Nationwide Clinical Practice Patterns of Anesthesiology Critical Care Physicians – A Survey to the Membership of Society of Critical Care Anesthesiologists

Abstract

Background:

Despite the growing contributions of critical care anesthesiologists to clinical practice, research and administrative leadership of intensive care units (ICUs), relatively little is known about the subspecialty-specific clinical practice environment. An understanding of contemporary clinical practice is essential to recognize the opportunities and challenges facing critical care anesthesia, optimize staffing patterns, assess sustainability and satisfaction, and strategically plan for future activity, scope and training. This study surveyed intensivists who are members of the Society of Critical Care Anesthesiologists (SOCCA) to evaluate practice patterns of critical care anesthesiologists, including compensation, types of ICUs covered, models of overnight ICU coverage, and relationships between these factors. We hypothesized that variability in compensation and practice patterns would be observed between individuals.

Methods:

Board-certified critical care anesthesiologists practicing in the United States were identified using the SOCCA membership distribution list and invited to take a voluntary online survey between May and June 2021. Multiple choice questions with both single- and multiple-select options were used for answers with categorical data, and adaptive questioning was used to clarify stem-based responses. Respondents were asked to describe practice patterns at their respective institutions, and provide information about their demographics, salaries, effort in ICUs as well as other activities.

Results:

490 participants were invited to take this survey and 157 (response rate 32.0%) surveys were completed and analyzed. The majority of respondents were white (73%), male (69%), and under 50 years of age (82%). The cardiothoracic/cardiovascular ICU was the most common practice setting, with 69.5% of respondents reporting time working in this unit. Significant variability was observed in ICU practice patterns. Respondents reported spending an equal proportion of their time in clinical practice in the operating rooms and intensive care units (median 40%, IQR 20-50%), whereas a smaller proportion – primarily those who completed their training before 2009 – reported administrative or research activities. Female respondents reported salaries that were $36,739 less than males; however, this difference was not statistically different and after adjusting for age and practice type, these differences were less pronounced (−$27,479.79, 95% CI: −$57,232.61 to $2,273.03; p = 0.07).

Conclusions:

These survey data provide a current snapshot of anesthesiology critical care clinical practice patterns in the United States. Our findings may inform decision-making around the initiation and expansion of critical care services and optimal staffing patterns as well as provide a basis for further work that focuses on intensivist satisfaction and burnout.

INTRODUCTION

In light of an aging and increasingly complex patient population, growing use of sophisticated life support technologies and increased healthcare costs, the value and relevance of critical care is increasingly important. Anesthesiologists participate in much of the perioperative critical care delivered in academic medical centers in the United States. Little is known, however, about current critical care anesthesia practice patterns. Such knowledge is relevant as it allows a better understanding of specialty-specific opportunities and challenges, helps guide strategic planning for expanding or organizing critical care services, and improves physician training and satisfaction. Large lacunae in our current understanding of national practice patterns within anesthesiology critical care practice include physician demographics, scope of practice, intensive care unit (ICU) staffing models, compensation, and balance of critical care delivery alongside anesthesiology provision and other activities.

Although previous surveys of the wider ICU community exist, a focused concentration on anesthesiology-trained intensivists with more granular and pertinent data acquisition above and beyond salary data has not been performed to date^1–3. While the number of anesthesia critical care medicine fellowship training positions has increased by 60% since 2014, the specialty continues to have unfilled positions within the fellowship match, suggesting that potential applicants may be uncertain about choosing a critical care fellowship ⁴. The availability of current and comprehensive descriptions of the current anesthesiology critical care medicine workforce would identify targets for strategies to enhance job satisfaction, mitigate burnout and ensure appropriate alignment of graduating physician numbers with market needs.

This survey of board-certified anesthesiology critical care physicians practicing in the United States sought to describe the current makeup of the anesthesiology critical care workforce and common themes of practice. By evaluating the types of intensive care units in which anesthesiologists practice, the composition of ICU teams, ICU coverage and salary models, this study evaluated several different aspects of anesthesia critical care practice. We hypothesized that there would be considerable variability in pay and practice patterns between individuals.

METHODS

Following study approval from the Committee on Clinical Investigations Institutional Review Board (IRB) at Beth Israel Deaconess Medical Center (BIDMC), survey invitations were sent via email to members of the Society of Critical Care Anesthesiologists (SOCCA). This voluntary, closed survey was available only to confirmed SOCCA members who received the survey through the email address registered to the SOCCA distribution list and to those who provided written informed consent. Surveys were distributed on May 11^th, 2021, with two additional reminders sent on May 26^th and June 2^nd. The survey closed on June 7, 2021. SOCCA members were also informed of the survey during the society’s annual meeting and at SOCCA-hosted webinars.

Survey responses were automatically collected using the Research Electronic Data Capture (REDCap) system hosted at BIDMC. The first section of the survey assessed eligibility for survey completion, asking respondents if they were board-certified critical care anesthesiologists and if they practiced in the United States. Respondents were also provided with a description of the study including goals and estimated length via an electronic written informed consent script, and were subsequently asked if they wished to participate in this research study. Consent was implied by clicking ‘yes’ to this question. The survey was fully anonymized, and no identifiable information was collected. Duplicate entries were not permitted. Results of the study are reported in accordance with the Checklist for Reporting Results of Internet E-Surveys (CHERRIES) guidelines⁵.

