Background: Development of diagnostic reasoning is a critical component of medical education, and feedback related to clinical outcomes is an important means of calibrating one’s diagnostic decision-making. The hospital is a key training ground for the development of skills in diagnostic reasoning. However, given the multiple provider handoffs that now occur during the care of hospitalized patients, trainees are often not present to witness the outcomes of the patients they admit. The LOOP Study is a national, multi-center, prospective initiative to create a mechanism for standardized feedback about clinical decisions made during overnight hospital admissions, and promote a culture of continuous improvement in diagnostic reasoning.
Methods: The study included seven training programs at six institutions. IRB approval was obtained at all sites. Residents were encouraged but not required to participate. Surveys were administered to determine trainees’ level of self-efficacy in diagnostic reasoning, giving and receiving feedback, and satisfaction with feedback before and after the study. Teams who assumed care of patients after night admissions were encouraged to send feedback forms to the admitting teams, describing what, if any, changes were made in patients’ diagnoses. The feedback forms were gathered, coded, and categorized for content to capture changes in diagnoses in the initial days of hospitalization.
Results: A total of 220 pre-intervention surveys and 87 post-intervention surveys were completed. Data from the pre-surveys revealed that 43% of residents were not confident in their ability to identify diagnostic errors or near misses in their own practice. A majority of residents stated that they did not get the same amount of feedback about patients that they admit at night compared to those during the day. 89% of residents liked receiving feedback, but only 47% liked giving it. Post-intervention survey results showed a statistically significant increase in learner satisfaction with frequency, quality, and overall value of the feedback. Additionally, there was a statistically significant increase in learner confidence to identify strategies to decrease or mitigate diagnostic errors in their own practice. Feedback forms were analyzed for diagnostic change rates, and revealed a shift over 40% of the time, with over 10% having major diagnostic changes.
Conclusions: The LOOP project provides a formalized mechanism to improve feedback regarding diagnostic reasoning and provide trainees an opportunity to track clinical outcomes for patients they admit to the hospital. Trainees improved their self-efficacy in diagnostic reasoning, and found the practice to be satisfying. With diagnostic changes occurring over 40% of the time, it is fundamental to implement feedback loops spanning critical transitions in care to ensure we have a deliberate, workflow congruent method of refining our diagnostic capabilities