Late-Career Pediatric Hospitalists: Programmatic Accommodations and Supports Free

In this cross-sectional study, we used a web survey administered to PHM division chiefs across the country to evaluate late-career hospitalist schedule accommodations and division chief knowledge about the existence of retirement programs and benefits.

We developed this survey via consensus of the study group on the basis of literature review^13,14  and a series of prospective respondent, expert, and cognitive interviews to ensure the validity of the content and response process. To optimize content validity, we obtained input from 5 late-career pediatric hospitalists (also prospective respondents) focused on content inclusivity, representativeness, clarity, and relevance of the tool at assessing our aims. To optimize response process validity, we performed 4 cognitive interviews using “think aloud” methodology with PHM division chiefs from different institutions and regions of the country. Finally, we pilot tested the instrument by asking 4 different PHM division chiefs to take the survey and provide feedback. The survey is shown in the Supplemental Information.

The electronic survey was administered by using REDCap, a secure, Health Insurance Portability- and Accountability Act-compliant, web-based application for electronic data capture hosted by the University of Arizona.¹⁵  Pediatric hospitalist leaders were surveyed anonymously from May 2020 to June 2020 through the American Academy of Pediatrics (AAP), Section of Hospital Medicine (SOHM) Division Chief listserv. Reminders to complete the survey were sent on three occasions. This listserv is open to members of the AAP Section of Hospital Medicine who are chiefs or associate chiefs of their PHM divisions. An AAP representative estimates enrollment at ∼300 members (N. Alexander, personal communication) but she could not comment on the number or types of programs represented.

Program and respondent characteristics were obtained via the survey for all respondents. This information was unavailable for nonrespondents.

The questionnaire contained yes or no, multiple-choice, and open-ended questions. (see Supplemental Information). Most survey questions included quantitative responses (eg, yes, no, or don’t know) first, followed by a free text opportunity to provide additional details or a description. Questions were formulated to capture information related to program size, location, and employer type. Additional questions investigated program considerations for late-career hospitalists, including the following: (1) variations in overnight call responsibilities, (2) modifications to the overall clinical schedule, (3) modifications to the clinical schedule because of COVID-19, (4) retirement planning services, and (5) retirement benefits. Surveys with less than 50% of the questions answered were excluded. Duplicate surveys from the same institution were also excluded, utilizing only the most complete survey or the first survey submitted if multiple surveys were complete.

Descriptive analyses summarized categorical data using count and percent. Association of program characteristics with schedule considerations and retirement benefits for late-career hospitalists was examined by using the Fisher exact test and its extension, the Fisher-Freedman-Halton test (estimated using the Monte Carlo simulations based on 10 000 random samples). Statistical analyses were performed by using the SAS software (Version 9.4, SAS Institute Inc, Cary, NC, USA). All statistical tests were 2-sided with significance evaluated at the 5% level. Example free-text descriptions reported by multiple division directors were identified, but a formal qualitative analysis was not performed.

The study was approved by the institutional review board at the study site. A survey introduction clarified that consent to use the submitted deidentified information was inferred by the completion and submission of the questionnaire.

There were 65 respondents to the survey (response rate = 22%, based on ∼300 listserv members), with 47 meeting inclusion criteria. The respondent program characteristics are presented in Table 1. These 47 programs encompassed 728 full-time equivalent (FTE) positions, employing 982 individual hospitalists with 253 (26%) employed by a medical school, 472 (48%) by a hospital, 250 (25%) by a medical group, and 7 (1%) with missing data. Program size varied from 2 to 63 FTEs, with 6 (13%) programs employing 4 or fewer FTEs, 25 (53%) with 5 to 14 FTEs, 9 (19%) with 15 to 29 FTEs, and 7 (15%) programs that employed 30 or more FTEs. Division directors estimated that the 982 individual hospitalists included 117 (12%) aged 50 to 64 years and 16 (1.6%) aged 65 years and older. Of the 117 hospitalists aged 50 to 64 years, 49 (42%) worked on a part-time basis, as did 8 (50%) of 16 hospitalists aged 65 or older. Of the 47 programs, 44 (94%) had at least 1 member over 50 and 13 (28%) had a member(s) aged 65 or older. One program had a member over 65 but nobody in the 50 to 64 group.

Of the responding programs, 35 (76%) reported that their hospitalists perform in-house overnight call responsibilities. Of these 35 programs, 22 (63%) employed at least 1 hospitalist over 50 but none over 65 years. Only 5 (23%) programs in this subgroup allowed opting out of any night call responsibilities. In contrast, of the 11 programs performing in-house night call which had members over 65 years, 6 (55%) allowed late-career hospitalists to opt out of some or all night call responsibilities. The 21 programs that made no allowances for late-career hospitalists indicated that age is not a factor in determining night shift assignments. As shown in Table 2, larger programs were more likely than smaller programs to allow late-career hospitalists to do fewer night shifts (P = .004). No statistically significant correlations were found between night call policies and type of employer (Table 3). In response to the question “Please provide any additional comments or information,” several division chiefs indicated concern that allowing late-career hospitalists to opt out of night call responsibilities would result in scheduling imbalances. One division chief expressed this sentiment as follows, “We would have too few people available for night shifts and too many people for the day shifts.” Several programs also indicated that, as more of their members approached age 50, they will need to develop a policy for night call. The comment “Allowing older faculty to opt out of night call works for us now, but when all of us currently in our 40’s hit 60, it will be harder to operationalize,” captures this concern. One program reported forming a “longevity committee” to develop a night call policy that considered age and length of employment to determine frequency of night call responsibilities.

