The Quality of Health Care
Medical Error and Patient Injury: Costly and Often Preventable
Research Report
Andrew H. Smith, AARP Public Policy Institute
September 1998
Table of Contents:
- Public Perception of Patient Safety and Medical Error
- Incidence of Medical Error and Injury
- Drugs and Medical Injury
- Costs Resulting from Medical Injury
- Why Do Medical Errors Happen, and How Should the Problem Be Addressed?
- Addressing the Problem from a Systems Approach
- Current Efforts to Address Medical Error From a Systems Perspective
- Conclusion
- Footnotes
Patient injuries that result from preventable medical errors are widespread and costly.1 One recent study found that more than one in six hospitalized patients suffered medical injuries that prolonged their hospital stays.2 It has been estimated that total annual costs associated with injuries resulting from medical error may be as high as $200 billion, the equivalent of nearly one out of every five dollars spent on health care in America.3 Estimates of the frequency of medical errors and injuries and the costs associated with them vary considerably, but even the most conservative estimates indicate that the problem is widespread, very costly, and requires serious attention.4
Preventable medical error and injury are of particular concern for older people because there is evidence that they are injured at a substantially higher rate than patients in other age groups. As Figure 1 indicates, patients age 65 and older experience medical injury two to four times as often as patients in age groups under the age of 45, according to a landmark study published in 1991, the most recent age-specific data available.5 Advancing age was the only demographic characteristic -- not gender, race, ethnicity, or income -- associated with a significantly increased incidence of medical injury and of injury due to "negligence."6 The evidence suggests that costs associated with preventable medical error and injury, both in terms of human suffering and dollars spent by the Medicare program to treat injured beneficiaries, are very significant.
Public Perception of Patient Safety and Medical Error
There is a substantial amount of public concern about patient safety.7 In a 1997 national survey, respondents rated the current health care system as only "moderately safe" -- safer than nuclear power and food handling, but less safe than airplane travel and the workplace.8 (See Table 1.) Forty-two percent of those surveyed said that they had been involved, either personally or through a friend or relative, in a situation where a medical mistake was made. Fifty-two percent of respondents stated that they were satisfied with the measures currently in place to prevent medical mistakes, but a large minority, 42 percent, said they were not satisfied.9 Not surprisingly, most of those who reported that they were not satisfied with current measures were those who had been involved in some way with a medical mistake.10
|
| Table 1. Perceived Safety of Various Environments | |
| Environment | Mean Scores |
| Airline travel | 5.2 |
| Workplace | 5.2 |
| Health care | 4.9 |
| Food handling | 4.4 |
| Nuclear power | 4.2 |
| Scores: 7=Safe, 1=Unsafe. Source: National Patient Safety Foundation at the AMA, "Public Opinion of Patient Safety Issues." Survey conducted by Louis Harris & Associates, September 1997. | |
Top of PageIncidence of Medical Error and Injury
As noted above, recent estimates of the incidence of medical errors resulting in injuries11 reach as high as 17.7 percent of hospitalizations.12 One important study of medical injury is the 1990 Harvard Medical Practice Study (Harvard Study), a population-based study of injuries resulting from medical care during hospitalizations in New York. This study found that nearly 4 percent of patients suffered an injury that caused their hospital stays to be prolonged, or resulted in measurable disability.13 The Harvard Study, which used reviews of medical records to detect medical injuries, found that almost 14 percent of those identified as having suffered medical injury died as a result of their injuries. If the rate of deaths resulting from medical error identified by the Harvard Study in New York were consistent with rates in the other 49 states, that would mean that 180,000 Americans die annually as a result of medical injuries.14 That figure would be comparable to the number of deaths that would occur if three jumbo-jets crashed every two days,15 and is approximately four times the number of traffic fatalities that occur annually in America.16
Consistent with other studies that have found that most medical injuries are due to errors, the Harvard Study determined that 69 percent of the medical injuries identified were due to error, and were, therefore, preventable.17
Studies conducted more recently indicate that medical injury may be substantially more common than suggested in the Harvard Study. Using a method more likely to capture incidents of medical error than the earlier study, Andrews and her colleagues found that 17.7 percent of patients whose care was observed experienced at least one serious adverse event per hospitalization.18 The frequency of medical injuries was linked to severity of illness and length of hospital stay, with the likelihood of experiencing a medical injury increasing by 6 percent per day of hospitalization. One or more causes19 of medical injuries were determined in just over one half of cases in the study. In 37.8 percent of cases, the adverse events were found to have been caused by an individual; 15.6 percent had interactive causes; and 9.8 percent were due to administrative decisions. Although 17.7 percent of patients experienced medical injuries that prolonged their hospital stays, the study found that only 1.2 percent filed claims for compensation for their injuries.
