Updated and reprinted from SpineLine, May/June 2001, pp.7-12.
Patient safety has been a basic tenet of medical practice since Hippocrates enunciated the principle of “first do no harm” more than a millennium ago. The limited availability of medical interventions and relatively primitive state of medical knowledge in Hippocratic times made such a decree relatively easy to implement. Modern advances in medical science have provided today’s health care providers with a myriad of sophisticated, effective treatment options. However, inherent in these advanced, potent and complex interventions is the potential for systemic side effects, treatment interaction and technical failure. The 21st century health care provider must be diligent in his or her pursuit of safe, effective treatment for patients.
However, implicit in this complicated, often multifaceted medical environment is the potential for errors in judgment, technical misadventures and systems failures that may cause physical harm to the patient. These concerns have been highlighted recently in the media and by the much-publicized Institute of Medicine (IOM) report, To Err Is Human.(1) This report stated that “at least 44,000” and possibly “as high as 98,000” die in the United States every year as a result of “medical errors.” These figures imply that medical errors should be considered the eighth leading cause of death in this country.(2 ) To put this number in perspective, the report cites the number of deaths from motor vehicle accidents as 43,458, breast cancer 42,297 and AIDS 16,516.(3) In this article, we will provide an analysis of the IOM report, highlight the issues raised and discuss NASS’ approach to the principle of patient safety.
Institute of Medicine
The Institute of Medicine is a branch of the National Academy of Sciences, which was given a congressional mandate in 1863 to “advise the federal government…on scientific and technical matters.” The IOM was chartered in 1970 to “secure the services of eminent members of appropriate professions in the examination of policy matters pertaining to the health of the public.”(1) To Err Is Human is the first in a series of reports anticipated from the IOM Quality of Health Care in America project. The IOM committee that generated the medical errors report includes physicians, health researchers and members of business and industry.
Basis for Statistics in the Report. One major criticism aimed at the IOM report is the origin of the numbers used to generate the statistics quoted. The national figures for deaths from medical errors were extrapolated from only 2 studies, 1 in New York(4,5) and the other in Colorado and Utah.(6) In these reviews, a limited sample of patient records was randomly selected from all hospital discharges in a 1-year period (1.7% in New York and 2.7% in Colorado and Utah) to look for “adverse events.” An adverse event was defined as an injury caused by medical management (rather than the disease process) that resulted in either a prolonged hospital stay or disability at discharge. A nurse or medical records administrator made the initial chart review for determination of an adverse event using a list of 18 screening criteria. These included such common occurrences as whether the patient was transferred from a general care unit to a special care unit and if there were subsequent admissions to any hospital. In the New York study, 2 physicians, and in Colorado and Utah, a single physician, evaluated whether the adverse event was a result of negligent or nonnegligent care. Adverse events associated with “negligent care” were deemed “medical errors.”
Determinations of “negligent care” or a “medical error” were fairly subjective. Negligence was defined as care that fell below the standard expected of physicians in their community. Such treatment evaluations are difficult in the best circumstances, even in the tort arena, much less a situation of postdischarge record assessment. However, in the Colorado and Utah study, family medicine physicians and general internists performed the analysis. Despite the lack of surgical input, operative adverse events were determined to constitute the majority (44.9%) of all adverse events, with 16.9% considered negligent. The highest proportion of operative events was technical (29.9% of total, 23.6% negligent), bleeding (17.1% of total, 9.8% negligent) and wound infection (11.4% of total, 20.8 % negligent).
The physician reviewers making this judgment in the New York study included internists and surgeons. Operative errors were also cited as the highest proportion overall (47.7% of the total, 17.0% considered negligent). The most common adverse events were wound infection (13.6% of total, 12.5% negligent), technical complications (12.9% of total, 17.6% negligent) and late complications (10.6% of total, 13.6% negligent).