Of the 43 programs employing hospitalists between 50 and 64, only 8 (19%) indicated that they make any modifications in daily schedules for late-career pediatric hospitalists and only 1 (2%) program indicated any reduced clinical productivity requirements for older physicians. Modifications to the daily schedule were more frequently reported (5 of 13 programs, 38%) for programs employing hospitalists aged 65 and older. Schedule modifications, such as part-time employment, were based on age, seniority, or the individual needs of some hospitalists. No statistically significant correlations were found between programs offering schedule modifications and employer type or program size (Tables 2 and 3).

Many programs that employed hospitalists over the age of 50 years made accommodations for older physicians in response to the COVID-19 pandemic. Fifty-two percent (52%) of programs employing hospitalists between ages 50 and 64 years modified clinical responsibilities, whereas those employing hospitalists aged 65 and older more commonly made schedule changes to protect older individuals (77%). No statistically significant correlations were found between COVID-19-related schedule modifications and employer type or program size (Tables 2 and 3). When asked to “describe the accommodations that you made for late-career physicians due to the Coronavirus pandemic,” division chiefs reported modifications to both clinical time and clinical exposure, with exposure-based changes being more common. Exposure-based modifications ranged from not having late-career hospitalists care for any known COVID-19-positive patients to removing late-career physicians from all or most clinical patient care responsibilities. “We offered all late career faculty the ability to designate in a higher risk group and, if selected, would assign to non-clinical duties,” stated one division chief. The programs that enacted time-based changes indicated that late-career physicians took some form of leave or decrease in FTE status, with plans to make up shifts later. “We offered those over 65 to move shifts into next year if desired,” stated one division chief.

Figure 1 outlines the retirement programs and benefits available at the respondent institutions. Notably, for many retirement items, the division chief was unsure whether their institution provided certain benefits. Financial planning programs for employed and retired hospitalists were offered by a total of 32 (68%) programs. For some programs, financial advisors were made available by the medical institution itself, whereas in other programs physicians could obtain financial advice and planning from the administrators of the retirement plan, such as TIAA-CREF or Fidelity.

Division chiefs of only 10 programs (21%) indicated that they were aware that their institution offers nonfinancial retirement counseling and only 15 (32%) division chiefs knew that their institution grants academic benefits to retired hospitalists. The benefits most offered were E-mail, access to conferences and the medical library, volunteer teaching or committee activities, and formal emeritus status. No program offered office space to retired physicians.

Health benefits for retirees were reported by only 9 (19%) programs. When available, health benefits for retirees were supported by a university or a governmental program for state employees. No programs indicated that they offered any incentives for early retirement. Medical school employers were more likely to provide academic and health benefits (P = .002) than were hospital or medical group employers (Table 3). No differences in provision were noted by program size (Table 2).

The medical literature contains many studies and editorials related to aging and retirement, but to our knowledge, this is the first study to directly examine the important issues of late-career pediatric hospitalists, their work schedules, and their retirement benefits. Pediatric Hospital Medicine is a young field, and our survey reveals that most pediatric hospitalists are in the early or middle stages of their careers, with the proportion of late-career pediatric hospitalists over the age of 60 years much lower than the published rate of 30% for all physicians. Although limited by a low response rate, in our study a majority of division directors endorsed that they have at least 1 member over 50 years old, highlighting the need to begin exploring the issues of aging and retirement in PHM. Our study also demonstrates that, although many programs were able to make temporary schedule modifications for late-career physicians during the COVID-19 pandemic, more permanent schedule policies allowing older hospitalists to opt out of some, or all night call responsibilities were mainly limited to large programs. Retirement counseling and support programs were rarely offered, and division chiefs were often unaware whether specific benefits and supports were available at their institutions.

