Tejal K. Gandhi, MD, MPH
Acknowledgment: The author thanks Dr. Eric Poon, Erin Graydon-Baker, Saila Basavaraju, and Zahra Ladak for their contributions to this manuscript.
Grant Support: Funding for the Quality Grand Rounds series is supported by the California HealthCare Foundation as part of its Quality Initiative. The authors are supported by general institutional funds.
Potential Financial Conflicts of Interest: None disclosed.
Requests for Single Reprints: Tejal Gandhi, MD, MPH, Brigham and Women's Hospital, 1620 Tremont Street, 3rd Floor, Boston, MA 02120; e-mail, firstname.lastname@example.org.
Missed follow-up of abnormal test results and resultant delays in diagnosis is a safety issue that is gaining increasing attention. Despite increases in the numbers and types of available diagnostic tests, current systems in health care do not reliably ensure that test results are received and acted upon by ordering physicians. This article examines the case of a patient whose diagnosis of tuberculosis was substantially delayed because of systems problems, including poor continuity (with multiple-provider involvement), lack of communication of test results and other clinical information, and several handoffs. Strategies to ensure adequate communication of critical information and follow-up of test results are discussed, such as explicit criteria for communication of abnormal results, test-tracking systems for ordering providers, and use of information technologies.
Timeline of events.
Communicating critical test results: Creating a fail-safe process.
Table. Breakdowns and Possible Solutions
Screenshot of results manager.
Eran Y Bellin
Montefiore Medical Center Bronx, N.Y.
March 3, 2005
Rare Human Error or Culture of Clinical Sloth?
To the editor: In quality improvement work system failures amenable to system correctives with information technology, new policy and procedures, engineering for redundancy is au courant and appropriate. However, it is essential that the contributing human failures must be thoroughly studied to see if they are merely chance unpredictable human error or more ominously evidence of a "culture of clinical sloth". Let us review the case presented by Gandhi(1) from a human error perspective. An elderly alcoholic presented with "Right upper lobe pneumonia" on chest x- ray and the treating physician did not order sputum induction for acid fast bacillus. Instead, he followed a recommendation by a radiologist to do another radiologic test "“ chest CT. Why, no immediately relevant diagnostic test? Were there no facilities for collecting the obviously necessary sputum specimen for acid fast bacillus?
The patient was treated with an unnamed antibiotic. Why does the case report not tell us what we suspect, that he was treated with a quinalone, usually a reasonable enough drug, but in this case would obscure the diagnosis of TB as TB is at least initially susceptible to monotherapy with a quinalone. The subsequent failure to review the CT compounds the error but is really beside the point.
In the hospital, the housestaff did not review the computer record for old chest x-rays or CT in a physically wasted elderly alcoholic. Does the writer believe that this is the only time such an error occurred? Or, is this common practice not to review the easily available old data? Who is supervising the housestaff? Twenty years ago, housestaff reviewed reams of charts. Now they have to click a mouse button to get relevant information. How often is this not done and why?
The x-ray was reported as normal by the housestaff. It is hard to believe from the case that it could have been reported as normal if actually reviewed. But, let us give the housestaff the benefit of the doubt. Is there a formal expectation that chest x-ray's final reports will be reviewed by the housestaff? Or, if change of service occurs before the final reading is available, that review of the chest x-ray will be listed on the to-do list at change of service? This is a cultural not technical failure.
We are collectively in the debt of the author for his courageous sharing of what is appalling medical care, but I fear not unique to his institution.
A culture of clinical sloth, shirking of personal responsibility, and overreliance on the magic of technology undermines quality clinical care. Analyses should highlight system failures but must independently assess the human ones as well for they require a different corrective.
(1)Gandhi TK. Fumbled Handoffs:One Dropped Ball after another. Ann Intern Med 2005:142;352-358
Darius A. Rastegar
Johns Hopkins Bayview Medical Center
March 4, 2005
Lost in transfer
Dr. Ghandi documents an example of what appears to be a growing problem: "fumbled handoffs" as patients navigate our increasingly fragmented health care system. The author rightfully notes that "diffused responsibility" is partly to blame and offers some suggestions for remedying this. However, I would submit that there is a simpler and more straightforward solution that was not proposed: to have one person who coordinates care over time and who follows the patient in the varied settings where care is provided. This person could see the "big picture" and would be ultimately responsible for ensuring that the patient receives adequate care. Of course, we have such individuals already: they are called "primary care physicians"; however, their role is rapidly diminishing and they are in danger of becoming just another worker on our health care assembly line. There are compelling forces behind specialization, but we need to find ways to preserve the continuity and accountability of the "old-fashioned" primary care doctor in our system.
I am a primary care physician.