Survey Design, Testing and Participation

This survey targeted anesthesiology critical care physicians practicing in the United States and collected information about the following domains: respondent demographics, training characteristics, practice patterns, environment, organizational structure(s), ICU activities/responsibilities and compensation. Survey questions were developed by members of the Data Subcommittee of the SOCCA Research Committee and subsequently vetted by the wider Research Committee for item generation and item reduction. Survey and pilot testing was conducted by members of the SOCCA Board of Directors and suggested changes were formally adjudicated and incorporated. The usability and technical functionality of the electronic REDCap survey was also thoroughly tested through a feedback process with the SOCCA Board of Directors and authors prior to widespread societal distribution of the questionnaire.

A link to participate in REDCap was sent directly from SOCCA to individual members email addresses. A total of 490 SOCCA members received the survey emails. No incentive was offered and participants were asked to consent immediately prior to survey completion. Questions included single-select, and multiple-select options for answers with categorical or nominal data. In addition, adaptive questioning was used to clarify responses based on previous answers. Five electronic pages of questions were presented in the survey, with a maximum of twenty-three questions per page, depending on answers and subsequent branching logic. Respondents were given the opportunity to review and change answers through a “Previous” button at the bottom of each page. An attention question was included 70% of the way through the survey to allow for assessment retrospectively of survey engagement and completeness. A copy of the final survey is included in Supplementary Appendix 1.

Statistical Analysis

Only surveys which correctly completed the attention question were included in the analysis. The overall response rate is reported as a frequency and proportion. No a priori power calculation was performed. Instead, the sample size was based on a fixed available sample of SOCCA members who responded after the maximum number of reminders.

Descriptive statistics of the data were generated for the entire cohort and stratified by fellowship graduation year. Fellowship graduation year was collected in categories representing increments of four years in the survey. In order to evaluate differences over time – and serve as a surrogate for years of experience – observed categories were further stratified into four groups for presentation, namely 1986-2004, 2005-2009, 2010-2014 and 2015-2020. These categories most closely represented quartiles of the data, noting the imbalance in the 2015-2020 category, from which the majority of respondents selected, without the ability to categorize this further. Continuous variables such as the years in practice was reported as median and interquartile ranges (IQR) given the skewed distribution. Normality of continuous variables was assessed with a Shapiro-Wilk test and confirmed by visually inspecting the distribution. Categorical variables were presented as frequencies and proportions. In a prespecified analysis, salary estimates were assessed using univariate and multivariable linear regression, including prespecified covariates for sex, age and practice type (academic versus community), all treated as categorical variables. Because salary was originally coded as a categorical variable, the median value for each category was abstracted for each respondent and used as a continuous variable in the regression models. Results are reported as mean differences and their associated 95% confidence intervals (CI). All inferences were evaluated using two-sided tests, with p-values < 0.05 considered statistically significant. Analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC).

RESULTS

Overall, 184 participants clicked the link to access the survey, one subject was excluded because of self-reported lack of board certification in anesthesiology critical care, and another was excluded because of practice outside of the United States. Of the remaining 182 subjects who consented to participate, 25 were excluded from analysis because the attention question was not completed and more than 50% of the survey was missing. A total of 157 completed surveys were thus received for an overall response rate of 32.0% (157/490) amongst all eligible SOCCA members (Supplemental Figure 1).

Characteristics of Survey Respondents

Respondents were predominantly male (68.8%) and less than 50 years of age (82.2%). When respondents were asked about race, 73.2% reported they were White, 1.9% Black or African American, 17.2% Asian and 1.3% multiracial with an additional 1.3% reporting their race as “other”. Of all survey respondents, 3.8% identified their ethnicity as Hispanic/Latino and 85.4% identified as Not Hispanic/Latino. When asked about race and ethnicity, 5.1% and 7.6%, respectively, preferred not to answer. Physicians surveyed reported practicing across a variety of geographical locations in the United States (Table 1). Most respondents completed residency training solely in anesthesiology (94.3%), but (42.0%) reported completing at least two fellowships. Over 60% of responding physicians reported having worked for only one employer following graduation, and this proportion was higher in recent fellowship graduates.

Table 1.

