Introduction
The COVID-19 pandemic led to significant, widespread impacts on surgical care for pediatric patients.1–3 As key agencies recommended delaying nonurgent cases after declaration of the pandemic, pre–hospital procedure cancellations became common.4 Expectedly, peak waves of COVID-19 infections were associated with a decrease in overall surgical volume.1,3 While most cancellations occurred prior to scheduled procedure date, same-day cancellations (SDCs) persisted. These SDCs affected children whose time-sensitive care was intentionally sanctioned despite the global health crisis; procedure cancellations, especially SDCs, are likewise known to represent significant financial, emotional, and employment burdens on patient families.5,6 Understanding that a sustained increase in SDCs would magnify these negative effects on patients and families, we sought to describe the impact of COVID-19 on SDC incidence, with a focus on SDCs due to acute illness, in pediatric patients during the prepandemic, early, and late waves at an urban, tertiary children’s hospital.
Methods
This retrospective study analyzed pediatric SDCs at a single institution from May 1, 2018, to January 1, 2022, representing approximately 22 months before and after the World Health Organization declared a global pandemic due to COVID-19. Patients aged 18 years or younger who had an SDC of a scheduled outpatient procedure were included for analysis. Patients with a cancellation prior to the day of surgery were excluded. Sociodemographic and surgical data were obtained for patients who met inclusion criteria from the electronic health record, including cancellation reason. The total number of completed cases each month was abstracted, but no other information on this group was collected. SDCs prior to March 2020 were defined as prepandemic, those March to November 2020 as early waves, and those November 2020 to January 2022 as late waves. Average SDC percentage per month was calculated as the number of SDCs per month divided by the total number of scheduled cases each month (completed cases and SDCs). The proportion of SDCs due to acute illness per month was calculated by dividing the number of SDCs due to acute illness by the total number of SDCs. The institutional review board of the Medical University of South Carolina determined that patient consent was not required for this study.
Results
A total of 808 SDCs occurred during the study period, of which 396 occurred prior to and 412 occurred during the pandemic. There were no notable differences in patient age, self-identified race or ethnicity, primary language, insurance status, or American Society of Anesthesiologists (ASA) status in the children who had SDCs prior to or after the pandemic began (Table 1). Notably, more than half of children had an unknown ASA status because this is not uniformly documented in cancelled cases due to an absent or incomplete preoperative note. Compared with the prepandemic period, there was a steep initial drop in operative case volume after the pandemic was declared followed by rebound of total number of completed cases during late-pandemic waves (Figure 1). There was no statistically significant difference in average SDC percentage between prepandemic and early-wave time periods (4.8%–3.9%; P = .063). The average SDC percentage during the early waves was significantly lower than the late waves, rising to an average SDC percentage of 5.9% at the end of the study period (P = .004). Similar trends were noted for the proportion of SDCs due to acute illness, in which no significant difference was noted in the average percentage prior to or in early waves of the pandemic (24.8% and 26.6%, respectively; P = .692). However, the proportion of SDCs for acute illness was higher for the late waves (37.1%) compared to prepandemic and early waves (P < .001 and P = .048, respectively).
Characteristics of Pediatric Outpatient Day of Surgery Cancellations
. | Prepandemica . | Intrapandemic . | P Value . |
---|---|---|---|
(n = 396) . | (n = 412) . | . | |
Age, years, median (IQR) | 4 (1–10) | 5 (1–11) | .165 |
Sex assigned at birth, n (%) | .003 | ||
Female | 138 (34.9) | 188 (45.2) | |
Male | 258 (65.1) | 228 (54.8) | |
Race, n (%) | .228 | ||
White | 157 (39.6) | 185 (44.5) | |
African American | 176 (44.4) | 173 (41.6) | |
Other/unknown | 63 (15.9) | 58 (14.1) | |
Primary language, n (%) | .349 | ||
English | 363 (91.7) | 389 (93.5) | |
Other | 33 (8.33) | 27 (6.49) | |
Insurance, n (%) | .694 | ||
Private | 101 (25.5) | 117 (28.1) | |
Public | 285 (72.0) | 288 (69.2) | |
Uninsured | 10 (2.53) | 11 (2.64) | |
ASA status, n (%) | .662 | ||
I | 57 (14.4) | 51 (12.26) | |
II | 86 (21.7) | 97 (23.32) | |
III | 35 (8.84) | 42 (10.1) | |
IV | 2 (0.51) | 5 (1.20) | |
Undocumented/unknown | 216 (54.6) | 221 (53.1) | |
Procedure service, n (%) | |||
General surgery | 97 (24.5) | 43 (10.3) | <.001 |
Plastics | 6 (1.52) | 4 (0.96) | .537 |
Otolaryngology | 121 (30.6) | 134 (32.2) | .650 |
Orthopedics | 23 (5.81) | 34 (8.17) | .217 |
Ophthalmology | 43 (10.9.) | 44 (10.6) | .910 |
Urology | 62 (15.7) | 60 (14.4) | .625 |
Gastroenterology | 19 (4.80) | 39 (9.38) | .014 |
Neurology | 6 (1.52) | 7 (1.68) | 1.000 |
Dental | 1 (0.25) | 18 (4.33) | <.001 |
Other specialtyb | 12 (3.03) | 15 (3.61) | .699 |
. | Prepandemica . | Intrapandemic . | P Value . |
