|
|||||
|
|
||||||
Journal of Oncology Practice, Vol 3, No 2 (March), 2007: pp. 66-70 © 2007 American Society of Clinical Oncology. DOI: 10.1200/JOP.0723501
Medical Errors: Focusing More on What and Why, Less on WhoRemember, there is nothing you can do to change [the past], but you can use its lessons to improve your future. Disclosure of medical errors and improvement in patient safety are inexorably linked, and provide one of the strongest reasons to report and disclose errors, including near misses in which no harm comes to the patient. As Rosner et al1 notes, "The paradox of modern quality improvement is that only by admitting and forgiving error can its rate be reduced." Error reports can be valuable learning tools as wellboth for those immediately involved and for the health care community at large. Nevertheless, US health care has not had a good record when it comes to reporting medical errors, even significant ones. For example, 20 states have mandatory reporting systems, but only six have received more than 100 reports in 1999. Yet the Institute of Medicine (IOM; Washington, DC) estimates that more than 1 million preventable adverse events occur each year in the United States, with up to 98,000 being fatal, a figure equivalent to one major airliner crash daily.2,3
Barriers to error reporting are found at many levels in the health care system. Moskop et al4 assert that US health systems are not generally designed to encourage error recognition, reporting, and remediation. Teaching hospitals have focused on the sequelae of errors rather than teaching ways to prevent them or the value of disclosing them. Physicians' training and attitudes place additional barriers to reporting errors. As the gatekeeper for a patient's care, the physician who commits an error, especially one that harms the patient, may feel deep shame, guilt, and a sense of failure. He or she may believe that disclosing the error to the patient will do irreparable damage to the physician-patient relationship and to the patient's trust in the health care system in general. Furthermore, physicians have historically received little or no training in how to communicate with patients and others about errors. Reporting systems have been relatively cumbersome. The process of completing detailed forms, submitting them up the chain of command, and attending meetings and interviews has deterred many health care professionals from reporting all but the most egregious errors. Both institutions and individual practitioners view the threat of malpractice liability as a significant barrier to error reporting. The disclosure could have a double impact, leading the patient to file suit once informed of the error and providing admissible evidence to the plaintiff's attorney. Amalberti et al have identified more fundamental barriers that health care must overcome before it can be an ultrasafe industry, such as civil aviation or nuclear power. To make substantial safety improvements, one suggestion is that the health care industry impose five types of constraints on its activities: limit maximum performance (ie, limit the level of risk in any given activity, such as high-risk surgical procedures), abandon professional autonomy, abandon the status of craftsman for that of an "equivalent actor" (as anesthesiologists have done), adopt system-level arbitration for errors, and simplify professional rules and regulations.5
Patient safety advocates have been working for more than a decade to identify best practiceswithin the health care industry as well as other industriesthat will encourage error reporting and, ultimately, make health care safer for patients. Experts have examined the Aviation Safety Reporting System, a voluntary effort that has helped decrease the risk of dying in a domestic jet flight by three-fold during the past 30 years. This confidential system focuses on reporting of near missesinstances that violated an established rule or safe practice, but didn't result in an accident. Each report of a near miss is analyzed for root causes, and the results and any recommendations are shared not only with those involved, but also with the entire aviation community through the Federal Aviation Administration.6 Charles Billings, the system's designer, suggests that the system has been successful because it doesn't punish pilots who report promptly, has a one-page report form that makes reporting simple, and gives expert feedback, which helps pilots view reporting as worthwhile.7 Similar systems have been put into place in other high-risk industries, such as nuclear power plant operation, petrochemical processing, and military operations. These systems focus on near missesthe designers have recognized that root causes and patterns among multiple causes are similar for both near misses and adverse events. The major difference in outcome results from the ability (or lack thereof) to recover from the error.