Characteristics of Survey Respondents

	Entire Cohort N = 157	Fellowship Graduation Year
		1986-2004 N = 21	2005-2009 N = 30	2010-2014 N = 36	2015-2020 N = 70
Biodemographic Characteristics
Female Sex	49 (31.2)	6 (28.6)	6 (20.0)	11 (30.6)	26 (37.1)
Age, years
30 - 40	70 (44.6)	---	1 (3.3)	8 (22.2)	61 (87.1)
40 - 50	59 (37.6)	4 (19.0)	20 (66.7)	26 (72.2)	9 (12.9)
50 - 60	22 (14.0)	11 (52.4)	9 (30.0)	2 (5.6)	---
60 - 70	3 (1.9)	3 (14.3)	---	---	---
70 - 80	3 (1.9)	3 (14.3)	---	---	---
Race
White	115 (73.2)	18 (85.7)	23 (76.7)	23 (63.9)	51 (72.9)
Black or African American	3 (1.9)	---	1 (3.3)	1 (2.8)	1 (1.4)
Asian	27 (17.2)	2 (9.5)	3 (10.0)	9 (25.0)	13 (18.6)
Multiracial	2 (1.3)	---	---	---	2 (2.9)
Other	2 (1.3)	---	---	1 (2.8)	1 (1.4)
Prefer Not to Answer	8 (5.1)	1 (4.8)	3 (10.0)	2 (5.6)	2 (2.9)
Ethnicity
Hispanic/Latino	6 (3.8)	---	1 (3.3)	---	5 (7.1)
Not Hispanic / Latino	134 (85.4)	17 (81.0)	23 (76.7)	32 (88.9)	62 (88.6)
Prefer Not to Answer	12 (7.6)	2 (9.5)	4 (13.3)	3 (8.3)	3 (4.3)
Missing	5 (3.2)	2 (9.5)	2 (6.7)	1 (2.8)	---
Training and Practice Characteristics
Location of Practice a
New England	33 (21.0)	5 (23.8)	4 (13.3)	9 (25.0)	15 (21.4)
Middle Atlantic	19 (12.1)	1 (4.8)	3 (10.0)	5 (13.9)	10 (14.3)
East North Central	18 (11.5)	2 (9.5)	2 (6.7)	4 (11.1)	10 (14.3)
West North Central	16 (10.2)	3 (14.3)	3 (10.0)	7 (19.4)	3 (4.3)
South Atlantic	12 (7.6)	1 (4.8)	3 (10.0)	1 (2.8)	7 (10.0)
East South Central	19 (12.1)	5 (23.8)	2 (6.7)	3 (8.3)	9 (12.9)
West South Central	6 (3.8)	---	---	2 (5.6)	4 (5.7)
Mountain	7 (4.5)	1 (4.8)	3 (10.0)	2 (5.6)	1 (1.4)
Pacific	27 (17.2)	3 (14.3)	10 (33.3)	3 (8.3)	11 (15.7)
Specialties
Anesthesiology Only	148 (94.3)	18 (85.7)	27 (90.0)	35 (97.2)	68 (97.1)
Internal Medicine & Anesthesiology	3 (1.9)	3 (14.3)	---	---	---
Anesthesiology & Other Specialty	5 (3.2)	---	3 (10.0)	---	2 (2.9)
Prefer Not to Answer	1 (0.6)	---	---	1 (2.8)	---
Jobs Since Training Finished
1	99 (63.1)	11 (52.4)	14 (46.7)	20 (55.6)	54 (77.1)
2-3	50 (31.8)	7 (33.3)	13 (43.3)	14 (38.9)	16 (22.9)
4-6	7 (4.5)	3 (14.3)	2 (6.7)	2 (5.6)	---
Missing	1 (0.6)	---	1 (3.3)	---	---
Years Practicing after Board Certification Fellowships	6 (3, 12)	23 (20, 27)	14 (12, 15)	9 (7, 10)	3 (2, 5)
No Other Fellowships	91 (58.0)	11 (52.4)	14 (46.7)	22 (61.1)	44 (62.9)
Adult Cardiothoracic Anesthesiology	34 (21.7)	5 (23.8)	6 (20.0)	5 (13.9)	18 (25.7)
Neurocritical Care	14 (8.9)	3 (14.3)	7 (23.3)	4 (11.1)	---
Pain Medicine	2 (1.3)	2 (9.5)	---	---	---
Other Fellowship	13 (8.3)	4 (19.0)	5 (16.7)	1 (2.8)	3 (4.3)
Compensation Characteristics
Annualized Full-Time Salary
< $250,000	5 (3.2)	2 (9.5)	1 (3.3)	---	2 (2.9)
$250,000-300,000	19 (12.1)	1 (4.8)	1 (3.3)	5 (13.9)	12 (17.1)
$300,000-350,000	32 (20.4)	2 (9.5)	3 (10.0)	8 (22.2)	19 (27.1)
$350,000-400,000	43 (27.4)	7 (33.3)	5 (16.7)	8 (22.2)	23 (32.9)
$400,000-450,000	29 (18.5)	3 (14.3)	10 (33.3)	7 (19.4)	9 (12.9)
$450,000-500,000	14 (8.9)	3 (14.3)	2 (6.7)	7 (19.4)	2 (2.9)
$500,000-550,000	9 (5.7)	2 (9.5)	5 (16.7)	---	2 (2.9)
> $550,000	5 (3.2)	1 (4.8)	3 (10.0)	---	1 (1.4)
Missing	1 (0.6)	---	---	1 (2.8)	---

Values are presented as n (%) or median (quartile 1, quartile 3) depending on variable type.

^aGeographic regions are defined as the following: New England (Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, Vermont), Middle Atlantic (New Jersey, New York, Pennsylvania), East North Central (Illinois, Indiana, Michigan, Ohio, Wisconsin), West North Central (Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota, South Dakota), South Atlantic (Delaware, District of Columbia, Florida, Georgia, Maryland, North Carolina, South Carolina, Virginia, West Virginia), East South Central (Alabama, Kentucky, Mississippi, Tennessee), West South Central (Arkansas, Louisiana, Oklahoma, Texas), Mountain (Arizona, Colorado, Idaho, Montana, Nevada, New Mexico, Utah, Wyoming), Pacific (Alaska, California, Hawaii, Oregon, Washington)

Practice Patterns

Over 80% of those who responded reported that ICU coverage in their institution was exclusively provided by board-certified intensivists (Table 2). Respondents most commonly reported supervising both physician trainees and advanced practice providers with a small minority (1.9%) in solo critical care practice. When covering the ICU, most intensivists (86.0%) did not have other responsibilities in anesthetizing locations. Intensivist responsibilities during times of ICU coverage frequently included - other critical care roles, most commonly in ICU bed triage/management (79.6% for units covered), hospital codes (47.8%) and rapid response systems (23.6%), with variable frequencies among respondents (Supplementary Table 1). The most frequently reported members of interdisciplinary teams on ICU rounds were pharmacists (96.8%) and respiratory therapists (66.9%). Critical care services were largely provided via the anesthesiology department with a unit director who was also from the anesthesiology department (Table 3). Unit type varied, with semi-closed units (42.7%) being the most common practice setting.

Table 2.

