Synergy, Salary, and Satisfaction: Benefits of Training in Critical Care Medicine and Infectious Diseases Gleaned From a National Pilot Survey of Dually Trained Physicians

A survey of physicians trained in critical care medicine (CCM) and infectious diseases (ID) suggested this combination is synergistic and satisfying. However, most respondents had to train in individual specialties at separate institutions. Avenues for CCM-ID training should be considered.

Abstract

Background. An increasing number of physicians are seeking dual training in critical care medicine (CCM) and infectious diseases (ID). Understanding experiences and perceptions of CCM-ID physicians could inform career choices and programmatic innovation.

Methods. All physicians trained and/or certified in both CCM and ID to date in the United States were sent a Web-based questionnaire in 2015. Responses enabled a cross-sectional analysis of physician demographics and training and practice characteristics and satisfaction.

Results. Of 202 CCM-ID physicians, 196 were alive and reachable. The response rate was 79%. Forty-six percent trained and 34% practice in the northeastern United States. Only 40% received dual training at the same institution. Eighty-three percent identified as either an intensivist with ID expertise (44%) or as equally an intensivist and ID physician (38%). Median salary was $265 000 (interquartile range [IQR], $215 000–$350 000). Practice settings were split between academic (45%) and community settings (42%). Two-thirds are clinicians but 62% conduct some research and 26% practice outpatient ID. Top reasons to dually specialize included clinical synergy (70%), procedural activity (50%), and less interest in pulmonology (49%). Although 38% cited less proficiency with bronchoscopy as a disadvantage, 87% seldom need pulmonary consultation in the intensive care unit. Median career satisfaction was 4 (IQR, 4–5) out of 5, and 76% would dually train again.

Conclusions. CCM-ID graduates prefer the acute care setting, predominantly CCM or a combination of CCM and ID. They find combination training and practice to be synergistic and satisfying, but most have had to seek CCM and ID training independently at separate institutions. Given these findings, avenues for combined training in CCM-ID should be considered.

(See the Editorial Commentary by Armstrong on pages 876–7, and the HIV/AIDS Major Article by Weiser et al on pages 966–75.)

More than 4 million patients are admitted to intensive care units (ICUs) in the United States each year [1]. Increases in life expectancy, comorbidities, immune suppression, and healthcare exposure in recent years have raised mean illness severity and complexity among ICU patients. Correspondingly, the ICU staffing model in the United States has undergone a paradigm change, with more “closed units,” multidisciplinary teams, greater diversity in intensivists' training backgrounds, and increasing demand for involvement of subspecialists [2]. More than half of all ICU admissions worldwide are related to infection, and infectious diseases (ID) consults are most frequently sought on ICU patients [3, 4]. Consultation by an intensivist or ID specialist has been shown to improve survival among the acutely ill [5, 6]. Because critical care medicine (CCM) and ID exhibit considerable overlap in disease epidemiology and involve interplay of multiple organ systems, dual training might offer synergistic benefit to patient care. Thus, it is unsurprising that a small, but growing, number of physicians are training in both CCM and ID in the United States [7]. Such dual training should, theoretically, diversify their skill set and foster cross-departmental clinical and research collaboration [8–10]. However, the training paths, career trajectories, and levels of career satisfaction among CCM-ID graduates remain unknown. This lack of data presents a missed opportunity to inform internal medicine residents who might otherwise consider dual training as well as recent CCM-ID graduates seeking a viable career niche. Well-conducted surveys gauge provider attitudes and satisfaction levels and may impact practice and policy [11, 12]. We conducted a survey of all CCM-ID physicians trained in the United States to catalogue their career paths and thoughts on this relatively rare combination.

METHODS

The questionnaire was prepared by the study investigators and distributed online using SurveyMonkey (Palo Alto, California). Questions were focused on subject demographics, practice characteristics, opinions on the combined specialty, and career satisfaction. The Office of Human Subject Research Protection at the National Institutes of Health (NIH) Clinical Center (Bethesda, Maryland) considered the study exempt from institutional ethics review. A data use agreement was signed between the NIH investigators and the American Board of Internal Medicine (ABIM), and a contact list of all US-trained CCM-ID physicians was generated from the ABIM data repository. Subjects were included regardless of board certification status in CCM and/or ID; also included were physicians without formal CCM fellowship training who obtained board certification in CCM between 1987 and 1989 by virtue of clinical experience. Physicians included on the basis of training required at least 1 year of training in each of the 2 specialties to be included in the ABIM dataset. No incentives were offered for completing the survey.

