Characteristics of Designated Institutional Officials and Sponsoring Institutions of ACGME-Accredited Training Programs

The designated institutional official (DIO) role has evolved since its inception in 1998.¹  Originally envisioned as a job to ensure substantial compliance with the Accreditation Council for Graduate Medical Education (ACGME) requirements, there are now 11 894 ACGME-accredited residency and fellowship programs in the United States during the 2019–2020 academic year that collectively enrolled 145 906 residents and fellows.²  The DIO role now extends beyond ensuring compliance with ACGME requirements and may involve managing graduate medical education (GME) budgets, addressing legal issues, working strategically in health care policy, and attending to professional development as health care leaders while fostering the growth of others to prepare them for succession.¹ 

Many factors have influenced evolution of the DIO role. Implementation of the ACGME Outcome Project that established the 6 major competencies, codifying limits on resident and fellow clinical and educational (duty) hours, and establishment of a single integrated system for accreditation of allopathic and osteopathic residencies are among the most significant changes. More recently, the ACGME Next Accreditation System, including implementation of the Clinical Learning Environment Review (CLER) process,³  profoundly influenced the role of the DIO. Yet, there remains a large gap in our knowledge regarding baseline sponsoring institutions (SIs) and DIO characteristics as well as a need for institutions wishing to increase diversity in the DIO role and other leadership positions within the GME community. The primary purpose of this study is to characterize the relevant characteristics of DIOs and the SIs they oversee. This will lead to a better understanding of the characteristics of DIOs and SIs at a single point in time, provide insight for others who may consider this career trajectory, and provide baseline data to assess the changing characteristics of DIOs and SIs over time.

The authors identified institutions sponsoring ACGME-accredited residency and fellowship programs on its website on February 15, 2020.²  We excluded SIs that reported no data on program number or filled resident positions on the ACGME website (eg, blank entry). If the ACGME website reported any numerical value (including 0) for number of programs or filled resident positions, these data were included for analysis. The ACGME website was used to identify DIO name, DIO appointment date, SI accreditation status, SI accreditation date, number of sponsored programs, and total filled resident/fellow positions. While there are no established parameters to define SI size, we utilized data from the ACGME website and from prior publications on CLER site visits to define the SI sizes as follows: large (> 30 programs), medium (3–30 programs), or small (0–2 programs) institutions.^4–6 

The authors used publicly available sources including SI websites, state licensure boards, and online curriculum vitae to identify DIO academic degree, primary specialty (physicians), age, gender, and academic rank. We defined senior academic rank as professor or associate professor. We determined the duration of DIO appointment as the difference in years between DIO appointment start date and the date data collection was initiated (February 15, 2020). We defined duration of practice as the duration of employment in the DIO's primary clinical profession post-residency (if DIO was a physician), recorded as the difference in years between employment start date and the date data collection was initiated. Three authors (R.S.D., B.L., S.D.) performed data collection in parallel, and another author (T.L.) adjudicated any discrepancies. Interrater reliability was excellent with a Cohen's kappa score of 0.95.

We present dichotomous categorical variables as numbers with percentages and continuous variables as medians with a 25% to 75% interquartile range (IQR). Categorical variables were compared using Fisher's exact test, and the Mann-Whitney U test was used to compare continuous variables. A P value < .05 was considered statistically significant. We adjusted significance thresholds based on the Bonferroni method⁷  separately for statistical comparisons of SIs and statistical comparisons of DIOs.

The authors performed a post-hoc comparison of institutional accreditation status comparing physician DIOs versus non-physician DIOs. A similar post-hoc analysis was performed comparing demographic characteristics between DIOs with MD degrees and DIOs with DO degrees. This study was exempt from Institutional Review Board review.

The authors identified 855 ACGME-accredited SIs at the time the review started (February 15, 2020). We included 831 SIs for analysis because 24 SIs had no data on program number or filled positions. The status of 715 SIs (86.0%) was continued accreditation and the status of 81 SIs (9.7%) was initial accreditation. Eighteen SIs (2.2%) had an accreditation status of initial accreditation with warning and 17 SIs (2.0%) had an accreditation status of continued accreditation with warning. The ACGME reported no SIs assigned probationary accreditation.

The number of ACGME-accredited programs per SI ranged from 0 to 170 with a median of 3 programs (IQR 1.0–11.0). The number of ACGME-accredited resident and fellow positions within SIs ranged from 0 to 2165 with a median of 47.0 positions (IQR 17.0–153.0). Large and medium SIs were more likely to have continued accreditation status and less likely to have initial accreditation status than small SIs (Table 1; P ≤ .001 for all comparisons). SIs with continued accreditation also had more residency and fellowship programs (median 4.0, IQR 1.0–13.0) compared to SIs with initial accreditation (median 1.0; IQR 0–1.5; P < .001).

The authors identify DIO characteristics in Table 2. Of the DIOs with known academic rank, 356 of 576 (61.8%) held senior academic rank (professor or associate professor). Two hundred eighty-three of 831 DIOs (34.1%) were women. Six hundred ninety-eight of 831 DIOs (83.9%) were physicians, with 618 of 831 (74.4%) having a MD degree and 80 of 831 (9.6%) having a DO degree.

The median duration of DIO appointment was 57.0 months (4 years, 9 months) and did not differ between men and women. Of those with known clinical practice duration post-residency, male DIOs were employed in their practice for longer (27.0 years, 19.5–34.0) than female DIOs (21.5 years; 16.0–29.0; P < .001).

