Autonomy Versus Control: Finding the Sweet Spot

Many emergency medicine faculty nostalgically recall nights as residents in the emergency department (ED) when their attendings would retire to their offices around midnight, saying “have a good shift, wake me up if anything bad happens.” Barring any significant badness, their staff would ultimately emerge (usually with bed head) at 6:45 a.m., ask groggily how things went, and proceed to sign off on the night's events sketchily documented in a stack of handwritten charts. In today's ACGME parlance, this might be considered, at best, “indirect supervision with direct supervision immediately available,” although how “immediate” faculty supervision can be after awakening from a dead sleep is questionable. Ah, the good old days … or were they? Residents, then as now, frequently did not know what they did not know and hence did not call for help when needed. Mistakes were made. Thankfully, most patients did not suffer, but some assuredly did. Today, much has changed. The advent of faculty-driven care, duty hour restrictions, and a rededication to the public trust placed in medical educators to train competent physicians has, at least in theory, created a safer environment for patients receiving care in our nation's teaching hospitals. No one would propose returning to the old model of care. But at what cost? Are our residents achieving the autonomy to develop the complex skills needed for emergency medicine practice? At the crux of the challenge is the ACGME requirement for graded and progressive responsibility, afforded to residents as they move along the arc of their professional development toward independent practice, the intermediate destination being the program director's stamp of approval that each graduating resident “demonstrates sufficient competence to enter practice without direct supervision.”1 But graded responsibility is one of the hardest goals to achieve. Knowing when an individual resident is ready to move to the next level of autonomy can be a challenge. Finding the sweet spot between permitting complete autonomy and maintaining complete control is both complicated and complex. The ACGME milestones will help in this effort. With standardized, competency-based, specialty-specific, developmental performance expectations, training programs and faculty will have a better chance of gauging where their residents are in the continuum leading to independent practice.2 Entrustable professional activities (EPAs) promise to provide even more guidance.3 By holistically integrating specific competencies and performance expectations into common physician tasks, EPAs will identify real-world physician responsibilities in a clinical context that carry tangible meaning for both educators and learners and will create a direct link between milestone achievement and the patient experience. But the phenomenon of entrustment is a complex one and extends beyond an individual faculty member's willingness to trust an individual resident. Practicing physicians must also have trust in system safeguards put in place to catch errors, trust in the health care system's tolerance of a clinical teaching model that carries intrinsic risk and cost, trust in the patient's willingness to accept care from a trainee, and trust in the medical profession's social contract with a public that wants high-end medical care while maintaining zero tolerance for mistakes. Poor management of these conflicting agendas puts in jeopardy the delicate relationship between ourselves and our trainees. Trust develops slowly over time, but in large programs, faculty experience with a particular resident can be limited. Faculty decisions may be biased by the last shift worked with that person 6 months earlier or the last bad outcome, continuous quality improvement review, or morbidity and mortality conference case. Faculty members, being human beings, also carry with them different tolerances for uncertainty.4 Some are intrinsically willing to turn over the controls to residents and allow them to fly solo, while those at the other end of the spectrum are more risk-averse, preferring to maintain control. The Joint Commission has attempted to address this uncertainty by requiring that all care providers on a clinical team be able to confirm the credentials of a trainee for a particular invasive procedure. But in a critical setting, it may be difficult or impossible to access a database for this information quickly. And what about a resident's sound decision-making, knowledge base, ability to adapt rapidly to a changing situation, interpersonal and communication skills, or ability to manage teams? These attributes are not as easily catalogued. Faculty development is key to addressing this issue. Proficiency as a medical educator goes beyond the ability to give a lecture, lead a small group, or teach the microskills of a bedside procedure. It includes a mindset that embraces shared responsibility and an appreciation for allowing residents to go to the limits of their ability, stepping in only when necessary. In the critical care example cited in the resident commentary, the faculty consciously stayed in the background until they were needed. Many of our most successful faculty have adopted a coaching model for teaching complex patient management skills, using scripted language that can help set up a resident for successful performance with appropriate levels of autonomy. Simply stating “This is your case, I'll be right here if you need me” can empower residents to push themselves just beyond their limits while maintaining an environment of safety and support. Residents should never find themselves having to read the tea leaves to decipher how a particular attending will want to manage a case, as described in the resident piece. Most of us teach the way we ourselves were taught, but is that good enough? As faculty, we are obliged to model evidence-based rather than anecdotal decision-making. We must be comfortable with our own uncertainty, exercise self-control rather than merely control, and be ready to justify our management strategies to learners in a respectful manner. Doing otherwise leads to unnecessary testing and inefficient work flow, models poor stewardship of resources and squanders the opportunity to teach and learn. Finally, assessment of resident performance lies at the heart of any clinical encounter that is co-managed by a board-certified physician and one on the way to certification. By developing and implementing standardized assessment tools, and providing vital feedback to trainees, we can support their progress along the path toward true independent practice.