The Effects of Coronavirus Disease 2019 on Pediatric Anesthesiologists: A Survey of the Members of the Society for Pediatric Anesthesia

Abstract

BACKGROUND: 

The coronavirus disease 2019 (COVID-19) pandemic has affected the personal and professional lives of all health care workers. Anesthesiologists frequently perform virus-aerosolizing procedures (eg, intubation and extubation) that place them at increased risk of infection. We sought to determine how the initial COVID-19 outbreak affected members of the Society for Pediatric Anesthesia (SPA) on both personal and professional levels. Specifically, we examined the potential effects of gender and age on personal stress, burnout, sleep deprivation, anxiety, depression, assessed job satisfaction, and explored financial impact.

METHODS: 

After receiving approval from the SPA Committees for Research and Quality and Safety and the Colorado Multiple Institutional Review Board, we e-mailed a questionnaire to all 3245 SPA members. The survey included 22 questions related to well-being and 13 questions related to effects of COVID-19 on current and future practice, finances, retirement planning, academic time and productivity, and clinical and home responsibilities. To address low initial response rates and quantify nonresponse bias, we sent a shortened follow-up survey to a randomly selected subsample (n = 100) of SPA members who did not respond to the initial survey. Response differences between the 2 cohorts were determined.

RESULTS: 

A total of 561 (17%) members responded to the initial questionnaire. Because of COVID-19, 21.7% of respondents said they would change their clinical responsibilities, and 10.6% would decrease their professional working time. Women were more likely than men to anticipate a future COVID-19–related job change (odds ratio [OR] = 1.92, 95% confidence interval [CI], 1.12-2.63; P = .011), perhaps because of increased home responsibilities (OR = 2.63, 95% CI, 1.74-4.00; P < .001). Additionally, 14.2% of respondents planned to retire early, and 11.9% planned to retire later. Women and non-White respondents had higher likelihoods of burnout on univariate analysis (OR = 1.75, 95% CI, 1.06-2.94, P = .026 and OR = 1.82, 95% CI, 1.08-3.04, P = .017, respectively), and 25.1% of all respondents felt socially isolated. In addition, both changes in retirement planning and future occupational planning were strongly associated with total job satisfaction scores (both P < .001).

CONCLUSIONS: 

The COVID-19 pandemic has affected the personal and professional lives of pediatric anesthesiologists, albeit not equally, as women and non-Whites have been disproportionately impacted. The pandemic has significantly affected personal finances, home responsibilities, and retirement planning; reduced clinical and academic practice time and responsibilities; and increased feelings of social isolation, stress, burnout, and depression/anxiety.

KEY POINTS

• Question: How has the coronavirus disease 2019 (COVID-19) global pandemic affected the personal and professional lives of pediatric anesthesiologists?
• Findings: The pandemic has significantly affected the personal finances, home responsibilities, job satisfaction, social isolation, and retirement planning of pediatric anesthesiologists with women and non-Whites disproportionally affected compared to males and Whites.
• Meaning: The effects of COVID-19 on the pediatric anesthesia workforce and its disproportionate effects on women and non-White physicians will require both private and academic practices to develop and institute programs to increase physician job satisfaction while decreasing attrition and expedited retirement.

Despite global awareness of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) since December 2019, coronavirus disease 2019 (COVID-19) became a global pandemic that caught the world unprepared in the early months of 2020.^1,2 Highly contagious, SARS-CoV-2 has infected ≥68.4 million people globally as of December 2020¹ and is capable of causing severe pneumonia, acute respiratory distress syndrome, and death, particularly in vulnerable populations such as older adults and those with chronic medical conditions (eg, cardiovascular disease, obesity, diabetes, respiratory disease, hypertension, and malignancy).^1,2 Anesthesiologists, anesthesiologist-intensivists, certified registered nurse anesthetists (CRNA), and certified anesthesiologist assistants (CAA) perform many procedures that place them at extremely high risk of COVID-19 infection in the operating room, radiology department, intensive care units, and wards of the hospital. Intubation and extubation, during which the SARS-CoV-2 virus can be aerosolized, present particular danger. Not surprisingly, these increased risks are associated with high levels of anxiety in health care providers, accompanied by shame and guilt for considering their own safety and welfare—rather than solely that of their patients—and putting their families at risk. The resultant profound effects include anxiety, burnout, depression, and moral injury.³

