Improving diagnosis by feedback and deliberate practice: one-on-one coaching for diagnostic maturation.

Abstract

Reflective practice is essential for the ongoing maturation of clinicians and requires regular self-evaluation in association with ongoing mentoring and feedback. Currently, most resident physicians do not have access to educational experiences that fulfill these needs. We present a novel model for structured one-on-one longitudinal coaching using the principles of deliberate practice to improve diagnostic skills. This is an easily implementable educational model that can be replicated in residencies across the country to improve clinical reasoning. Skills learned through this program have the potential not only to bolster the academic approach to patients but to also directly improve the clinical assessment and care of patients under the trainee’s care.

Resident physicians spend their days pre-rounding, rounding, putting in orders, calling consultants, responding to emergencies, writing progress notes, compiling discharge summaries, updating families, and signing out to the next team. Rounds can be rushed. With all these tasks, where is the time to delve into the intricacies of clinical reasoning and mature as diagnosticians?

Most formal diagnostic education offered in residency is passive, often through a lecture-based format. While morning reports can offer active learning opportunities, many residents are restrained due to the presence of their colleagues and supervisors regardless of the collegiality of the teaching environment. Not all trainees are equally comfortable with responding to probing questions by their chief residents and program directors during morning report who are ultimately responsible for the promotions and recommendations. Additionally, residents are not consistently challenged to demonstrate their facility with fundamental diagnostic skills such as identifying salient data, recognizing clinical syndromes, and generating differential diagnoses. Nor does the structure of morning report provide a mechanism for receiving longitudinal and personalized feedback. Finally, the unceasing beckons of pagers can disrupt the trainee from deep uninterrupted learning as one might be thinking of orders that need to be placed for sick patients or how to face a family member who just arrived for a difficult conversation.

Clinical care is necessary to improve diagnostic skills, but it is not sufficient. Practicing a skill repeatedly without reflection or feedback can lead individuals to repeat the same errors with minimal or no improvement in performance. In 1993, Ericsson et al introduced the notion of deliberate practice, defined as highly motivated learners setting goals of well-defined objectives or tasks of realistic difficulty achieved through focused and repetitive practice with reliable measurements and with external feedback working towards those goals.(1) With deliberate practice, learners decrease errors and hone a particular skill before addressing additional dimensions of their performance.(2) Coaching is imperative in deliberate practice.(3) It provides learners the longitudinal and personalized feedback that is crucial for skill improvement.

This approach has been demonstrated to be an effective way of improving diverse skillsets such as mastery in athletics, chess, and music. For example, Ericsson used the principles of deliberate practice to train a college student to increase his recall of number sequences. Initially, the student could memorize and recall a span of seven digits, but through practice and coaching, he was soon able to recall numbers with up to eighty digits.(3) Ericsson identified opportunities to improve discrete skills among medical trainees through deliberate practice as well, including in laparoscopy, radiograph interpretation, and interviewing skills.(4) Through deliberate practice, trainees develop superior mental frameworks that allow them to consistently perform at a higher level.(3) Dr. Atul Gawande has also proposed coaching for surgeons to improve their operative performance which has been well-received by surgical trainees.(5)

The integration of deliberate practice has been extensively discussed in multiple specialties and usually geared toward the acquisition of discernible individual skills especially regarding surgical and emergency medicine resident training.(6, 7) The ideas of deliberate practice have also been incorporated into optimizing specific skills such as reading an electrocardiogram, navigating difficult clinical encounters, safe pharmaceutical prescribing practices, and performing critical care resuscitation.(8, 9) (10–12) There has been some literature on longitudinal, infrequent resident observation and coaching during which primary care interns had three clinic visits directly observed by faculty over the course of a year that was directly followed by a self-assessment and 30-minute feedback session regarding the learner’s behavior. Interns who were coached had significantly higher behavioral scores at the end of the assessment compared to those that were not coached. (11, 13)The idea of deliberate practice has been incorporated into developing clinical competencies for the ACGME as well.(14, 15) Although deliberate practice has often been focused on specific skills, it has not been harnessed for improving diagnostic capacity which involves numerous inter-related skills such as identifying salient data, systematically identifying problems and syndromes, creating a differential diagnosis, and prioritizing testing and treatment with precision. To be sure, some elements of deliberate practice for diagnostic mastery have been incorporated into online case-based simulations.(16)

Towards the end of their residency training, two trainees looked to improve their diagnostic skills with a senior clinician through one-on-one sessions using the principles of deliberate practice. For both learners this was a self-identified area of focus, motivated by the perception that the trajectory of their improvement as diagnosticians had plateaued. The specific skills worked on were as follows: reviewing information presented in complex cases, identifying salient findings using a standard framework, generating a differential diagnosis, and prioritizing diagnostics appropriately to arrive at an accurate diagnosis.

The trainee and senior clinician met fortnightly for structured one-on-one sessions that lasted 45–60 minutes in which we read through cases from the Massachusetts General Hospital in the New England Journal of Medicine (Figure 1). Neither the trainees nor the senior clinician read the case beforehand. The trainees read the case details aloud which ordinarily took 15–20 minutes. Before the final diagnosis was revealed, the pair stopped and discussed the case in the same structured manner each time. The data from the case was systematically divided into the following categories: host, epidemiology, and clinical presentation. The trainee then generated a list of problems and identified a clinical syndrome. Next, trainees generated a differential diagnosis that would explain the identified problems.

