Life is short, art is long, opportunity fleeting, experience deceiving, and judgment difficult. - —HIPPOCRATES
Increasing attention is being paid to the occurrence of medical error in the lay and medical literature. Efforts to limit mistakes through ongoing refinements in care continue. However, internal medicine does not appear to have established models for discussion of medical error within internal medicine practice or training. Questions remain on how to design educational experiences that are effective in reaching this goal. This article focuses on the morbidity and mortality conference as a potential forum for discussing and learning from medical error. While our comments are focused on internal medicine, we hope they will be of use to those in other specialties as well.
Despite being universally familiar to physicians, the morbidity and mortality conference (M&MC) lacks a precise definition, a standard format, and identified goals. In a recent survey, we found it to be widespread in internal medicine training programs, but heterogeneous in format (see below). Its historical roots (discussed in the sidebar) appear to spring from the efforts of individuals who were attempting to improve medical practice through examination of adverse outcomes and errors. The modern M&MC has had limited examination, but may be straying from the precepts from which it evolved. The literature of internal medicine contains predominantly indirect discussions of the M&MC, yet the occurrence of errors, and the need to evaluate and learn from them, remain. This paper reviews what we believe to be the origins and modern applications of M&MCs and discusses relevant literature on medical error. From this review, we attempt to identify goals for responding to medical error that we believe can be incorporated into the M&MC, but may be lacking in many educational and practice settings.
MODERN MORBIDITY AND MORTALITY CONFERENCE
Limited Modern Descriptions
The limited modern descriptions of the M&MC come from the disciplines of surgery and anesthesiology. Bosk examined a surgical M&MC as part of a sociologic study of an academic surgical service.1 He saw it as a setting in which the attending surgeon publicly “puts on the hair shirt” and models the ethical standards expected of peers and subordinates. He identified a number of elements that serve to inculcate social norms: a hierarchy of authority, open discourse, face-to-face interaction, and a public forum for the analysis of error and allocation of blame.1,p.192
Other authors criticize the M&MC as a crude means of identifying and evaluating adverse patient care events in order to improve patient care.2,3 Many cases identified by occurrence screening and other risk-management mechanisms are not submitted for evaluation to the M&MC, and surgeons are often not present when the cases' complications are discussed.
Recent commentators have highlighted deficiencies in the modern M&MC.4 Biddle reviewed 105 cases presented at the weekly anesthesia M&MC at his institution and found that 72% of the cases presented did not involve morbidity or mortality.5 Cases were selected because of their educational interest, because of an unexpected event, or because of a management error with or without complication. Since the majority of cases were chosen for presentation because of their teaching value he suggests that the M&MC would be more appropriately named a “case conference.”
Campbell recently reviewed his own experience and polled surgeons in Great Britain about the format and conduct of the surgical M&MC.6 He advocates regularly scheduled meetings during hours convenient for clinicians. Treating surgeons should be present and prepared to discuss their patients and complications. A moderator should select cases, guide discussion, and intervene if discussion becomes acrimonious or unreasonable. Campbell emphasizes that surgical departments should conduct concurrent review of adverse events, and argues that all problems should be presented for potential discussion and review at the M&MC. His thoughtful recommendations are the first modern reformulation of principles for this conference, and could be adapted for an internal medicine M&MC designed to address medical error.
Harbison and Glenn recently surveyed staff surgeons and surgical housestaff nationally to determine their perceptions of the M&MC.7 Both groups had favorable views of the conference, with staff responses slightly more positive than those of trainees. Both groups perceived education to be the primary focus. Residents believed that the M&MC could be best improved by decreasing defensiveness and blame.