Evaluation of clinician wellness involves consideration of a number of factors, including stress and burnout.^4–6 These factors can negatively impact patient care and lead to increased medical errors, malpractice risk, and early retirement.^4,6 Greater clinician stress may lead to higher rates of drug and alcohol addiction, divorce, and suicide.^4,6 Clinicians are more likely to experience burnout symptoms than the general US workforce and are more likely to be dissatisfied with work-life balance.⁷ Before the pandemic, burnout rates among US physicians overall were estimated to be approximately 46%.⁷ Female physicians, who now compose more than 36% of academic anesthesiologists,⁸ have higher rates of burnout and may therefore be more significantly affected by COVID-19 because they bear a greater proportion of childcare and household responsibilities than men.^9–11

We sought to determine how the initial COVID-19 pandemic has affected members of the Society for Pediatric Anesthesia (SPA), a group representing 3245 pediatric anesthesiologists. Specifically, we sought to identify the effects of COVID-19 on changes in retirement planning and factors contributing to these changes including years in practice. In addition, we wanted to ascertain how COVID-19 may lead to alterations in future occupational plans and whether the outcome is gender-related. Finally, we attempted to surmise how stress, burnout, sleep deprivation, anxiety, and depression are associated with job satisfaction during the pandemic.

METHODS

Survey Design and Administration

This article was written in accordance with the Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines. The study was approved by the Colorado Multiple Institutional Review Board (IRB), and the requirement for written informed consent was waived. After approval from both the Research and the Quality and Safety committees of SPA, we sent an e-mail questionnaire to all SPA members (n = 3245) between June 27, 2020 and July 10, 2020 (Supplemental Digital Content, Table [Survey] 1, https://links.lww.com/AA/D396). The survey was created with input from SPA’s Research and Quality and Safety committees as well as a survey and database specialist. We piloted the survey with attendings and fellows of the Children’s Hospital Colorado to assess for language and comprehension. The questionnaire consisted of 35 questions. Of these, 22 were related to well-being and included measures of social isolation, perceived stress, job satisfaction, anxiety and depression; the Positive Emotions, Engagement, Relationships, Meaning, Accomplishment, Health (PERMAH) well-being survey^12,13; and the modified 2-question summative Maslach Burnout Inventory (MBI).^14,15 An additional 13 questions measured the effects of COVID-19 on current and future practice by assessing changes in finances,^12,13 retirement planning, academic time, academic productivity, and clinical and home responsibilities. We also asked whether the respondent felt safe at work (eg, was adequate personal protective equipment [PPE] provided, were patients tested for COVID-19) and if a family member had been stricken by the virus. Finally, we collected provider demographics, including age, gender, fellowship training, type of practice (private, academic, military, or other), site of practice (freestanding pediatric hospital, adult hospital, pediatric hospital within an adult hospital, ambulatory surgery center, or other), location of practice by time zone (East, Central, Mountain, Pacific, or outside of North America), and years in practice (Supplemental Digital Content, Table [Survey] 1, https://links.lww.com/AA/D396). All data were deidentified and collected with Research Electronic Data Capture (REDCap; Vanderbilt University, Nashville, Tennessee [https://redcap.vanderbilt.edu]) electronic data capture tools. A reminder e-mail was sent biweekly for 1 month to those members who did not complete the survey during the study collection period.

Nonrespondent Survey

Because the initial response rate was low, a method was needed to quantify nonresponse bias. Therefore, a shortened follow-up survey was sent to a randomly selected subsample (n = 100) of SPA members who had not responded to the initial survey. These surveys were used to determine response differences between the 2 cohorts. Members were selected through the Society’s membership directory and were contacted via telephone or e-mail address based on their listed information.