Drugs have been found to be among the most common causes of medical injury. In the Harvard study, 19.4 percent of the injuries detected were related to the use of drugs, while the Andrews study determined that 9.3 percent of injuries were medication-related.20
A large percentage of adverse drug events (ADEs) have serious consequences, and many of them are preventable. Bates and his colleagues found that of all ADEs identified in their study, 1 percent were fatal, 12 percent life-threatening, 30 percent serious, and 57 percent significant. Of ADEs that were determined to have been preventable, 20 percent were life- threatening, and 43 percent were serious. (See Figure 2.) Overall, 28 percent of the ADEs were judged preventable, but of life-threatening and serious ADEs, 42 percent were determined to have been preventable.21 Bates found rates of 6.5 ADEs and 5.5 potential ADEs per 100 non-obstetrical admissions to tertiary-care hospitals. Classen and colleagues found that adverse drug events complicated 2.43 percent of hospital admissions, adding significantly to length of hospital stays and to costs.22
|
Costs Resulting from Medical Injury
The costs associated with injuries resulting from medical error are quite substantial. As noted above, one recent estimate placed the total costs associated with medical injury at as much as $200 billion annually.23
Most studies that attempt to estimate costs associated with medical error have focused on injuries resulting from the use or misuse of medications. In their 1995 study, Johnson and Bootman estimated that costs associated with drug-related illness and death that resulted primarily from patient non-compliance, and inappropriate prescribing, and/or monitoring by health care professionals equal $76.6 billion annually.24 The costs calculated for drug-related illness and death were limited to those that arose from medication use or misuse in an outpatient setting, with the largest component of costs resulting from drug-related hospitalizations.25
The ADEs identified in the Classen study, half of which were identified as preventable, added 1.91 days to the mean length of hospital stays and resulted in increased costs per stay of $2,262.26
In a follow up to their earlier study, Bates and colleagues determined that an additional 2.2 days of hospitalization were required for patients experiencing an ADE, at an average added cost of $3,244. For ADEs identified as preventable, patients stayed in the hospital an average of 4.6 extra days, at an average additional cost of $5,857.27
Why Do Medical Errors Happen, and How Should the Problem Be Addressed?
1. Negligent and/or incompetent providers
As a recent survey reveals, many people believe that medical errors and injuries occur because there are just too many "bad doctors" and other health care professionals performing in a negligent manner.28 Medical injury is viewed as primarily the result of allowing incompetent and/or careless providers to continue in the practice of medicine, and of hospital under-staffing and other cost-cutting practices.29 It has frequently been observed that relatively few providers are sanctioned by the medical profession and/or state entities charged with enforcing standards of medical practice despite evidence of widespread negligence.30
Those who believe that medical negligence and an ineffective oversight system are largely responsible for medical error and injury have responded in a number of ways. For example, they promoted the development and use of a practitioner databank. As a result, the National Practitioner Data Bank (NPDB) was created. The NPDB collects and releases information (to authorized entities) relating to medical malpractice payments, adverse licensure actions, certain types of professional review actions, and reports of Medicare and Medicaid sanctions taken against physicians, dentists, and some other health care practitioners.31 They have also defended the laws that govern medical malpractice actions against a strong effort from the medical community to enact legal reforms that would curtail malpractice litigation.32
2. Inevitable human error and systems failures
A contrasting view holds that the problem of medical error and injury results primarily from systems failures. Proponents of this view acknowledge that there are incompetent and impaired providers who commit errors that result in patient injury, and that few physicians face disciplinary actions. However, they observe, there is little evidence that negligence is the major cause of medical error, or that rooting out negligent and incompetent providers would solve the problem.
