Abstract
Hospitals adapt to surges in a variety of ways, including expediting admitting and discharge procedures, hiring temporary staff and locums to meet increased demand, and temporarily expanding or creating facilities. To mitigate negative effects associated with surges, a small community-based hospital’s Internal Medicine (IM) residency program and hospitalist program collaborated by shifting senior residents into admitting physician roles with indirect supervision to meet increased surge needs. This shift allowed an admitting attending physician to shift to a newly created rounding service and, on average, the number of rounding services was increased from seven to nine. Over a fifteen-month period (five quarters), there was a substantial increase in patient census and rounding physicians, paired with minimal differences in individual physician loads, suggesting that this approach helped alleviate strain on the hospital system. Additionally, this approach was well-received by residents, with residents reporting that the workload during the surge was appropriate, that the balance between service and education was appropriate, and that the surge experience was beneficial for their education and ability to practice independently. In summary, this innovative surge management technique has the potential to help mitigate the risk of burnout among hospitalists, while prioritizing resident wellness and education. Importantly, this novel approach may be reproducible in other community hospitals and scalable in larger institutions with residency programs.
Keywords: COVID-19 Surge, Medical residents, Patient census
1. Introduction
The COVID-19 pandemic had a profound impact on hospitals in many ways, including resident education. Specifically, the surge in inpatient admissions forced hospitals to care for patients beyond their typical capacity.1,2 At teaching hospitals, residents were frequently re-assigned to meet the growing crisis in hospital capacity.3–5 Increased utilization adversely influenced resident medical education and wellness, with many residents reporting increased rates of burnout and clinical education disruption.3,4 Consistent with previous surges in 2020–2022, models forecasted a surge in late December 2022 and January 2023. Hospitals adapted to subsequent surges in a variety of ways, including expediting admitting and discharge procedures, hiring temporary staff and locums to meet increased demand, and temporarily expanding or creating facilities to meet needs.6,7
In an effort to mitigate negative effects associated with surges, the hospital’s Internal Medicine (IM) residency program and hospitalist program collaborated to shift senior residents into admitting physician roles with indirect supervision to meet increased surge needs. By shifting admitting and rounding positions, the hospital was able to help alleviate pressure associated with COVID-19 surges.
2. Methods
Corewell Health Lakeland is a community-based health system in Southwest Michigan with a 220-bed acute care flagship hospital. In addition, the IM residency had 23 residents at the time of the pandemic, including seven senior residents. Prior to the COVID-19 surge, admissions performed by residents were staffed by attending physicians on the same day of admission. There were typically two admitting physicians during daytime hours. When staffing an admission prior to the surge plan, all admissions seen before midnight would be billed on the initial day of hospitalization. A typical rounding service would have one attending physician working with a senior resident and two interns (Fig. 1). During times of high patient census, a third-year senior resident from the rounding hospitalist service was shifted to the admitting service and worked with a previously scheduled admitting senior resident. One of the two admitting attending physicians was shifted to a newly created rounding service. The two admitting team residents, who previously would have staffed patients with an attending on admitting duty, admitted patients independently. Additional supervision was provided if requested at time of admission via the remaining admitting attending, who was on-site at all times during the admitting shift. A rounding physician received the patient on the next calendar day, attested resident notes, and billed for the admitting service on the second day of hospitalization. In addition, the hospitalist service added an additional surge service, based on rotating hospitalists, subject to provider availability. The goal was to increase the number of rounding services from seven to nine. The COVID surge plan was implemented in early January 2022.
Fig. 1.
Hospitalist team structure.
Prior to implementation of the surge plan, compliance with existing ACGME Common Program8 and Internal Medicine Requirements was reviewed with Graduate Medical Education (GME) leadership. The hospitalist service operates a night float system and rounding services do not work overnight. The surge plan was deemed compliant by GME leadership in terms of duty hours, patient caps, and service/education balance. In addition, as residents’ admitting shifts were considered indirect supervision and billed by the rounding attending physician on day #2, their practice was based on their training license and not an independent practice license.
Hospitalist census per team, physician counts, and individual physician loads over a fifteen-month period using a single-center observational study were retrospectively reviewed and examined. Information was analyzed using arithmetic means, standard deviations, and one-way ANOVAs. All statistical analyses were conducted using statistical package for social sciences (SPSS) version 29.0.0.0. All IM rounding attendings were included in the study, except an Infectious Disease/Hospitalist physician who consistently carried an additional hospitalist census of five. The total hospitalist census was adjusted to account for those five patients. In addition, senior residents at the time of the hospitalist surge were surveyed to gauge their educational experiences.