Characteristics of Coverage

	Entire Cohort N = 157	Fellowship Graduation Year
		1986-2004 N = 21	2005-2009 N = 30	2010-2014 N = 36	2015-2020 N = 70
ICU Coverage
Board Certified Intensivists Only	129 (82.2)	18 (85.7)	24 (80.0)	30 (83.3)	57 (81.4)
Intensivists and Non-Intensive Physicians	15 (9.6)	1 (4.8)	3 (10.0)	4 (11.1)	7 (10.0)
24/7 In-House Intensivist Coverage	82 (52.2)	11 (52.4)	19 (63.3)	18 (50.0)	34 (48.6)
Intensivists In-House Daytime & Home Night Call	76 (48.4)	11 (52.4)	10 (33.3)	17 (47.2)	38 (54.3)
Providers Under Their Supervision
Interns	92 (58.6)	15 (71.4)	16 (53.3)	20 (55.6)	41 (58.6)
Residents	133 (84.7)	20 (95.2)	28 (93.3)	29 (80.6)	56 (80.0)
Fellows	121 (77.1)	17 (81.0)	23 (76.7)	28 (77.8)	53 (75.7)
Nurse Practitioners	125 (79.6)	18 (85.7)	24 (80.0)	29 (80.6)	54 (77.1)
Physician Assistants	105 (66.9)	12 (57.1)	20 (66.7)	24 (66.7)	49 (70.0)
None (Work Independently)	3 (1.9)	---	1 (3.3)	---	2 (2.9)
Daytime Coverage
Primary Physician of Record	93 (59.2)	11 (52.4)	23 (76.7)	18 (50.0)	41 (58.6)
Consultant Seeing All Unit Patients	86 (54.8)	15 (71.4)	9 (30.0)	21 (58.3)	41 (58.6)
Consultant Seeing Mechanically Ventilated Patients	4 (2.5)	---	1 (3.3)	1 (2.8)	2 (2.9)
Consultant Giving Recommendations on Specific Patients	7 (4.5)	1 (4.8)	1 (3.3)	2 (5.6)	3 (4.3)
Worked Hours / Week When Covering the ICU
< 40	3 (1.9)	1 (4.8)	1 (3.3)	1 (2.8)	---
40	2 (1.3)	1 (4.8)	---	---	1 (1.4)
41-50	7 (4.5)	2 (9.5)	1 (3.3)	2 (5.6)	2 (2.9)
50-60	25 (15.9)	2 (9.5)	7 (23.3)	5 (13.9)	11 (15.7)
60-70	33 (21.0)	3 (14.3)	3 (10.0)	8 (22.2)	19 (27.1)
70-80	40 (25.5)	6 (28.6)	10 (33.3)	7 (19.4)	17 (24.3)
80-90	34 (21.7)	1 (4.8)	6 (20.0)	11 (30.6)	16 (22.9)
> 90	12 (7.6)	5 (23.8)	2 (6.7)	2 (5.6)	3 (4.3)
Missing	1 (0.6)	---	---	---	1 (1.4)
Consecutive Days/Nights Working in ICU
1-3	12 (7.6)	2 (9.5)	5 (16.7)	3 (8.3)	2 (2.9)
4-7	114 (72.6)	11 (52.4)	22 (73.3)	28 (77.8)	53 (75.7)
7-14	28 (17.8)	6 (28.6)	3 (10.0)	5 (13.9)	14 (20.0)
14+	2 (1.3)	1 (4.8)	---	---	1 (1.4)
Missing	1 (0.6)	1 (4.8)	---	---	---
Weeks Per Year Coving ICU
0-5	3 (1.9)	1 (4.8)	---	1 (2.8)	1 (1.4)
5-10	29 (18.5)	5 (23.8)	6 (20.0)	4 (11.1)	14 (20.0)
10-15	71 (45.2)	9 (42.9)	14 (46.7)	18 (50.0)	30 (42.9)
15-20	26 (16.6)	---	5 (16.7)	7 (19.4)	14 (20.0)
20-25	12 (7.6)	4 (19.0)	1 (3.3)	2 (5.6)	5 (7.1)
25-30	9 (5.7)	1 (4.8)	3 (10.0)	1 (2.8)	4 (5.7)
30+	5 (3.2)	1 (4.8)	---	2 (5.6)	2 (2.9)
Missing	2 (1.3)	---	1 (3.3)	1 (2.8)	---
ICU Shifts Covered / Month
< 5	22 (14.0)	6 (28.6)	5 (16.7)	3 (8.3)	8 (11.4)
