Abstract
Background
During the COVID-19 pandemic, elective cases across the nation were suspended, leading to major decreases in operative volume for surgical trainees. Surgical resident operative autonomy has been declining over time, so we sought to explore the effect COVID-19 had on resident autonomy within VA teaching hospitals.
Methods
A retrospective analysis of surgical cases across specialties was performed using the VA Surgical Quality Improvement Program database from September 2019 to September 2021 at VA teaching hospitals. Supervision codes are recorded prospectively: attending surgeon performs the operation (AP), resident completes majority of the case with the attending scrubbed (AR), and resident is primary surgeon without attending scrubbed (RP).
Results
20,457 cases pre-COVID decreased to 11,035 during peak-COVID (P < .001). Overall, RP cases increased from 6.5% to 7.6% during the peak (P < .001) and trended back downwards during the recovery periods. AP decreased initially (29.9%-27.7%, P < .001), but regressed back to pre-pandemic numbers. In general surgery RP cases, urgent cases such as laparoscopic cholecystectomies increased from 18.8% to 27.5%, while elective repairs decreased during the peak. Similar changes were noted across specialties.
Discussion
Operative cases dropped by half from pre- to peak- COVID and remained 20% below pre-pandemic volume the following year. Interestingly, RP rates increased for several specialties during the peak of the pandemic, which may have resulted from a relative higher ratio of resident personnel:case volume and shift in case distribution from elective to urgent. The increase in RP rate has begun to regress to pre-COVID levels which need to be readdressed.
Key Takeaways
• Operative volume decreased significantly during the peak of the COVID-19 pandemic and did not recover to previous numbers, leading to ongoing reduction in resident case exposure. • Although operative autonomy was higher during peak-pandemic for trainees, it has begun to regress, generating concern that resident independence will continue to fall as affected classes progress in their training. • Tools such as simulation, practice kits, and virtual education can be used to bridge the gap in deficiencies that resulted from the decreased volume experienced during the pandemic.
Introduction
The Coronavirus Disease 2019 (COVID-19) took the world by storm after initial cases were discovered in China in December 2019. By January 20, 2020, the first case was diagnosed in the United States (US) and on March 11, 2020 the World Health Organization declared COVID-19 a pandemic after more than 118,000 cases in 114 countries were reported. 1 In anticipation of the surge in the US and the predicted strain to healthcare systems, the American College of Surgeons issued a statement urging healthcare systems, hospitals, and surgeons to minimize elective procedures. 2 Hospitals across the country suspended elective cases and precautions were initiated to conserve resources such as personal protective equipment, ventilators, and available beds. Nationwide, Veterans Affairs (VA) medical centers stopped elective cases on March 16, 2020 with no uniform date to resume cases. 28 million elective cases were estimated to be canceled worldwide during the peak that was expected to last 12 weeks. 3 At the time, it was unknown how long emergency protocols would continue.
Surgical residents were among the foundation of essential workers who continued to care for patients throughout COVID-19. Surveys of resident experiences during the pandemic demonstrated burnout and concerns about progression in their surgical training.4,5 Education became mostly virtual or self-study and opportunities for hands-on training in the operating room were diminished. Several residency programs underwent emergency restructuring to reassign trainees to different rotations, with several being non-operative such as intensive care units, to help manage the overwhelming volume of patients with COVID.6-8 Efforts were directed to managing critically ill patients while operative volume declined.
Prior to the pandemic, concerns of dwindling resident operative autonomy was rising. 9 Despite studies demonstrating no difference in morbidity and mortality with resident independence, a significant decline in resident autonomy in the last 2 decades was observed.10-12 The decrease in case volume during the pandemic continued to fuel concerns regarding training competent surgeons for independent practice. One study demonstrated that 14%-18% of trainees expected a negative impact in the progression of operative autonomy due to the pandemic. 5 This study explores the effect of the COVID-19 pandemic on resident autonomy within VA teaching hospitals.