3. Results
Hospitalist census, average number of physicians, and individual averages were collected over the span of 15 months or five quarters—from the beginning of 2021 to the end of the first quarter of 2022. Over those five quarters, the mean hospitalist census was 114.59 (SD = 14.11; min = 77; max = 171), the mean average number of physicians was 7.31 (SD = 0.58; min = 6; max = 9), and the mean individual average was 15.70 (SD = 1.51; min = 11; max = 21) patients per attending physician. Quarterly means, standard deviations, minimums, and maximums are reported in Table 1.
Table 1.
Descriptive statistics by quarter.
| Mean | SD | Min | Max | ||
|---|---|---|---|---|---|
| Hospitalist Census | 2021 1st Quarter | 105.29 | 11.91 | 77 | 126 |
| 2021 2nd Quarter | 108.38 | 8.11 | 88 | 132 | |
| 2021 3rd Quarter | 111.68 | 8.98 | 87 | 136 | |
| 2021 4th Quarter | 122.31 | 9.57 | 98 | 140 | |
| 2022 1st Quartera | 125.12 | 18.02 | 87 | 171 | |
| Average # of Physicians | 2021 1st Quarter | 6.99 | 0.36 | 6 | 8 |
| 2021 2nd Quarter | 7.07 | 0.25 | 7 | 8 | |
| 2021 3rd Quarter | 7.05 | 0.23 | 7 | 8 | |
| 2021 4th Quarter | 7.41 | 0.49 | 7 | 8 | |
| 2022 1st Quartera | 8.03 | 0.67 | 6 | 9 | |
| Individual Averages | 2021 1st Quarter | 15.06 | 1.56 | 11.00 | 18.0 |
| 2021 2nd Quarter | 15.41 | 0.97 | 12.60 | 17.0 | |
| 2021 3rd Quarter | 15.83 | 1.20 | 12.40 | 19.4 | |
| 2021 4th Quarter | 16.63 | 1.39 | 13.40 | 19.9 | |
| 2022 1st Quartera | 15.55 | 1.83 | 11.80 | 21.1 |
Period of added surge capacity.
To assess whether hospitalist census, average number of physicians, and individual averages differed in surge quarters (2021 4th Quarter and 2022 1st Quarter) relative to more typical quarters (2021 1st, 2nd, and 3rd Quarters), three one-way ANOVAs were conducted. Hospitalist census, average number of physicians, and individual averages were all significant at the 0.001 level (F(4, 445) = 48.422, p < 0.001, F(4, 445) = 91.907, p < 0.001, F(4, 445) = 15.501, p < 0.001, respectively). Three separate post hoc Games—Howell tests were conducted, with the following results indicated in Table 2.
Table 2.
Post-hoc comparisons.
| Dependent Variable | Quarter (I) | Quarter (J) | Mean Difference (I–J) | SE | p |
|---|---|---|---|---|---|
| Hospitalist Census | 2021 1st Quarter | 2021 2nd Quarter | −3.097 | 1.534 | 0.262 |
| 2021 3rd Quarter | −6.397 | 1.583 | <0.001 | ||
| 2021 4th Quarter | −17.020 | 1.624 | <0.001 | ||
| 2022 1st Quartera | −19.836 | 2.297 | <0.001 | ||
| 2021 2nd Quarter | 2021 3rd Quarter | −3.300 | 1.265 | 0.073 | |
| 2021 4th Quarter | −13.923 | 1.315 | <0.001 | ||
| 2022 1st Quartera | −16.739 | 2.091 | <0.001 | ||
| 2021 3rd Quarter | 2021 4th Quarter | −10.623 | 1.372 | <0.001 | |
| 2022 1st Quarter | −13.439 | 2.127 | <0.001 | ||
| 2021 4th Quarter | 2022 1st Quartera | 2.816 | 2.158 | 0.689 | |
| Avg # Physicians | 2021 1st Quarter | 2021 2nd Quarter | −0.77 | 0.046 | 0.456 |
| 2021 3rd Quarter | −0.66 | 0.045 | 0.590 | ||
| 2021 4th Quarter | −0.418 | 0.064 | <0.001 | ||
| 2022 1st Quartera | −1.045 | 0.080 | <0.001 | ||
| 2021 2nd Quarter | 2021 3rd Quarter | 0.012 | 0.035 | 0.997 | |
| 2021 4th Quarter | −0.341 | 0.058 | <0.001 | ||
| 2022 1st Quartera | −0.968 | 0.075 | <0.001 | ||
| 2021 3rd Quarter | 2021 4th Quarter | −0.352 | 0.057 | <0.001 | |
| 2022 1st Quarter | −0.979 | 0.074 | <0.001 | ||
| 2021 4th Quarter | 2022 1st Quartera | −0.627 | 0.087 | <0.001 | |
| Individual Averages | 2021 1st Quarter | 2021 2nd Quarter | −0.351 | 0.196 | 0.380 |
| 2021 3rd Quarter | −0.767 | 0.209 | 0.003 | ||
| 2021 4th Quarter | −1.563 | 0.222 | <0.001 | ||
| 2022 1st Quartera | −0.488 | 0.256 | 0.317 | ||
| 2021 2nd Quarter | 2021 3rd Quarter | −0.416 | 0.161 | 0.077 | |
| 2021 4th Quarter | −1.212 | 0.177 | <0.001 | ||
| 2022 1st Quartera | −0.138 | 0.219 | 0.970 | ||
| 2021 3rd Quarter | 2021 4th Quarter | −0.796 | 0.192 | <0.001 | |
| 2022 1st Quarter | 0.279 | 0.231 | 0.747 | ||
| 2021 4th Quarter | 2022 1st Quartera | 1.074 | 0.242 | <0.001 |
Period of Added Surge Capacity.