The survey was distributed in 2 phases, via email with the survey Web link and 2 subsequent weekly email reminders. Subjects were only contacted via telephone if email addresses were not initially available or if email requests bounced back. Respondents were permitted to skip questions and remain anonymous. Individual responses that were implausible (eg, 80 weeks/year of clinical service) were excluded. Aggregate responses were reported as means (±SD) for continuous variables, proportions for categorical variables, and medians (interquartile range [IQR]) for ordinal variables. Because the study was aimed at understanding training and practice characteristics and satisfaction levels among CCM-ID graduates, our analyses were descriptive. A copy of the survey and additional detail on survey phases are provided in the Supplementary Data.

RESULTS

Two hundred two CCM-ID trained or certified physicians were identified in the ABIM data repository, of whom 2 had died and 4 were unreachable. Overall, 155 of 196 physicians contacted completed all or part of the survey (response rate 79%) (Figure 1).

Figure 1.

Flowchart demonstrating the distribution of graduates with training in critical care medicine (CCM), infectious diseases (ID), and both, as well as the breakdown of respondents. ^aReceived at least 1 year training; ^blncludes trainees that received training for at least 1 year in CCM training programs or critical CCM/pulmonary disease combined programs; ^cDual certification implies CCM and ID certification by the American Board of Internal Medicine. Not everyone received formal CCM training; diplomates were allowed to be “grandfathered” ie, earn certification through a practice pathway for the first three rounds of exam administration (1987–89); ^dAs of November 2014; ^eSurvey phase-1 “response rate = 94/108 (87.0%): This phase consisted of respondents with dual active certification as of November 2014. Survey phase-2 response rate = 61/88 (69.3%): This phase consisted of respondents without dual active certification, which included those without initial or maintenance of certification in one or both specialties as of November 2014. 25% of the respondents in phase-2 reported having recently completed fellowship training.

Demographics and Training Characteristics

Demographic and training characteristics are presented in Table 1. The mean age was 46.3 ± 10 years and 77% of the respondents were men. Forty-six percent of CCM and 42% of ID fellowship training occurred in the northeastern United States. Forty percent of physicians completed their fellowships at the same institution. The number of fellowship graduates dually trained in CCM and ID has remained small, but steadily increased: 9 before 1987, 22 in 1987–1996, 38 in 1997–2006, and 78 in 2007–2015 (counts based on 144/155 [93%] responses). Some CCM-ID graduates also maintained board certifications in other subspecialties: 27 in pulmonology, 7 in neurocritical care, 2 in hospice and/or palliative medicine, 2 in sleep medicine, and 1 in geriatrics. The 202-physician CCM-ID cohort included 25 (12%) physicians who were never enrolled in an official CCM or pulmonary and critical care medicine (PCCM) training program, but were allowed to earn certification through a practice pathway that acknowledged their clinical experience in CCM (source: ABIM).

Table 1.

Critical Care Medicine/Infectious Diseases Physician Survey: Demographics and Training Characteristics (n = 155)

Characteristic	CCM-ID Respondents	Responses per Question
Age, y, mean±SD	46 ± 10	127 (82)
Sex		152 (98)
Male	117 (77)
Female	35 (23)
Ethnicity		152 (98)
Not Hispanic or Latino	136 (90)
Hispanic or Latino	16 (11)
Race		150 (97)
White	96 (64)
Black or African American	9 (6)
Asian	41 (27)
Native Hawaiian or other Pacific Islander	1 (1)
Mixed race	3 (2)
Location of CCM fellowship		146 (94)
Northeast US	67 (46)
Midwest US	21 (14)
South US	20 (14)
West US	26 (18)
Outside the US	12 (8)
Location of ID fellowship		151 (97)
Northeast US	64 (42)
Midwest US	29 (19)
South US	25 (17)
West US	25 (17)
Outside the US	8 (5)
CCM and ID fellowship at same institution		151 (97)
No	91 (60)
Yes	60 (40)
Board certifications in other specialties		152 (98)
Internal medicine	134 (88)
Pulmonary	27 (18)
Pediatrics	3 (2)
No other certifications	7 (5)
Other	14 (9)
Neurocritical care	6 (4)
Sleep medicine	2 (1)
Emergency medicine	2 (1)
Palliative care	1 (1)
Geriatrics	1 (1)
Tropical medicine	1 (1)
Palliative care and neurocritical care	1 (1)

Data are presented as No. (%) unless otherwise indicated.