Analysis of reporting systems inside and outside of health care has identified seven characteristics shared by successful systems8:
Valley Hospital, a 451-bed acute care facility in Ridgewood, New Jersey, is one example. According to Michael Mutter, MS, RPh, director of patient safety, in the hospital's drive to reduce medication errors, "encouraging the reporting of Level 1 errors was a critical preliminary step in gaining the trust of caregivers ... reporting Level 1 errors is a heroic deed. When you prevent something from happening, you experience a positive emotion." This is quite different from the guilt or shame someone feels over committing an error that has caused harm to a patient. Valley Hospital has found that staff members become caught up in the positive emotions and are eager to report near misses. Near-miss reports at Valley Hospital rose from one in 1996, when the initiative began, to 158 in 2002. Today, nearly 80% of all error reports are for near misses. Mutter believes this is not only because people are eager to feel heroic, but also because they recognize that reporting an error is a valuable learning experience. "The value of learning is equal whether a near miss or an error that reaches the patient has occurred," comments Mutter, which is another way near-miss reporting helps engender trust with staff. Valley Hospital's system is now used for all error reporting, from the laboratory to patient falls. The hospital has created a simple one-page reporting form that can be accessed and filed online. Error reports do not go into caregiver files, but they are reviewed by the Patient Safety Committee, which analyzes process failures and recommends changes. Technology has played a significant role in many hospitals' efforts to establish confidential reporting in a nonpunitive environment. For example, Dartmouth-Hitchcock Medical Center (Lebanon, New Hampshire) launched a Web-based reporting system first in the acute settings and then in ambulatory settings. The form includes optional identification of the reporter and 15 standard categories of incidences, such as an incomplete discharge form, an allergic reaction to food, and a medication omission. When the report is submitted, a designated department staff member and the occurrence manager responsible for the occurrence category are simultaneously and immediately notified via electronic mail. This allows timely response and follow-up. Furthermore, aggregate data have helped identify trends, which have led to new procedures, medication labeling revisions, and other system-related issues.
Perhaps no effort has embraced technology and incorporated all seven characteristics of successful programs as completely as has that developed by researchers at Johns Hopkins University School of Medicine (Baltimore, Maryland). The Intensive Care Unit Safety Reporting System (ICUSRS) initially involved about 30 intensive care units (ICUs) nationwide, including those at Johns Hopkins. Through ICUSRS, any staff member can submit an error report from any personal computer with Internet access through a password-protected secure site, using a confidential hospital code. Reporters do not have to identify themselves. The error report goes to a central database, where it is analyzed by an interdisciplinary team of experts. The team sends a monthly report that includes quantitative data specific to each site and combined data for all participants, as well as a quarterly newsletter with safety tips and profiles of ICU front-line staff working on safety efforts. As is the case with many institutions that have developed error-reporting projects, the ICUSRS at Johns Hopkins is one component of a larger effort to eliminate errors and improve patient safety. At Hopkins, a medication reconciliation process ensures that doses go through multiple checks from physician to pharmacist to nurse. Care of hospitalized patients is tracked using daily goal forms, completed by both physicians and nurses. Peter Provonost, MD, medical director of the Johns Hopkins Center for Innovation in Quality Patient Care (Baltimore, Maryland), was one of the designers of the ICUSRS and has helped disseminate its success to others. Working with the Michigan Health & Hospital Association (Lansing, Michigan), for example, Provonost applied the Hopkins model to develop the Keystone Center for Patient Safety & Quality (Lansing, Michigan) in 2003. This voluntary program involves more than 125 ICUs in Michigan and five in other states. Participants agree to assemble a team, including senior management, ICU management and staff, physician, and pharmacist, and to implement a series of interventions in line with available resources. The interventions include the ICUSRS, along with efforts to improve communications, reduce catheter-related bloodstream infections, reduce ICU mortality, and improve the care of patients on ventilators. Monthly conference calls, a dedicated Web site, and frequent electronic communication among participants have encouraged a strong collaborative spirit and exchange of useful information. Kathy Schumacher, MSA, data outcomes manager at Keystone participant William Beaumont Hospital (Royal Oak, Michigan), comments, "The setting of daily goals has had the greatest impact" at Beaumont, which lowered its bloodstream infection and ventilator-associated pneumonia rates within the first six months of implementation.7 In fact, Keystone Executive Director Chris Goeschel, RN, MHA, notes that 26 ICUs that had been participating for more than six months had no bloodstream infections reported since implementing the project. Officials in New Jersey, Maryland, and Rhode Island have spoken with Goeschel about implementing similar programs in their states.9 Since the Institute for Healthcare Improvement (Cambridge, Massachusetts) launched the 100,000 Lives Campaign, more than 3,100 hospitals have participated, increasing the likelihood that hospital-based oncologists and other physicians have taken part in one or more campaign-related projects. The campaign's goal was to eliminate an estimated 100,000 patient deaths in US hospitals during the course of 18 months. The campaign focused on six strategies to reduce errors and improve patient safety10:
At one time, it might have been acceptable to have a benchmark of 80% or 90% success implementing guidelines and protocols in health care, but that is no longer the case, notes Donald M. Berwick, MD, chief executive officer of the Institute for Healthcare Improvement. Consumers who have come to expect excellence in electronics or automobiles should expect the same of health care.