The highest rate of nonoperative adverse events was related to drug treatment in both studies (19.4% of total, 17.7% negligent in New York and 19.3% of total, 35.1% negligent in Colorado and Utah). Antibiotics universally constituted the most common drug class involved in adverse events.
Another obvious criticism of these articles is the retrospective nature of the reviews. The New York study was based on 1984 data and reported in 1991. The Colorado and Utah review used 1992 discharges and was ultimately published in a peer-reviewed journal in 2000.
In order to arrive at the estimates for patient deaths from medical errors, the IOM report extrapolated data from a subsequent random year, 1997, when there were 33.6 million admissions to hospitals in the United States. The New York study implicated 3.7% of all hospital admissions to be associated with an adverse event and 13.6% of those incidents to result in the patient’s death. When these figures are extrapolated over 33.6 million admissions, the upper estimate of 98,000 patient deaths per year related to medical error is generated. The corresponding statistics for Colorado and Utah are 2.9% of hospitalizations incurring an adverse event with 6.6% causing patient death.
Incorporating these figures determines the lower-end estimate of 44,000 patient deaths from medical errors annually.
Are the Estimates Believable? These startling and very worrisome statistics evoked a major media splash and induced cries of shock and horror from the public and Capitol Hill. Congress began hearings on the topic of medical errors within 2 weeks of the release of the IOM report. A federal interagency task force (Quality Interagency Coordination Task Force-[QuIC]) was formed by order of President Clinton to report on the status of medical errors and suggest government actions to improve patient safety.(7) Senators Charles Grassly, Joseph Lieberman and Robert Kerrey introduced a bill, S.2378, Stop all Frequent Errors (SAFE) Act of 2000. This bill would require hospitals and other providers to report medical errors as a condition of participation in the Medicare and Medicaid programs. Senators Arlen Specter and Tom Harkin proposed The Medical Error Reduction Act of 2000. This legislation would provide funds to states establishing reporting systems. In addition, the bill would fund 15 demonstration projects for health care facilities and organizations to research the causes of medical errors and to use technology to develop models and methods to reduce medical errors.(8) The budget of the Agency for Healthcare Research and Quality (AHRQ) was increased by $20 million to establish a Center for Patient Safety and provide research dollars to projects in the medical errors field. Six grants totaling approximately $2 million have already been awarded.(9)
As with any statistic drawn from a somewhat questionable methodology, there are both detractors and supporters of the figures. McDonald et al(10) have suggested that the deaths due to medical errors in the IOM report are exaggerated. They have contended that the screening criteria for identification of an adverse event (eg, patient transfer from a general care unit to a specialty care unit) tends to highlight a group of patients with more severe illnesses. This “high-severity” group would be expected to have a certain mortality rate, even with flawless care. As the studies serving as a basis for the IOM report did not specifically address whether the patient’s death came about as a direct cause of an adverse event, the actual rate of death from medical error should only be the difference between the observed rate of mortality in the study group and the expected rate of death for a high severity of illness population. Using a model based on rates of death in New York for the same year as the medical errors study (1984), McDonald et al(10) calculated an expected death rate for a high severity of illness group which was almost identical to the actual observed death rate in the medical errors study group. Their conclusion, therefore, was that the actual rate of patient deaths occurring as a direct result of a medical error was quite small. Leape, a coauthor of the original New York study, has written a rebuttal of the McDonald theory but also stated that their review did have “signifi-cant limitations and methodological weaknesses.”(11)
Troyen Brennan, another coauthor of the New York study, has written a perceptive review of the IOM report.(12) He applauded the spotlight that the report has cast on a significant quality of care issue, but cautioned that this interest and momentum should be channeled into further research to clarify some of the concerns generated by the original limited study. In particular, he raised the direct causality issue as requiring further evaluation, along with more refinement of the criteria which should be used by review physicians to make judgments concerning whether adverse events were in fact preventable “medical errors.” Brennan further pointed out that contrary to the impression of apathy and inattention generated by media reports, the medical profession, hospitals and regulatory agencies have been devoting significant time and resources to improving patient safety. He cites specific examples of the decreased rates of mortality in cardiac and intracranial surgery to illustrate this point. As his final commentary, Brennan emphasizes the need for revamping the litigation system and applying firm peer review confidentiality protection before any meaningful voluntary reporting system is likely to be effective.
