Introduction
Challenging conversations are ubiquitous in medicine, perhaps none more so than providing unsolicited feedback to a colleague with whom there is clinical disagreement. These conversations are common in antimicrobial stewardship, as many programs utilize prospective audit-and-feedback—the process of reviewing real-time antimicrobial use and highlighting opportunities for improvement in the clinical care provided by a peer (eg, de-escalation of unnecessarily broad antibiotics or optimization of duration of therapy). Reference Barlam, Cosgrove and Abbo1 For this reason, the context of antimicrobial stewardship will be used to demonstrate a framework to approach challenging conversations, though this structure can be applied to any domain or specialty.
Fostering a productive and collegial conversation is challenged by several factors, including perceived and real differences in training and expertise between the steward and provider, lack of alignment in values and priorities (eg, a steward’s emphasis on antibiotic appropriate use versus a clinician’s need for efficiency on a busy clinical service), and the emotional beliefs that often inform antibiotic prescribing habits. Reference Szymczak and Newland2 In addition, providers often abide by an unspoken culture of “noninterference”—the mutual posture of not questioning the decision-making autonomy of colleagues—that limits their willingness to intervene on the prescribing habits of peers. Reference Charani, Castro-Sanchez and Sevdalis3
Despite these challenges, the benefits of creating a positive and collegial conversation are myriad. Most immediately, a successful stewardship intervention improves the quality and safety of care for patients. Long-term, productive conversations facilitate strong relationships between peers and create a safe space for teaching—thus reducing the need for repeated interventions and promoting future collaboration.
Elements of a productive conversation & the advocacy-inquiry framework
Fostering a productive conversation requires attention to delivery and content, recognizing that a successful “difficult” conversation still produces a positive learning experience. First, the conversation must be non-confrontational to avoid triggering defensiveness. Second, the conversation must be direct and concise to maintain the colleague’s focus and attention—especially given that these conversations often occur in busy clinical settings. Third, the steward must accurately identify and teach to a colleague’s specific knowledge gap to avoid appearing condescending or patronizing by sharing concepts that are already known. Finally, a productive conversation creates a sense of shared ownership of a patient, with emphasis on the function of both individuals as members of one team with collective responsibility.
The advocacy-inquiry framework serves as one approach to cultivate a constructive conversation. This methodology originated in the business and organizational behavior literature and was later adopted in the medical simulation sphere as part of “debriefing with good judgment”—an approach designed to encourage open dialogue, reflective thinking, and accurate assessment of a learner’s thought process. Reference Senge4–Reference Sawyer, Eppich, Brett-Fleegler, Grant and Cheng6 In this framework, an advocacy statement is utilized to make an objective observation and is followed by an inquiry designed to reveal a learner’s assumptions and identify the “invisible frames” that inform their actions. Reference Rudolph, Simon, Dufresne and Raemer5
Psychological safety—defined as “a shared belief that the team is safe for interpersonal risk taking” and a sense of “confidence that the team will not embarrass, reject, or punish someone for speaking up”—is of particular importance in highly functioning clinical teams. Reference Edmondson7,Reference Rotman Devaraj, Cooper and Simpkin Begin8 Notably, the objective and candid nature of advocacy–inquiry statements promotes a culture of psychological safety by encouraging strategies that are associated with safe learning environments—including normalizing discussion of mistakes, fostering curiosity, seeking new perspectives, and recognizing the valuable contributions of all team members. Reference Rotman Devaraj, Cooper and Simpkin Begin8
Steps 1 and 2: Advocacy-inquiry
Advocacy. The advocacy statement highlights a discrepancy between a clinical situation and a perceived inconsistent clinical response with factual observations. Successful advocacy statements often begin with self-oriented language: “I noticed…” or “I saw….”
Inquiry. The inquiry subsequently probes this discrepancy with an open-ended question to query the thought process yielding the colleague’s decision or action. Successful inquiries often begin with “Can you help me understand…?” “Can you share more about…?” or “Can you teach me…?”
A paired advocacy-inquiry statement is a powerful way to open a conversation. It is concise and direct, highlighting a specific issue in two sentences. Despite its brevity, it is also impartial—effectively calling attention to this discrepancy without making assumptions about what occurred, why it happened, or who is responsible for a decision. Finally, it rapidly opens the door for bidirectional conversation, promoting a sense of collegiality and shared decision-making regarding a patient’s care.
Additionally, an advocacy-inquiry statement allows stewards to open the conversation with a posture of humility. Antimicrobial stewards are often not at the bedside and therefore make recommendations based on chart review alone; as such, the open-ended inquiry creates space for stewards to glean new information from the provider that may influence their recommendations (eg, undocumented symptoms).
The juxtaposition of a clinical situation with a perceived incompatible clinical action through a paired advocacy-inquiry statement may be sufficient for a colleague to self-identify an opportunity for improvement. Thus, pausing after the advocacy-inquiry statement allows a colleague to spontaneously correct their error; in these cases, stewards may simply affirm the colleague’s revised action, and the subsequent identify-teach intervention may not be required.
Steps 3 and 4: Identify-teach
Identify. The third step, identify, is active listening. The goal of this step is to not simply hear the colleague’s response, but to understand the thought process that yielded their clinical decision. Correctly diagnosing the underlying reason guiding their actions allows the steward to identify what concepts to teach to specifically address the colleague’s cognitive error(s).
Teach. In this final step, the steward provides targeted teaching to address the framework error. It is often helpful to offer an alternative path forward—as doing so opens the door to collaborative, shared decision-making and demonstrates shared investment in the patient’s outcome. Providing resources (eg, clinical guidelines or scientific articles) can also substantiate the steward’s recommendation and serve as affirmation and reassurance for providers, especially when patients are critically ill and clinical decisions may be guided by emotion and concern.
The open-ended and unbiased approach of the advocacy-inquiry statement sets the stage for a non-confrontational and targeted identify-teach dialogue, as illustrated in Table 1. By correctly identifying the framework error, the steward creates a concise and high-yield teaching opportunity—and avoids the pitfall of inadvertently teaching content that is already known. Most importantly, the identify-teach framework allows the steward to correct an underlying misunderstanding and behavioral habit—thus improving both the care of the individual patient and avoiding future errors related to the same issue.
Though this framework is designed to create open dialogue and minimize defensiveness, clinical disagreement may remain even after targeted teaching. It is important to acknowledge that not every conversation will lead to success the first time, nor should that be a realistic expectation. The act of initiating a collegial conversation creates an opportunity for a thoughtful pause and may reiterate concepts that can carry momentum for the next conversation.
Conclusion
Effective communication in clinical medicine requires initiating difficult conversations and providing unsolicited feedback to peers, both of which are crucial for ongoing professional growth and maintaining patient safety and high-quality care. To facilitate productive conversations, clinicians must create an open and psychologically safe atmosphere in which concise, targeted teaching can occur. The paired advocacy-inquiry statement followed by the identify-teach framework creates an organized approach to initiating such conversations and fostering productive relationships with colleagues across specialty and rank.
Acknowledgements
The authors would like to acknowledge Paul Currier, MD, the Learning Lab Medical Simulation Center at Massachusetts General Hospital, and Zahra Kassamali Escobar, PharmD.
Financial support
Alyssa Castillo, MD, would like to acknowledge the National Institutes of Health T32 Training Grant for its financial support in the development of this work.
Competing interests
The authors report no conflicts of interest relevant to this work.