Mean hospitalist census was significantly lower in the first quarter of 2021 relative to the third quarter of 2021 (p < 0.001), the fourth quarter of 2021 (p < 0.001), and the first quarter of 2022 (p < 0.001); the second quarter of 2022 was significantly lower relative to the fourth quarter of 2021 (p < 0.001) and the first quarter of 2022 (p < 0.001); and the third quarter of 2021 was significantly lower than to the fourth quarter of 2021 (p < 0.001) and the first quarter of 2022 (p < 0.001).
The average number of physicians was significantly lower in the first quarter of 2021 relative to the fourth quarter of 2021 (p < 0.001) and the first quarter of 2022 (p < 0.001); the second quarter of 2022 was significantly lower relative to the fourth quarter of 2021 (p < 0.001) and the first quarter of 2022 (p < 0.001); the third quarter of 2021 was significantly lower than to the fourth quarter of 2021 (p < 0.001) and the first quarter of 2022 (p < 0.001); and the fourth quarter of 2021 was significantly lower than the first quarter of 2022 (p < 0.001).
Individual averages were significantly lower in the first quarter of 2021 relative the fourth quarter of 2021 (p < 0.001); the second quarter of 2022 was significantly lower relative to the fourth quarter of 2021 (p < 0.001); the third quarter of 2021 was significantly lower than to the fourth quarter of 2021 (p < 0.001) and the first quarter of 2022 (p < 0.001); and the fourth quarter of 2021 was significantly higher than the first quarter of 2022 (p < 0.001).
Residents each averaged five to seven admissions per day during the surge period. Duty hours and patient caps were tracked with no reported violations from admitting or rounding residents. In terms of residents’ feedback regarding their experiences (Fig. 2), six of seven senior residents completed the survey. The residents felt that the workload during the surge was appropriate (50 % strongly agree, 50 % agree). The majority felt that the service and education balance was appropriate (83 % agree, 17 % neutral). The residents also unanimously reported that the surge experience was beneficial for their education and ability to practice independently (33 % strongly agree, 67 % agree), despite feeling overwhelmed by the responsibility (83 % strongly agree, 17 % agree).
Fig. 2.
Resident survey of shifting duties.
4. Discussion
Since patient surges significantly impact hospitalist census, physician staffing, and individual workloads, hospitals frequently employ different predictive models to anticipate future surges.7,9 The anticipated COVID-19 surge necessitated adaptive strategies to meet the demands of an increased inpatient census. During the COVID-19 pandemic, a small community hospital experienced a substantial increase in patient census and physician numbers, especially during the fourth quarter of 2021 and the first quarter of 2022 relative to earlier quarters. The IM residency program adapted by shifting senior residents into admitting physician roles to facilitate additional support of attendings to hospitalist rounding roles. This was done in conjunction with the hospitalists using an additional rotating service, consisting of hospitalists who otherwise would not have been working on a given day.
Statistical analyses demonstrated that this model was helpful in managing an increased hospitalist census. The significant increase in the average number of rounding physicians during the surge, paired with the less significant differences across quarters in average individual physician loads, suggests that this approach helped alleviate strain on the hospital system. In addition to patient care, the strain related with COVID-19 surges increased the risk of burnout among physicians.9 This intervention was designed to mitigate the risk of burnout among rounding teams by stabilizing the individual team census.