Kenneth M. Kessler
University of Miami School of Medicine
March 14, 2005
Fumbled handoffs revisited
To the Editor: After reading Dr. Gandhi's discussion (1) of "fumbled handoffs," I was concerned that I had a very different perspective of the case. I asked my wife, a pediatric radiologist, her thoughts about an elderly male with a history of alcohol abuse who presents with weight loss and cough. She immediately replied that her major concern was tuberculosis. It was reassuring that she concurred with my initial thoughts but a bit troubling that someone remote from clinical adult medicine for over two decades could "nail" the diagnosis when the clinicians caring for the patient did not and that the author did not identify an error in initial diagnosis as the key mistake in this case. To consider that the diagnosis of tuberculosis in this patient was dependent on the first radiologist suggesting a CT scan and the second radiologist suggesting the possibility of tuberculosis is an example of the muddy thinking that has led to the explosive overuse of medical imaging. Redundancies are already built into the system including mandatory old chart review, old radiograph review, review of radiographs with the attending physician and radiologist. Furthermore, the public health implications of a delay in diagnosis of an active case of tuberculosis spending weeks in the hospital and nursing home are significant. Clearly, the issue of "fumbled handoffs" would not be a consideration if the diagnosis of tuberculosis was considered from the outset. While better test reporting mechanisms are clearly needed, blame placed on an inadequate system begs the issue of individual responsibility and accountability. Unlike the author, I believe that more vigilance is the answer because doing it right the first time must remain the key or sloppiness is rewarded and perpetuated.
References: 1. Gandhi TK. Fumbled handoffs: one dropped ball after the other. Ann Intern Med. 2005;142:352-358.
Tejal K Gandhi
Brigham and Women's Hospital
April 29, 2005
Like Dr. Rastegar, I too am a primary care physician (PCP) and agree that this role is essential in ensuring continuity and quality of care. However, even if a primary care physician was the only provider involved, there is potential that the test result may not have been seen, especially since many PCPs are dissatisfied with their systems for tracking test results (1). If the patient had been taken care of by his own PCP in the hospital, perhaps his PCP would have remembered the prior outpatient evaluation. However, having systems to ensure that the outside records are available is still be essential, as it is difficult for any PCP to remember the details of every patient seen. Furthermore, we have to acknowledge that many PCPs now use the hospitalist model, due to time constraints, competing demands, and evidence for improved quality of care (2). Therefore, we need strategies to improve the hospitalist model and the resulting handoffs in care (3).
In response to Dr. Kessler's comments, I agree tuberculosis (as well as lung cancer) should have been on the differential when the patient first presented. This is why the chest CT scan was ordered- to further clarify the diagnosis. The misdiagnosis occurred because the ordering provider never saw the result of the test, and the patient subsequently presented to the hospital without any pulmonary complaints. Errors in diagnosis are one of the most common types of errors in the ambulatory setting (4), and clearly work needs to be done to better understand how these errors can be prevented. However, I take exception to Dr. Kessler's argument that we can blame the patient's outcome on sloppiness and that increased vigilance is the answer. Patient safety and human factors literature clearly state that even the best-trained individuals will make potentially serious errors, and vigilance is a very weak error prevention strategy (5). Physicians practice in a healthcare system where test results are not easily tracked, patients are sometimes poor historians, multiple handoffs exist, and information gaps are the norm. Most human errors are induced by these kinds of systems failures. Therefore, we need to re-design systems to ensure that physicians' clinical decision making and workflow are facilitated to make it easier to achieve the highest quality of care, and that errors, which are guaranteed to happen, are caught and mitigated.
1. Poon EG, Gandhi TK, Sequist TD, Karson AS, Murff HJ, Weber D et al. 'I wish I had seen this result earlier': dissatisfaction with test result management systems in primary care. Arch Intern Med 2004; 164(20):2223-2228. 2. Auerbach AD. Wachter RM. Katz P. Showstack J. Baron RB. Goldman L. Implementation of a voluntary hospitalist service at a community teaching hospital: improved clinical efficiency and patient outcomes. Ann Intern Med 2002;137(11): 859-65.
3. Goldman L. Pantilat SZ. Whitcomb WF. Passing the clinical baton: 6 principles to guide the hospitalist. American Journal of Medicine 2001;111(9B):36S-39S.
4. Phillips RL, Bartholomew LA, Dovey SM, Fryer GE, Miysohi TJ, Green LA. Learning from malpractice claims about negligent, adverse events in primary care in the United States. Qual Saf Health Care 2004; 13:121-126.
5. Institute of Medicine. To err is human. Building a safer health system. Washington, D.C.: National Academy Press, 1999.
Gandhi TK. Fumbled Handoffs: One Dropped Ball after Another. Ann Intern Med. 2005;142:352–358. doi: https://doi.org/10.7326/0003-4819-142-5-200503010-00010
Download citation file:
Published: Ann Intern Med. 2005;142(5):352-358.
Hospital Medicine, Infectious Disease, Mycobacterial Infections, Pulmonary/Critical Care.
Results provided by:
Copyright © 2020 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use