Figure 1:

Structure of education sessions followed the approach described in this process graph.

At this time, the senior clinician challenged the trainees to expand or contract the differential diagnosis, pointed out any salient case details that the trainee overlooked, identified, and filled in gaps in clinical knowledge. The senior clinician shared their own diagnostic approach. Before checking the final diagnosis, the pair declared their top two or three diagnoses which fostered active learning and engagement. After checking the final diagnosis, the pair reviewed the explanatory section to compare their diagnostic approach to that of the experts in the Journal. Cognitive errors and knowledge gaps of both the trainee and senior physician were reviewed. The senior clinician also provided readings as needed to strengthen areas where the trainees felt weak.

From the perspective of the educator, the senior clinician felt that the clinical coaching was a very useful and satisfying teaching exercise. The process emphasized useful skills that not only assisted in approaching current and future written cases but taught a useful approach to patients encountered in the hospital. It also taught a disciplined system of thought to the trainees that can be passed on to future learners.

Both trainees feel that these sessions help them internalize a framework for approaching challenging cases. Even when a diagnosis seemed readily apparent, the differential diagnosis was still reviewed systematically to avoid premature closure and exercise “slow thinking” over “fast thinking.(17) The goal was to develop and practice the process of developing a differential as opposed to celebrating having deduced the correct diagnosis. Individualized coaching allowed them to not only have their ingrained and perhaps subconscious cognitive errors laid bare in the open but also to have those cognitive errors corrected in real time. This experience allowed for the trainees to make errors during a time of practice when not taking care of patients, and then to reflect on this error without fear of repercussion. The longitudinal experience allowed for teaching points to be revisited week after week so that knowledge and approach could be cemented. Both trainees continue to use the framework taught more than one year after concluding their one-on-one training sessions. For both trainees, this was a unique individual experience that was not attained through morning report, noon conference, journal clubs, clinic precepting, inpatient precepting, or academic mentorship. The longitudinal follow up with attention to diagnostic acumen was unique and allowed them to grow with deliberate coaching. This experience also allowed the trainees to develop a disciplined diagnostic approach that they continue to use, greater confidence in their diagnostic skills, as well as a life-long mentorship with a teacher with whom the trainees continue to discuss difficult or challenging cases.

We believe this kind of clinical coaching can be easily replicated and implemented in residencies across the country and offer some suggestions for implementation. We propose the third year of training as an ideal time to challenge residents with such coaching since second year residents are still learning to be team leaders and teachers. Program directors could consider creating pilot coaching programs with a select group of enthusiastic senior clinicians and residents. Given the preponderance of distinction programs and tracks within residencies, such tutorial-style sessions could form the backbone of a certificate or distinction in clinical diagnosis.

In terms of logistics, we recommend flexibility in scheduling to meet the needs of faculty and trainees and suggest approximately 15–20 meetings per year. Trainees will benefit from keeping a log of learning objectives that were identified. To ensure that the discussion is full-throated and that the trainee doesn’t worry about disgracing themselves before evaluators, we recommend that the selected faculty members are not responsible for resident evaluations, promotions, and chief-residency selections. In our case, the senior clinician was a faculty in a related, but different training program than the trainees.

Multiple means of a long-term assessment could be developed to assess if there are other downstream benefits of such as a program. A prospective cohort study could be performed of residents enrolled in the program to study if they were less likely to make diagnostic errors, if they had different antibiotic prescription practices, or if they spent less money on testing. Further, such residents could be followed over time and interviewed to see if they had greater satisfaction with their work or a greater feeling of diagnostic confidence.

Certainly, there might be some challenges in implementing such one-on-one coaching sessions in residencies. For instance, faculty are already overburdened and may be hesitant to meet one-one-one with residents. As both residents and attendings have extensive responsibilities and constraints on their time, a concerted effort would need to be made to align their schedules. If a residency program has clinic block, this would be targeted as a time to meet as usually the residents will have greater control of their schedules. Further, given the rise of medical education tracks in residencies, clinician educators interested in furthering diagnostic acumen could have such a program integrated into their medical education time. For our participation, we found it easiest to meet at the discretion of the learner and educator as this allowed for the greatest flexibility. Each individual residency program could determine when this could most easily fit for both parties and PDSA (Plan-Do-Study-Act) cycles could be enacted to optimize this. For instance, the first cycle could plan for a 45-minute coaching session, enact this intervention, study it using learner and faculty surveys, and then adjust the length as suggested by the survey results.

Another potential barrier would be that the trainee and senior clinician might not develop a strong working relationship due to differences in personality, learning style or a variety of factors. If this is the case, an opportunity to switch mentors should be offered to the trainee. One final potential barrier may be that some faculty could be wary of exposing their own knowledge gaps or feel unqualified to act as a sole critique of clinical acumen. However, senior physicians modeling that they too have something to learn can be immensely helpful for trainees struggling with feelings of inadequacy. Such sessions can then help both trainees and faculty embrace a growth-mindset.

Most patients will be the subject of at least one diagnostic error at least once in their life.(18) Further, the absence of a well-honed diagnostic approach may result in over-testing and overtreatment, which tax patients, health systems, and nations economically.(19) Deliberate practice through one-on-one longitudinal coaching has the potential not only to bolster the medical education but to also directly improve the clinical assessment and care of patients by the trainees.