The Valid Issues
Although it is easy to be critical of the IOM report, NASS members are all aware that medical errors do occur. For example, an undeniable medical error occurs with wrong site surgery. The Joint Commission on Accreditation of Health Care Organizations (JCAHO) has mandated reporting and a root cause analysis of “sentinel events,” which include wrong site surgery. Between January 1995 and January 2001, JCAHO reports 107 instances of wrong site surgery, constituting 9.7% of all sentinel events.13 The American Academy of Orthopaedic Surgeons (AAOS) has been reviewing the issue of wrong site surgery for a number of years. AAOS issued its first report on this topic in September of 199714 before the IOM’s Quality of Health Care in America project had even begun. The AAOS report cited 331 instances of wrong site surgery obtained from closed claims data, 11 involving spinal surgery at the wrong level. The cases all involved discectomy surgery to be performed at one level. In all but one instance, the wrong level operated on was at the segment above the actual pathology.
In addition to highlighting the topic of medical errors, the IOM report brings out several issues regarding reporting, analysis and reduction of medical errors.
Reporting of Medical Errors. The IOM report recommended a mandatory reporting system for medical errors.1 The report of the Quality Interagency Coordination Task Force, which reviewed the IOM recommendations, suggested a reporting system with both mandatory and volunteer components.(7) The Task Force also recommended protection for the peer-review “root cause analysis” but also noted “individuals should have access to information leading up to and including the occurrence of a preventable error that caused their serious injury or the death of a family member.” The Task Force further noted, “Any legislation or administrative intervention in this area should not undermine individuals’ rights to redress for criminal activity, malpractice or negligence.” Fear of malpractice litigation still remains a significant impediment to meaningful reporting systems according to Brennan.(12) Mohr(15) has pointed out significant factors, which argue against major tort reform as an element of a better quality improvement system. Gostin(16) also views the present state of medical malpractice as a barrier to a public health approach to reducing error. He advocates a system of mandatory and volunteer reporting systems similar to those that have successfully reduced errors in aviation(17) and the motor vehicle industry.(18) The keys to these systems, however, are that they are confidential and nonpunitive.
Six-Sigma Quality. Another initiative that has been touted as a goal that should be taken from industry and adopted by health care is the concept of 6-sigma quality. Mark Chassin, a member of the IOM’s Committee on Quality of Health Care in America and a coauthor of the To Err Is Human report, has written on the idea of 6-sigma quality(19) and its relevance to health care. The term was initially coined by the Motorola Corporation to designate its strategy for setting tolerance limits in the manufacturing process of its products. Six-sigma quality relates to an error limit set 6 standard deviations above the mean when measuring statistical variations on a normal distribution curve or bell curve. Chassin notes that this high-quality standard translates into fewer than (3.4) defects per million units or opportunities. If the error tolerance levels are set at lower sigma (fewer standard deviations above the mean), then more defects or errors can be expected to occur. An example cited was the airline industry operating at about 5 sigma for fatal errors (230 deaths per million opportunities) and at a 4-sigma rate for baggage handling (6,210 lost bags per million opportunities).
