Interventions to Support Resident and Fellow Well-Being During the COVID-19 Pandemic: A Scoping Review

The COVID-19 pandemic led to rapid and wide-scale changes in graduate medical education. While the pandemic spurred the adoption of telemedicine and virtual learning platforms and encouraged multidisciplinary and interprofessional collaboration in education and patient care, the large volumes of critically ill patients, uncertainties around disease management, and fears of infection challenged health care professionals worldwide, making efforts to address their well-being a global priority.^1,2  Residents and fellows functioned at the front lines of care for patients with COVID-19 and experienced unique vulnerabilities. Multiple reviews and high-quality studies prior to the pandemic showed that resident physicians are at higher risk for burnout and depression than the general population.^3-5  Residency training is characterized by high stress, long hours, disrupted sleep, social isolation, and extreme physical and mental demands.⁶  Data collected since March 2020 suggest that COVID-19 exacerbated preexisting high rates of burnout and depression in residents, with the negative effect on well-being particularly pronounced during the pandemic’s first surge phase and for individuals redeployed to the front lines of patient care.^7,8  However, there remains a gap in the literature in examining the well-being strategies used and the effectiveness of these efforts. In this review, we report on interventions to promote resident and fellow well-being during the COVID-19 pandemic. Lessons learned can guide how graduate medical education should function when faced with future crises.

We used the biopsychosocial model to categorize well-being into 3 dimensions.⁹  Physical well-being focuses on bodily health and functioning and encompasses factors such as nutrition, exercise, and sleep. Psychological well-being emphasizes mental and emotional aspects. Lastly, social well-being involves the quality of social connections, social support, and engagement. This framework interconnects physical, psychological, and social domains and provides a comprehensive model for understanding and promoting well-being.⁹  Through the review, we sought to answer 3 research questions:

1. What interventions were implemented to support the well-being of residents and fellows during the COVID-19 pandemic?

2. What domains of well-being were addressed?

3. What were the key findings for the effectiveness of interventions on well-being?

We also aimed to identify knowledge gaps and limitations to inform future work in this area.

A scoping review methodology was chosen because it identifies the types of interventions implemented, the populations studied, and the outcomes measured. The review is based on the scoping review framework by Arksey and O’Malley.¹⁰  We used 5 of the 6 steps of the framework: identifying the research question, searching the literature, selecting studies, organizing and charting the data, and summarizing and reporting the results. We did not use the sixth step (consulting with stakeholders to validate the findings), which is an optional component of the framework.¹⁰  The literature search was conducted according to the standards and guidelines established in the Preferred Reporting Items for Systematic Reviews and Meta-Analysis with extension for Scoping Reviews (PRISMA-ScR).¹¹ 

Unlike other published reviews of resident well-being, we exclusively reviewed studies of interventions implemented during the COVID-19 pandemic. A medical librarian (J.S.) and the 4 coauthors (S.A., H.I., L.K., I.P.) developed a comprehensive search strategy. The search query utilized search terms related to “well-being,” “residents,” “medical residents,” “fellows,” “medical fellows,” “residency,” “graduate medical education,” “mental health,” and “COVID-19.” The databases searched were PubMed, Scopus, Embase, PsycINFO, CINAHL, and ERIC. The PubMed search strategy (online supplementary data) was adapted for the other databases. Database searches were initially conducted in December 2023 and repeated in June 2024 to check for new literature. In addition to peer-reviewed journal articles, we examined funded trials, peer-reviewed conference papers, and grey literature with sufficient information about methodologic approach and results. We manually reviewed the reference lists of studies that met inclusion criteria to identify other relevant works. The Figure details the screening workflow.

Inclusion criteria were determined to address our research aim and key questions. Studies needed to be written in English, be published between January 1, 2020, and June 25, 2024, and describe well-being interventions targeting residents and/or fellows in the context of the COVID-19 pandemic. We included studies that provided sufficient information about their methodologic approach to allow others to replicate the intervention. We excluded reviews, commentaries, perspectives, editorials, and studies that did not report an intervention.

Data extraction was conducted in 2 stages. First, 2 authors (J.S., I.P.) independently screened titles and abstracts to determine their inclusion against criteria. The same authors read the full text of selected manuscripts, and the entire review team independently screened 11 of the 225 retrieved publications (4.9%) to ensure consistency. During data cleaning, keywords were consolidated in the case of plurals (“intervention” vs “interventions”). Discrepancies between the 2 primary reviewers were resolved by 2 additional authors (S.A.R., H.I.); L.K. served as a tiebreaker when needed.