5-7	47 (29.9)	4 (19.0)	5 (16.7)	13 (36.1)	25 (35.7)
8-10	40 (25.5)	3 (14.3)	9 (30.0)	11 (30.6)	17 (24.3)
10-12	21 (13.4)	1 (4.8)	6 (20.0)	4 (11.1)	10 (14.3)
13-15	21 (13.4)	5 (23.8)	4 (13.3)	5 (13.9)	7 (10.0)
Other	4 (2.5)	1 (4.8)	1 (3.3)	---	2 (2.9)
Missing	2 (1.3)	1 (4.8)	---	---	1 (1.4)
Most Common Night Coverage Model
Shift with 24/7 In-House Coverage	85 (54.1)	8 (38.1)	18 (60.0)	18 (50.0)	41 (58.6)
Home Call with Pager After Day Shift	63 (40.1)	10 (47.6)	11 (36.7)	15 (41.7)	27 (38.6)
Handover Fully to Tele-ICU	4 (2.5)	2 (9.5)	---	2 (5.6)	---
No Intensivist Coverage	2 (1.3)	---	---	1 (2.8)	1 (1.4)
Other	3 (1.9)	1 (4.8)	1 (3.3)	---	1 (1.4)
In House Coverage for Faculty on Home Call
In-Unit Housestaff	38 (24.2)	9 (42.9)	5 (16.7)	9 (25.0)	15 (21.4)
Advance Practice Providers	23 (14.6)	1 (4.8)	6 (20.0)	6 (16.7)	10 (14.3)
Hospitalists	1 (0.6)	---	---	---	1 (1.4)
Tele-ICU Coverage	1 (0.6)	---	---	---	1 (1.4)
Missing	94 (59.9)	11 (52.4)	19 (63.3)	21 (58.3)	43 (61.4)
Other Responsibilities When Covering ICU
No Other Formal Responsibilities	135 (86.0)	20 (95.2)	26 (86.7)	29 (80.6)	60 (85.7)
Remote Anesthetizing Locations	8 (5.1)	---	2 (6.7)	3 (8.3)	3 (4.3)
Main Operating Room Anesthesiology Provision	1 (0.6)	---	---	---	1 (1.4)
Other	11 (7.0)	---	3 (10.0)	4 (11.1)	4 (5.7)
ICU Coverage Model
Day Time In-Person Coverage	153 (97.5)	20 (95.2)	30 (100.0)	34 (94.4)	69 (98.6)
Nighttime Coverage In-House	88 (56.1)	9 (42.9)	22 (73.3)	17 (47.2)	40 (57.1)
Nighttime Pager Call	67 (42.7)	10 (47.6)	12 (40.0)	14 (38.9)	31 (44.3)
Tele-ICU	13 (8.3)	5 (23.8)	3 (10.0)	2 (5.6)	3 (4.3)
Other	2 (1.3)	1 (4.8)	1 (3.3)	---	---
N/A	1 (0.6)	---	---	1 (2.8)	---
Beds Covered
Day Shift	14 (12, 18)	15 (12, 18)	15 (10, 21)	12 (10, 16)	12 (10, 16)
Night Shift	26 (20, 40)	32 (16, 45)	24 (18, 26)	27 (20, 40)	27 (20, 40)
Nighttime Pager Call	19 (12, 24)	22 (17, 24.5)	19 (15, 24)	20 (12, 26)	20 (12, 26)
Tele-ICU	100 (80, 140)	85 (80, 100)	80 (80, 100)	150 (140, 185)	150 (140, 185)
Allied Specialties Involved in Rounds
Pharmacists	152 (96.8)	20 (95.2)	30 (100.0)	35 (97.2)	67 (95.7)
Social Workers	48 (30.6)	6 (28.6)	9 (30.0)	11 (30.6)	22 (31.4)
Occupational Therapists	29 (18.5)	5 (23.8)	6 (20.0)	4 (11.1)	14 (20.0)
Physical Therapists	36 (22.9)	5 (23.8)	9 (30.0)	7 (19.4)	15 (21.4)
Respiratory Therapists	105 (66.9)	17 (81.0)	22 (73.3)	23 (63.9)	43 (61.4)
Billing Personnel / Coders	6 (3.8)	1 (4.8)	2 (6.7)	1 (2.8)	2 (2.9)
Other	12 (7.6)	2 (9.5)	3 (10.0)	3 (8.3)	4 (5.7)
None of the Above	2 (1.3)	---	---	1 (2.8)	1 (1.4)