Those who subscribe to a "systems approach" to medical error, drawing on psychological and human factors research, argue that human beings, no matter how careful and conscientious they are, will make mistakes.33 They also note that because the practice of medicine is complex, there are a great many opportunities for mistakes to occur, and that the high level of complexity makes it unrealistic to depend on promoting individual perfection as the method to avoid mistakes that result in patient injury. For example, in one study of an intensive care unit, it was determined that patients received an average of 178 "activities" each day.34 The average number of errors per patient per day was 1.7, or slightly less than 1 percent. Thus, the unit was functioning correctly 99 percent of the time.35 Leape notes, however, that even an accuracy rate of 99.9 percent may not prove adequate, noting that a 99.9 percent accuracy rate would translate to:
- Two unsafe landings at O'Hare airport each day;
- 16,000 pieces of lost mail per hour; and
- 32,000 bank checks deducted from the wrong account every hour.36
Addressing the Problem from a Systems Approach
One medical specialty, anesthesiology, has already made significant improvements in its safety record. Mortality resulting from errors in anesthesia has been reduced by 95 percent over the past 15 years.37
Recognizing system factors, rather than carelessness or incompetence as the most important causes of medical error, anesthesiologists designed fail-safe systems and developed and implemented training programs to avoid errors.38
The success story in anesthesiology illustrates the possibilities and problems for other areas of medical practice. Errors and the resulting injuries in anesthesiology, unlike those in many areas of medical practice, tend to be dramatic and severe.39 Information about incidents and the circumstances surrounding them were, therefore, available to those attempting to understand the problems, and the reasons the errors occurred were often transparent. These factors were conducive to understanding the problems and developing approaches to correct them.
A number of scholars believe that the most important reason that medicine has failed to develop more effective ways to prevent error is that, except in the case of the practice of anesthesiology, there has been little opportunity to study the reasons that errors occur. Information about medical error is inadequate for researchers because most errors go unreported. Unlike errors in anesthesiology, which, as noted above, cannot easily be hidden, errors occurring in other areas of medical practice tend to be less frequently obvious and dramatic in effect. In what some call medicine's culture of blame, there is good reason not to volunteer information that an error has occurred when it might otherwise remain undiscovered. In the medical culture, error cannot be accepted; physicians are taught in medical school and during residency to learn and practice error-free medicine, i.e., to be perfect. Error is treated as a moral failing,40 and it is not surprising that mistakes are driven "underground."
Advocates of the systems approach argue that, for medicine to enjoy the success observed in anesthesiology, it is essential to overcome the barriers to full reporting of medical errors. For researchers to devise ways to prevent and/or to absorb41 errors and prevent injuries, they must learn precisely how and why errors and their resulting injuries take place. They must have access to detailed and comprehensive information on errors, and full information can be obtained only if there is full disclosure of errors.
Current Efforts to Address Medical Error From a Systems Perspective
A number of initiatives have been developed to study and address the problem of medical error using a systems approach. Examples include:
-
The National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP), an organization of pharmacy and health care professional groups, the U.S. Food and Drug Administration, the U.S. Pharmacopoeia, and consumer organizations, among others, has developed numerous recommendations to prevent medication errors. These recommendations, addressed to pharmaceutical manufacturers, packagers and repackagers, hospitals and hospital pharmacies, outpatient pharmacies, physicians and other health care personnel, should lead to the safer use of drugs in all settings.
Among NCC MERP's recommendations: (1) print warnings only on caps and ferrules of injectables; (2) make intravenous drug names visible on both sides of the container; and (3) print drug names in type that is at least as large as company names and logos.42
The organization is also encouraging the use of its "Medication Error Index for Categorizing Errors," a new indexing system that will help researchers to track medication errors in a consistent, systematic manner.43 Widespread use of the index should result in the efficient collection and compilation of data on medication error, and thereby allow the development of recommendations that could lessen the chance for patient injury.
- The National Patient Safety Foundation at the AMA (NPSF) and the National Patient Safety Partnership (NPSP) constitute two major initiatives to (1) study medical error and (2) develop systems-based responses to reduce the incidence of medical error and absorb errors when they do occur so that the errors do not reach the patient.
- The NPSF was founded by the American Medical Association in 1997, but is now an independent foundation supported by a broad range of organizations, including health care professional organizations, consumer organizations, insurance companies, managed care organizations, and academicians. The NPSP was founded by the U.S. Veterans Administration, and like the NPSF, has a broad range of participating organizations. The NPSF and NPSP have recently linked their efforts to promote research into the causes and cures for medical error and injury. Among the projects they are working on together are:
- (1) an effort to design a voluntary, confidential, non-punitive system that would promote the reporting of essential data that would allow researchers to learn the nature of systems failures that lead to injury; and
- (2) a survey of health care providers and the medical culture as it relates to patient safety.