In 2017, Rimsza found 20% of clinically active members of the AAP were 61 years old or older.¹⁶  According to our survey respondents, PHM appears to have far fewer late-career physicians in practice, with only 1.6% of members aged 65 years or older and 12% aged 50 to 64. It appears many programs have at least 1 member over the age of 50; however, programs without older physicians may not have responded, resulting in nonrespondent bias. In the future, as the pediatric hospitalist workforce ages, it is likely that the issues of late-career pediatric hospitalists, especially regarding a program’s clinical manpower requirements, will need greater focus. It will be important to have adequate numbers of seasoned clinicians to deal with the increasing complexities of caring for hospitalized patients and to mentor younger, less experienced hospitalists. As stated by McDonald, “There is no substitute for a sage, wise physician in every practice and the patient benefit that flows from their accessible consultation.”³ 

The programs responding to our survey make a variety of accommodations for older late-career pediatric hospitalists. Some programs, especially large programs and programs with hospitalists 65 and older, limit night call responsibilities for older physicians. Although there is some evidence that some older physicians are more prone to medical error and that sleep deprivation can be problematic for all physicians,^4–10  we found no research specifically revealing that sleep deprivation is more problematic for older physicians. We did not explore the reasons why some programs limit night call responsibilities for older physicians. It may be based on anecdotal input that sleep deprivation is harder for those of advanced years. Another possibility is that this represents a lifestyle reward enacted to retain physicians with seniority rather than a decision made for safety and reduction of medical error. The validity of making accommodations for older physicians is supported in the literature. For example, one study revealed that the main reasons for retirement in physicians aged 60 to 69 were excessive workload and burnout.¹⁷  In another study of practicing late-career physicians, factors that would extend their careers were decreased workload, shorter hours, and decreased work-related bureaucracy.¹⁸  As the pediatric hospitalist workforce ages, it may not be possible for programs to allow for night call modifications and still have enough individuals to provide nighttime coverage. Strategies, such as hiring more nocturnists or offering higher pay for younger physicians willing to cover night shifts, may need to be entertained. Other strategies designed to improve circadian adaptation, such as rotating and bunching night shifts to allow recovery from sleep debt, should be considered.⁹ 

It is clear from the current COVID-19 pandemic that pediatric hospitalist divisions recognize that certain accommodations may be needed to protect older, late-career physicians. The consequence of special accommodations for older physicians is that younger physicians need to work more (and less desirable) shifts. For these younger physicians, this situation is a potential source of resentment toward their older colleagues. In a study of members of the American Pediatric Society, several late-career respondents indicated that they felt they still had contributions to make, but they had been treated in a way that made them feel “unwanted and undervalued.”¹⁹  If we are to foster a culture that recognizes the importance of senior physicians, programs need to find ways to foster mutual respect and a shared understanding of the importance of the unique roles of the late-career physician.

Progression to retirement can be difficult for physicians.²⁰  There are many competing factors that can push a physician toward or away from considering retirement. Financial needs, as well as the desire for camaraderie and self-esteem, must be balanced against factors such as burnout, deteriorating health, need to care for a spouse or other family member, concern about competence and ability to keep up with medical advances, or frustration with bureaucracy and “paperwork,” just to name a few.^16,21,22  To help physicians move toward a successful retirement, some institutions, primarily medical schools, provide some benefits and supports including (1) financial planning, (2) retirement transition counseling and mentoring to help with emotional issues of retirement, (3) postretirement health benefits, and (4) academic benefits such as access to E-mail and library and ability to teach and serve on committees.^17,19,22,23  Two recent studies of late-career faculty at US medical schools indicated that, despite a rapidly increasing number of older faculty, there are few formal retirement policies or supports in place to help ease older physicians to a successful retirement.^24–26  Therefore, it is not surprising that our study did not identify significant programmatic support for retiring or retired pediatric hospitalists. Interestingly, many hospitalist division chiefs were unaware whether such supports and benefits exist at their institutions.

Additional studies are needed, using qualitative interviews with individual hospitalists themselves, to explore opinions and attitudes regarding such issues as night call responsibilities, productivity requirements, and support programs for pre- and postretirement. In addition, a survey of younger pediatric hospitalists, focusing on attitudes regarding aging physicians and accommodations made for late-career physicians, could help inform staffing models and effective means of developing collaboration between all practitioners.

We were unable to obtain information about listserv subscribers, so we do not know if their programs are representative of the PHM universe. Additionally, significant differences may exist between responding and nonresponding programs. If programs without older hospitalists did not respond, our finding that most programs employ late-career physicians could be the result of nonrespondent bias. This limits the ability to generalize this data to assess the total number of programs employing late-career pediatric hospitalists. The small percentage of responding programs calls into question the generalizability of our data to all hospitalist programs and conclusions must be made cautiously. Nevertheless, we did receive responses from large and small programs, from a variety of employer types (medical school, hospital, practice group), and from all regions of the country. Finally, the lack of division chief knowledge about retirement supports and benefits does not indicate that these programs do not exist.

The number of pediatric hospitalists in the late-career stage is small. Most programs do not allow aging physicians to opt out of any night call responsibilities, but most programs did recognize the need to offer older physicians schedule and assignment modifications related to COVID-19 exposure. Other than financial counseling and some academic benefits, most programs did not provide retirement counseling or other supports for late-career physicians. Our research is preliminary, but divisions should now begin the next steps of evaluating the pros and cons (and divisional costs) of making various accommodations and providing supports for late-career pediatric hospitalists.

We thank Hanna Riggins for her contributions to this study.