Values are presented as n (%) or median (quartile 1, quartile 3) depending on variable type.

Table 3.

Characteristics of Organizational Practice

	Entire Cohort N = 157	Fellowship Graduation Year
		1986-2004 N = 21	2005-2009 N = 30	2010-2014 N = 36	2015-2020 N = 70
Primary Practice Composition
Academic	129 (82.2)	18 (85.7)	24 (80.0)	28 (77.8)	59 (84.3)
Private	8 (5.1)	1 (4.8)	2 (6.7)	2 (5.6)	3 (4.3)
Hybrid	18 (11.5)	1 (4.8)	4 (13.3)	5 (13.9)	8 (11.4)
Tele-ICU	1 (0.6)	1 (4.8)	---	---	---
Other	1 (0.6)	---	---	1 (2.8)	---
Practice Type
Tertiary / Quaternary Medical Center	129 (82.2)	18 (85.7)	24 (80.0)	28 (77.8)	59 (84.3)
Affiliated ICU Community Practice	7 (4.5)	---	---	1 (2.8)	6 (8.6)
Unit Type
Open	18 (11.5)	3 (14.3)	1 (3.3)	3 (8.3)	11 (15.7)
Semi-Open	30 (19.1)	1 (4.8)	8 (26.7)	8 (22.2)	13 (18.6)
Semi-Closed	67 (42.7)	12 (57.1)	12 (40.0)	16 (44.4)	27 (38.6)
Closed	41 (26.1)	5 (23.8)	9 (30.0)	9 (25.0)	18 (25.7)
Missing	1 (0.6)	---	---	---	1 (1.4)
Unit Director Department
Anesthesiology	125 (79.6)	17 (81.0)	24 (80.0)	29 (80.6)	55 (78.6)
Surgery	80 (51.0)	13 (61.9)	14 (46.7)	20 (55.6)	33 (47.1)
Medicine	26 (16.6)	5 (23.8)	4 (13.3)	7 (19.4)	10 (14.3)
Neurology	20 (12.7)	4 (19.0)	4 (13.3)	6 (16.7)	6 (8.6)
Independent Critical Care	11 (7.0)	2 (9.5)	4 (13.3)	1 (2.8)	4 (5.7)
Other	2 (1.3)	1 (4.8)	---	---	1 (1.4)
Parent Department for ICU Services
Anesthesiology	137 (87.3)	17 (81.0)	23 (76.7)	32 (88.9)	65 (92.9)
Surgery	4 (2.5)	---	2 (6.7)	---	2 (2.9)
Medicine	11 (7.0)	2 (9.5)	3 (10.0)	4 (11.1)	2 (2.9)
Other	4 (2.5)	2 (9.5)	2 (6.7)	---	---
Missing	1 (0.6)	---	---	---	1 (1.4)
ICU Provision Performed by Intensivists
0-24%	19 (12.1)	2 (9.5)	3 (10.0)	8 (22.2)	6 (8.6)
25-49%	15 (9.6)	2 (9.5)	5 (16.7)	4 (11.1)	4 (5.7)
50-74%	58 (36.9)	11 (52.4)	9 (30.0)	13 (36.1)	25 (35.7)
75-100%	64 (40.8)	6 (28.6)	13 (43.3)	11 (30.6)	34 (48.6)
Missing	1 (0.6)	---	---	---	1 (1.4)
Anesthesiology Critical Care Provision Within Critical Care
0-24%	47 (29.9)	3 (14.3)	9 (30.0)	15 (41.7)	20 (28.6)
25-49%	60 (38.2)	7 (33.3)	11 (36.7)	12 (33.3)	30 (42.9)
50-74%	45 (28.7)	9 (42.9)	9 (30.0)	9 (25.0)	18 (25.7)
75-100%	3 (1.9)	1 (4.8)	1 (3.3)	---	1 (1.4)
Missing	2 (1.3)	1 (4.8)	---	---	1 (1.4)
Average CPT 99291 / 99292 Charges Per Day
< 10	45 (28.7)	4 (19.0)	8 (26.7)	9 (25.0)	24 (34.3)
11-15	71 (45.2)	10 (47.6)	15 (50.0)	17 (47.2)	29 (41.4)
16-20	13 (8.3)	2 (9.5)	2 (6.7)	3 (8.3)	6 (8.6)
21-25	3 (1.9)	---	---	---	3 (4.3)
> 25	2 (1.3)	---	1 (3.3)	---	1 (1.4)
Unsure	23 (14.6)	5 (23.8)	4 (13.3)	7 (19.4)	7 (10.0)
Pay Structure
Solely per Shift Payment	4 (2.5)	1 (4.8)	---	---	3 (4.3)
Pay Per Shift	2160 (1830, 2480)	2160 (2160, 2160)	---	2160 (1500, 2800)	2160 (1500, 2800)
Solely Salaried with No Incentives for Calls	30 (19.1)	3 (14.3)	12 (40.0)	5 (13.9)	10 (14.3)
Salary with Bonus Structure	34 (21.7)	5 (23.8)	4 (13.3)	12 (33.3)	13 (18.6)
Salary with Additional Pay Per Call	39 (24.8)	5 (23.8)	6 (20.0)	7 (19.4)	21 (30.0)
Salary with Pay Per Call and Bonus Structure	55 (35.0)	8 (38.1)	9 (30.0)	12 (33.3)	26 (37.1)
Hourly Rate Reimbursement for Critical Care
< $200	50 (31.8)	6 (28.6)	10 (33.3)	15 (41.7)	19 (27.1)
$200-215	16 (10.2)	3 (14.3)	2 (6.7)	3 (8.3)	8 (11.4)
$216-230	6 (3.8)	2 (9.5)	2 (6.7)	---	2 (2.9)
$230-250	6 (3.8)	---	2 (6.7)	3 (8.3)	1 (1.4)
> $250	3 (1.9)	---	2 (6.7)	---	1 (1.4)
Unsure	75 (47.8)	9 (42.9)	12 (40.0)	15 (41.7)	39 (55.7)
Missing	1 (0.6)	1 (4.8)	---	---	---
Stipend to Provide Critical Care
Yes	27 (17.2)	4 (19.0)	5 (16.7)	8 (22.2)	10 (14.3)
No	112 (71.3)	16 (76.2)	24 (80.0)	21 (58.3)	51 (72.9)
Unsure	17 (10.8)	1 (4.8)	---	7 (19.4)	9 (12.9)
Missing	1 (0.6)	---	1 (3.3)	---	---
Non-Clinical Compensation Following ICU Shifts
No	56 (35.7)	8 (38.1)	11 (36.7)	12 (33.3)	25 (35.7)
Yes	101 (64.3)	13 (61.9)	19 (63.3)	24 (66.7)	45 (64.3)
Non-Clinical Compensation Days for Every 7 Worked
1	31 (31.00)	5 (38.46)	5 (26.32)	6 (26.09)	15 (33.33)
2-3	40 (40.00)	6 (46.15)	8 (42.11)	8 (34.78)	18 (40.00)
4-6	17 (17.00)	---	4 (21.05)	4 (17.39)	9 (20.00)
7+	12 (12.00)	2 (15.38)	2 (10.53)	5 (21.74)	3 (6.67)

Values are presented as n (%) or median (quartile 1, quartile 3) depending on variable type.

