Delay in Initiating Antibiotics Results in Fatal Error
- Understand the importance of ongoing patient re-evaluation to confirm initial clinical impression.
- Define the attending role with respect to "remote supervision."
- Outline the role of the program director in processing medical errors.
- List the different forums for processing medical errors.
Case & Commentary
A 21-year-old woman with a history of systemic lupus erythematosus (SLE), on long-term prednisone, presented to the emergency department (ED) with a few hours of fever, chills, myalgias, and vomiting. On arrival to the ED, she was hypotensive, but responded to IV fluid resuscitation. Laboratory evaluation revealed an elevated white count. The medical housestaff evaluated her, contacted the admitting attending by phone, and admitted the patient to a medical ward with a presumptive diagnosis of viral syndrome versus food poisoning. She continued to require fluid resuscitation for blood pressure support. No antibiotics were given.
In the morning, 10 hours after admission, her condition began to deteriorate. She developed shock refractory to fluid resuscitation, and a subtle petechial rash (Figure) was noted. At that time, she was examined by the attending physician. Suspecting meningococcemia, the attending started antibiotic therapy and transferred the patient to the intensive care unit (ICU). Despite initiation of antibiotics and full supportive treatment, the patient had a cardiac arrest and died.
The error in this case reflects poor clinical judgment and a fund of knowledge deficit. The first premise in clinical care is to consider and treat the most life-threatening conditions, while waiting for patients' illnesses to declare themselves. Given her chronic prednisone use, this patient should have been recognized as an unstable, immunosuppressed patient. The differential diagnosis of hypotension in a patient on chronic prednisone must include early sepsis and adrenal insufficiency. Neither of these life-threatening conditions was apparently considered. Concerns about early sepsis should have resulted in the ordering of empiric broad-spectrum antibiotics and admission to an intermediate care unit. Consideration of adrenal insufficiency should have prompted the administration of intravenous hydrocortisone. In this instance, the hypotension was ascribed to volume depletion on the basis of a few hours of vomiting, an unlikely explanation. Additionally, the patient continued to require fluids for blood pressure support; the admitting team should have re-evaluated her for this ongoing hypotension. Continual patient re-evaluation is a critical skill, both to follow the progression of underlying illness and to ensure that the team is working with the correct diagnosis.
In this case, the attending physician was contacted by phone, although the nature of that contact is not specified. I refer to this type of supervision as "remote supervision." It applies anytime an attending is not physically present in a patient care unit to personally evaluate and manage patients. Remote supervision of residents is the most common mechanism of housestaff supervision, whether overnight or during the day. Oftentimes, faculty physicians are admitting patients (remotely) while simultaneously seeing outpatients. They make rounds late at night or early in the morning and thus rely on remote communication of clinical changes. Additionally, most institutions do not require overnight faculty presence. If institutions are going to care for patients this way, then standards must be set for quality of care. New admissions must be presented in their entirety to the attending. If the attending has any concerns regarding the clinical skills or decision making of the team, then he or she must evaluate the patient personally. This standard should be applied regardless of time of day.
Changes in inpatient medicine over the past 10 years challenge the concept of remote supervision. Given the managed care revolution, the need to manage patients effectively has become a fiscal imperative. Cost containment and demands for improved quality of care have led to the birth of a new specialist in medicine: the hospitalist. Defined as individuals who practice at least 25% time in the inpatient setting (1), hospitalists hold the advantage of having the inpatient ward as their practice venue. They are present in the hospital more often (around the clock in some institutions), enabling the timely evaluation of patients. Supervision is no longer remote. Although there are no data to confirm that fewer errors occur on hospitalist services, two studies at teaching hospitals showed that hospitalists led to reduced lengths of stay, cost of care, and mortality.(2,3) In an analogous way, the on-site presence of intensivists (who are, in essence, "ICU hospitalists") appears to improve outcomes (4), and has been promoted by the Leapfrog group as one of its quality standards.(5)
Although the data are limited, it makes intuitive sense that the more timely involvement of attending physicians such as hospitalists and intensivists would lead to less expensive and better quality care. Nevertheless, a recent study on the presence of in-house attending trauma surgeons showed no impact on mortality or length of stay.(6) Studying the impact of different organizational models of care is notoriously difficult, and institutions will need to decide on staffing and supervision models based on imperfect data.