Abbreviations: CCM, critical care medicine; ID, infectious diseases; SD, standard deviation; US, United States.

Practice Characteristics

Practice characteristics are presented in Table 2. Respondents most commonly identified themselves as intensivists with ID expertise (44%) or equally as intensivists and ID physicians (38%). Only 10% identified as ID physicians with CCM expertise. The average difficulty associated with finding a position with clinical responsibilities in both specialties was rated at 3.0 ± 1.3 on a scale of 1 to 5 (5 indicating most difficult). Nearly a third of the respondents were unable to find a position with dual practice components. Respondents reported a mean salary of $281 000 ±$106 000 and a median of $265 000 (IQR, $215 000–$350 000). Thirty-four percent currently practice in the northeastern United States and 12% currently practice overseas. Employment settings were almost equally divided between academic (45%) and community or private practice (42%). Federal positions accounted for 9% of employment positions.

Table 2.

Critical Care Medicine/Infectious Diseases Physician Survey: Practice Characteristics and Career Satisfaction (n = 155)

Characteristic	CCM-ID Respondents	Responses per Question
Primary self-identification		138 (89)
Intensivist with expertise in ID	61 (44)
ID physician with expertise in CCM	13 (10)
Intensivist and ID physician equally	53 (38)
Neither	1 (1)
Other	10 (7)
Current employment location		148 (96)
Northeast US	50 (34)
Midwest US	21 (14)
South US	28 (19)
West US	31 (21)
Canada	9 (6)
Outside the US and Canada	9 (6)
Employment setting		140 (90)
Academic	63 (45)
Community or private practice	59 (42)
Federal	13 (9)
Administrative	2 (1)
Policy	1 (1)
Industry	1 (1)
International or global health	1 (1)
Other hospital roles		138 (89)
Antibiotic stewardship	61 (44)
Quality control or quality improvement	52 (38)
Hospital epidemiologist/infection control	39 (28)
ICU director	30 (22)
Code team	26 (19)
Residency or fellowship program director	12 (9)
Ethics committee	7 (5)
Other	18 (13)
None	29 (21)
Average weeks on service, mean ± SD
In ICU	19 ± 12
On ID service	16 ± 14
Percentage of time spent in various roles, mean ± SD
Clinical	67 ± 27
Research	13 ± 22
Education	8 ± 8
Administrative	9 ± 15
Industry	1 ± 2
Editorial	0.2 ± 2
Other hospital	3 ± 7
Other	0.2 ± 2
Primary research focus		140 (90)
Clinical	61 (44)
Epidemiology	7 (5)
Translational	10 (7)
Basic science	9 (6)
No research	53 (38)
Area of clinical work		145 (94)
Both CCM and ID	69 (48)
CCM only	60 (41)
ID only	11 (8)
Neither CCM or ID	5 (3)
Type of ICU employed in		140 (90)
Medical	42 (30)
Surgical/trauma	4 (3)
Mixed medical/surgical	69 (49)
Cardiothoracic	9 (6)
Neurologic	2 (1)
Not applicable—does not work in ICU	14 (10)
Satisfaction with career path (0–5), median (IQR)	4 (4–5)	142 (92)
Receptivity of departments to CCM-ID		140 (90)
CCM more than ID	55 (46)
ID more than CCM	12 (9)
CCM and ID equally	43 (31)
Neither were receptive	9 (6)
Not applicable (I do not practice CCM or ID)	11 (8)

Data are presented as No. (%) unless otherwise indicated.

Abbreviations: CCM, critical care medicine; ICU, intensive care unit; ID, infectious diseases; IQR, interquartile range; SD, standard deviation; US, United States.