Anesthesiology critical care physicians practiced across multiple types of ICUs. The cardiothoracic/cardiovascular ICU was the most common setting, with approximately 70% of subjects reporting working some proportion of clinical time in a CTICU (Supplementary Table 1). The self-described percent effort spent in each ICU or between ICUs varied widely among survey respondents (Figure 1). Coverage of cardiovascular and surgical ICUs were the most common. When assessing the effort spent on a particular activity, 90% of survey respondents spent time in both the operating rooms and intensive care units but the actual percent effort devoted to each of these activities varied (Figure 1A). On average respondents reported spending equal time in the ICU and OR (median 40% [IQR 20%, 50%]) and fewer time on administrative (median 10% [IQR 0%, 20%]), education (median 10% [IQR 0%, 10%]) and research activities (median 0% [IQR 0%, 10%]; Supplementary Table 1). Less than 10% of respondents spent time in dedicated Burns, Transplant or Medical ICUs, and tele-ICU practice was reported for approximately 10% of respondents (Figure 1B). Percentage of overall ICU effort varied by ICU type. (Figure 1C). Notably, of those respondents spending time attending on cardiovascular units, more than 70% spent the majority of their overall ICU time in the cardiovascular ICU. Additionally, anesthesiology critical care physicians who completed their training before 2009 spent more time in leadership and administrative positions when compared to more junior staff. More than 70% of surveyed intensivists actively participated in ECMO management (Supplementary Table 1).

Figure 1. Analysis of Effort Levels.

A breakdown of respondents work activities are described in Panel A. In Panel B, ICU units that respondents indicate working in are described, along with the effort levels reported. Respondents primarily reported working in a cardiovascular unit. Further, in Panel C, the breakdown of ICU type is reported among those who report working in that unit. Participants who did not report working in a unit are excluded from Panel C.

Intensive Care Unit Coverage

The majority of respondents described an academic practice in a tertiary medical center (82.2%) (Table 3). Respondents were also asked to estimate the percentage of total critical care provided by anesthesiology-intensivists versus other critical care providers, with 38.2% reported 25-49% anesthesia-intensivist coverage and 28.7% reporting 50-74% anesthesia-intensivist coverage across their practice.

Data collected with respect to intensivist workload is reported in Tables 2 and 3. The number of weeks per year spent covering the ICU was reported in ranges beginning with 0-5 weeks per year to over 30 weeks per year. Most intensivists (45.2%) reported covering the ICU 10-15 weeks per year. The majority of intensivists (75%) reported working more than 60 hours per week, more than half of respondents (54.8%) reported working more than 70 hours per week, and more than 25% of respondents reported working more than 80 hours per week, when covering the ICU. The number of ICU shifts per month varied widely between respondents, with the majority reporting less than ten. 72.6% of intensivists reported working 4-7 consecutive day/night shifts.

During the daytime, intensivists were either the physician of record (59.2%) or serving as a consultant to the primary service (54.8%). The most frequently reported model (54.1%) of night-time coverage was shift-based work with 24-hour in-house intensivist coverage, followed by home pager call at night after a day shift (40.1%). Among those on home call, most did not detail who provided in-house coverage. Night-time coverage varied significantly by primary ICU in which the respondents worked, as detailed in Figure 2.

Figure 2. Most Common Night Coverage Model.

Data are presented on the most common nighttime coverage model for each of the ICUs. Respondents were categorized based off the ICU they reported working in most commonly. A significantly different coverage model was reported across units (p<0.001).

Compensation

Financial compensation for critical care services varied from less than $250,000 to over $550,000 among our respondents. With regards to compensation there was a wide range, with respondents reporting an annual income ranging, with 20%, 27% and 18% of survey respondents reporting annual salaries between $300,000-$350,000, $350,000-$400,000 and $400,000-$450,000 respectively. Compensation levels did not differ between geographical locations. In unadjusted models, age, gender and practice type were all associated with salary levels (Figure 3, Supplementary Table 2). Respondents from private practice reported higher compensation levels than individuals in academia (mean difference $87,891, 95% CI: $24,490 to $151,291; p = 0.02). Further, mean compensation for female physicians was $36,739 less (95% CI: −$67,061 to −$6,417) than males (p = 0.02; Figure 3). After adjusting for age category and academic vs non-academic practice type however, this difference was no longer significant (mean difference −$27,480, 95% CI: −$57,233 to $2,273; p = 0.07). Compensation structures varied also, with the two most frequent being a fixed salary with additional pay per call both with and without a bonus structure. The majority of respondents did not receive an ICU stipend. Most physicians surveyed generated under 15 critical care billing Current Procedural Terminology (CPT) codes (99291/99292) per day. Almost half (47.8%) were unsure of the exact hourly rate for time-based critical care billing.

Figure 3. Salary Effect Estimates.

The mean difference in salaries (purple dots) and the associated 95% confidence intervals (error bars) are presented in three unadjusted models. Further, estimates are presented from a multivariable model include age, sex and practice type. In order to generate salary levels from the reported categories, the mean values were used for analysis. White points represent the referent category.

DISCUSSION

In this survey of members of the Society of Critical Care Anesthesiologists, we present a current description of the clinical practice environment for critical care anesthesiologists. We found that intensivists work in a variety of critical care unit types and often lead interdisciplinary teams. We observed considerable variability in hours worked per week, shifts per month, and weeks in an ICU per year. While coverage models varied considerably, respondents were most likely to be part of in-house 24/7 coverage when providing care in a cardiovascular or medical ICU. Although we observed no geographical difference in compensation was present, we did find that unadjusted salaries for female physicians were lower than for males. Although this difference was attenuated after adjustment for age and academic vs non-academic practice type, further work is needed to clarify salary inequities among critical care anesthesiologists.

