Where Does the Week Go? Parsing Trends in Program Director Workload Open Access

Program directors (PDs) need to follow several sets of regulation imposed by different governing bodies (eg, the Accreditation Council for Graduate Medical Education [ACGME], specialty boards, local institutional graduate medical education offices) as well as to be in charge of the admission process, resident promotion, resident graduation, and preparation for residents who opt for pursuing a fellowship program.^1,2 

Although the influence of PDs on the success and well-being of individual residents is undoubtedly significant, it is difficult to quantify which characteristics and behaviors of PDs are specifically associated with resident outcomes. Because of the job complexity and demanding learning curve of becoming a PD, turnover has been hypothesized to be detrimental to residents and is a topic of interest in many specialties.^3,4  For example, within general surgery residency programs, greater PD turnover is associated with higher failure rates on board examinations for graduating residents.⁵  In addition, a significant proportion of PDs have tenure shorter than the term required by the ACGME.^4,5  Turnover can affect faculty, residents, program quality, and recruitment.⁶ 

Surgical specialties have been shown to have higher resident attrition rates than other specialties,^7-10  which may be related to a variety of factors, including lifestyle issues and resident and program characteristics.^11-13  Better understanding of the interplay between PDs and resident retention may aid in the development of strategies to mitigate attrition, particularly for residents in specialties at high risk of attrition. In this study, we were interested in elucidating the tenure and workload of residency PDs and exploring the relationship between PD factors and resident attrition rate.

This is a retrospective study performed using publicly available data obtained from the ACGME Data Resource Book from 2012 to 2021.¹⁴  All ACGME-accredited anesthesiology, emergency medicine, family medicine, internal medicine, obstetrics and gynecology (OB/GYN), pediatrics, psychiatry, general surgery, plastic and integrated plastic surgery, orthopaedic surgery, otolaryngology (ENT), integrated thoracic surgery, urology, and integrated vascular surgery programs were included in the analysis, as these are the largest residency specialties based on number of residents. Variables obtained directly from the ACGME Data Resource Book include mean number of hours per week spent by PDs on administration, clinical supervision, teaching/didactics, and total hours; number of programs with new PDs; number of residency programs; percent of programs with new PDs (turnover); number of residents; number of residents who withdrew from residency; and average tenure in years of PDs. Resident attrition rate was calculated as the number of residents who withdrew from residency divided by total number of residents within a given specialty each year. Data are aggregated at the specialty level, with one observation per specialty per year.

Descriptive statistics were used to examine characteristics related to PDs, including percent change in PD tenure over the study period, for each specialty. To examine changes in average PD tenure and total hours per week spent by PDs on program-related activities over the study period, linear regression models were used, with year and specialty as predictor variables. Three additional linear regression models were used to examine the relationship of year and specialty with proportion of weekly hours spent on clinical, administrative, and teaching duties, respectively. Binary logistic regression analysis was used to examine the effect of program factors (specialty, year, weekly hours spent by PDs on program-related activities, and PD turnover) on resident attrition rate. General surgery was selected as the reference group because there is a large existing body of literature about resident attrition in general surgery. We used SAS Studio software, version 3.81 (SAS Institute Inc), to perform the analysis for this article.

The average tenure of PDs has decreased between 2012 and 2021 for all included specialties, except anesthesiology, family medicine, OB/GYN, and integrated plastic surgery (Table 1). On average, the tenure of a PD decreased from 7.0 years to 6.6 years in this interval. On linear regression analysis, length of tenure was significantly associated with year (2021 compared to 2012, B=-0.3, P=.01; see Table 2). Tenure length was also related to specialty. Relative to the reference group (general surgery), anesthesiology, OB/GYN, integrated thoracic, and integrated vascular surgery PDs had significantly shorter average tenure, while ENT, pediatrics, and urology PDs had longer tenure on average (Table 2).

Among all specialties, the total number of weekly hours spent by PDs on program-related activities, classified as administrative, teaching, or supervising, has decreased over time (from 51.3 hours in 2012 to 44.2 hours in 2021), mainly at the expense of teaching (Figure 1A and supplementary data Table 1). Proportionally, this represents an increase in time devoted to administrative work (34.9% in 2012 to 36.9% in 2021: Figure 1B). Linear regression analysis showed a significant association between year (2021 compared to 2012) and proportion of time spent on teaching (B=-0.02, P<.0001), clinical (B=0.01, P=.03), and administrative (B=0.02, P<.0001) activities (supplementary data Tables 2-4). This trend is most pronounced among internal medicine PDs, where the proportion of time devoted to teaching has dropped by over 40% (17.5% in 2012 vs 10.1% in 2021), with a corresponding increase in the proportion of time devoted to clinical supervision (from 27.0% in 2012 to 34.6% in 2021; supplementary data Figure 1). The steepest decline presented during the COVID-19 pandemic.