Methods
Study Design and Population
This is a retrospective analysis utilizing surgical data from the VA Quality Improvement Program (VASQIP) database. Cases performed between September 16, 2020 and March 31, 2021 in general surgery, neurosurgery, orthopedic surgery, otolaryngology (ENT), plastic surgery, urology, and vascular surgery at teaching VA medical centers were included in this study. VA medical centers with an affiliated general surgery residency since 2010 were classified as teaching hospitals. The data were securely stored on the VA Informatics and Computing Infrastructure (VINCI). This study was determined to be exempt by the institutional review board of a VA medical center and approved by VA Informatics and Computer Infrastructure (DART 2020-05-042-D). A waiver of informed consent was granted for the use of de-identified data.
Surgical procedures were excluded if they were not in the listed subspecialties, did not have documented level of resident supervision, if they involved patients younger than 18 years of age, and if they were performed on patients with American Society of Anesthesiologists physical status classification VI (indicating the patient was declared brain dead).
Cases were categorized into 3 supervision groups based on level of resident and attending involvement. Attending primary (AP): the procedure is performed by the attending with or without the assistance of a resident; attending-resident (AR): the resident performs more than 50% of the case with the attending scrubbed; resident primary (RP): the operation is completed by the resident without the attending scrubbed. In RP cases, the attending is either supervising directly in the operating room or immediately available in the operating suite for supervision or consultation as needed. These last 2 groups are separately coded in VASQIP, but were combined due to a very low number of cases where the attending surgeon was not present in the operating room. 12 RP cases signify resident autonomous operations. Supervision codes are prospectively recorded by the circulating nurse in the operating room at the time of surgery and confirmed with the attending surgeon at the completion of the case.
Variables and Statistical Analysis
Variable definitions and criteria for VASQIP are publicly available. 13 Six-month blocks were used to represent pre-COVID (Sept. 16, 2019 to March 15, 2020), peak-COVID (March 16, 2020 to Sept. 15, 2020), early recovery (Sept. 16, 2020 to March 15, 2021), and late recovery (March 16, 2021 to Sept. 31, 2021). Our primary outcome was the proportion of RP, AR, and AP cases within the peak-COVID timeframe compared to pre-COVID. Secondary outcomes included proportion of emergency cases overall and for each subspecialty we examined: case volume, the proportion of RP, AR, and AP cases during peak-COVID vs pre-COVID, and changes in the overall trend in supervision including early and late recovery. Current Procedural Terminology (CPT) codes were reviewed to determine the most commonly performed RP operations during the time frame of the study.
All statistical analyses were performed using SAS software, version 9.4 (SAS Institute Inc) for Windows (Microsoft, Inc). Categorical variables were compared with chi-squared methods and reported as counts (percentage). Continuous variables were compared with ANOVA or Mann-Whitney U test and reported as means (standard deviation) or median (interquartile range) as appropriate. Trends were analyzed by chi-squared for trends. The significant threshold was 2-sided P < .05.
Results
The final population included 61,683 cases after exclusions were performed (Figure 1). During the pandemic, overall operative volumes were significantly reduced from 20 457 pre-COVID to 11,035 at the peak (P < .001). Rates of emergency cases increased from 4.6% to 7.6% (P < .001). All specialties experienced a fall in case volume at the height of the pandemic: general surgery by 21.5%, neurosurgery by 41.8%, orthopedic surgery by 57%, ENT by 48%, plastic surgery by 56.8%, urology by 39.1%, and vascular surgery by 27.6% (all P < .001). Case volume started to rise again, but even a year afterwards in late recovery, the total volume was still 20% below pre-pandemic numbers (Table 1). None of the specialties recovered to their previous levels during the time frame of the study (all trends P < .001). Inclusion and exclusion flowchart. VASQIP, Veterans Affairs Surgical Quality Improvement Program; VA, Veterans Affairs; ASA, American Society of Anesthesiologists. Change in Operative Volume Over Time. COVID, Coronavirus Disease; ENT, Otolaryngology.
Comparison of Supervision Level Through the COVID Pandemic.
COVID, Coronavirus Disease.
Bold P-value significant at P < .05.
aComparison of pre-COVID to peak-COVID.
bAnalysis for entire trend.