Statistical Analysis

All deidentified REDCap data were exported as an Excel spreadsheet (Microsoft Corporation, Redmond, Washington). Univariate analyses were performed with either the Wilcoxon Rank Sum test or the Fisher exact test based on the distributions of the variables. To limit confounding, we created 2 multivariable logistic regression models and 1 multivariable linear regression model. The first was performed with change in retirement plans due to COVID-19 as the response variable and years in practice (≥11 vs ≤10 years), financial impact due to COVID-19 (yes versus no), change in home responsibilities due to COVID-19 (yes versus no), academic practice (yes versus no), and ethnicity (non-White versus White) as the predictors. The second model included change in future clinical job responsibilities due to COVID-19 as the response variable and change in clinical responsibilities due to COVID-19 (yes versus no), financial impact due to COVID-19 (yes versus no), change in home responsibilities due to COVID-19 (yes versus no), gender (male versus female), and ethnicity (non-White versus White) as the explanatory variables. Finally, the multivariable linear regression model included the job satisfaction total score as the response variable while social isolation (yes versus no), Perceived Stress Scale score, PERMAH score, depression/anxiety total score, sleep quality rating, 2-question summative MBI score, and age were included as predictors. All predictors in all multivariable regression models were chosen based on expert opinion. The R statistical software package (Version 4.0.0) was used for this investigation, and 2-sided P values of <.05 were considered statistically significant.

Power Analysis

Based on a preliminary analysis, 15% of anesthesia providers stated that they were changing their retirement plans as a result of COVID-19. We hypothesized that pediatric anesthesiologists who have been practicing longer (≥11 vs ≤10 years) would display a 2-fold increase in the odds of changing their retirement plans. Given the approximate 60:40 predominance of pediatric anesthesia providers who have been in practice for ≥11 years over those with ≤10 years in SPA, a total sample size of 422 (n = 169 in the ≤10-year cohort) was needed to show this difference with a type I and II error rate of 5% and 20%, respectively, based on Fisher exact test.

RESULTS

Of the 3245 SPA members surveyed, 561 (17%) responded to the initial online questionnaire and most answered all of the questions (response missingness ranged from 0% to 15.5% depending on the question). Primary respondent demographics, years in practice, practice setting and type, and level of training are listed in Table 1. A considerable number of individuals stated that their experience with COVID-19 would lead to a change in their retirement plans, with 14.2% stating that they would retire earlier and 11.9% planning to retire later (Table 2). There was an association between years in practice (≥11 vs ≤ 10 years) and COVID-19–related changes in retirement plans (odds ratio [OR] = 2.63, 95% confidence interval [CI], 1.64-4.30; P < .001), which continued after multivariable adjustment (adjusted odds ratio [aOR] = 2.78, 95% CI, 1.73-4.55, P < .001; Table 3). In addition to years in practice, respondents were more likely to change retirement plans if they were financially impacted by COVID-19, their home responsibilities increased because of COVID-19, or they did not work at an academic institution (Table 3). Further, 60% of respondents reported a significant financial impact due to COVID-19 (Table 3).

Table 1. - Respondent Demographics

Characteristic	Value
Age, y, median (first quartile, third quartile)	45 (37, 58)
Gender
Female	266 (56.2)
Male	199 (42.1)
Transgender female, gender nonconforming, different identity	0 (0.0)
Transgender male	1 (0.2)
Different identity	0 (0.0)
Decline to state	7 (1.5)
Ethnicity
White	353 (75.1)
Black	19 (4.0)
Asian	56 (11.9)
Native Hawaiian/Pacific Islander	2 (0.4)
American Indian/Alaska Native	2 (0.4)
Mix of the above	18 (3.8)
None of the above	20 (4.3)
Years in practice
0–5	105 (22.2)
6–10	95 (20.0)
11–15	76 (16.0)
16–25	85 (17.9)
>25	108 (22.8)
Retired	5 (1.1)
Region of Unites States
Pacific time zone	76 (16.1)
Mountain time zone	40 (8.5)
Central time zone	106 (22.5)
Eastern time zone	230 (48.8)
Hawaii/Alaska	4 (0.8)
Outside the United States	15 (3.2)
Practice type
Academic	363 (77.4)
Private	90 (19.2)
Other	16 (3.4)
Primary site of practice
Freestanding pediatric hospital	299 (63.2)
Adult hospital	24 (5.1)
Pediatric hospital within adult hospital	131 (27.7)
Ambulatory surgical center	9 (1.9)
Clinic setting	1 (0.2)
Other	9 (1.9)
Level of training
MD/DO/MBBS	466 (98.5)
CRNA/CAA	5 (1.1)
Other	2 (0.4)

All values are presented as count (percentage) unless otherwise indicated.

Abbreviations: CAA, certified anesthesiologist assistant; CRNA, certified registered nurse anesthetist.