The health care field has some excellent examples of high sigma quality. The most notable relates to patient deaths under anesthesia. Deaths related to anesthesia occurred at a rate of between 25 to 50 per million opportunities in the 1970s and 1980s.(20) After adoption of clinical guidelines for intraoperative monitoring (including a pulse oximeter for most cases), more recent estimates peg the rate as around 5.4 deaths per million opportunities. This rate is very close to the 3.4 per million numbers to attain 6-sigma quality.(21,22)
Other areas of medicine do not fare so well. For example, use of beta blockers has been shown to save lives after acute myocardial infarction. The evidence for a therapeutic benefit is so strong that the American Medical Association issued one of its few Quality Care Alerts on the subject.(23) A study of actual beta blocker usage in this patient population showed only 21% of the eligible patients were prescribed the medication.(24) This translates to a defect or failure rate of approximately 790,000 per million opportunities or a tolerance of under 1 sigma quality.
Medical Errors as a Systems Problem. Another model borrowed from industry may be crucial to the evaluation of medical errors and the development of solutions for the problem. This is the concept that systems failures—not individual human failures—are to blame for many of the adverse events occurring in health care and furthermore, that systems can be designed to back up human error. Leape(25) points to medication errors as an example of a systems and human problem having a systems solution. In his large study of adverse drug events, about 78% of the medication errors were found to be systems problems, encompassing issues such as transcription errors, faulty drug identification and inadequate monitoring (blood levels, renal function, etc).(26) However, it was noted that about 20% of physicians’ written orders to the pharmacy (human error) contained mistakes such as wrong dosage or frequency. A computerized order entry system cut this error rate by two thirds.(26,27) A systems intervention was thus able to address the systems issues directly and provide added value to the physician and the patient, serving as a backup for the sometimes imperfect human memory. It should be stressed, however, that the clinical judgment of the physician remains intact as the primary therapeutic decision-making entity.
Dr. John Eisenberg, Director of the Agency for Healthcare Research and Quality (AHRQ), has proposed a reasonable and insightful approach to patient safety.(28) He notes that the long-standing approach of name, blame and shame directed toward an individual has not resulted in a reduction in errors. Eisenberg proposes that the key to an overall improvement in health care quality and patient safety is a “learning health care system.” He notes 8 critical components to this systems approach to patient safety which are rooted in continuing education: (1) informatics for information, (2) guidelines as learning tools, (3) learning from opinion leaders, (4) learning from the patient, (5) decision support systems, (6) the team learning together, (7) learning organizations and (8) just in time and point of care delivery.
NASS Approach to Patient Safety
NASS has already demonstrated a firm commitment to quality patient care, including patient safety, for many years. Efforts by NASS to develop the components of a continuous quality improvement (CQI) program date back more than a decade. In cooperation with a number of other professional societies, multidisciplinary panels have developed clinical guidelines to use as common treatment protocols to facilitate research and use as educational tools. Outcomes instruments are in place to allow evaluation of treatment. Analysis of outcomes data then leads the cycle back full circle to modify the treatment guideline according to evidence from the outcomes study.
Physician education to upgrade diagnostic and technical skills has always been a cornerstone of NASS’ mission. This has been fulfilled by the clinical focus of our annual meeting and continuing medical education (CME) courses. NASS has reached out to the Pacific Rim, Central and South America, and Europe to foster communication and exchange of ideas through combined meetings with other spine societies.
Patient education efforts have included educational brochures on a number of clinical problems such as herniated disc and spondylolisthesis. Additional information for patients is available on the NASS Web site (www.spine.org).
The release of the IOM report and the subsequent flurry of media attention and legislative initiatives stimulated NASS and many other professional medical organizations to institute a more formal evaluation of the area of patient safety. Accordingly, at its April 2000 meeting, the NASS Board created the NASS Task Force on Patient Safety. This group was charged with reviewing NASS efforts on patient safety and recommending an integrated, coordinated effort to increase efforts in the area of patient safety through NASS’ mission to “advance quality spine care through education, research and advocacy.”