The authors also compared DIO characteristics by gender (Table 3). Of those with known academic rank, a higher proportion of male DIOs had senior academic rank (67.5% vs 52.2%; OR 1.90 [95% CI 1.34–2.70]; P < .001) and academic rank of full professor (39.3% vs 24.4%; OR 2.01 [95% CI 1.38–2.94]; P < .001) compared to female DIOs. There was also an association between female gender and a greater number of ACGME-approved programs and filled resident/fellow positions (P < .001 for both associations).

Physician DIOs were more likely to lead SIs with continued accreditation status compared to non-physician DIOs (87.7% vs 77.4%; OR 2.07 [95% CI 1.30–3.30]; P = .002). DIOs with a MD degree had a longer duration of employment in their primary profession compared to DIOs with a DO degree (27.0 years vs 19.0 years, P < .001; more information provided as online supplementary data).

Although the majority of DIOs are men, 283 of 831 (34.1%) DIOs are women and female gender was associated with leadership of SIs with more programs and more residents and fellows in training. The predominance of male DIOs is consistent with the existing body of literature documenting gender inequity in academic medicine.^8,9  International studies similarly confirm gender disparities in the recruitment and advancement of women in academic leadership roles.^8,10  Despite leading larger SIs, female DIOs were less likely to have senior academic rank or academic rank of full professor than male DIOs; however, this association may be confounded by the significantly younger age of female DIOs observed in our study and the corresponding significantly fewer number of years spent in their profession. We report a higher proportion of female DIOs who had no academic rank.

The median DIO appointment duration was 4 years and 9 months. Our study did not provide empiric data regarding DIO turnover, which may be an area for future study. We speculate potential reasons for DIO turnover may include an increasing number of newly accredited SIs and preference for limited terms to allow other candidates opportunities to serve in a leadership role. We also hypothesize DIO turnover may result from the increased accountability associated with expanding DIO responsibilities, increased regulatory burden, and a lack of resources to implement change. The rapidly changing health care environment, including the prevalence of mergers and acquisitions, is a further challenge to DIOs. Common challenges affecting GME when mergers and acquisitions occur include variable organizational culture, mistrust of intentions, lack of communication and shared vision, and substantial changes that may erode the learning environment.¹¹  The DIO possesses a history with the organization, knowledge about the ACGME, and leadership that can foster a successful transition.

There are potential advantages and disadvantages for the DIO to be a physician. We observed an association between physician DIOs and SIs with continued accreditation. Physician leaders have the advantage of being experienced in health care delivery.¹²  They may lead by example and demonstrate the intended change through their own clinical practice. By being at the frontline of health care delivery, physician executives may be able to implement more effective change. A systematic review demonstrated evidence supporting the importance of including physicians on governing boards to improve organizational performance.¹³  However, the impact of physician leadership in GME is unknown. Notably, much of the increasing burden on DIOs comes from responsibilities outside clinical education. For instance, studies suggested DIOs should have expertise in state law related to licensing and visa issues, as well as legal proficiency in formulating resident contracts and complying with labor laws.¹  Non-physician DIOs with expertise in business administration, law, and finance may be advantaged by this experience that physicians often lack.

The authors note that 59 of 772 (7.6%) DIOs report an MBA degree, 38 of whom are physicians. There has been considerable growth in physician MBAs in the United States, with many employed in the clinical sector or within hospital administration.¹⁴  Studies have demonstrated that physician executives who pursued a MBA degree viewed it as a crucial step toward leadership, team-building, awareness of public health issues and strategies, enhancement of negotiation skills, and understanding the financial aspects of medicine.¹⁵  The value of an MBA degree to DIOs is unproven, but this training may equip them with skills to face the challenges of a dynamically evolving health care and clinical education landscape.¹⁵ 

Our study demonstrated that large and medium-sized SIs were more likely to have continued ACGME accreditation and less likely to have initial ACGME accreditation compared to small SIs. The rapid growth of training in new subspecialties may lead to an increased number of applications from new SIs at smaller specialized academic medical centers and community practices in the development phase.¹⁶  This is consistent with our finding that SIs with more programs and filled positions are associated with a higher likelihood of continued accreditation status.

To the best of our knowledge, this is the first study to review demographic, academic, and clinical practice-related characteristics of DIOs in the United States. Additional studies might identify prior leadership and educational roles (eg, residency program director) that prepare DIOs for the challenge of this appointment. These and similar data should inform career paths for future DIOs and better identify experiences and development programs to ensure DIOs are well-prepared to meet the challenges of this important role.

Our study has several limitations. We assessed the characteristics of DIOs at a single point in time. Missing data for certain variables, notably DIO appointment duration and highest academic degree obtained, may impact outcomes. We were unable to determine DIO academic rank in over 30% of DIOs, and individual DIOs were not contacted to obtain missing data. The missing academic rank information was mostly applicable to community-based programs; thus, our analysis likely overrepresents DIOs from academic and university-based programs. Since we abstracted information from public databases, our study may be limited by reporting bias. There were several SIs excluded that had no data on resident and program number in the public Accreditation Data System, likely because they are newly accredited this academic cycle. We did not abstract data on leadership or educational roles held prior to DIO appointment. This may be an important component in preparing the DIO for leadership responsibilities.^17,18  Information on DIO race and/or ethnicity is important. However, these data were not reported because they were not publicly available.

Ongoing collection of DIO and SI information would be of interest and may identify trends over time. Certain objective data on SI and DIO characteristics over the last 12 years are published online,¹⁹  and future studies should analyze these data.

This cross-sectional study describes characteristics of DIOs and SIs at a single point in time and offers insights for those pursuing a DIO position and for institutions wishing to increase diversity. Female DIOs are underrepresented, are less likely to have senior academic rank or full professorship, and are more likely to lead larger SIs.