The systems approach has been successfully employed in non-health care settings that are, like health care, high risk enterprises. Both the airline industry's Aviation Safety Reporting System (ASRS) and the National Aeronautics and Space Administration's (NASA) "Close-Call" reporting system were developed through use of the systems approach.44
As noted above, the success achieved in anesthesiology through the use of a systems approach to improve patient safety strongly suggests that applying that approach would be appropriate in other areas of medical practice. Before systems changes to prevent medical error and patient injury can be devised and implemented, the weaknesses in the complex systems of medical care that allow, or even promote, medical errors must be identified and understood. A great deal of research must be performed before the goal of substantially reducing rates of preventable injury can be realized.
The systems approach promises significant reductions of preventable medical error and injury in the future. It cannot, however, eliminate current and future needs for patient compensation when a preventable injury does occur, despite systems improvements. Neither can it adequately address errors/injuries that arise from provider incompetence and/or impairment. Those are matters that must continue to be addressed through legal and administrative mechanisms.
The work of the NPSF, NPSP, and NCC MERP, among other organizations, to coordinate and support research and disseminate its results, should lead to safer medical practice, fewer patient injuries, and reduced health care costs. Success in preventing or absorbing medical error should prove beneficial to Medicare beneficiaries, who most frequently suffer medical injuries, and could save the Medicare program billions of dollars currently devoted to treating preventable medical injuries.
1 "Medical error" may
be defined as "an unintended act (either of omission or
commission) or one that does not achieve its intended
outcomes." Leape, Lucien. "Error in Medicine."
Journal of the American Medical Association
272(23):1851-57 (Dec. 21, 1994).
2 Andrews, Lori B., Carol Stocking,
Thomas Krizek, et al. "An Alternative Strategy for
Studying Adverse Events in Medical Care." Lancet
349:309-13 (Feb. 1, 1997).
3 Perrone, J. "Designing a
Safer, Smarter Health Care System: AMA Foundation Looks at
Ways to Prevent Mistakes," American Medical News
40(40):1 (Oct. 27, 1997).
4 Reduction of medical error is listed
as one of "Six National Aims" in the Report of the
President's Advisory Commission on Consumer Protection and
Quality in the Health Care Industry (March 1998).
5 Patients, Doctors, and
Lawyers: Medical Injury, Malpractice Litigation, and Patient
Compensation in New York. The Report of the Harvard Medical
Practice Study to the State of New York. Harvard Medical
Practice Study, 1990, 6-23.
6 Ibid.
7 "Public Opinion of
Patient Safety Issues: Research Findings," National
Patient Safety Foundation at the AMA, September 1997.
8 Ibid.
9 Ibid.
10 Ibid.
11 "Medical injuries"
here refer to "iatrogenic injuries," i.e., injuries
or conditions resulting from treatment by physicians or
surgeons.
12 Andrews, et al.
(1997).
13 Harvard Medical Practice
Study (1990).
14 Leape (1994).
15 Ibid.
16 There were 43,910 deaths in
1997 resulting from motor vehicle accidents. National Center
for Health Statistics. "Births, Marriages, Divorces, and
Deaths for February 1997. Monthly Vital Statistics
Report." 46: 2. (1997).
17 Leape (1994).
18 Andrews and her colleagues
used a prospective, observational approach that followed the
care of all patients admitted over a period of time to three
units of a teaching hospital, as opposed to the Harvard
Medical Practice Study that used retrospective reviews of
medical records. Andrews, et al. (1997).
19 "Interactive
causes" refers to "interactions between individuals,
or between individuals and hospital entities, or between
hospital entities, such as the failure of a consultant team to
communicate adequately with the requesting team."
Andrews, et al. (1997) at p. 311.
20 Harvard Medical Practice Study
(1990).
21 Bates, David W., David J.
Cullen, Nan Laird, et al. "Incidence of Adverse Drug
Events and Potential Adverse Drug Events: Implications for
Prevention." Journal of the American Medical
Association 274(1): 29-34 (July 5, 1995).
22 Classen,, David C., Stanley
L. Pestotnik, R. Scott Evans, et. al. Adverse Drug Events in
Hospitalized Patients," Journal of the American
Medical Association 277(4):301-06 (Jan. 22/29,
1997).