In terms of resident well-being, the COVID-19 pandemic nationally adversely affected both resident wellness and education in internal medicine and other specialties.4,5 Residents reported both increased burnout and a disruption in their clinical schedule.4 Specifically, residents were often required to perform duties outside of their typical environment.5 There may be potential concerns of disrupting the service and education balance in medical education, as residents may perceive an increase in service responsibility as adversely affecting their education.10 In 2020, ACGME emphasized the need to protect residents and highlighted the importance of complying with duty hours and patient caps during the COVID-19 pandemic.11 This novel approach for managing patient census surges attempted to balance both clinical responsibilities and medical education, which was reflected in the survey of affected senior residents. Both admitting and rounding residents were able to avoid duty hour and patient cap violations during the intervention period. Since the intervention period, ACGME has adopted an Emergency Program Categorization, which allows relief when all other measures have failed to meet a potential crisis. Fortunately, the program was able to mitigate the patient surge during the intervention period, and did not need to seek relief after that time period.12
Similar to national trends,5 the residents reported their experience with COVID-19 to be beneficial for independent clinical practice. Regarding teams relying on interns without a senior resident, there were no reports of patient harms due to the downsized teams. However, the change in team structure likely required both increasing autonomy by the interns and enhanced attending supervision. Although this novel approach was designed to reduce burnout among the rounding teams, including residents, it likely increased the stress load for the admitting residents. Future surge plans should continue to find ways to mitigate resident burnout.
A strength of the current study was that it provides a solution to a real-world situation which hospitals will continue to encounter. Another important aspect of these findings is that they may be reproducible in other community hospitals with residency programs. Limitations of the study include being a single-center observational study and that this model would require revision to be feasible in hospitals without resident physicians. These findings underscore the importance of flexible staffing models in times of patient surges in a community-based hospital setting. Further research could explore the potential effects of this model on resident education, burnout, and patient outcomes. Additionally, evaluating similar adaptive strategies in other institutions and assessing their scalability could help refine best practices for surge preparedness at larger facilities.
In conclusion, a collaborative effort between the hospitalist group and IM residency to shift residents to an independent admitting service was successful in reducing the risk for burnout during the COVID surge. Maintaining a balance between service and education was prioritized, and the overall effect on resident education was positive.
Footnotes
Conflicts of interest: All authors: no reported conflicts of interest.
Financial support: None.
Data availability statement
The data that support the findings of this study are available from the corresponding author, MW, upon reasonable request.
References
- 1. Holthof N, Luedi MM. Considerations for acute care staffing during a pandemic. Best Pract Res Clin Anaesthesiol. 2021;35(3):389–404. doi: 10.1016/j.bpa.2020.12.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Rahman NA, Guidry K, Brining ED, et al. Community hospital response to COVID-19 outbreak. West J Emerg Med. 2022;23(2):129–133. doi: 10.5811/westjem.2021.9.52294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Devitt J, Malam N, Montgomery L. A family medicine residency program’s response to an impending COVID-19 Surge. J Am Board Fam Med. 2021;34(Supplement):S217–S221. doi: 10.3122/jabfm.2021.S1.200193. [DOI] [PubMed] [Google Scholar]
- 4. St-Pierre F, Petrosyan R, Gupta A, et al. Impact of the COVID-19 pandemic on internal medicine training in the United States: results from a national survey. BMC Health Serv Res. 2023;23(1):1285. doi: 10.1186/s12913-023-10237-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Mazandi V, Gordon E. Resident education and redeployment during a disaster. Anesthesiol Clin. 2021 Jun;39(2):353–361. doi: 10.1016/j.anclin.2021.02.007. . Epub 2021 Mar 24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Watson SK, Rudge JW, Coker R. Health systems’ “surge capacity”: state of the art and priorities for future research. Milbank Q. 2013;91(1):78–122. doi: 10.1111/milq.12003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Klein MG, Cheng CJ, Lii E, et al. COVID-19 models for hospital surge capacity planning. Syst Rev. 2022 Feb;16(1):390–397. doi: 10.1017/dmp.2020.332. . Epub 2020 Sep 10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Accreditation Council for Graduate Medical Education. Common program requirements. https://www.acgme.org/globalassets/dh_dutyhourscommonpr07012007.pdf .
- 9. Keniston A, Patel V, McBeth L, Bowden K, Gallant A, Burden M. The impact of surge adaptations on hospitalist care teams during the COVID-19 pandemic utilizing a rapid qualitative analysis approach. Arch Public Health. 2022 Feb 17;80(1):57. doi: 10.1186/s13690-022-00804-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Galvin SL, Buys E. Resident perceptions of service versus clinical education. J Grad Med Educ. 2012 Dec;4(4):472–478. doi: 10.4300/JGME-D-11-00170.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Accreditation Council for Graduate Medical Education. The ACGME Common Program Requirements and COVID-19. https://www.acgme.org/newsroom/2020/3/the-acgme-common-program-requirements-and-covid-19/
- 12.Accreditation Council for Graduate Medical Education. Accreditation Council for Graduate Medical Education; https://www.acgme.org/newsroom/2022/11/programs-may-now-request-emergency-categorization/ [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author, MW, upon reasonable request.