The rapid growth of the hospitalist model—both within and outside academic hospitals—seems to indicate that leaders are convinced of its benefits.(7) The introduction of hospitalists into academic medical centers is likely influencing graduate medical education.(8) One great challenge is to balance resident autonomy with the appropriate level of supervision when hospitalists are integrated into training programs. At least one study supports that their presence does not compromise resident autonomy.(9)
The duty-hour regulations imposed by the ACGME in July 2003 are also likely to impact and change the level of attending involvement. These regulations require that, when averaged over 4 weeks, housestaff work no more than 80 hours per week and have 1 day in 7 off. They also require that every duty period be separated by 10 hours and that no shift exceed 24 continuous hours with an additional 6 hours for education and transfer of care.(10) To meet these requirements, many programs have implemented or expanded night-float programs.(11,12) The number of handoffs between providers has certainly increased. This discontinuity of care by housestaff places more reliance on the attending physicians for the details of patient care. It is likely that duty-hour reform will improve resident fatigue; however, it may compromise patient safety. New systems will need to be adopted to improve continuity. One such mechanism would be to have continuous presence of hospitalists throughout the day and night.
At least one study demonstrates that medical errors among internal medicine residents are not uncommon. One hundred fourteen internal medicine residents completed an anonymous questionnaire describing their most significant mistake and their response to it.(13) Mistakes included errors in diagnosis (33%), prescribing (29%), evaluation (21%), communication (5%), and procedural complications (11%). Serious adverse outcomes occurred in 90% of the cases, including death in 31% of cases. Most importantly in this study, only 54% of house officers discussed the mistake with their attending physicians, and only 24% told the patients or families. Those who accepted responsibility for the mistake and discussed it were more likely to report constructive changes in practice.(13)
Given that errors are not uncommon among residents and accountability is less than adequate, program directors and those responsible for medical education play a critical role in patient safety. The overall role of the program director is to help residents turn into independent practicing physicians. Errors related to fund of knowledge deficits, inadequate clinical skills, poor clinical judgment, and problem solving must be addressed from the perspective of the individual, the program, and the health care system. Errors never occur in a vacuum: usually, system-based issues are contributing factors.
Program directors have three roles when it comes to dealing with medical errors. The first relates to the providers: the responsible faculty must discuss the error with the house officers involved. Oftentimes, the program director can facilitate this. Many housestaff worry when their program director is notified of their mistakes, concerned that such information will harm their fellowship or job prospects, or that they'll be sued or suffer personal embarrassment. Regardless, the program director is the only individual who can determine whether an error is an isolated circumstance or represents a problematic pattern of performance. If such a pattern is present, then it needs to be carefully examined. The pattern may reflect basic fund of knowledge and clinical skills deficits that are easily remediable. It may also reflect underlying depression, attention deficit disorder, substance abuse, etc. All of these issues must be addressed for residents to successfully negotiate training.
The second role of the program director is in defining the educational curriculum. If the error is felt to be common within the program, then the program director should develop an educational initiative designed to prevent similar occurrences, such as a resident report, clinicopathological conference (CPC), or Morbidity and Mortality (M&M) conference devoted to discussing the case. The third role of the program director is to serve as a liaison between the program and the health care system. If significant systems issues are identified, then that information needs to be communicated to the appropriate individuals in the hospital administration. Too often, there is a disconnect between residency issues and the institutional quality apparatus. Residents, by operating at the sharp end of care, are often the ones best positioned to identify major systems flaws that require action.
