How best to train the next generation of medical specialists has become a question of significant concern as our profession comes under increasing scrutiny on issues of performance, patient safety, and return on public investment.1–4 Traditional systems of graduate medical education (GME) have focused heavily on cognitive and technical competencies, which has led to patchy and incomplete training in domain-independent skills such as communication, interdisciplinary teamwork, and awareness of the professional’s place in a complex system of health care. Without a mechanism to drive transformational change, educators often struggle to provide consistency in programmed learning encounters and assessment. Inconsistencies in the quality of training can build over long periods of time, with some programs lagging far behind whilst others flourish even within the same institution or country. This situation is now an urgent concern, with reform seen as essential to improve the fidelity of the training process and build a workforce competent to practice in a time of rapid change in health care demand and delivery.5
Transformational Change in GME
In the past decade, national stakeholders have turned to regulation and “topdown” implementation of reform based on the principles of competencybased training.6–9 Reforms are usually context specific, driven from within national institutions and agencies with reference to best practices and stakeholder consultation. Rapid, wholesale implementation of an external accreditation system (taken from one national jurisdiction and applied to another) is without precedent in GME and represents a novel approach to educational reform. All of Singapore's public hospitals recently engaged in a phased realignment to the requirements of the Accreditation Council for Graduate Medical Education International (ACGME-I), a new subsidiary organization of the United States Accreditation Council for Graduate Medical Education (ACGME).
Changes in any health care environment take place within complex, integrated, and culturally nuanced systems of funding, service delivery, and professional development. Transformational change in GME is fraught with the risk of disturbing the delicate balance between these components, thereby slowing reforms or distracting from their primary objective.8, 9 For example, detailed critiques of the Modernising Medical Careers program in the United Kingdom have highlighted many such threats including inadequate justification of reforms in terms agreed on by all stakeholders, inadequate time to pilot new processes, and difficulty managing the consequences of new regulations for hospital staffing and medical career progression.8, 9 With respect to these difficulties, one author proposed that “adherence to simple change-management theory and practice might have improved the acceptance of change and reduced the confusion.”8
Here, we describe implementation of ACGME-I standards at a single academic medical center in Singapore from 2009 to 2012. We have organized our discussion using the change-management framework laid out by Kotter and colleagues, which has been described in detail in several publications and shared in the public domain.10–12 This model describes eight key processes necessary to conceptualize and execute transformational change within a large organization:
* Create a sense of urgency.
* Form a powerful coalition.
* Create a vision for change.
* Communicate the vision.
* Remove obstacles.
* Create short-term wins.
* Build on the change.
* Anchor the changes in corporate culture.
Awareness of this framework together with sensitivity to local antecedent, coincident, and contemporary issues may help others succeed in circumstances similar to ours. Finally, we offer a glimpse of the effort and expense required at our institution to transition completely from one GME model to another.
National Considerations: Singapore’s Existing Training System
Singapore has transitioned rapidly from low to high per-capita gross domestic product through a process of industrialization, urbanization, investment in tertiary education, and enthusiastic engagement in global trade. Citizens and permanent residents enjoy universal access to public health care through a system of heavy government subsidy, within a financing system anchored on the principles of individual responsibility, charitable endowment, and affordable health care for all citizens.13 Growth in health care and biomedical research sectors is a stated priority of the government, and significant funds for infrastructure, staffing, research, and education have followed.14 Although a substantial private health care system operates in Singapore, public hospitals are exclusively responsible for GME.
Until recently, the Yong Loo Lin School of Medicine at the National University of Singapore (first established as the Singapore Medical School in 1905) remained the sole training institution for the degree of bachelor of medicine and surgery (MBBS). Successful completion of a five-year undergraduate medical degree led to a one-year internship at an assigned government hospital. After this, trainees were registered as medical practitioners with the Singapore Medical Council, allowing further training in government hospitals or independent practice in private community clinics. Doctors remaining in hospital practice after internship were designated as medical officers, rotating every six months within or between institutions. Separate administrative bodies provided oversight: one for employment and service provision (Ministry of Health Holdings, a state-owned holding company for public health care assets), another for accreditation of training (the specialist accreditation boards of the Singapore Medical Council), and others for summative evaluation of medical knowledge and skills (usually the Royal Colleges of the United Kingdom).