Establishing a nonpunitive culture as a key step in eliminating errors is not easy, despite many success stories. "It takes work to develop a culture of no-fault," says Peter D. Eisenberg, MD, an oncologist with California Cancer Care (Greenbrae, California). "Our practice culture is one of collaboration, cooperation, and communication. We talk about no-fault and work at it constantly." As a practice accredited by the Accreditation Association for Ambulatory Health Care, California Cancer Care has an error-reporting system in place. Any member of the staff is encouraged to bring errors and near misses to the attention of an administrator, who then initiates a review. Findings are discussed in weekly meetings, and all staff involved are notified of changes in the process, if any. The focus is on process and systems. "We generally assume that when something doesn't go the way it's supposed to, our systems are at fault or we didn't follow the system correctly," explains Eisenberg. Many health care systems have found the issue of change management to be part of the challenge moving toward a nonpunitive culture and enhanced error reporting. When Lehigh Valley Hospital and Health Network (Allentown, Pennsylvania) attempted to initiate a series of quality improvements designed to reduce medical errors, it found significant staff resistance, largely because of fear of disciplinary action and the belief that near misses were not really important enough to report or eliminate. The hospital designed an interactive workshop that was incorporated into the continuing education program for the nursing staff. The training highlighted the value of near-miss reporting, demonstrated nonpunitive coaching and feedback skills, and provided information on the hospital's new Nonpunitive Patient Safety Policy. Pre- and postworkshop surveys found significant changes. After attending the workshop, staff were more likely to report near misses, felt less fearful of reprisal, and understood that error reports were primarily sought to increase patient safety.12 "Issues surrounding change management cannot be overestimated," says Mutter of Valley Hospital. He suggests that to achieve a change in culture, a health care system must have a well-defined message that is consistently and frequently repeated. The message needs to be one that everyone will buy into and find worthwhile. Stakeholders at every level must embrace the need for change. And there must be a commitment to use the information provided for positive change.13 Organizers of the 100,000 Lives Campaign suggested using the Model for Improvement as a strategy to drive change. The model includes two components: key questions to evaluate a potential change and a four-step process to test a change. In the model, a series of three questions should be answered before a change is implemented: What are we trying to accomplish? How will we know that the change is an improvement? What change can we make that will result in improvement? The four-step processplan, do, study, and acthelps staff conduct small-scale pilot tests of a change, in the work setting, to identify problems and allow for revisions before a change is made facilitywide. Repeatedly applying the cycle in a series of pilot tests that work out the kinks in a change before implementation helps the staff overcome resistance to change. Staff are much more likely to participate when they know that a change can and will be revised, if necessary. The results of the 100,000 Lives Campaign and others have clearly shown it is possible to significantly reduce medical errors, thereby reducing morbidity and mortality. In fact, the Institute for Healthcare Improvement has launched its second phase, 5 Million Lives, to protect patients from 5 million incidents of medical harm over the next 2 years. Every error has at least one root cause, and every cause can be eliminated, but only if the error is revealed. As key players on the patient care team in inpatient and outpatient settings, physicians must be willing to champion efforts to create a nonpunitive culture and implement error-reduction strategies.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1935-469X. Print ISSN: 1554-7477
|