Compared to pre-COVID, there were differences in trends during peak-COVID for each specialty (Figure 2). Urology (16.4%-18.6%, P = .04), neurosurgery (7.4%-10.4%, P = .04) and vascular surgery (3.3%-4.8%, P = .01) all experienced significant increases in RP cases. General surgery, orthopedic surgery, ENT, and plastic surgery all had no significant changes in RP cases. Orthopedic surgery (60.6%-64.5%, P < .001) had a significant increase in AR cases while ENT (79.3%-70.0%, P = .002) had a significant decrease. ENT (15.1%-26.3%, P < .001) conversely had an increase in AP cases while neurosurgery (P = .04), orthopedics (P < .001), and vascular surgery (P = .02) experienced a decrease. General surgery and plastic surgery were the only specialties without any changes in any supervision level during the peak of pandemic. (A) Change in proportion of Resident Primary cases through COVID pandemic. (B) Change in proportion of Attending Resident cases through COVID pandemic. (C) Change in proportion of Attending Primary cases through COVID pandemic. COVID, Coronavirus Disease; ENT, Otolaryngology.
When examining the entire trend, most specialties returned to pre-COVID levels. Only urology (18.1% early recovery, 19.0% late recovery, P = .01) demonstrated an increase in RP cases over the full course of the study, while none of the other subspecialties had significant trends in RP cases in either direction. This increase for urology correlates to a subsequent significant decrease in AR cases (67.5% pre-COVID, 66.7% peak COVID, 64.9% early recovery and 65.0% late recovery, P = .02). Vascular surgery also experienced a significant decrease in AR cases (61.1% vs 62.9% vs 60.3% vs 58.9%, P = .04). Plastic surgery underwent a significant decrease in AP cases (21.5% vs 27.7% vs 22.8% vs 15.0%, P = .04), while general surgery had a significant increase in AP cases (34.7% vs 33.6% vs 34.9% vs 36.7%, P = .03).
Frequency of the Top 5 Common Resident Primary (RP) Cases in General Surgery Over Time, CPT Code in Parentheses.
COVID, Coronavirus Disease; Lap, Laparoscopic.
Discussion
The COVID-19 pandemic has left a lasting impact on all medical education, with surgical resident training being particularly affected. Operative volume was reduced by half during the height of the pandemic. Although this was expected from the implementation of emergency protocols, case volume was not restored and remained one-fifth less than pre-pandemic volume a year later. Additional waves of the virus, testing prior to elective cases, and patient caution in returning to healthcare settings likely contributed to the slow recovery in volume. Diminished operative cases were reported by several surgical residency programs, with one program reporting as high as a 74% reduction.14-16 Junior trainees demonstrated higher concerns of progressing in their training from the reduction in volume compared to seniors and graduating chiefs.4,5,15,16 The immediate effect of the pandemic was recognized by the American Board of Surgery and temporary solutions were put into place. Graduation requirements were modified for the class of 2020 and the minimum requirement of 250 cases by PGY2 was extended into the first 6 months of PGY3. 17 These provisional modifications allowed surgical residents to progress in their training, but the lasting impact on junior classes may be overlooked. Early trainees lost opportunities to build the foundation of their core surgical skills. While we have shown a persistent decrease in case volume 1 year after the peak of COVID, other studies have not looked at whether cases have currently returned to prior rates outside the VA system.
Suspension of nonemergent operations leads to a shift in the types of cases that residents encountered during and after the pandemic. Specialties that have a higher proportion of elective cases compared to emergency cases such as plastic surgery, ENT, and orthopedic surgery experienced a larger drop in volume. Common RP cases that were core procedures, shifted from elective towards urgent operations during the pandemic. General surgery experienced a decrease in RP cases such as umbilical hernia repairs by 80% and knee arthroplasties for orthopedic surgery were reduced by 68%. Modified radical neck dissection was the most common RP ENT case prior to the pandemic and was decreased by 53%. Fortunately, most specialties returned to near pre-pandemic ratios during the late recovery period, but the loss of exposure to those cases had already occurred. Ellison et al described similar findings with a reduction of 87% in elective operations vs a 20% decrease in emergency procedures. 5 Educators teaching the residents who worked through the COVID pandemic need to understand that the trainee's case exposure and hands on training will likely be less than that of their predecessors.