Our findings have clinical implications. As demonstrated here, anesthesia-trained critical care physicians may provide critical care in a variety of clinical environments and represent a flexible resource for hospitals. Surveys conducted in recent years have focused on a range of concerns for practicing intensivists. A 2021 survey by the American Society of Anesthesiologists Committee on Critical Care Medicine and Society for Education in Anesthesiology focused on factors that impact satisfaction in practice, and highlighted burnout, work-life harmony and lack of respect as factors that impact satisfaction in anesthesia practice.⁶ Other surveys have focused on clinical practice amongst intensivists in Latin America⁷ and staffing models in cardiac surgery ICUs⁸, or on benchmarks in neurocritical care³. On the other hand, there is a paucity of data describing the practice patterns and reimbursement structure of anesthesia trained intensivists in the US on the attitudes and practices characterize.

Following the publication of a 2000 report predicting a significant shortage of critical care physicians in the coming decades⁹, intensivists have worked to address current and predicted shortages of intensive care physicians as the size of an aging and the increasingly ill population continues to grow^10–15. Strategies to mitigate the shortage of intensivists have been hampered by inadequate data on both current clinical practice environment of critical care medicine physicians and outcomes of different organizational approaches. Similarly, how to best staff ICUs overnight and the most cost-effective mix of providers is unclear ^{16, 17}. Although many strategies have been suggested^18–23 the real world applicability of these approaches has been limited given the dearth of knowledge on current contemporary overnight staffing practices. Our survey establishes that critical care anesthesiologists practice in a variety of ICUs and night time coverage models. In addition, we found that nighttime coverage models vary possibly with the type of ICU given that amongst our respondents, those who practice in medical and cardiovascular ICUs are more likely to be part of a model with in-house intensivist coverage overnight. Existing literature is mixed regarding whether 24/7 in house intensivist coverage in cardiac surgery ICUs improves outcomes^{24, 25} implying that other factors may influence this staffing decision.

Our survey also suggests that compensation for critical care anesthesiologists varies considerably. Appropriate compensation for these subspecialty trained anesthesiologists must take into account all value offered by anesthesia trained intensivists, with a recent article hypothesizing that potential salary cute disincentivize the pursuit of critical care training for anesthesiologists¹³. A 2013 Society of Critical Care Medicine survey reported compensation for physicians in critical care compared to other critical care disciplines² and an annual online Medscape survey¹ also reports average compensation for intensivists compared to other medical specialties. However, these surveys do not report specific compensation structures for critical care anesthesiologists. Our survey describes self-reported compensation for SOCCA members who are board-certified anesthesiologists and observes that more than half of respondents receive a fixed salary with additional compensation for overnight call both with and without a bonus structure; however a majority of respondents did not receive an ICU stipend. We found that female intensivists had lower unadjusted salaries than their male counterparts and younger physicians less than their older counterparts. These findings echo those reported by a recent survey of neurointensivists by the Neurocritical Care Society³. We also found that while respondents did have a sense of how many critical care CPT codes they submitted per day, almost half were unsure of the reimbursement rate for time based critical care billing.

Our survey has several limitations. First, our response rate was 32% and thus it is possible that our findings may not be generalizable to all critical care anesthesiologists practicing in the United States. However, this response rate is somewhat higher than described previously, with prior surveys of intensivists describing response rates of 25.2%⁶ and 31%³. We also note here that our denominator was all SOCCA members, which may lead to a predilection towards academic practice and additionally include members who may not currently practice critical care. In addition, since respondents largely reported practicing in academic/tertiary medical centers, our findings may not accurately portray practices in other health system settings. More than half the respondents are under 50 years old, and this survey may not be an accurate reflection of intensivists who have been in practice for longer periods. Notably, responses may be more indicative of physician rather than organizational-level practice as ICU sections with more members are likely to be overrepresented. Further limiting our analysis was the inability to comment on exact salaries of respondents or their graduation year. In the survey participants were asked to indicate their responses by selecting a single category, therefore salaries were averaged across the range for each category for analysis, which may have affected the results we observed. For fellowship graduation year this resulted in a higher proportion of respondents indicating 2015-2020, thus not allowing us to formally report quartiles of respondents and instead arbitrarily categorizing fellowship year based off the observed distributions of the data.

However, our findings lend themselves to further work regarding critical care anesthesia practice. In the future, additional granular information about factors affecting compensation, temporal trends in ICU coverage and clinical practice patterns can help inform decisions on staffing, clinical organization and compensation. Trends highlighted in these results will also inform strategies to reduce burnout, which continues to be a serious concern amongst intensivists^26–31.

Anesthesiologists trained in critical care medicine represent a flexible critical care provider resource well-equipped cover a variety of intensive care units. Results of this survey show that critical care anesthesiologists are comfortable practicing in a variety of ICUs and coverage models. This survey accomplishes an important first step in describing current practice of critical care anesthesiologists and generating hypotheses for further work.

KEY POINTS.

Question:

What is the current demographic composition of critical care anesthesiologists and what are the most important themes across their aggregate practice nationwide?

Findings:

Critical care anesthesiologist members of SOCCA currently practice in a variety of organizational structures and staffing systems and, although practice locations involved a variety of intensive care units, the most commonly reported was a cardiothoracic ICU followed by a surgical ICU.

Meaning:

Recognition of the wide variety of organization and staffing systems used by critical care anesthesiologists in the United States will assist clinical managers in developing sustainable staffing and coverage models that optimize patient care, and further work may better identify drivers of satisfaction among anesthesiologist intensivists in the United States and subsequently factors that mitigate burnout.

Glossary Terms:

CCM

critical care medicine

ICUs

intensive care units

IRB

Institutional Review Board

BIDMC

Beth Israel Deaconess Medical Center

SOCCA

Society of Critical Care Anesthesiologists

REDCap

Research Electronic Data Capture

IQR

interquartile range

CHERRIES

Checklist for Reporting Results of Internet E-Surveys

CI

confidence intervals