The M&MC has received much less attention in the literature of internal medicine. No reference explicitly describes or defines a medical M&MC, though there are scattered comments about the negative aspects of the conference and its adverse impact on attendees.8,9,10,11 Bosk compares the medical version of the M&MC to the surgical M&MC unfavorably:
Only one or two cases of interest are reported on. They are chosen for their heuristic value. They are intensively researched. Their presentation is separated from their occurrence by six to eight weeks. There is no attempt at this conference or elsewhere to make public and demand accounts for each death and complication that occurs as there is in the surgical conference.1,p.220
In 1998 we conducted a survey of all internal medicine training programs listed in the directory of the 1997 Association of Program Directors of Internal Medicine (APDIM). The findings give a more comprehensive view of the M&MC as it currently exists in departments of medicine. Two hundred and ninety-five questionnaires were returned. Ninety percent of the respondents reported having a conference named “morbidity and mortality” or its equivalent. Conferences were usually held monthly, and most had mandatory attendance for housestaff and students. Nearly all sites assigned a leader or moderator, a key faculty member or chief resident. Cases were most often selected because of unexpected adverse events or suspected error, with most of the remainder being selected for teaching value. At two thirds of the sites the conference served the purpose of meeting administrative requirements such as quality assurance. Over 80% of these programs reported that when present, medical error was discussed with moderate to high success. However, only half had an established method or procedure for handling the discussion of errors. The amounts of time dedicated to open discussion were highly variable, with one fifth allotting less than ten minutes. Overall, respondents rated their M&MCs more highly than they rated the other conferences in their departments.
This brief review of the modern M&MC reveals a prevalent conference with a role in internal medicine training programs that is not clearly defined. The conference lacks explicit goals, methods, or formats, and it is not described in the internal medicine literature as having an important educational or institutional role in addressing medical error. Cases are chosen such that error is likely to be found, but there is no general consensus about how the conference should be structured in order to ensure that discussion of mistakes is fruitful.
CONFRONTING ERRORS IN MEDICINE
Our survey found that medical error is identified and addressed during M&MCs. The literature on medical error offers a context for considering how this can be done effectively.
The expected norms, work routines, and rituals in our profession offer a paradox: physicians strive for—and patients expect—perfection, yet this standard cannot be consistently achieved. There is a “necessary fallibility” inherent in clinical practice that can never be absolutely eliminated.10,12,13 Reconciling educational goals with the confrontation of mistakes presents a particular challenge: education is more effective when enjoyable, not painful, and adult learning occurs best through collaboration, not prescription.14
How well do modern medical systems confront medical mistakes? Not at all well, according to the scant literature on the subject. Barriers to public discussion include the anxiety inherent in exposing individual fault, potential loss of respect, and fears of legal action. In a study of errors made by internal medicine house officers, while mistakes were sometimes addressed in rounds and conferences, in about half of these meetings “the tough issues were not addressed.”15
The “tough issues” are often avoided because it is inherently difficult to face mistakes; a misguided approach to this problem can undermine the M&MC and cause it to miscarry. The nature of learning from error is a delicate issue. For example, the moderator may try to avoid the discomfort of confronting mistakes by choosing cases that raise controversies in management rather than clear errors. This improves the educational environment of the M&MC but undercuts its effectiveness in establishing and reinforcing norms and standards. In contrast, if the style of the conference is unduly blunt and criticism is directed towards individual persons rather than towards procedures and patterns of decision making, then participants may fear public humiliation and the result may be aversive conditioning rather than the forthright evaluation of poor medical practices and outcomes. In fact, these latter reactions have been observed to occur. Although mistakes may be discussed in a conference setting, the current process and the culture in which this occurs are reported to contribute to feelings of worthlessness and humiliation, reinforcing avoidance behaviors, denial, and cover-ups by the physicians involved.16,17,18,19
Professionals improve their skill through reflection on their actions and the subsequent outcomes.20 An M&MC assures that such reflection occurs in critical cases. The danger and the irony in this is that an ineffectual conference not only fails to achieve its goals, but can convey the message that facing medical error either cannot be done effectively or is unpleasant and to be avoided. Thus, the M&MC is a double-edged sword. Especially careful attention should be given to making it successful.
A MODEL FOR THE M&MC
We believe that the M&MC can provide a unique and important adjunct to the training and continued education of all internists. The overarching goals of such a conference are to make better doctors and a stronger medical profession. Hence, we stress the introspective nature of our conception of the M&MC: a conference in which we review ourselves without outside influences. We contrast this with that of external, more public, and institutional review that is inherent in modern quality assurance.21,22,23,24
Our suggestions for the M&MC do not flow from experience with a particular model that has enjoyed success. Rather, they have evolved from our experience with, and reflection on, pseudo-M&MCs, such as case conferences, or CPCs.
