve provider-patient interactions and organization-wide culture. The aim in implementing shadow coaching for the organization was to build a quality improvement program to support providers in meeting performance targets with the ultimate aim of providing high-quality care and patient experiences and to improve provider morale and daily work life. This rationale was based on evidence that providers must be engaged and buy-in to the coaching and improvement process for shadow coaching to be effective.³⁷ Research suggests that an organization’s culture needs to be centered around continual improvement for all providers and include the use of data to monitor quality, and target quality improvement strategies.³⁸

Our findings support this evidence as the reported levels of engagement from providers and the voicing of their desire for more coaching, more training and additional methods for improving provider-patient interactions indicated that shadow coaching was important to them, seen as effective, and a critical part of the organization’s culture of improvement. Also engaged providers were willing to make changes based on coaching recommendations underscoring that, even in the face of reported barriers, leadership should consider continuing or strengthening the program to further improve their organizational culture.

Research suggests several barriers to improving physician patient communication: high workload for physicians, low medical literacy of patients and physician’s low awareness of communication skills.^39-42 Our findings identified several similar implementation barriers for organizations to address related to time pressures within a given patient encounter^43-45 and language/translation issues.^46,47 Providers in our study also pointed to the barrier of patient transportation which hindered providers’ ability to provide continuity of care and follow up care.^48,49 Often, any initial improvement from communication training dissipates as providers return to overloaded schedules and variable patient expectations.¹² Organizations may want to embed shadow coaching into long-term professional development, as providers indicated coaching identified tangible recommendations for behavior change. This would mean not only providing initial coaching sessions but also adding follow-up sessions. Future research could determine when follow-up coaching sessions should occur.

According to the coaches, CAHPS data were seen as an effective metric for selecting providers eligible for coaching, for measuring provider-patient interactions and for tracking overall organization-wide trends of patient experience. CAHPS metrics were also fully integrated into the performance management approach by medical directors and leadership of individual providers. Coaches most often provided specific tangible recommendations about three aspects of communication that are captured within CAHPS survey items: engaging and spending time with patients, listening to patients, and providing information to patients and families that is easy to understand. CAHPS items may not capture several subtle changes and/or recommendations for providers such as sitting with patients, positioning in the exam room, making eye contact, and asking open-ended questions. This may be in part because by design CAHPS items have been designed to both capture information where the patient is the best source of the information and areas of patient experience most valued by patients and families. Items such as making eye contact with patients may be more important to providers than patients or be too subtle or nuanced to be captured by CAHPS measures. Research shows clinicians more satisfied with their patient interactions make fewer medical errors and are less likely to depart from an already understaffed workforce;^50-52 however, this research speaks generally about provider job satisfaction and interacting with patients and not about specific aspects and behaviors of providers and their communication with patients. Coaches can observe, identify and provide recommendations to individual providers on specific, modifiable aspects of communication. Only with observed performance and feedback do clinicians get an accurate sense of their own behavior, what they already do well and what they need to improve.⁵³

Limitations.

We studied one organization’s experience with shadow coaching, limiting the generalizability of our findings. Implementing shadow coaching in healthcare organizations serving less vulnerable populations may involve different challenges. We also were not able to interview all coached providers and as such may have only included those with more positive feelings toward coaching. However, our sample was purposively sampled to pull individuals from strata by provider type and CAHPS performance, so a variety of provider perspectives were included. The results of non-response bias analysis indicated no difference across those interviewed/not interviewed. A broader, multi-site evaluation of shadow coaching may identify additional insights, lessons learned and challenges.

Regular messaging by leadership about the priority and purpose of shadow coaching – that is, as a mechanism to support high-quality care and provider-patient interactions as well as to improve provider morale and work life – was considered essential for both physician engagement in coaching and its implementation as an organization-wide effort. CAHPS data was seen as an effective performance management metric for benchmarking, measuring change, counseling providers, and evaluating provider-level efforts, though it was not always considered to be specific enough for providing tangible recommendations about modifiable provider behaviors. Based on respondent recommendations, one suggestion to consider is embedding shadow coaching into an organization’s long-term strategy of professional development by having initial shadow coaching sessions be followed up with regularly planned re-coaching. Re-coaching sessions could be targeted on handling issues raised by providers, such as dealing with difficult patients or specific clinic settings. Further research is needed to assess these implementation suggestions.

DATA AVAILABILITY STATEMENT:

Data available on request due to privacy/ethical restrictions. The data that support the findings of this study are available in an aggregated, de-identified format upon request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.