Overall, the proportion of weekly hours devoted to teaching ranges from 8% to 18% for all specialties. When averaged across the 10-year study period, clinical supervision comprises the largest proportion of work-related hours for PDs in most procedural specialties (anesthesiology, ENT, general surgery, orthopaedic surgery, plastic surgery, thoracic surgery, urology, and vascular surgery), while administration is the primary activity for all other PDs (Table 3).

A logistic regression model of program-related factors on resident attrition did not show any significant association between PD turnover and attrition. Resident attrition rate was significantly associated with specialty, with all specialties having lower odds of attrition relative to general surgery, except plastic, thoracic, and vascular surgery. Hours spent each week by PDs on clinical supervision were significantly associated with resident attrition rate, with a decreased odds of attrition associated with increased hours spent on clinical supervision (OR 0.93, P<.0001; Table 4).

In this article, we examined the current work environment for PDs in surgical and nonsurgical fields. We found a decrease in the average length of tenure and the total number of weekly hours devoted to residency-related activities by PDs across specialties. There is a significant association between time spent by PDs on clinical supervision and resident attrition rates.

A previous survey among Canadian PDs showed that the amount of protected time does not correlate with job time requirements. Although the relationships and fulfillment of PDs was high, it was counterbalanced by a lack of administrative support and compensation. The net effect was that many PDs consider leaving their position.¹⁵  In the current study, we found that total weekly hours spent by PDs on PD-related responsibilities has been declining in recent years. This may result in decreased job satisfaction and ultimately impact tenure, as we have shown that tenure has decreased for PDs overall in the past decade. Thus, the looming effect of this decrease in work hours may signify a future exacerbation of the trend in decreased PD tenure, which may have pernicious effects on board passage rates, resident recruitment, faculty experience, and indirectly on patient care. An alternate explanation is that these changes in weekly workload reflect a delegation of responsibilities to other core faculty within the residency program.

In 2000, a survey of PDs revealed that 18% of time was spent teaching¹⁶ ; we found a similarly low proportion of time devoted to teaching among PDs in all fields in our study. Emotional exhaustion was reported frequently in the above-referenced survey and, importantly, was more prevalent among younger PDs than those who were more experienced.¹⁶  This may further contribute to increased future turnover. In 2020, 34% of internal medicine PDs met burnout criteria, and 45% contemplated leaving the position in the past year.¹⁷  The administrative burden placed on PDs, especially in nonprocedural specialties, coupled with the decreased support in terms of work hours may play a critical part in the development of burnout among PDs.

Although we did not find a statistically significant relationship between PD turnover and resident attrition rates in the current study, we did show an association between an increased number of hours per week spent by PDs on clinical supervision and lower rates of resident attrition. This may reflect a relationship between programs that place a priority on appropriate resident support in clinical settings and resident satisfaction and retention. Lower attrition rate residency programs have been shown to be more likely to use resident remediation¹⁸ ; thus, PDs who spend more time on clinical supervision may also be better able to directly mentor or remediate residents at risk of attrition. It is also possible that a greater abundance of hospital resources and support may serve as a moderator variable underlying the relationship between clinical supervision and resident attrition.

This study was limited by using aggregated retrospective data, which does not contain detailed information about individual PDs that may help to explain some of the trends we have observed in the current article. Additionally, the outcome measure of this study is resident attrition rate; while attrition is important, it does not fully capture resident well-being or success. Other outcomes such as resident satisfaction, competency, or post-residency performance could provide a more nuanced understanding of the impact of PD characteristics on residents.

While these findings offer valuable insights into the current state of PDs, future work should explore additional PD characteristics and their impact on a broader range of resident outcomes to provide a more comprehensive understanding of the complex dynamics at play within residency training programs. Such knowledge is crucial for informing strategies to support PDs in their roles and optimize resident experiences.

In this study, we found a decrease in PD tenure and total weekly hours spent by PDs on program-related activities from 2012 to 2021. Additionally, we observed a significant association between time spent by PDs on clinical supervision and resident attrition rates.

The online supplementary data contains further data from the study.

Funding: The authors report no external funding source for this study.

Conflict of interest: The authors declare they have no competing interests.

This work was previously presented as a poster at the Association for Surgical Education Surgical Education Week, April 2024, Orlando, Florida, USA.