Entry into formal postgraduate training was largely at the discretion of the medical officer, allowing continuing workplace experience whilst he or she decided on a particular career or specialty. Once enrolled to a period of basic specialist training (equivalent to residency in the United States), a medical officer could express attachment preferences through a ranking system based on career interests and training gaps, but would usually continue to rotate between institutions to gain required clinical exposure. Entry to advanced specialist training (equivalent to U.S. fellowship training) was delayed until the appropriate college examination was successfully completed. Until 2008, therefore, GME in Singapore was a rotating apprenticeship of varying duration with standards set internally by national bodies and externally through the Royal College system. Apprenticeship-style training with close links to the U.K. Royal Colleges reflected the historical development of GME. Singapore did not gain full independence from Britain until 1959. Under colonial rule, local graduates had not been encouraged to undertake formal postgraduate training. This situation changed with the inauguration of the Singapore Academy of Medicine in 1957, which promoted specialization and encouraged postgraduate study overseas to develop local expertise and training capacity.15
Making the Case for GME Reform
In 1993, a government white paper proposed regulating the supply of medical manpower tightly for fear of creating supply-induced demand. For a long period, the university medical class was limited to 150 students. Increases in health care use caused by population growth, aging, chronic metabolic disease, and increasing health care complexity were not foreseen, and a medical staffing shortage developed.16 The high-stakes nature of college examinations, the associated cost of repeated attempts to pass, and the early option of exit to private practice led many local MBBS graduates to practice in the community without recognized postgraduate qualifications. The problem of growing public hospital workload was, therefore, compounded by poor retention of staff. The time devoted to clinical work (and, for many, the competing demands of research and publishing) made it difficult to protect faculty and trainee time for education and ensure that central disbursements of training funds were spent for the purpose intended. In summary, GME in Singapore faced the following difficulties:
* Significant proportions of the workforce practicing without formal postgraduate training or qualifications
* Allocation of postgraduate funds to a training system with substantial variation in duration and rates of completion
* Poor correlation between investments in education and improvements in assessment and training “on the ground” (i.e., adoption of best practices)
* Difficulty reconciling the competing demands of clinical service, research, and education at undergraduate and postgraduate levels, with a lack of accountability to the latter
This was the national background when the ACGME was approached by the Ministry of Health to oversee the external accreditation of GME at all public hospitals in Singapore. The competency model and the process of individual hospital accreditation held appeal as a direct method to address the concerns given above.17 A formal letter of agreement establishing the relationship between the ministry and ACGME-I was signed in the fall of 2009. A vice president for accreditation services was appointed to lead ACGME-I in November 2009.18 Offering an example of the pace of reform, plans to transition to ACGME-I-accredited GME had actually been announced to hospital leadership during a series of meetings in late 2008 with the stated intent of accepting a proportion of the graduating class of May 2010 into seven provisionally accredited programs. External consultation for “the most fundamental and comprehensive revamp of our specialist training system in the last 50 years” had been limited.19
Establishing a Sense of Urgency
The change deadline itself and the inevitability of a shift in national policy were reinforced by the Ministry of Health from the outset, but the degree to which reforms were prioritized and implemented in each hospital depended on the support of hospital executives. It became clear with time that many aspects of the new accreditation system offered solutions to three major strategic institutional concerns. Firstly, a merger between the National University Hospital and the National University of Singapore schools of Medicine, Dentistry, and Public Health was established in 2008. This formed a National University Health System with the principal aim of promoting academic medicine. The reach of the new organization’s educational mission was, therefore, broad, extending from undergraduate to early and advanced postgraduate levels. The intellectual, material, and human resources operating in the field of medical education needed to be managed with structure and accountability. This was offered by the principles of administration in GME laid out in the ACGME-I accreditation requirements. Secondly, the hospital needed specialists with the broad competencies required to manage clinical complexity and think across traditional boundaries to deliver patient-centered management in an integrated system of health care delivery. A GME system emphasizing competence in clinical quality improvement (i.e., practice-based learning and improvement) and systems-based practice met this need. Thirdly, rotation between hospitals thwarted the possibility of taking a longitudinal approach to training in clinical quality, orientated to institutional systems and policy. All of these concerns could be addressed as a consequence of the changes proposed. The hospital executive affirmed the reforms as a top priority and set out to form the coalition responsible for leading the necessary changes.
Forming a Powerful Coalition
A group responsible for managing the accreditation process was drawn from existing hospital faculty and staff. Without the time required to develop a formal recruitment process, the hospital relied on appointment by invitation with a promise of protected time for education, discussed below. Memorably, the designated institutional official (DIO) accepted her position after a short telephone conversation, without a detailed understanding of the responsibilities the job entailed. Although this may appear unseemly in retrospect, appointment by invitation was a necessity given the short time scale involved, and key positions were only offered to staff who were well known for their specific interest, ability, or experience in postgraduate education and leadership. New external recruitment was limited to the appointment of administrators and program coordinators through the hospital’s human resources department.