Respondents reported spending a majority of their time doing clinical work (67%), followed by research (13%), administrative (9%), and education (8%). However, 62% are involved in research in some capacity, which is distributed among clinical (44%), translational (7%), basic science (6%), or epidemiological (5%) categories. Other roles included committees for antibiotic stewardship (44%) and quality control (38%). Some assumed leadership roles as hospital epidemiologists (28%), ICU directors (22%), and residency/fellowship program directors (9%). A majority practice both CCM and ID (48%) or CCM alone (41%), with only a few currently practicing ID alone (8%) or neither (3%). Respondents who attend in the ICU do so for 19 ± 12 weeks per year and those who provide in-patient ID consultations do so for 16 ± 14 weeks per year. Respondents practice in mixed medical-surgical ICUs (49%), medical ICUs (30%), cardiothoracic ICUs (6%), and surgical/trauma ICUs (3%). Fifty-three of 144 (37%) respondents maintain an outpatient practice component, of whom 38 (26%) maintain an ID clinic.

Opinions Regarding CCM-ID

Qualitative responses pertaining to training and practice characteristics of CCM-ID are presented in Figure 2A–D and Table 2. CCM departments were the most receptive of respondents' dual training (46%), while nearly a third of respondents (31%) perceived both CCM and ID departments as equally receptive. The top 3 reasons to specialize in CCM-ID were perceived clinical synergy (70%), increased procedural activity over isolated ID training (50%), and lack of comparable interest in pulmonology (49%). Opportunities for hands-on care (77%), the ability to do invasive procedures (76%), and higher compensation compared to ID practice alone (70%) were advantages offered by CCM to ID practice. Fifty-nine percent of respondents believe that dual training better enabled them to consult on ID cases in the ICU. Advantages offered by ID training to CCM included a better ability to identify multidrug-resistant infections (87%), improve antibiotic stewardship (86%), manage unusual infections (86%), and identify sepsis (71%). Perceived disadvantages of training in ID in lieu of pulmonary training included less proficiency with bronchoscopy (38%), increased difficulty in obtaining an intensivist job (24%), and less proficiency managing lung disease (18%). When attending in the ICU, ID consults are most commonly called when the need for long-term follow-up is anticipated (36%). Despite lacking formal pulmonary training, 63% of the respondents claim they never request pulmonary consults and 24% request them at most 1–2 times a week while attending in the ICU.

Figure 2.

A–D, Qualitative responses pertaining to training and practice characteristics of critical care medicine (CCM) and infectious diseases. Respondents were allowed to select >1 option. Abbreviations: ID, infectious diseases; MDR, multidrug resistant.

Respondents Without Active Dual Certification

Forty-eight percent of respondents completed dual training but did not hold active board certification in both specialties. Twenty-five percent of respondents were not eligible to take one or both certification examinations because they only recently finished fellowship. In the remainder, stated reasons for not obtaining dual initial certification included onerous effort (16%), isolated CCM practice (13%), unnecessary for fellowship completion (11%), and isolated ID practice (5%). For those who obtained dual initial certification and were eligible for maintenance of certification in one or both, reasons for not doing so included onerous effort (19%), isolated ID (16%) or CCM practice (16%), lack of recertification requirements for continuing current practice (14%), or prohibitive cost (10%) (Supplementary Table E1).

Respondent Satisfaction

Respondents rated satisfaction with their career path at a median of 4 (IQR, 4–5) out of 5 (5 indicating most satisfied). Among 142 of 155 respondents, 45% were extremely satisfied, 38% very satisfied, 10% satisfied, 6% somewhat satisfied, and 1% dissatisfied. If respondents had to reselect a career path, 76% claimed they would retrain in CCM and ID, 2% would retrain in CCM alone, and none said they would train in ID alone (Figure 3).

Figure 3.

Fellowships that respondents would pursue if they, hypothetically speaking, had to repeat their training. *“critical care medicine (CCM) and other specialty” was subclassified into “CCM, infectious diseases (ID), and pulmonary” and “CCM and other.” The latter subgroup included anesthesiology (2), nephrology (2), and medical informatics as specified by respondents in free text.