NASS Strategic Plan Initiatives. At its October 2000 meeting, the NASS Board of Directors adopted as part of its strategic plan the committment to patient safety discussed in the recommendations from the IOM report pertaining to professional societies. This portion of the IOM Report’s recommendation 7.2 states:
Professional societies should make a visible commitment to patient safety by establishing a permanent committee dedicated to safety improvement and promoting a culture of safety. This committee should:
- Develop a curriculum on patient safety and encourage its adoption into training and certification requirements;
- Disseminate information on patients’ safety to members through special sessions at annual conferences, journal articles and editorials, newsletters, publications and Web sites on a regular basis;
- Recognize patient safety con-siderations in practice guidelines and in standards related to the introduction and diffusion of new technologies, therapies and drugs;
- Work with the AHRQ Center for Patient Safety to develop community-based collaborative initiatives for error reporting and analysis and implementation of patient safety improvements;
- Collaborate with other professional societies and disciplines in a national summit on the professionals’ role in patient safety.
NASS Patient Information Projects. In addition to its diagnosis-based patient information brochures, NASS has developed a patient handout on non-steroidal anti-inflammatory medication (NSAIDs). This was done to address specifically IOM concerns that the most common nonsurgical adverse events involved medications. NSAIDs are probably the most commonly prescribed family of medications by NASS’ multidisciplinary membership. This topic thus appeared to be an appropriate first undertaking to provide a patient education document for routine office use.
Also developed by NASS is a patient education brochure specifically outlining tips for generalized patient safety. This publication incorporates suggestions from the AHRQ’s 20 Tips for Patient Safety(29) and the 5 suggestions from the Quality Interagency Coordination Task Force (QuIC).(30)
Also available is an office spine diagram. This take-home sheet includes a diagram for the physician to outline the site(s) of pathology to the patient during office discussions as well as space to specify the differential diagnosis, plan of treatment and any follow-up. The patient would leave the office with a hard copy summary of the visit to which he or she and other health care providers, such as physical therapists, could refer. If the treatment plan includes surgery, the side and level(s) of the proposed operation could be specified. When brought to surgery by the patient, the handout would serve as an additional check of side and level to avoid wrong site surgery.
Physician Education Projects. The Task Force plans to disseminate information on patient safety via NASS CME programs.
The major initiative is the “Sign, Mark and X-Ray” campaign which was launched at the 2001 Annual Meeting. This program provides tools (such as the patient education diagram) and outlines a checklist series of steps (sign the appropriate side of surgery, mark an easily identified anatomic structure and obtain an intraoperative X-ray film for confirmation of level) to avoid wrong-site surgery.
Summary
The Institute of Medicine report To Err Is Human has focused national attention on the area of patient safety. There is heightened awareness on the part of the public, our patients, physicians and other health care providers, professional societies, hospitals, regulatory agencies and legislators.
NASS has reaffirmed its commitment to quality patient care. Specific initiatives to address patient safety issues were launched at the Patient Safety symposium at the 2001 NASS Annual Meeting in Seattle. In addition, medical errors, and in particular prevention of errors, will be examined in future SpineLine articles throughout the remainder of the year.
Acknowledgment
Pamela M. Hayden also contributed to this article. Pamela is Senior Manager of Scientific Affairs at the North American Spine Society.
References
1. Kohn LT, Corrigan JM, Donaldson MS, eds. To Err Is Human. Institute of Medicine, Washington, DC: National Academy Press; 1999.
2. Centers for Disease Control and Prevention (National Center for Health Statistics). Deaths: Final Data for 1997. National Vital Statistics Reports. 1999;47:27.
3. Centers for Disease Control and Prevention (National Center for Health Statistics). Births and Deaths: Preliminary Data for 1998. National Vital Statistics Reports. 1999;47:6.
4. Brennan TA, Leape LL, Laird NM, et al. Incidence of adverse events and negligence in hospitalized patients: results of the Harvard Medical Practice Study I. N Engl J Med. 1991;324:370-376.
5. Leape LL, Brennan TA, Laird NM, et al. The nature of adverse events in hospitalized patients: results of the Harvard Medical Practice Study II. N Engl J Med. 1991;324:377-384.
