We examined the allocation of self-reported time spent by Internal Medicine fellows across subspecialties at a single academic medical center. This is the first study to our knowledge which describes how time is spent by trainees in medicine subspecialty fellowships. In addition, we describe the barriers and facilitating factors to fellows’ experience on the inpatient consult service, which represents the largest portion of the trainees’ experience in the first two years. Changing clinical rotations and improving workflows to improve efficiency are frequently used by fellowship program directors to combat burnout.4 Therefore, a better understanding of fellows’ experience is an important step in enhancing the fellowship learning environment and reducing burnout.
Our study suggests there are shared themes across all subspecialties in how fellows spend their time. The largest self-reported proportion of time spent by fellowship trainees is in direct patient care, however clinical documentation comprised a significant proportion of fellows’ time, particularly during first year of fellowship. This is common across nearly all subspecialties and is most pronounced during the first year of training. The amount of time spent on documentation in relation to other activities that represent important learning opportunities, such as patient care, studying and preparing presentations was striking. Our data suggest that note-writing burden may be higher for fellows than attending physicians.21 Further, recent studies suggest that the burden of electronic health record burden is associated with fellow burnout.3 Therefore, interventions that reduce the note writing burden may have a significant positive impact.
Similarly, fellows spent considerable time responding to patient messages. This is particularly notable as the burden of patient messages and other asynchronous communication has increased significantly since the COVID-19 pandemic.22 Fellowships with larger outpatient components may be particularly affected and the impact of high message volumes on fellows may disproportionate compared to attending physicians. For example, fellows may not only have more uncertainty in answering messages, but the in-person support from faculty that fellows have during inpatient consultation or outpatient clinic may not be as readily available. Further, fellows are likely less efficient in their message handling as compared to attending physicians. This area deserves further study to better understand its impact.
Understanding fellow experience during inpatient consultation rotations is important as inpatient consultation constitutes the largest share of fellow training time. Moreover, the inpatient consultation service is a unique experience for subspecialty fellows and differs significantly from experiences on primary teams and outpatient care that fellows have during residency training. We found that the highest positive impact on experience during consultation include timing of the consultation request, sufficient time to learn, and clarity of the consult question, while unpredictability of workload and perceived low necessity of the consult had the most negative impact. These data highlight that pace of work, predictability of the workload and ability to add value and to learn may be particularly important to fellow wellness. Ideally, consult services would be structured to maximize these aspects, however multiple systems barriers make this a challenge.19 Specifically, consult requests arise throughout the day making it difficult to predict or pace fellow workload. Moreover, smaller fellowship programs with only a single consult team may not have the ability to shift workload to other providers during busy days. Finally, limitations in optimal communication around consult requests by the primary team can impact fellow perception of their ability to add value to patient care.20
Our study suggests several interventions which have the potential to improve fellows’ experience. Exploring whether time spent on documentation can be decreased without compromising learning or patient care and providing education and support to fellows in order to make documentation more efficient has the promise of creating more time for learning and personal activities.23 Recognizing the burden of patient messages by both providing training and support for handling them and incorporating the increasing time burden of this activity when considering fellow workload will be important. Structuring consult services such that excess workload can be shifted to other providers, making work hours more predictable and a mechanism for shifting non-urgent consults to the outpatient setting or to e-consultation models has the potential to further enhance the learning environment. Finally, improving communication around consults between primary and consulting teams, where several models have been proposed, holds further promise.20,24,25
Our study has several limitations. This was a single center study which may not be generalizable to other institutions. The fellows represented here are primarily in research-focused training programs, with consolidation of clinical time in the first one to two years and extended research time in additional years, including non-ACGME accredited time. Our data were collected prior to the Covid-19 pandemic. Fellowship structure and the inpatient consultation experience for fellows stayed largely the same since the pandemic. However, the outpatient clinical experience may have evolved, particularly with respect to telemedicine and asynchronous communication with patients, which may have added to fellow workload. Finally, given the voluntary and self-reported nature of the study, it is subject to selection and recall bias.