Table 2. - Impact of COVID-19 and Hospital Response on Respondents

COVID-19–related effect	N (%)
Retirement change
No change	349 (73.9)
Retire earlier	67 (14.2)
Retire later	56 (11.9)
Future occupation change
No change	288 (66.5)
Change in responsibilities	94 (21.7)
Reduce time working	46 (10.6)
Leave profession	5 (1.2)
Academic time
No change	147 (40.6)
Increase	60 (16.6)
Decrease	155 (42.8)
Academic productivity
No change	106 (29.3)
Increase	52 (14.4)
Decrease	204 (56.4)
Change in clinical job responsibilitiesa
None	331 (70.3)
Working with different patients	48 (10.2)
Working in different unit	19 (4.0)
Assuming different role	57 (12.1)
Other	47 (10.0)
Financial effectsa
None	193 (40.7)
Furlough days	80 (16.9)
Change in call money	85 (17.9)
Change in retirement matching	81 (17.1)
Change in base salary	87 (18.4)
Change in incentive pay	153 (32.3)
Change in expense funds	99 (20.9)
Change in vacation accrual	29 (6.1)
Other	27 (5.7)
Adequate PPE from institution
Yes	310 (65.8)
No	128 (27.2)
Other	33 (7.0)
Adequate testing for operating room population
Yes	353 (74.8)
No	101 (21.4)
Other	18 (3.8)

Data are presented as count (percentage).

Abbreviations: COVID-19, coronavirus disease 2019; PPE, personal protective equipment.

^aMultiple answers accepted; percentages do not equal 100%.

Table 3. - Retirement Changes Due to COVID-19

Variable	Odds ratio	95% CI	P a
Years in practice (≥11 vs ≤10 y)	2.78	1.73-4.55	<.001
Financial impact from COVID-19 (yes versus no)	1.95	1.23-3.14	.005
Change in home responsibility from COVID-19 (yes versus no)	1.82	1.16-2.86	.010
Academic practice (yes versus no)	0.42	0.30-0.78	.003
Ethnicity (non-White versus White)	0.81	0.47-1.35	.42

Abbreviations: CI, confidence interval; COVID-19, coronavirus disease 2019.

^aBased on multivariable logistic regression model with all predictors being modeled as binary variables. P < .05 was considered statistically significant.

As with changes in retirement plans, a substantial number of respondents indicated that COVID-19 will likely lead to alterations in their future occupational plans, including a change in clinical responsibilities (21.7%), reduced time working (10.6%), and leaving the profession (1.2%; Table 2). Among respondents who worked in an academic practice, 42.8% reported a reduction in academic time, and 56.4% reported a decrease in academic productivity (Table 2). Almost 40% of respondents had changes in their home responsibilities as a result of COVID-19, including homeschooling, care of dependents, and other responsibilities (Table 4). Very few respondents had been infected by SARS-CoV-2 or had experienced hospitalization, loss of a family member, or loss of a colleague to the disease (Table 4). However, 27.2% of respondents had inadequate PPE, and 21.4% reported insufficient testing of operating room personnel or patients (Table 2).

Table 4. - Personal Impact of COVID-19 on Respondents

COVID-19–related effect	N (%)
Change in home responsibilitiesa
None	279 (59.5)
Homeschooling of children	118 (25.2)
Additional childcare	78 (16.6)
Closure of daycare	39 (8.3)
Caretakers not showing up	49 (9.8)
Care of elderly family members	33 (7.0)
Other	25 (5.3)
Loss of family member or friend from COVID-19
Yes	32 (6.8)
No	440 (93.2)
Loss of colleague from COVID-19
Yes	24 (5.1)
No	449 (94.9)
Infected by COVID-19
Yes	9 (1.9)
No	463 (98.1)
Immediate family hospitalized and/or critically ill from COVID-19
Yes	7 (1.5)
No	462 (98.5)

Data are presented as count (percentage).

Abbreviation: COVID-19, coronavirus disease 2019.

^aMultiple answers accepted; percentages do not equal 100%.