Information was extracted at the full-text review stage using a workbook in Microsoft Excel 365. It included: (1) publication details (title, year, country of corresponding author, publication type, and keywords); (2) study population (residents and/or fellows); (3) study design (quantitative, qualitative, or mixed methods); (4) primary study focus/foci; (5) outcomes examined; and (6) feasibility and sustainability data reported in the primary literature. Data were checked for consistency by the authors. Articles were grouped based on the study population, type of intervention, and the well-being domain(s) targeted (physical, psychological, or social) as determined by the entire research team. We summarized the data in 2 tables and through narrative descriptions of the intervention characteristics to generate findings specific to our research questions. The study was reviewed and approved by the Sheikh Khalifa Medical City Research Ethics Committee (RS-735). Informed consent is not applicable for this review.

The search identified 1744 articles, with 1731 through database searches and 13 from other sources. After removing duplicates, 1110 articles remained for title/abstract review. During title/abstract screening, 885 articles (79.7%) were discarded as not meeting inclusion criteria. Based on the full-text assessment of the remaining 225 articles, 207 (92.0%) did not meet inclusion criteria. The majority of excluded articles described the negative effect of the COVID-19 pandemic on residents but did not report on any interventions. Ultimately, 18 of the 225 (8.0%) published studies were included in this review.^12-29 

Most of the studies (14 of 18, 77.8%) described interventions implemented in the United States.^{13-18,20-25,28,29}  The remaining 4 studies were conducted in Canada (n=2),^19,27  Israel (n=1),¹²  and the Netherlands (n=1).²⁶  Seventeen studies (94.4%)^12-26,28,29  described an intervention at a single institution/health system, and 12 (66.7%)^{12,13,16,19-26,28}  targeted residents in a single specialty. Represented specialties included general surgery and surgical specialties (n=3),^17,19,25  anesthesiology (n=3),^16,22,28  family medicine (n=3),^13,21,24  internal medicine (n=2),^20,26  pediatrics (n=1),¹²  and emergency medicine (n=1).²³  The majority of interventions were instituted in the spring or summer of 2020. Study characteristics are shown in Table 1.

The included articles describe changes to trainee schedules and deployments,^{12,17,19,21,25,28}  increased mental health support services,^{14,15,17,20,26,29}  and virtual programs to build and support peer communities.^{13,16,17,19,23,27,29}  Many described a combination of several interventions that were simultaneously implemented. Several hospitals expanded existing well-being interventions, such as resident counseling services, often adapting them to virtual delivery during the pandemic.^{13-15,18,25,26}  Most program directors increased communication with trainees and incorporated their feedback to improve the well-being initiatives.

Interventions targeted physical, psychological, and social dimensions of well-being. While most studies (16 of 18, 88.9%) included strategies to support psychological well-being,^13-24,26-29  9 studies (50.0%) described multiple interventions that targeted several well-being dimensions.^{13,17,19-21,23,25,27,28}  Four studies (22.2%) focused on 2 dimensions,^13,23,25,28  and 5 studies (27.8%) targeted 3 dimensions of well-being.^17,19-21,27 Table 2 describes the interventions.

Strategies to address physical well-being and safety focused on schedule changes to maintain clinical training and resources to care for patients with COVID-19 while ensuring social distancing,^{12,17,19,21,25,28}  addressing basic needs (lodging away from families, transportation, and food) during pandemic surges,^17,19,20,25  promoting good nutrition and exercise,^19,27  and providing personal protective equipment.^17,19,25  Interventions targeting psychological well-being included increasing access to mental health support and counseling,^{13-17,20,26,29}  peer support,^22,25,26,29  enhancing personal resilience through synchronous and asynchronous well-being curricula,^21,27,28  mandatory check-ins,¹⁹  and mindfulness training.^18,24  Strategies to promote social well-being included regular communication from program directors,^16,17,20,21  debriefings via video conferencing,^16,23,26,28  use of social media,^17,25  virtual social events,²⁵  and a wellness competition.²⁵  The evaluation period for most interventions was limited to 2 to 3 months during the peak phase of the COVID-19 pandemic; one study collected data over an 8-month period.²⁷ 

We assessed study quality by looking at which outcomes were reported and if sufficient information was provided to allow replication of the intervention. Seven studies did not provide any data on program effectiveness.^{13,16,17,19-21,28}  Of the interventions with outcome measures, 6 used self-reported satisfaction surveys,^{12,22,23,25-27}  and 5 reported utilization metrics.^{14,15,18,22,29}  Only 2 studies used well-being instruments with established validity evidence.^18,24  We planned to use the Medical Education Research Study Quality Instrument (MERSQI)³⁰  to assess study quality but found that most studies did not report sufficient data to facilitate scoring.