The barriers mentioned above are very real and impact our ability to learn from our mistakes. At our institution, we have adopted several venues for error reporting that have helped the program directors carry out all three roles. We now have monthly patient safety discussions at residents' report. In these sessions, we discuss cases where errors or potential errors were thought to occur due to systems-related issues in the process of care. Importantly, the vice president for hospital quality and patient safety moderates these sessions. The format has enabled the identification of many systems issues that have subsequently been improved. While initially skeptical, the housestaff have embraced this format as a constructive way to have a voice in the larger process of care.
In addition to reporting within these and other conferences (eg, M&M conferences), we have also implemented an anonymous web-based reporting system called Penn Occurrences Reporting and Tracking System (PORTS). This system allows any provider to submit an online report of any situation that created a near miss for an adverse event or actually caused an adverse event. This information is collected and collated centrally by the institution-based Clinical Effectiveness and Quality Committee. Their role is to identify not only individual events but also look for patterns of events that can lead to systems-based improvements. Residents in our program have embraced this anonymous reporting tool as a way to improve the system of care at our hospital.
If error reporting is done in a non-biased, non-confrontational format with an opportunity for learning, it can lead to substantial improvements in both education and patient care. As physicians, we have a responsibility to ourselves and our patients to develop systems to minimize errors. As educators, we have an obligation to help trainees understand the importance of their roles as both providers of care for individual patients and as leaders in improving the systems of care in which they work. Recognizing the complex factors that contribute to these errors is necessary to prevent future occurrences.
Lisa M. Bellini, MD Associate Professor of Medicine Vice Chair for Education and Inpatient Services Department of Medicine University of Pennsylvania Medical Center
Faculty Disclosure: Dr. Bellini has declared that neither she, nor any immediate member of her family, has a financial arrangement or other relationship with the manufacturers of any commercial products discussed in this continuing medical education activity. In addition, her commentary does not include information regarding investigational or off-label use of pharmaceutical products or medical devices.
1. Wachter RM. An introduction to the hospitalist model. Ann Intern Med. 1999;130:338-42.[ go to PubMed ]
2. Auerbach AD, Wachter RM, Katz P, et al. Implementation of a voluntary hospitalist service at a community teaching hospital: improved clinical efficiency and patient outcomes. Ann Intern Med. 2002;137:859-65.[ go to PubMed ]
3. Meltzer D, Manning WG, Morrison J, et al. Effects of physician experience on costs and outcomes on an academic general medicine service: results of a trial of hospitalists. Ann Intern Med. 2002;137:866-74.[ go to PubMed ]
4. Pronovost PJ, Angus DC, Dorman T, et al. Physician staffing patterns and clinical outcomes in critically ill patients: a systematic review. JAMA. 2002;288:2151-62.[ go to PubMed ]
5. Fact sheet. The Leapfrog Group. October 2003. Accessed January 13, 2004.
6. Arbabi S, Jurkovich GJ, Rivara FP, et al. Patient outcomes in academic medical centers: influence of fellowship programs and in house on call attending surgeon. Arch Surg. 2003;138:47-51.[ go to PubMed ]
7. Wachter RM, Goldman L. The hospitalist movement 5 years later. JAMA. 2002;287:487-94.[ go to PubMed ]
8. Shea JA, Wasfi YS, Kovath KJ, et al. The presence of hospitalist in medical education. Acad Med. 2000;75:S34-6.[ go to PubMed ]
9. Chung P, Morrison J, Jin L, et al. Resident satisfaction on an academic hospitalist service: time to teach. Am J Med. 2002;112:597-601.[ go to PubMed ]
10. Resident duty hours language: final requirements. Accreditation Council of Graduate Medical Education Web site. February 13, 2003. Accessed January 22, 2004.
11. Morelock JA, Stern DT; Association of Professors of Medicine. Shifting patients: how residency programs respond to residency review committee requirements. Am J Med. 2003;115:163-9.[ go to PubMed ]
12. Resident work hours benchmarking project summary. University Health System Consortium Web site. June 2003. Accessed February 9, 2004.
13. Wu AW, Folkman S, McPhee SJ, et sl Do house officers learn from their mistakes? JAMA. 1991;265:2089-94.[ go to PubMed ]
Figure. Petechial Rash