Definition of the M&MC. Though the M&MC conference may take a number of different forms, one can consider its defining characteristic to be that it is designed to identify medical errors in order to learn from them to improve medical practice. As such, it is an institutional expression of our responsibility to face and profit from our mistakes, both as individuals and as a profession. An effective M&MC should
* identify events resulting in adverse patient outcomes,
* foster discussion of adverse events,
* identify and disseminate information and insights about patient care that are drawn from experience,
* reinforce accountability for providing high-quality care, and
* create a forum in which physicians acknowledge and address reasons for mistakes.
Guiding principles. The conference should facilitate the open discussion of medical error. The stance of the conference toward medical error should be explicit. An example might be:
* Medicine is difficult.
* Errors are inevitable, but they give us a tool to improve our skill as physicians.
* The goal of the conference is not to criticize, but to profit by sharing and examining our experience.
The type of error reviewed should be one from which others can learn; cases should not be chosen to demonstrate gross mismanagement. The M&MC should not be viewed as the same as or a substitute for quality improvement activities for accreditation.
Case selection. All serious adverse patient outcomes should be identified on a continuous basis, and cases should be chosen from among them for discussion at the M&MC. Attendees should have an understanding of how cases are screened and chosen for review and discussion. Cases should not be stale. Choosing recent cases emphasizes the importance of facing medical error directly and in a timely manner. Cases should be discussed in advance with the treating physicians so that the details of the error can be clarified.
Moderator. The conference should be managed by a senior physician who is skilled at creating a supportive atmosphere. The moderator should call on one or more prepared discussants, whose comments can model a middle ground between minimization and magnification. The moderator may call upon additional members of the audience, but should be prepared to rephrase/moderate comments that are not supportive or sympathetic in tone. The moderator and/or senior members of the staff should selectively recount similar or relevant errors that they have made and the lessons/benefits that they have drawn from reflecting upon them.
Attendance. Wide representation should be sought from all levels in the institutional hierarchy, from chief of service to student or intern, and from all relevant departments. Physicians and students involved in the care of the patients presented should be present at the conference. The treating physician(s) should be given the opportunity (but not be required) to present the case, the circumstances leading to the outcome, and the lessons they have drawn. This puts the locus of control with the treating physicians more than with the moderator. The conference should be conducted with a tone that would be appropriate for the treating physician(s) to hear, whether or not they are actually present.
Conclusions drawn. The moderator should summarize conclusions to be drawn from the discussion, and implications for changes in practice and behavior. If systems problems or generic mistakes are identified, discussion should ensue about how to prevent future similar adverse events, and subsequent changes should be reported at later meetings of the M&MC.
The M&MC brings doctors together to examine cases that have gone badly in an effort to increase their skill. The conference is also implicit acknowledgment of the importance of confronting error, and it inevitably models the ways in which this can be done. These two objectives, learning from mistakes and confronting error, are central to the medical profession and form the basis of our belief that the M&MC has importance in medical training.
The modern M&MC has an uncertain role in the training and continuing medical education of internists. This uncertainty stems from an absence of well-articulated goals and methods for such a conference. The set of objectives we offer above may be helpful in creating an M&MC that can address medical mistakes effectively and productively. The goals we describe are intended to keep the focus of the conference on error, not allowing it to stray into more comfortable areas, while remaining consistent with theories of education referable to adults and professionals. The methods we propose are chosen to minimize the threatening nature of public discussion of error and to avoid harming participants through unintentional public humiliation.
The M&MC should not be viewed as the same as, or a substitute for, quality improvement activities required for accreditation. Such dual purposes require concessions (such as publicly-available documentation) that may unintentionally lead participants toward maladaptive defense mechanisms—such as avoidance, minimization, or magnification of errors—and negatively affect future professional development.
These ideas are offered not as an attempt at a blueprint, but rather to stimulate a debate through which we can identify a working model, and refocus a tradition of self-analysis and critical thinking. There will always be difficulties associated with such an approach, given the delicate nature of learning from error, but confronting such risks should benefit both physicians and patients.