One of the first tasks of the DIO and her new staff was to inaugurate a department of medical affairs for education (MAE). The organization chart for this new department is shown in Figure 1. MAE was provided with dedicated office space, meeting rooms, and newly seconded senior administrative staff with a strong track record in managing complex projects in previous positions in the organization. MAE amalgamated the office of the associate dean (responsible for undergraduate education) with the office of the DIO, ensuring equal representation of undergraduate and postgraduate education to the hospital’s board. New programs in faculty development, resident welfare, curriculum design, evaluation, internal review, human resource management, and resident-led research were established, drawing more staff with an interest in education to committee positions under the umbrella of the GME committee. Realignment of undergraduate and postgraduate administration created a “virtuous cycle,” whereby ACGME-I accreditation could be overseen whilst harmonizing activities across the continuum of medical education described above. For example, undergraduate curriculum and assessment systems were purposefully reorganized with graduating competencies and residency requirements in mind.
The position of the DIO was reinforced by vocal support from hospital executives and by the clear and concisely articulated requirements for accreditation outlined by the ACGME-I in its institutional, foundational, and advanced specialty requirements. The residency model placed the DIO in a position of oversight to trainees who would now stay with the institution for a minimum period of three years. The opportunity for continuous supervision and development of local talent was particularly appealing to senior clinicians operating with the longer-term development of their departments in mind. The DIO and her coalition were therefore equipped with a clear mandate to deliver change, with willing partners amongst senior clinicians and the financial means to make change stick.
Funding subventions for MAE were substantial; ACGME-I guidelines for protected time dedicated to GME were used to benchmark salary requirements for core faculty and program directors. The Ministry of Health agreed to fund these requirements directly to the hospital, which could in turn use the money flexibly, recruiting new staff if necessary to maintain clinical workload whilst others (i.e., new program directors and core faculty) devoted more time to GME. “Back-fill,” as this process was known, allowed the DIO and MAE to offer sufficient assurance to clinical leaders and administrators that increasing time for GME would not adversely affect the ability of clinical departments to reach existing targets in noneducational domains. A sum of 9.4 million Singapore dollars (S$) was spent in the financial year to April 2011 from the hospital’s operating margin of S$300 million. Seventy-five percent of this was allocated to medical salaries and a further 6% to faculty development.
Creating and Communicating a Vision of Competency-Based GME
We make the following observations of our efforts to communicate the ACGME-I model of competency-based medical education to colleagues during planning and implementation for accreditation. “People skills” were crucial. The DIO and program directors spent considerable time selling the message of competency-based education to get buy-in from clinicians of all seniority. Regular presentations were made at the hospital’s medical board, ensuring inclusion of the hospital’s most senior staff in the decision-making process and allowing material concerns and criticisms to influence the implementation effort. The six competencies and the progress of our efforts were discussed at special grand rounds, department meetings, and open briefing sessions. The competencies were also marketed using posters placed around the hospital to familiarize staff, students, and patients. We found that the competencies were easy to market to clinicians, who instinctively understood their worth. The emphasis on non-domain-specific areas such as communication skills and systems-based practice provided reaffirmation to many faculty who believed that the traditional emphasis on knowledge and technical skill led to inadequate emphasis on other professional attributes.
Other forms of communication were used to maintain the flow of information at three important levels:
* To the project coalition as a whole
* To staff within the institution
* To prospective residents and external stakeholders (including local media)
Staff retreats were held by MAE and by individual departments to organize responses to the accreditation deadline. E-mail was used extensively to resolve day-to-day problems and refine plans and ideas. Public Internet and private intranet pages allowed access to important updates and to the prospectus of each new residency program. Social networking using Facebook and Twitter was included in our communication strategy to reach out to prospective residents, many of whom were planning to apply from overseas. Given the rapid pace of change, the ability to broadcast detailed updates to those following the process was particularly important.
Removing Obstacles and Empowering Others to Act
A significant outlay was made on faculty development under oversight of the University Medical Education Unit with leave and funding prioritized for MAE staff and new program directors. Local, national, and international training experiences were chosen for their relevance to the change process and quickly provided an important “shared language” of best practices in GME through which problems in the design and implementation of our programs could be understood. Although it is beyond the scope of this article to list every training event, we would point out that the visiting programs organized by the Harvard Macy Institute and ACGME faculty were particularly influential.
Lessons learned during faculty training were used to drive local programs of change, often refined and finalized during staff retreats. The way was opened for staff to pursue improvements in their own programs and teaching. Any improvement that better met ACGME-I requirements and philosophy tended to be met with encouragement. Curricula were drafted and assessment tools chosen. The content, layout, revision, and use of resident portfolios were discussed at length. A centrally coordinated core education program for all first-year residents was implemented to avoid duplication of effort within individual departments.20 Capital investment was made in an online residency management system to reduce the time spent completing and compiling assessments and other documentation. A two-day residential course for new residents and faculty was arranged to build relationships, morale, and rapport amongst educators and trainees. An objective structured clinical examination was administered to document trainees’ clinical competence on entry into the program and provide residents with early experience of assessment and performance feedback.