On univariate analysis, female gender was associated with higher odds of a future job change due to COVID-19 (OR = 1.92, 95% CI, 1.12-2.63; P = .011) than was male gender. This result is most likely related to our finding that home responsibilities increased more for women than for men during the pandemic (OR = 2.63, 95% CI, 1.74-4.00; P < .001). Based on multivariable logistic regression, this latter predictor was strongly associated with future occupational change (aOR = 1.83, 95% CI, 1.19-2.83, P = .006, Table 5), whereas gender was not significantly associated (aOR = 1.41, 95% CI, 0.90-2.22, P = .14, Table 5). Other explanatory variables that were correlated with future occupational change included alteration in clinical responsibilities as a result of COVID-19, financial impact from COVID-19, and ethnicity—with non-Whites being more impacted (Table 5).

Table 5. - Future Occupational Changes Due to COVID-19

Variable	Odds ratio	95% CI	P a
Change in clinical responsibility from COVID-19 (yes versus no)	2.69	1.73-4.18	<.001
Change in home responsibility from COVID-19 (yes versus no)	1.83	1.19-2.83	.006
Financial impact from COVID-19 (yes versus no)	1.65	1.06-2.58	.027
Gender (female versus male)	1.41	0.90-2.22	.14
Ethnicity (non-White versus White)	2.05	1.28-3.29	.003

Abbreviations: CI, confidence interval; COVID-19, coronavirus disease 2019.

^aBased on multivariable logistic regression model with all predictors being modeled as binary variables. P < .05 was considered statistically significant.

Multiple measures of well-being were also analyzed in this sample (Supplemental Digital Content, Table 2, https://links.lww.com/AA/D396). A high proportion of respondents (25.1%) felt socially isolated (Supplemental Digital Content, Table 2, https://links.lww.com/AA/D396). Individuals who intended to change their retirement plans felt more socially isolated than those who did not expect to change their plans (OR = 1.96, 95% CI, 1.22-3.13; P = .004). In addition, both changes in retirement planning and future occupational planning were strongly associated with total job satisfaction scores (both P < .001). On multivariable analysis, total job satisfaction scores were linearly associated with Perceived Stress Scale scores, PERMAH scores, and 2-question summative MBI scores (Table 6). Both women (versus men) and non-Whites (versus Whites) had higher likelihoods of burnout on univariate analysis (OR = 1.75, 95% CI, 1.06-2.94, P = .026 and OR = 1.82, 95% CI, 1.08-3.04, P = .017, respectively) with burnout defined as scores of >4 on 2-question summative MBI questions (eg, “I feel burned out from my work” and “I have become more callous toward people since I took this job;” Supplemental Digital Content, Table 1, https://links.lww.com/AA/D396).

Table 6. - Job Satisfaction

Variable	Coefficient	95% CI	P a
Social isolation (yes versus no)	0.228	−0.574 to 1.030	.58
Perceived stress total score	−0.166	−0.330 to −0.001	.049
PERMAH total score	0.089	0.038-0.141	<.001
Depression/anxiety total score	−0.027	−0.212 to 0.158	.77
Sleep quality rating	0.009	−0.382 to 0.399	.97
Maslach Burnout Inventory score	−0.604	−0.740 to −0.468	<.001
Age	−0.022	−0.053 to 0.008	.15

Abbreviations: CI, confidence interval; PERMAH, Positive Emotions, Engagement, Relationships, Meaning, Accomplishment, Health.

^aBased on multivariable linear regression model with social isolation being modeled as binary variables while the other predictors were modeled as continuous variables. P value <.05 was considered statistically significant.

The demographics of the nonrespondents were similar to those of SPA members who responded to the primary survey (Supplemental Digital Content, Table 3, https://links.lww.com/AA/D396). In addition, the nonrespondents did not differ significantly from the initial respondents in their answers to questions regarding the impact of COVID-19 (Supplemental Digital Content, Table 4, https://links.lww.com/AA/D396).