23 Perrone (1997).
24 Johnson, Jeffrey A. and J.
Lyle Bootman. "Drug-Related Morbidity and Mortality: A
Cost-of-Illness Model," Archives of Internal
Medicine 155:1949-56 (Oct. 6, 1995). This estimate
includes all types of medication error, both preventable and
non-preventable. It does not include costs associated with
injuries that are the result of unforseeable
allergic/idiosyncratic responses or those that occur when the
provider knows that there are risks associated with a drug but
prescribes it anyway because, in his/her judgment, the
potential benefits outweigh the risks.
25 When indirect costs due to
non-compliance are added to the direct cost figures, total
economic costs rise to approximately $100 billion. Berg, J.S.,
J. Dischler, J.J. Raia, and N. Palmer-Shevlin,
"Medication Compliance: A Healthcare Problem,"
Annals of Pharmacotherapy 27(9):S3-S22 (1993).
26 Ibid.
27 Bates, David W., Nathan
Spell, David J. Cullen, et al. "The Costs of Adverse Drug
Events in Hospitalized Patients," Journal of the
American Medical Association 277(4):307-11 (Jan. 22/29,
1997).
28 See Richards, Edward P. and
Katharine C. Rathbun, Law and the Physician: A Practical
Guide. Little, Brown, and Co.:New York (1996).
29 Ibid.
30 See, for example, Public
Citizen, "16,638 Questionable Doctors." (March
1998). It is noted that, although there have been more
disciplinary actions taken against physicians recently, few
have been required to stop practicing medicine, even for a
short time. In 1996, 16,638 physicians were disciplined by
state boards or federal agencies. The rate of "serious
disciplinary actions" was 3.96 per 1,000 doctors (2,731
actions).
31 Title IV of the Health Care
Quality Improvement Act of 1986 (P.L. 99-660) established the
National Practitioner Data Bank (NPDB). Regulations governing
the NPDB may be found at 45 CFR Part 60. The information in
the NPDB is available only to state licensing boards,
hospitals and other health care entities, professional
societies, certain Federal agencies, and others as specified
in the law. Only hospitals are mandated by law to query the
Data Bank.
32 Nonetheless, many states
passed "tort reform" measures in the wake of the
alleged medical malpractice insurance crisis of the late
1980s. They included such measures as placing caps on possible
damage awards (particularly on awards for "pain and
suffering"), restrictions on statutes of limitations,
limitations of plaintiff attorneys' fees, and other
measures to discourage potential complainants from filing
malpractice actions.
33 For a brief overview of
relevant developments in cognitive psychology and human
factors research, see Leape, p. 1853 (1994).
34 An "activity" is
defined as any interaction between health care personnel and
patients that presents an opportunity for an adverse patient
outcome.
35 Leape (1994).
36 W.E. Deming, written
communication quoted in Leape (1994).
37 Orkin, P.K. "Patient
Monitoring During Anesthesia as an Exercise in Technology
Assessment." In Saidman, L. J. and N.T. Smith, eds.
Monitoring in Anesthesia 3rd Ed. London, England: Butterworth
Publishers, Inc. (1993).
38 See Gaba, D.M., "Human
Errors in Anesthetic Mishaps," International
Anesthesiology Clinics 27(3):137-47 (Fall 1989). Also see
Cooper, J.B., R.S. Newbower, and P.J. Kitz, "An Analysis
of Major Errors and Equipment Failures in Anesthesia
Management: Considerations for Prevention and Detection,"
Anesthesiology 60(1):34-42 (Jan. 1984).
39 Leape, p. 1856 (1994).
40 Ibid.
41 It is recognized that errors
are inevitable in any human endeavor, including the provision
of health care. Error "absorption" refers to the
notion that well-designed error prevention systems will
"absorb" errors, keeping them from reaching the
patient and causing injury.
42 See U.S.P.,
"Medications Errors Council Recommends Changes to Medical
Product Packaging and Labeling," The Standard (Sep. 16,
1997).
43 U.S.P., "Medication
Errors Council Promotes Categorization Index," The
Standard (October 1996).
44 See Helmreich, R.L.
"Managing Human Error in Aviation," Scientific
American 276(5):62-67 (May 1997).
Written by Drew Smith, AARP Public Policy Institute
September 1998
©1998 AARP
May be copied only for noncommercial purposes and with
attribution; permission required for all other purposes.
Public Policy Institute, AARP, 601 E Street, NW, Washington, DC
20049
Pub ID: IB35