Many improvements in GME had already been made within departments, and adapting these changes to the context of the ACGME-I framework was crucial in meeting the accreditation deadline. Within the department of medicine, for example, chief resident positions had already been introduced to help improve the working life of the department and its trainees.21, 22 This had led to introduction of a night float system, which ultimately helped to bring duty hours within limits compatible with ACGME-I requirements. To improve supervision and mentoring, a physician advancement program had been established to retain select trainees for the duration of their basic specialist training. The longitudinal evaluation of these trainees provided experience easily applied to residency training.
Short-Term Wins, Building on Change, and Anchoring Within Institutional Culture
We made efforts to publicize early gains to colleagues. A key short-term win was the mock external audit undertaken in May 2010 by expert site visitors experienced in ACGME audits in the United States. The successful results of this audit boosted the confidence of those involved and greatly renewed the enthusiasm of the team. The positive words of the auditors were used to publicize the strengths of the programs as they existed at that stage and helped close remaining shortcomings where problems were pointed out. New processes and learning activities were critically examined for their worth within the first few months of their introduction through a mixture of formal survey, feedback, and faculty opinion. We presented findings at local and international meetings to develop a professional focus on education and increase awareness of GME as a subject for research and publication.20, 23
Recognizing and developing systems of reward for early gains built persistence by embedding them in the culture of the institution and its departments. Performance indicators for all staff have been reorientated to take into account overall contributions to education as well as faculty feedback from residents. A “clinician educator” academic career track that was developed shortly prior to the ACGME-I reforms is increasingly used for key residency faculty, with well-defined criteria for clinical and academic progression. In the expectation that a self-perpetuating process of improvement has been introduced through educational reform, concrete outcomes relating to clinical care, staff involvement in patient safety activities, patient satisfaction surveys, and staff retention rates are being watched carefully.
Looking Back, Looking Forward
Although we have sacrificed much of the detail to provide a coherent narrative, we have illustrated how a GME system can be reengineered to meet an external standard. Although we would commended Kotter’s framework of change management, we have used it here simply to make sense of our efforts in retrospect. We would emphasize the importance of the following steps in the context of our experience:
* Establish a sense of urgency: The motivation for early efforts to establish funding, strategic initiative, and administrative reorganization were defined from a clear sense of urgency driven by the internal and external factors described.
* Form a powerful coalition: As part of the coalition rather than bystanders, senior executives and clinical leaders were influential in setting the tone and strategy for change, allocating resources, resolving organizational bottlenecks, and promoting parallel changes in other systems (such as a well-managed academic career track for educators). A well-funded program of faculty development for key staff was pivotal, providing a shared language of best practices in medical education.
* Create and communicating a vision for change: The extent to which most faculty and prospective residents were unfamiliar with the concepts of competency-based training as articulated by ACGME requirements cannot be overemphasized. The addition of a Web site and the use of social media to traditional forms of communication proved essential to managing information and expectations.
As of the spring of 2012, 13 programs of residency training are accredited, 6 being accredited by October 2010 and a further 7 following a series of site visits undertaken in 2011. This brings an end to the implementation phase and a further transition from the task force approach to the steady work of the committee. Senior staff retention in leadership positions within the programs is high. Bonds have formed between generations of doctors in a traditionally hierarchical system of leadership and responsibility. Residents are increasingly turned to for advice on quality and process improvement, for clinical leadership in patient care, and to gauge sentiment toward change efforts in noneducational domains. Our experience suggests that external accreditation, if appropriately funded, led, and supported, can act as a powerful stimulus for change in GME programs.
Acknowledgments: The authors wish to acknowledge the wide cast of program directors, administrators, and faculty colleagues involved in residency implementation from 2008 to 2010. In particular, thanks to Professor Ho Khek Yu for guidance and support and the 2010 NUHS pioneer residents themselves, for whom this work was carried out and whose input was gratefully received. Finally, the Kotter Principles (and by extension the idea for this article) were introduced to the authors by Professors Elizabeth Armstrong and Tom Aretz of the Harvard Macy Institute.
Other disclosures: None.
Previous presentations: Elements of this work were presented by the authors in poster format at the Eighth Asia Pacific Medical Education Conference, Singapore, January 26–30, 2011; and at the 2011 ACGME Annual Educational Conference and preconference in Nashville, Tennessee, March 3–6. The Asia Pacific Medical Education Conference abstract was published as Khoo SM, Huggan PJ, Olszna DP et al. Transitioning to the ACGME competency-based internal medicine residency program—The experience of a university hospital in Singapore. Med Educ. 2011;45(3 suppl):11.