DISCUSSION

Our survey revealed that the initial COVID-19 pandemic has affected many aspects of SPA members’ personal and professional lives. Consistent with other studies,^11,16–18 women and under-represented minorities (URM)—defined as Black, Hispanic, and Native American—have had a disproportionately negative impact. Indeed, for women in academic medicine, many gains of the last 20 years are in danger of being lost.^10,11,19 Additionally, for all respondents, we found that COVID-19 has had significant effects on personal finances; home responsibilities, including child and elder care; and retirement planning. Changes in home responsibilities have affected women to a far greater extent than men and are the key determinant in how the COVID-19 pandemic is changing their current and future career plans.^10,19 These career changes will reduce clinical practice time and responsibilities and, when combined with changes in retirement planning, may result in pediatric anesthesia workforce projections that are inaccurate.^20,21 Finally, and not surprisingly, the pandemic has caused significant changes in job satisfaction and social isolation. Respondents who were less satisfied with their work and more socially isolated had all the major signs of stress, burnout, and depression/anxiety—findings consistent with other well-being–related research.^22–24

In our survey, we asked respondents to identify their gender in terms of male and female. Historically, gender identification was binary. Recognizing that some individuals have a nonbinary gender identity, we designed our survey to offer nonbinary choices. Our results showed that 56.2% of respondents identified as female, 42.1% as male, 0.2% as transgender male, and 1.5% declined to answer (Table 1). Additionally, 19% of respondents did not answer this question. Because the number of nonbinary respondents was so small, we were unable to make conclusions about this population and recognize that further study is required.

Our finding that the COVID-19 pandemic is affecting female pediatric anesthesiologists more profoundly than male pediatric anesthesiologists is consistent with the research of others.^10,19 Despite the many social changes that have occurred over the past century, women still bear a disproportionate burden of home responsibilities, particularly related to child and elder care.¹¹ Our results confirm this additional burden. Since the COVID-19 pandemic began, female pediatric anesthesiologists have had to balance home and professional responsibilities to a much greater extent than in the past because childcare has become unpredictable. They have had to deal with quarantine, homeschooling, and the sustained loss of or decreased hours of daycare. Further, the COVID-19 pandemic has ravaged elder care, resulting in closures of assisted living and nursing homes and further disproportionately stressing female caregivers. Although the proportion of female attending anesthesiologists in private and academic practices has been steadily increasing over the past 2 decades, at higher levels of leadership, gender disparity remains a persistent and pervasive problem.^17,18,25,26 Women now make up >50% of all medical school matriculants and are the majority of pediatric anesthesiology fellows.^17,18,25,26 In fact, women now make up the majority (56.4%) of the younger pediatric anesthesiology workforce (<35 years old) in the United States.^20,21 Thus, our finding that women may be forced to decrease their clinical hours, leave the profession entirely, or retire early has profound future workforce implications.^16,17

Before COVID-19, women were under-represented in academic anesthesiology, especially in positions of leadership and influence.^17,18,27 This under-representation is demonstrated in the following categories: academic positions and promotion, journal authorship, editorial board membership, award recipients, and number of department chairs. Women also continue to be reimbursed at lower rates than their male colleagues.²⁶ Our survey findings that indicate less academic time and productivity for both male and female pediatric anesthesiologists may prove particularly devastating to the gains women have made over the last 20 years. Indeed, Andersen et al¹⁹ found that the proportion of medical papers with female first authors and a US affiliation was 19% smaller in 2020 than it was in 2019.

In addition to gender disparities, the finding that non-Whites are more likely to make a future occupational change, including changing clinical responsibilities, decreasing time, or leaving the profession entirely is highly concerning. Like all fields of medicine, anesthesiology and pediatric anesthesiology have a paucity of URMs.^16,28,29 URM represent over one-third of the US population but constitute only 6%–9% of practicing physicians and only 5% of pediatric anesthesiologists.²⁹ Sadly, these numbers have not changed significantly in the past 3 decades. Over the past 20 years, the absolute number of trainees in both anesthesiology and pediatric anesthesiology has steadily increased. However, this trend results primarily from an increase in the number of White and Asian anesthesia residents and pediatric anesthesiology fellows. The proportion of URMs has not changed at all during this same period.^16,29 These disparities are stark and represent centuries of complex, multifactorial systemic issues that have only begun to be identified.¹⁶ Our finding of non-Whites being more likely to change their clinical practice of pediatric anesthesiology could have catastrophic effects on a specialty that already has a low percentage of URMs. We implore leaders to act to prevent further attrition.