DISCUSSION

This is the first report to provide insight into the national experience of physicians specializing in CCM and ID in the United States. The survey respondents represent more than three-quarters of everyone who has trained in CCM and ID in this country to date. Our results show that CCM-ID graduates have gone on to pursue a variety of health-related careers in academic, community, and other settings and assume multiple professional roles. Most CCM-ID trained physicians are practicing clinicians, although nearly two-thirds are involved in some research activity. They enjoy high levels of career satisfaction and feel they enjoy greater hands-on care and compensation than if they had trained in ID alone.

Despite the perceived advantages of CCM-ID training observed in the survey, such dual graduates comprise only 1% of all CCM physicians nationwide, and most have trained only recently [7]. Several factors might account for their underrepresentation. At present, most training programs nationwide lack dedicated CCM-ID tracks. A lack of data around the feasibility of this combination has likely been the major impediment to generating new training initiatives. Program directors who may have intended to create such tracks are faced with the difficult question: What is the appropriate division-level distribution of salary and mentorship responsibility? Such allocations are more complex at institutions where CCM and pulmonary divisions are combined and training is intertwined, which is the case at most institutions across the country. It is not surprising that 60% of CCM-ID graduates sought training at separate institutions for each specialty. Motivated trainees are required to individually explore and negotiate combined training opportunities, which generally involves applying to 2 separate fellowship programs over different application cycles. Individualized training tracks may foster professional sequestration and disconnection from PCCM or ID training departments. Further, CCM and ID professional societies and international forums currently lack CCM-ID assemblage. Despite these factors, a sharp uptrend has been observed in dual graduates over the last decade at fellowship programs across all geographic regions. If this trend persists, we are likely to see more CCM-ID providers over the next decade, regardless of whether official combined tracks operationalize. Workability of such tracks could be determined by establishing a CCM-ID training pilot at institutions offering independent CCM training.

The proportion of ID training programs that fail to fill first-year positions via the match has progressively risen each year, up to an alarming 58% this year [13, 14]. With waning interest in joining ID training via the match and the projected nonexpansion of per capita full-time equivalent supply, the ID physician pipeline is in jeopardy [15]. According to the 2015 Physician Compensation Report by Medscape representing the opinion of nearly 20 000 physicians from 26 specialties (with 1% CCM and 1% ID specialists), the average ID physician salary occurred in the lowest quartile for medical subspecialists at $213 000 [16]. Fifty-two percent of the ID physicians in that report claimed not feeling fairly compensated. A task force convened by the Infectious Diseases Society of America (IDSA) has been commissioned to comprehensively examine the decline in interest in ID, including at dual programs, and to propose potential remedies to reverse this trend [17]. The promotion of combined training in ID and CCM has been suggested as one measure to renew interest in training in ID [7, 18]. The findings of our survey may stimulate discussions among program leadership nationwide and provide pilot data to gauge the utility of combined CCM-ID training and stimulate trainees to consider this nontraditional training path. While the combination may incentivize training in ID, it should not be perceived as a means to generate more outpatient ID providers. Candidates who train in critical care prefer the acute care setting. Most respondents did not think of themselves as primarily ID physicians, but rather as intensivists with an expertise in ID or as equally an intensivist and ID physician. Hence, it is not surprising that less than a fourth practice outpatient ID. The average salary of the CCM-ID–trained physician in our survey closely matches the national average for CCM physicians, which is $281 000 [16]. Notably, 89% of respondents in our survey have a CCM practice component, suggesting that their salary is weighted toward CCM.

To be well-rounded ID providers, trainees must acquire core competency in all facets of the specialty. Notwithstanding, the need for compartmentalization of ID provider roles in the developed world is inevitable. The information base on management of human immunodeficiency virus (HIV)/AIDS continues to expand at a rapid pace and has prompted the establishment of HIV-centric training opportunities [19–21]. Similarly, with rising case complexity, the expanding role of ID-savvy physicians in the management of acutely ill patients is being increasingly recognized [22]. These paradigm shifts suggest that ID divisions, which respondents reported as being less receptive to combined training than CCM divisions, may stand to gain from embracing the concept.