Resident operative autonomy overall was higher during the peak of the pandemic. One possible etiology for this finding is a previous study showing that VA hospitals with lower cases per resident afforded more autonomy. 18 Lower operative volume relative to resident complement allows for sufficient coverage and is an opportunity for more than one resident to be present in a case—a senior resident to take a junior through the case. Amman et al described a slightly increased rate of teaching assistant cases because residents were able to double-scrub more during the pandemic. 15 In addition, resident involvement in general and vascular surgery has been found to be higher for emergency cases relative to elective surgery. 19 This could have contributed to the increase in RP and AR cases during peak-COVID when there was a shift towards urgent and emergent procedures. Unfortunately, resident autonomous cases were declining again at the end of the study. In the late recovery period, AP cases were higher than the pre-COVID periods despite a drop during the peak. Technical skills and competency are critical components of affording autonomy, the growth of which may have been disrupted by the pandemic. At a single training program, 75% of attendings and more than half of their residents perceived a reduction in resident technical skills during COVID-19. 16 It is possible that near the end of the study, residents did not have the proficiencies to be provided autonomy, which contributed to increased rates of AP cases and the reduction in resident autonomous procedures.
This data is important for residency programs and surgical educators who may see deficiencies in operative skills and patient management resulting from lower volume experienced during the pandemic. If resident competency is not reaching appropriate levels in this time immediately post-pandemic, then a decrease in autonomy will occur and be further detrimental to resident training. Residents would need more intensive practice and case review to bridge the gap in technical skills and expand limited case exposure. Virtual teaching, simulations, and practice kits cannot fully imitate real patients and operative scenarios, but are valuable tools in this phase of recovery now and if a similar event were to occur again. 20 Development of low-cost training kits that residents can practice with outside of the simulation lab may promote further practice if operative volume remains a concern. Case volume does not always equal competency, but lessons in the operating room are invaluable and cannot be replicated. Further studies with competency based evaluations will help us to better understand and characterize the full effects of COVID. Although the pandemic has passed, we cannot overlook the continued recovery period as a critical time in implementing change in surgical education to improve training for future generations.
The retrospective nature of the study limits the variables and factors that are evaluated. Difficulties of a large database study include concerns about quality, however the VASQIP database has been found to be highly accurate. 21 As this is a combination of over 90 facilities with trainees, there is heterogeneity in the practices and patterns at the different centers which impacts the results, but also may allow more generalizability of the results. Although the designation of RP is clear because the attendings are not scrubbed, there is a risk of misclassification bias despite the definition of 50% of the procedure. Additionally, it is difficult to capture conversions from RP to AR and what the reasons for conversions were. No other forms of bias are believed to be applicable for this study and database. The pandemic affected the whole nation, but the burden to each medical center and training program varied. Despite this, the VA health care system is affiliated with many surgical residency programs across the nation and contributes to a significant portion of their training. There was not a uniform date that elective cases were resumed in the VAHCS, so it is difficult to assess the status of recovery in each hospital. Lastly, we are unable to describe resident reallocation during the pandemic, which can contribute to the resident to case ratio and affect resident autonomy.
The full impact of COVID-19 on surgical education has yet to be determined, but this study demonstrates that the changes are not short-lived. With the decrease in operative volume for surgical residents and a shift in their case exposure, attention must be focused on bridging the gap in both knowledge and surgical skills to return to pre-pandemic levels. Only then will we be able to maintain and hopefully improve resident operative autonomy after the unexpected pandemic ran its course. Future studies examining resident competency in the operating room may identify deficiencies and better tailor surgical education to post-COVID needs.
Footnotes
Author's Note
This work was presented as a podium presentation at the 2023 annual Association For Surgical Education meeting held in San Diego, California on April 15, 2023.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.