COVID-19 is affecting retirement plans for both male and female pediatric anesthesiologists. One chooses to be a physician for many reasons; it is not a career decision so much as a life choice and a calling. Because many sacrifices are made along the way, the decision to retire from medicine is not undertaken lightly or without great consideration. Historically, many pediatric anesthesiologists practice well into their seventh and sometimes eighth decade. In fact, over 7% of pediatric anesthesiologists are over the age of 65.²⁰

It is always important to take into consideration changing population needs in an attempt to train approximately the “right” number of physicians. Training of physicians is costly to the individual and time consuming. It is imperative from a societal perspective that we follow population trends by altering medical school class sizes and residency and fellowship position numbers to best meet the projected needs of the country. Many complex confounding variables determine workforce need. The best pre–COVID-19 data suggested that we were overtraining pediatric anesthesiologists and that trainees would outpace retirees by 2035.^20,21,30,31 Our findings that approximately 15% of pediatric anesthesiologists will likely retire earlier than planned and that 12% may retire later than planned challenges these previous workforce projections. Change in retirement planning was associated with number of years in practice, with those in mid-to-late career being the most affected. Pediatric anesthesiologists who work in private practice environments reported being more affected than physicians working in academic medicine. Retirement planning may also be altered based on the financial impact of COVID-19. Our respondents reported changes in incentive pay, base salary, call reimbursement, and retirement matching as the top factors that affected their personal finances. At the time the survey was administered, the fiscal year had not ended for many organizations; therefore, the full financial impact of COVID-19 may not fully be recognized in our results. It is difficult to predict how COVID-19 will affect the pediatric anesthesiology workforce supply and demand, but clearly, these changes will have relevance for years to come.³²

Finally, the pandemic has affected physician burnout and well-being. Physician burnout is a recognized public health crisis that affects nearly 50% of all practitioners in the United States.^7,33 It has enormous consequences for both the affected physician and the safe and effective delivery of health care. Burnout results in reduced self-care, medical errors, reduced productivity, and increased job turnover and health care costs.³⁴ The COVID-19 pandemic has exacerbated preexisting burnout and moral injury in health care professionals. Evidence from previous epidemics and disasters underscores the high risk for clinicians of developing long-term mental health issues and emphasizes the need for continued support during and after the pandemic.^35–37 Although we do not know the incidence of burnout among pediatric anesthesiologists before the pandemic, our results revealed that 25% currently feel socially isolated. Further, those pediatric anesthesiologists who were experiencing low job satisfaction were also experiencing higher levels of burnout, perceived stress, anxiety, and depression, and lower measures of well-being as reported by the PERMAH profiler.²³ These results are extremely concerning, as the lower job satisfaction scores are most associated with changes in retirement planning and clinical responsibilities. This finding would suggest that those pediatric anesthesiologists who are most dissatisfied with their job during the COVID-19 pandemic are experiencing the most burnout, stress, anxiety, and decrease in well-being. To the best of our knowledge, this has never been reported before and has implications for pediatric anesthesia practices across the country. We believe that the COVID-19 pandemic will require both private and academic practices to develop and institute programs on physician well-being to increase physician job satisfaction and decrease attrition and expedited retirement.

Finally, the feelings of social isolation reported by 25% of our respondents are extremely worrisome. Loneliness can have serious consequences for the individual and subsequently for the institutions in which they practice. It has been reported that loneliness can reduce longevity by 70%. Comparing this to smoking, which reduces longevity by 50%, the serious implications of social isolation become clear.³⁸ Loneliness is experienced by the brain as physical pain and was shown to be associated with increased absenteeism, increased errors, and decreased productivity.³⁸ Conversely, strong social connections are associated with healthier immune systems, lower rates of depression and anxiety, and longer lives.³⁸ The COVID-19 pandemic has forced social isolation on many levels with mandated social distancing and fewer interactions with family, friends, and colleagues. Constant mask wearing—necessary to decrease the spread of the virus—has also resulted in decreased visibility of smiles and positive nonverbal interactions. Our data suggest that a quarter of our specialty is experiencing loneliness, the effects of which may be detrimental to our colleagues, patients, and institutions. Similar to other reports, our data show associations between loneliness and burnout.³⁹ Those who report job dissatisfaction experience high levels of both loneliness and burnout.

