High-Quality Learning for High-Quality Health Care: Title and subTitle BreakGetting It Right

In July 2003, the Accreditation Council for Graduate Medical Education (ACGME) established common residency program requirements that, among other provisions, restricted the number of hours the 100 000 residents in the United States can be on duty. In the 3 years that followed, the duty hour sections of the common requirements have received considerable attention from the media, the public, and the academic community. Comments have ranged from praise of the limits as critical to better patient care and learner safety and well-being,^{1 - 2} to predictions that they will spell doom for professionalism and continuity of care.³

The ACGME established duty hour limits with the goal of ensuring high-quality learning and safe, effective care in teaching settings.⁴ At the same time, the ACGME accreditation focus extends far beyond resident hours. Its review committees use information on the curriculum, the quality and engagement of faculty, institutional resources, supervision, educational outcomes, and the results of evaluations of residents, the faculty, and the program to determine accredited programs' substantial compliance and suitability to educate residents. Information about programs' performances includes input from more than 30 000 residents annually via a confidential Web-based survey (with response rates of about 90%) and confidential interviews of 10 000 to 12 000 residents during the more than 2000 accreditation visits ACGME conducts.⁵

The ACGME accredits 120 specialties and subspecialties, each with varying patterns of patient care and service demands as well as particular and time-tested learning environments. Every specialty has responded to the challenge of a single standard for duty hours by examining the resident hours that needed to be reduced to come into compliance, the extent to which activities important to attainment of competence could occur within a new model of learning, and the individual program's ability to replace the clinical contributions of residents after the institution of the limits.

However, duty hour reform is not an end unto itself. Three aims unite the ACGME and the GME community in the redesign of the learning environment: safe and effective patient care, high-quality resident learning, and resident safety and well-being.⁴ Changing only one variable (duty hours) in a complex system may, in fact, detract from achieving these aims. Comments from residents and others underscore the dual nature of the residency, highlighting that elements of the clinical environment and the learning model are linked, with both affected by the limits. A poignant example of this comes from a resident who completed the narrative section of the anonymous ACGME survey: “I’m sure we are in compliance with all of your requirements and yet both patient care and my education have gotten worse. Now I am here alone at night – there used to be two of us; one had to go home. I am looking after his patients as well as my own; I don't know his patients. The faculty is busy doing work I used to do.” (ACGME Resident Survey 2005-2006, unpublished data). The health care system has depended for decades on the vigilance of overworked residents; changing duty hours calls for a different model of patient care and a different model of resident learning. It is a large undertaking.

While there is a great deal of evidence that sleep deprivation compromises all 3 aims, there is little evidence that complying with duty hour requirements either increases sleep or improves the 3 aims. Most residents report that they use the extra time made available by duty hour reform to study or simply to have a more normal life; they do not necessarily spend the time sleeping.^{6 - 7} The continued heavy reliance on residents for clinical services, the added intensity of the fewer hours they now work, and the associated reduced availability of residents as caregivers require that other changes be made in both the patient care and educational systems. Changing duty hours, in the absence of other changes in the learning environment, may make patient care less safe.⁸ A systematic review examined interventions to reduce resident hours without other changes in the education and patient care system. It found that common approaches to reduce duty hours solely through scheduling changes such as night float and cross-coverage systems produced identical or worse outcomes of mortality, adverse events, and medication errors.⁹

Several residency programs and institutions have redesigned both patient care and resident learning. Examples include a neurosurgical program in Florida that has reduced call to every seventh night, using faculty and other types of health professionals to fill in the gaps.¹⁰ Another institution has redesigned its patient care support systems, resulting in a reduction of 12 hours of work time from every resident's weekly schedule.¹¹ A third is using “lean production” (improving system performance through the identification and elimination of waste and continual reduction of resources)¹² to move away from a traditional on-call system with its attendant fluctuations in admissions.¹³ Others are using the deep knowledge residents and others involved in direct day-to-day patient care have about how clinical work is performed to identify problems and to suggest interventions that enhance the safety and effectiveness of their clinical systems.^{14 - 15} These examples suggest that the GME community is trying to comply with duty hour regulations while improving patient care and education. Rules and strict enforcement of rules may be helpful in regulating hours, but consideration of both rules and context is needed to improve patient care and resident education. The focus must not solely be on duty hour reform, but also must be on designing a system in which both patient care and education can thrive.

In 2004, the ACGME reconstituted its Duty Hour Committee to a Committee on Innovation in the Learning Environment, to expand the focus from duty hours to the larger set of attributes important to safe patient care and effective learning. One year prior to instituting limits on duty hours, the ACGME implemented a requirement for the use of 6 general competencies considered important for physician practice (patient care, medical knowledge, practice-based learning and improvement, interpersonal and communication skills, professionalism, and systems-based practice).¹⁶ These competencies and their assessment make resident learning achievements explicit and begin to move education away from tradition-bound relationships between hours spent and presumed learning. The experience of a few truly innovative programs and institutions has shown that uncoupling learning from service demands can optimize the educational experience under constrained hours while offering superior care.¹⁷ For years residents have been socialized to cope; now, in novel programs, they are being socialized to master and lead safe systems of care.¹⁸

The public fully expects that every patient should have an awake, alert, and competent physician at all times. This underlies the focus on hours worked by resident physicians. Less well understood are the enormous gains in competence that occur during the 3 to 7 years of residency. Chief residents are extraordinarily competent. Yet, even with limits placed on their hours, residents early in their training are by definition not yet competent. Their relative inexperience makes them error-prone and potentially dangerous to patients and themselves if they are functioning alone and in the absence of supervision. Inexperience is especially dangerous when coupled with fatigue. High-quality care and good learning require immediately available competent supervision in the form of various levels of supervising residents and faculty. In a survey study, residents who experienced an adverse event were equally likely to implicate inadequate supervision (20%) and excessive duty hours (19%) as possible causes, followed by problems with patient handoffs (15%).¹⁹

Safe care and effective learning require consideration of this inexperience as well as work hours. Approaches that may help include simulation and other rehearsals of clinical skills to gain experience in a safe setting; closer coordination of care among nurses, physicians, and others to reduce variation and ensure that standards of care are always met; transparency of the system so that patients, nurses, and others know the level of training of each resident and how long these residents have been on duty; and an institutional culture that allows caregivers to say “I need help” without resultant stigma. Major cultural changes are under way as health care attempts to comply with duty hours and other expectations on the part of regulators and the public and with the expectations of more informed and engaged learners.

The academic community continues to contribute to the debate about resident duty hours and the broader elements of the learning environment to work toward ensuring high-quality learning for high-quality health care. Accreditation is not static; examples of programs and institutions that have successfully modified their patient care and learning environments inform the development of standards and enable the dissemination of new approaches to the problem of delivering safe and effective care in an environment that fosters good learning. Adapting to limited resident hours and the 6 general competencies as the organizing principles for resident learning will necessitate change, including a reconceptualization of professionalism for all health professionals, not just physicians. Professional standards suited to multidisciplinary teams, handoffs, and shift-based approaches to care are emerging and need further refinement; some of this is occurring already.²⁰ The ACGME, through its Committee on Innovation in the Learning Environment, is exploring the attributes of the ideal learning environment for residents. High-quality learning is impossible in the absence of high-quality patient care; likewise, high-quality patient care is impossible without high-quality learning. Attention to both is needed.