---|---|---|---|
(n = 396) . | (n = 412) . | . | |
Age, years, median (IQR) | 4 (1–10) | 5 (1–11) | .165 |
Sex assigned at birth, n (%) | .003 | ||
Female | 138 (34.9) | 188 (45.2) | |
Male | 258 (65.1) | 228 (54.8) | |
Race, n (%) | .228 | ||
White | 157 (39.6) | 185 (44.5) | |
African American | 176 (44.4) | 173 (41.6) | |
Other/unknown | 63 (15.9) | 58 (14.1) | |
Primary language, n (%) | .349 | ||
English | 363 (91.7) | 389 (93.5) | |
Other | 33 (8.33) | 27 (6.49) | |
Insurance, n (%) | .694 | ||
Private | 101 (25.5) | 117 (28.1) | |
Public | 285 (72.0) | 288 (69.2) | |
Uninsured | 10 (2.53) | 11 (2.64) | |
ASA status, n (%) | .662 | ||
I | 57 (14.4) | 51 (12.26) | |
II | 86 (21.7) | 97 (23.32) | |
III | 35 (8.84) | 42 (10.1) | |
IV | 2 (0.51) | 5 (1.20) | |
Undocumented/unknown | 216 (54.6) | 221 (53.1) | |
Procedure service, n (%) | |||
General surgery | 97 (24.5) | 43 (10.3) | <.001 |
Plastics | 6 (1.52) | 4 (0.96) | .537 |
Otolaryngology | 121 (30.6) | 134 (32.2) | .650 |
Orthopedics | 23 (5.81) | 34 (8.17) | .217 |
Ophthalmology | 43 (10.9.) | 44 (10.6) | .910 |
Urology | 62 (15.7) | 60 (14.4) | .625 |
Gastroenterology | 19 (4.80) | 39 (9.38) | .014 |
Neurology | 6 (1.52) | 7 (1.68) | 1.000 |
Dental | 1 (0.25) | 18 (4.33) | <.001 |
Other specialtyb | 12 (3.03) | 15 (3.61) | .699 |
Abbreviation: ASA, American Society of Anesthesiologists.
Patients with cancellations before March 1, 2020, were considered prepandemic.
Other specialties included cardiac, hematology/oncology, radiology (including interventional), cardiology, pulmonology, nephrology, transplant, dermatology, and burn.
Temporal trends in the percentage of same-day cancellations (SDCs) overall (blue) and percentage of SDCs due to acute illness only (red) relative to total scheduled cases by month. Evaluation of the percentage of cancellation (ie, the probability of cancellation over time) was evaluated using a logistic regression model with cancellation (yes/no) as the dependent variable and the natural cubic spline of month from the initiation date (initiation month = 1) as the independent variable; a similar model was used to assess percentage of SDCs due to acute illness alone.
Temporal trends in the percentage of same-day cancellations (SDCs) overall (blue) and percentage of SDCs due to acute illness only (red) relative to total scheduled cases by month. Evaluation of the percentage of cancellation (ie, the probability of cancellation over time) was evaluated using a logistic regression model with cancellation (yes/no) as the dependent variable and the natural cubic spline of month from the initiation date (initiation month = 1) as the independent variable; a similar model was used to assess percentage of SDCs due to acute illness alone.
Discussion
This retrospective analysis of over 800 clinical encounters at a single institution identified a concerning trend within late waves of the pandemic in which increasing SDCs and rising percentage of cancellations due to acute illness were noted. While a time-concordant peak wave of COVID-19 infections could have impacted these cancellation trends, the tail end of our study period occurred during a local nadir in terms of daily positive cases.7 Increasing SDCs for acute illness may reflect heightened awareness and concern for unacceptable perioperative respiratory risk at a provider level. Unilateral decisions to cancel surgery for recent respiratory infection (whether COVID-19 or a routine respiratory virus) may not be fully supported by available evidence on outcomes in children undergoing surgery with recent symptoms.8–11 Although the pandemic has concluded, pediatric surgical care remains significantly affected by respiratory viral infection rates and seasonality. Prior work has shown likewise significant variation in interprovider decision-making regarding cancellation for acute illness.12 Investigating potential changes in clinical decision-making and cancellation rates thus represents an important element in caring for the more than 3 million children who receive surgical treatment in the US every year.13 Targeting SDCs—which has an exaggerated negative effect compared with earlier cancellation—will allow the greatest minimization of the medical, financial, and emotional impacts on these children and their families.5,6 Given this study’s limitations via its retrospective and single institutional nature, future work should investigate multi-institutional trends of SDCs and cancellations for acute illness across different pediatric populations and geographic areas. A more nuanced understanding of factors affecting SDC rates will allow institutions to implement triage strategies, mitigate short-notice cancellations, and address potential disparities within perioperative care.
Dr Tanious conceptualized and designed the study, collected data, carried out the initial analysis and critically reviewed and revised the manuscript. Dr Bunnell collected data, carried out initial analysis, and critically reviewed and revised the manuscript. Dr Barnett designed the data collection instrument, collected data, and drafted the initial manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
CONFLICTS OF INTEREST DISCLOSURES: The authors have no conflicts of interest relevant to this article to disclose.
FUNDING: Support was provided solely from institutional and/or departmental sources.