This study has many of the limitations associated with online surveys. Although 561 pediatric anesthesiologists completed the questionnaire, this number represents only 17% of potential respondents—a response rate that is consistent with survey research.⁴⁰ A low response rate is a potential source of bias as results are only representative of those who replied. In addition, because the physician anesthesiologists surveyed practice primarily at North American medical institutions and belong to SPA, our results may not be reflective of non-SPA members, physicians in training, nurse anesthetists, or practitioners in other countries, thereby limiting the generalizability of our findings. On the other hand, the large survey sample size ensured that the demographic profile of survey respondents reflected the survey population and provided a sufficiently large data set for analysis. Additionally, as with all survey studies, respondents may not be 100% truthful in their responses and may not have carefully read the questions or thought through their responses before answering. Despite using a survey specialist, seeking advice from SPA’s Quality and Safety Committee and Research Committee members, and piloting the survey with attendings and fellows of the Children’s Hospital Colorado to assess for language and comprehension, we confused race and ethnicity in our survey and did not delineate Hispanic versus non-Hispanic respondents. Hispanics, who represent a portion of URMs, would not be counted in our survey. Thus, our findings concerning the effects of the pandemic on URMs must be interpreted cautiously.

In conclusion, the COVID-19 pandemic has affected many aspects of the personal and professional lives of pediatric anesthesiologists, albeit not equally. Women and URMs have had a disproportionately negative impact, potentially erasing many of the professional and academic gains of the last 2 decades. The pandemic has significantly affected the personal finances, home responsibilities, and retirement planning of pediatric anesthesiologists with resultant reductions in clinical and academic practice time and responsibilities. Although predicting how COVID-19 will affect pediatric anesthesiology workforce projections is difficult, the impact will clearly be substantial. Finally, the pandemic has caused significant changes in job satisfaction and revealed a troubling level of social isolation. Respondents who were less satisfied with their work and more socially isolated had all the major findings of stress, burnout, and depression/anxiety.

Thus, it is imperative that as a collective group, we act now more than ever to encourage well-being interventions that combat loneliness and burnout in pediatric anesthesiologists. Studies have suggested combatting loneliness by developing and supporting a culture of empathy and inclusion, fostering social connection, and celebrating collective successes.^38,39 SPA has a unique opportunity to develop and institute well-being support measures with interventions that encourage professional and social relationships between pediatric anesthesiologists across the country. These collaborative relationships will not only help to advance the care of our patients, but they may also help to support, retain, and improve the well-being of pediatric anesthesiologists.

DISCLOSURES

Name: Rebecca D. Margolis, DO, FAOCA.

Contribution: This author helped conceptualize and design the study and data collection instruments; acquire, analyze, and interpret the data; draft the initial manuscript; and review and approve the final manuscript.

Name: Kim M. Strupp, MD.

Contribution: This author helped conceptualize and design the study and data collection instruments; acquire, analyze, and interpret the data; draft the initial manuscript; and review and approve the final manuscript.

Name: Abbie O. Beacham, PhD.

Contribution: This author helped conceptualize and design the study and data collection instruments; acquire, analyze, and interpret the data; draft the initial manuscript; and review and approve the final manuscript.

Name: Myron Yaster, MD.

Contribution: This author helped conceptualize and design the study and data collection instruments; acquire, analyze, and interpret the data; draft the initial manuscript; and review and approve the final manuscript.

Name: Thomas M. Austin, MD, MS.

Contribution: This author helped conceptualize and design the study and data collection instruments; acquire, analyze, and interpret the data; draft the initial manuscript; and review and approve the final manuscript.

Name: Andrew W. Macrae, BS.

Contribution: This author helped acquire, analyze, and interpret the data and review and approve the final manuscript.

Name: Laura Diaz-Berenstain, MD.

Contribution: This author helped conceptualize and design the study and data collection instruments; acquire, analyze, and interpret the data; draft the initial manuscript; and review and approve the final manuscript.

Name: Norah R. Janosy, MD.

Contribution: This author helped conceptualize and design the study and data collection instruments; acquire, analyze, and interpret the data; draft the initial manuscript; and review and approve the final manuscript.

This manuscript was handled by: James A. DiNardo, MD, FAAP.

ACKNOWLEDGMENTS

The authors thank Claire Levine, BS, MS, for her editorial assistance, Kim Battle of the Society for Pediatric Anesthesia for her assistance in the electronic distribution of the survey, Randy Clark, MD, for his assistance, and the Society for Pediatric Anesthesia’s research and quality and safety committees for their assistance in survey design.