Cover Image

PAPERBACK
$59.75



View/Hide Left Panel

13
Confronting Special Implementation Issues: The Epidemiology of Quality Problems

R. Heather Palmer

One full chapter of the Institute of Medicine (IOM) report concerns the "epidemiology of quality," in other words, the pattern of occurrence of quality problems (IOM, 1990). This paper expands upon the implications of those patterns.

THREE TYPES OF QUALITY PROBLEMS

The IOM report analyzed the extent to which harm created for patients by quality problems is caused by overuse of services, by poor technical quality, or by underuse. An example of overuse would be performing an unnecessary coronary angioplasty. An example of poor technical quality might be perforating the bowel during an abdominal operation or administering a drug other than the drug ordered by the physician. An example of underuse might be neglecting to prescribe an effective drug, that is, a drug that could cure a patient.

The IOM committee sought to define the magnitude of each of these because it believed that the three different types of problems must be detected and corrected by using different methods. For example, overuse can be identified by reviewing procedures performed on Medicare beneficiaries that are believed to be subject to overuse. The corrective intervention would be denial of payment for the procedure.

Poor technical quality is also studied among recipients of care. In the Medicare Peer Review Organization (PRO) program it is detected primarily by the use of written criteria called "generic quality screens." Nurse reviewers read patients' records, using these criteria to identify adverse events occurring to patients. A physician reviews all cases of adverse events to determine whether the cause was poor quality of care. The intervention used for



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 96
Medicare: New Directions in Quality Assurance 13 Confronting Special Implementation Issues: The Epidemiology of Quality Problems R. Heather Palmer One full chapter of the Institute of Medicine (IOM) report concerns the "epidemiology of quality," in other words, the pattern of occurrence of quality problems (IOM, 1990). This paper expands upon the implications of those patterns. THREE TYPES OF QUALITY PROBLEMS The IOM report analyzed the extent to which harm created for patients by quality problems is caused by overuse of services, by poor technical quality, or by underuse. An example of overuse would be performing an unnecessary coronary angioplasty. An example of poor technical quality might be perforating the bowel during an abdominal operation or administering a drug other than the drug ordered by the physician. An example of underuse might be neglecting to prescribe an effective drug, that is, a drug that could cure a patient. The IOM committee sought to define the magnitude of each of these because it believed that the three different types of problems must be detected and corrected by using different methods. For example, overuse can be identified by reviewing procedures performed on Medicare beneficiaries that are believed to be subject to overuse. The corrective intervention would be denial of payment for the procedure. Poor technical quality is also studied among recipients of care. In the Medicare Peer Review Organization (PRO) program it is detected primarily by the use of written criteria called "generic quality screens." Nurse reviewers read patients' records, using these criteria to identify adverse events occurring to patients. A physician reviews all cases of adverse events to determine whether the cause was poor quality of care. The intervention used for

OCR for page 96
Medicare: New Directions in Quality Assurance repeated or flagrant instances of poor quality is to sanction the physician who is responsible. Poor technical quality, of course, may also provoke a malpractice suit. Underuse is more difficult to measure. It may occur among recipients of care, but to get a complete picture one must also study nonrecipients who may have been unsuccessful in getting access to care. The Medicare program currently addresses underuse only in capitated settings, and it does not pursue the issue vigorously even there. Presumably if underuse of a serious nature were discovered in a health maintenance organization (HMO), the result of the intervention would be a sanction. Underuse, including omission or delay in matters of diagnosis or treatment, is also becoming a common cause of malpractice suits. Evidence of Quality Problems Evidence for the prevalence of these three types of problems is provided in the IOM report. Information on overuse comes primarily from data comparing rates of use of procedures and admissions in apparently comparable populations, although there is seldom any assurance that the populations are truly comparable. Wide variations in the use of procedures are often found, especially for procedures and types of admissions where there is little scientific evidence to guide physician decisions. It was widely believed that the difference in rates of use between a high-use and a low-use area was created by overuse in the high-use area, but we know now that the situation is more complex. Chassin et al. (1987) and Leape et al. (1990) report studies in which criteria chosen by expert panels to define appropriateness of a procedure were applied to data from medical records of Medicare beneficiaries. They show that overuse certainly occurred for procedures where clinical indications are in dispute. For instance, Chassin et al. (1987) found that 32 percent of carotid endarterectomies were unnecessary. Leape et al. (1990) found small geographic areas where up to 67 percent of carotid endarterectomies were unnecessary. Surprisingly, however, the differences in overuse between high-use and low-use areas were small. In other words, the proportion of inappropriate use did not explain most of the difference between high- and low-use areas. Overuse is a major target in Medicare because its elimination can save money without lessening quality. Providers generally agree that overuse is a problem for the Medicare program, although the IOM committee found that elderly beneficiaries seldom complained of it. The committee found evidence of poor technical quality from several sources. These included reports of sanctions initiated by PROs against physicians or hospitals, reports of license withdrawals by state licensing boards, and reports of malpractice claim settlements. None of these sources