Demand for intensive care services is also likely to increase as the US population aged >65 years is expected to grow from 12% in 2000% to 19% by 2030 [23]. While the adequacy of the CCM provider pipeline is debatable, enhancing quality and diversity of the current workforce is imperative [24, 25]. Programmatic reform and innovation attract high-quality trainees. A Web-based survey of emergency physicians demonstrated a surge in emergency medicine–CCM physicians in recent years despite a lack of pathway formalization and board eligibility [26]. Provisions for emergency physicians to train and certify in CCM have been implemented in response and represent milestones of programmatic innovation [27]. A similar pattern was noted in our survey. Creation of CCM-ID tracks is likely to generate a pool of satisfied and motivated intensivists with a unique and valuable skill set that extends beyond the standard ID knowledge base that is developed during a CCM fellowship. The CCM community can leverage this skill set to enhance hospital and ICU-specific infection control practices, responses to antimicrobial resistance, and preparedness against global threats such as Ebola virus disease and anthrax, as well as enhance the research work force for diagnostic biomarkers, epidemiologic surveillance, and novel therapies for sepsis. Currently, 42% of CCM-ID trained physicians are in private practice. Addition of ID-trained intensivists to community-based intensivist group practices will enhance diversity in knowledge base and experience and enable easy peer-to-peer consultation.

Career satisfaction has direct ramifications on the physician workforce, patient satisfaction, and outcomes [28, 29]. The practices of CCM and ID individually involve long work hours, systematic multiorgan assessments, and a need for attention to detail and have been independently implicated in physician burnout [30, 31]. Notwithstanding, most respondents demonstrated high degrees of career satisfaction. A previous satisfaction survey suggested high levels of job satisfaction even among non-CCM–trained ID physicians (5.53 on a 7-point Likert scale) [30]. In the aforementioned Medscape report, 53% of CCM and 54% of ID physicians claimed overall career satisfaction [16]. In our study, 93% respondents claimed overall satisfaction. In the report, 44% of CCM and 50% of ID physicians claimed they would choose the same specialty again, which was claimed by 76% of respondents in our study. As such, these do not present an apples-to-apples comparison, and measuring the precise increment in satisfaction afforded by dual training would entail comparisons using the same survey instrument. However, certain aspects of CCM-ID observed from the survey likely mitigate the dissatisfaction that often accompanies physician burnout.

More than a third of the respondents claim less proficiency with bronchoscopy. Trainees in PCCM programs enjoy greater opportunities for bronchoscopy through ambulatory and inpatient consultation. However, 87% of the respondents rarely need a pulmonary consult while attending in the ICU, and only 12% would include pulmonary fellowship if they had to retrain. Fiber-optic bronchoscopy is a core curricular requirement for nonpulmonary CCM training, and the lesser need for pulmonary consultation may indicate respondents' comfort level with bronchoscopy on intubated patients. Procedural electives and simulation-based training, known to enhance bronchoscopy skills, could be incorporated into the training curriculum of formal CCM-ID tracks [32].

Our study is not without limitations. First, we did not survey physicians with other combinations of training, limiting direct comparisons with other specialties. However, the study was designed to elicit pilot data about an underrecognized specialty combination rather than generate comparative analyses among specialties or specific patterns of training within CCM-ID. Second, it is possible that dually trained physicians who are more satisfied with their careers were more likely to complete the survey. However, the overall response rate was 79%, which far exceeds response rates obtained in similar physician online questionnaires in the past and mitigates nonresponder bias, if any [33]. Third, the option to respond anonymously precluded us from eliciting specific differences between respondents and nonrespondents. We provided question-specific response rates wherever possible. Fourth, as with social desirability bias, use of the survey as a promotional opportunity for the specialty combination would likely have been alleviated by self-administered online questionnaires in lieu of live interviews [34]. Fifth, the response rate was higher among those who carried dual active certification than among those who did not. However, the latter group included recent graduates, who may have been hesitant to respond to practice characteristics and experience. Last, our results do not reflect the characteristics and opinions of those who received subspecialty training outside the United States.

CONCLUSIONS

Graduates dually trained in ID and CCM find the combined training and practice opportunities synergistic and satisfying. They reported more hands-on care opportunities and higher compensation than those trained in ID alone. Yet the majority of respondents have had to seek fellowship training in the individual subspecialties at separate institutions. Programmatic efforts to encourage combined training in both specialties could enhance physician satisfaction and meet unique and increasingly complex patient care needs as our population ages.

Supplementary Data

Supplementary materials are available at http://cid.oxfordjournals.org. Consisting of data provided by the author to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the author, so questions or comments should be addressed to the author.