Origins of the Modern Morbidity and Mortality ConferenceThe first antecedents of the morbidity and mortality conference(M&MC) are hard to trace. The rise of the modern hospitalas a center of clinical practice and education stimulatedthe development of systems to review, standardize, and improvemedical procedures. Hospital committees were establishedto examine adverse outcomes. These committees appearto have been the most immediate precursors of the M&MC.The following discussion traces several disparate influences ontoday's M&MC. McIntyre and Popper's survey of the evolutionof the medical audit provides further detail of relatedthemes for the interested reader.25
The End Result System
Ernest Amory Codman, an early 20th-century surgeon and outspokenreformer, championed the evaluation of clinical practiceby introducing the end result system. Dr. Codman applied themethods of the natural scientist and the industrial manager tosurgical practice at the Massachusetts General Hospital. He developedthe concept of the end result card, which documentedeach patient's symptoms, clinical diagnosis, treatment plan,complications, final diagnosis, and annually updated outcome.Whenever perfection was not achieved, an analysis of the causewas detailed on the card, based on a taxonomy of errors hedeveloped.26 Codman called for open acknowledgment of these “end results” to physicians and to the public, and challengedhospitals to hire clerks to record these data, to publish results,and to establish efficiency committees empowered to correcterrors identified by this process.
Codman's ideas elicited intense opposition from physicians,who resisted the loss of autonomy inherent in his model. However,his ideas influenced the standardization of hospital practicesby the American College of Surgeons in 1916, and haverelevance even today for both the M&MC and quality assurance practices.
The Anesthesia Study Commission
Codman's work reflected a desire to improve medical practiceby the examination of experience. In 1935, the PhiladelphiaCounty Medical Society continued in this tradition by forminga group initially called the Anesthesia Mortality Committee,which was an early precursor of the M&MC. Its objective wasto facilitate discussion and to share knowledge about fatalitiessecondary to anesthesia, and “other interesting anesthetic situations.” 27 This multi-institutional review group attemptedto disseminate new information and to improve communitystandards of care.
In 1940, the name of the organization was changed to theAnesthesia Study Commission and its mandate was expandedto review a variety of topics, not limited to fatalities. It metmonthly and was composed of anesthesiologists, surgeons, andinternists representing a variety of institutions. Its meetingswere open to “all physicians, residents and interns … as wellas numerous nonresident visitors.” The commission generatedperiodic public reports of its activities. In his 1945 review ofthe commission's history and data, Ruth reports that at least two thirds of fatalities reviewed were classified as preventable, and that the commission's conclusions often differed from the causes of death included on death reports.27
The goals and format of the Anesthesia Mortality Committeehave been highly influential and widely replicated. Itis instructive to review those elements of this antecedent ofthe M&MC that remain relevant to our current conferences.The dual nature of the meeting as a forum for education and system improvement was explicit: the committee was founded to improve anesthesia practice by an open review of cases likely to reflect medical error. Cases were selected in such away that error was likely to be found. Initially, only those casesin which mortality was immediately related to anesthesia were chosen, but eventually a variety of cases of morbidity and mortality related to surgery and anesthesia were selected for review. Meetings were held monthly, and error was confronted directly albeit anonymously. The style was participatory; discussion proceeded with comments from the floor. Physicians caring forthe patient were not identified by name, but Ruth reports that they not infrequently chose to identify themselves during the discussion of the case.
Ruth's work stressed the instructional value of open discourse of problematic cases while protecting participants through optional participation. Nevertheless, Ruth describes tension within the Mortality Committee between its educational goals and a fear of incrimination on the part of participants. Reporting of fatalities was voluntary, the treating physicians did not always participate, and weaker departments of anesthesia did not consistently file reports. The more florid orobvious errors were not considered. These factors tended to focus the debate on the more minor errors of management.
The historical tensions inherent in the roots of the M&MCremain relevant today. Consideration of these issues should be incorporated into the design of a modern formulation of the conference if one's goals include improvement through the discussion of medical error.