6. Thomas EJ, Studdert DM, Burstin HR, et al. Incidence and types of adverse events and negligent vare in Utah and Colorado. Med Care. 2000;38:261-271.
7. Department of Health and Human Services. Doing What Counts for Patient Safety: Federal Actions to Reduce Medical Errors and Their Impact. Report of the Quality Interagency Coordination Task Force (QuIC) to the President. February 2000.
8. Ponzer K. Background Paper Medical Error. National Orthopedic Leadership Conference. Washington DC, May 2000.
9. Agency for Healthcare Research and Quality. AHRQ Awards Six Grants in Systems-Related Best Practices to Improve Patient Safety (press release). Rockville, MD; November 9, 2000.
10. McDonald CJ, Weiner M and Hui SL. Deaths due to medical errors are exaggerated in Institute of Medicine report. JAMA. 2000;284;93-95.
11. Leape L. Institute of Medicine medical error figures are not exaggerated. JAMA. 2000;284: 95-97.
12. Brennan TA. The Institute of Medicine report on medical errors-could it do harm. N Engl J Med. 2000;342:1123-1125.
13. Schyve PM. Facts about patient safety. Paper presented at: Practice Guidelines Partnership Meeting; February 23, 2001; Chicago, IL.
14. Canale ST, DeLee J, Edmonson A, et al. Report of the American Academy of Orthopaedic Surgeons Task Force on Wrong-Site Surgery. AAOS, 1998.
15. Mohr JC. American medical malpractice litigation in historical perspective.JAMA. 2000;283:1731-1737.
16. Gostin L. A public health approach to reducing error: medical malpractice as a barrier. JAMA. 2000;283:1742-1743.
17. Maurino D, Reason J, Lee R. Beyond Human Aviation Factors. Aldershot, England; Avery Press; 1995.
18. Bonnie R, ed. Reducing the Burden of Injury: Advancing Prevention and Treatment. Washington DC: National Academy Press; 1999.
19. Chassin M. Is health care ready for six sigma quality? Millbank. 1998;76:565-591.
20. Ross AF, Tinker JH. Anesthesia risk. In: Miller RD, ed. Anesthesia. 4th ed. New York: Churchill-Livingston; 1994.
21. Eichhorn JH. Prevention of intraoperative anesthesia accidents and related severe injury through safety monitoring. Anesthesiology. 1989;70:572-577.
22. Orkin FW. Patient monitoring during anesthesia as an exercise in technology sssessment. In: Saidman LJ, Smith NT, eds. Monitoring in Anesthesia, 3rd ed. London, UK; Butterworth-Heineman; 1993.
23. Beta blocker prophylaxis after myocardial infarction. AMA Quality Care Alert. American Medical Association; Chicago, IL: 1998.
24. Soumeri SB, McLaughlin TJ, Spiegelman D, et al. Adverse outcomes of underuse of B-blockers in elderly survivors of acute myocardial infarction. JAMA. 1997;277:115-121.
25. Leape L. The fallacies of failure. Hospitals and Health Networks. 1998;July 20:14.
26. Leape LL, Bates DW, Cullen DJ. Systems analysis of adverse drug events. JAMA. 1995;274:35-43.
27. Bates DW, Leape LL, Cullen DJ, et al. Effect of computerized physician order entry and a team intervention on prevention of serious medication errors. JAMA. 1998;280:1311-1316.
28. Eisenberg JM. Continuing education meets the learning organization: the challenge of a systems approach to patient safety. Journal of Continuing Education in the Health Professions. 2000;20:197-207.
29. 20 Tips to Help Prevent Medical Errors. Patient Fact Sheet. Rockville, MD:Agency for Healthcare Research and Quality, US Dept of Human Services; AHRQ Publication No. 00-PO38. 2000.
30. Five Steps to Safer Health Care. Patient Fact Sheet. Rockville, MD: Quality Interagency Coordination Task Force; 2001.