Our scoping review focused exclusively on studies describing well-being interventions for residents and fellows implemented during the COVID-19 pandemic. As the pandemic disrupted routine patient care and clinical training, residency programs worldwide quickly implemented support strategies and well-being interventions.⁸  There was convergence in these interventions, including schedule changes to manage workload, facilitate social distancing, and reduce exposure; enhanced mental health services; increased communication from leaders; and activities to promote social connections.

Although the COVID-19 pandemic affected resident training worldwide, most studies originated in the United States, which is consistent with the overall predominance of studies from the United States in the medical education literature.³¹  Additionally, most interventions targeted a single institution or a single residency program. This may be because workflows, exposure risks, and educational program designs differ among institutions and programs, highlighting the need for tailored interventions. It is notable that some interventions focused on structural or organizational changes, such as schedule modifications, while others focused on individual-level initiatives, such as efforts to promote mindfulness and personal resilience. One study explicitly mentioned the use of both types of interventions.²⁵  Prior systematic reviews of programs to address well-being and reduce burnout have found that both individual and structural/organizational changes were effective.^6,32 

Most studies in our review targeted surgical and anesthesiology residency programs. Possible explanations include the specialty-specific effect of the pandemic on clinical disruptions and exposure risk. A systematic review of approaches to maintaining resident education during the pandemic showed surgical residents’ medical education was greatly disrupted by the cancelation of elective surgeries.³³  This may have provided some specialties with more time for well-being interventions and/or scholarly publications. Anesthesiology trainees were commonly redeployed to critical care settings where they cared for the sickest patients and often performed procedures with a high risk of infection,²⁸  likely contributing to the increased need for well-being interventions in this group. Only one study targeted emergency medicine residents,²³  who were at the forefront of patient care. It is unclear whether emergency medicine residency programs did not implement well-being programs or were too busy with clinical workload to disseminate their initiatives. However, the overall number of interventional studies is too low to provide a robust analysis of observations within study subgroups.

The volume of initial articles screened suggests the importance placed by educators and researchers on resident well-being during the COVID-19 pandemic. The small number of articles meeting inclusion criteria likely demonstrates the obstacles that educators faced in developing, implementing, and studying the outcome of interventions during a global health crisis. Position statements, perspectives, and other noninterventional publications on well-being dominated the literature. Additionally, of the 18 studies identified, few evaluated the effectiveness of the interventions or provided enough information for generalizability or quality assessment. It is possible that clinical demands, workforce shortages, and social distancing mandates hindered the ability to design and implement well-being interventions. This is consistent with other studies of well-being initiatives during global crises. For example, a review of 117 well-being studies of health care professionals during viral outbreaks (SARS, MERS, Ebola, H1N1, and H7N9) found only 4 that addressed mental health.³⁴  A Cochrane systematic review also found that studies conducted during earlier epidemics provided limited evidence to inform well-being interventions for frontline workers during the COVID-19 pandemic.³²  Notably, resident well-being studies prior to the pandemic have also been criticized for lack of methodologic rigor and limited use of instruments with established validity evidence.³⁵ 

While our findings do not allow us to make specific recommendations regarding the effectiveness of the well-being interventions studied, they do provide important information on approaches taken and address gaps in the current literature. Substantial efforts are needed to design, implement, and evaluate research on resident well-being interventions in global crises. This can be done prospectively by individual institutions, through resource and expertise sharing among institutions, and/or by partnering with organizations that promote research on trainee well-being, such as the Accreditation Council for Graduate Medical Education, which offered guidance to teaching institutions for promoting resident and faculty well-being during the COVID-19 pandemic.³⁶  Our study suggests that future research on trainee well-being interventions should improve the methodologic rigor of research study designs, increase follow-up periods to enable longitudinal observation of effectiveness, and establish assessment protocols and consensus outcome measures.

Limitations of our study include the exclusion of non-English language manuscripts, which may limit generalizability and adoption or adaptation to different environments and contexts. For studies that reported multiple interventions, it was not possible to determine which interventions were associated with the reported outcomes. Although we searched the grey literature, it is likely that initiatives were launched during the COVID-19 pandemic but were never published in the literature or disseminated beyond their program or institution.

This review found that studies on interventions to support residents and fellows during the COVID-19 pandemic were limited to a single program or institution and were rarely sustained after the peak weeks of the pandemic. Although multiple interventions are described, the studies are limited by the lack of methodological rigor and long-term outcome measures. Research is still needed to better understand the components of effective interventions to safeguard resident well-being during times of crisis.

The online supplementary data contains the search protocol used in the study.

* These authors contributed equally to this work.

Funding: The authors report no external funding source for this study.

Conflict of interest: The authors declare they have no competing interests.