OCR for page 96
Medicare: New Directions in Quality Assurance permits comparing the distribution of poor quality across states because the procedures for invoking a sanction, withdrawing a license, or settling a malpractice claim are highly variable from state to state. From these data, however, it emerged that a small number of physicians account for a large proportion of quality problems. Information on poor technical quality comes also from record reviews done by the PRO program or by hospital risk management personnel to identify patients who experience adverse events. Adverse events are common. The California Medical Insurance Feasibility Study of 1977 showed that in 4.65 percent of hospital admissions, adverse events occur that are "potentially compensable" because they were caused in part by poor technical quality (Mills, 1977). Adverse events occur most commonly with more complex procedures and with sicker patients. Underuse was the problem for which the committee found the least information, although Medicare consumers complained often of it. Some of the complaints of failure to provide service concerned services that the Medicare program has made a conscious decision not to cover. As noted earlier, Medicare concerns itself only with underuse in capitated settings. In fee-for-service settings, underuse may also occur, less for financial reasons than for physical, psychological, and logistical reasons. Perhaps the federal government and third-party payers are not anxious actively to promote use of services, even those deemed necessary, because that would result in their paying for more care. If so, this is a shortsighted point of view. Investment in early intervention and prevention services may save costs in the long run. We do not know, and desperately need to know, the extent of these potential savings. It is noteworthy that the committee found evidence of underuse primarily for what we might call "nonglamorous" conditions, such as incontinence, depression, or gait problems, and for simple types of services, such as the time and attention of providers, particularly nonphysicians including nurses. The IOM report did not find enough information to decide that any one of the three quality problems should take precedence. An interesting pattern emerges in the report: overuse and quality problems are primarily cited in the area of "high-technology" medicine and underuse in the area of "low-technology" medicine. This must surely derive from the present pattern of benefits coverage and reimbursement, which favors performance of high-technology procedures. Three other issues deserving comment include the overlap between overuse and poor technical quality, ways to discourage overuse while overcoming underuse, and ways to improve technical quality. The Overlap Between Overuse and Poor Technical Quality First, let us consider the relationship between overuse and poor quality. Table 13.1 provides a framework for discussing this relationship. Unnecessary

OCR for page 96
Medicare: New Directions in Quality Assurance TABLE 13.1 Probability of Good Outcome for Patients According to Competent Implementation and Necessity of Health Services Is the Service Competently Implemented? Is the Service Necessary?     Yes No Yes Good Bad No Bad Worst care, even if perfectly implemented, worsens the patient's outcome (net benefit) because it provides little or no gain in health while potentially causing unavoidable complications and side effects, discomfort, anxiety, and use of patient time. Obviously, then, overuse is bad for both cost and quality reasons. The report quotes data to show that Americans have more procedures than any other nation on earth. For instance, the U.S. rate for hysterectomy is three times higher than that in England and Wales. As another example, coronary bypass surgery is done at the rate of 19 per million in France and 483 per million in the United States. Many more American than French individuals, then, have complex procedures that put them at risk for poor quality. Overuse may also adversely affect outcomes because a high volume of output, and the spread of procedures or other types of care to less well-prepared provider teams, may lead to less competent implementation of those services. Poor implementation in turn harms patients and necessitates still more services to repair the damage; for instance, high rates of surgery bring with them an increased need to treat nosocomial infections. By doing many procedures and therefore having to do them in less well-prepared settings, problems with poor technical quality increase. Obviously, poor technical quality is most disturbing or troubling if the procedure was unnecessary in the first place. The patient will not benefit and is put at risk for serious harm. An inevitable conclusion seems to be that problems with poor quality in the United States would decline substantially if overuse could be controlled. Discouraging Overuse and Overcoming Underuse Second, to control overuse we must first define and detect it. The two scenarios in Table 13.2 show how complex this can be.

OCR for page 96
Medicare: New Directions in Quality Assurance TABLE 13.2 Two Scenarios for Considering "Necessity" (Probability of Net Benefit) of a Procedure Scenario Dimensions Heroic Procedure Perfectionist Procedure Patient status Very sick Very healthy Probability of benefit Low High Risk High Low Outcome likely with or without procedure Bad Good In the first scenario we have what is described as "the heroic procedure": the patient is very sick, there is a low probability of benefit, but the benefit could be very great. The procedure is much more risky in so sick a patient and in fact the outcome is likely to be bad whether or not the procedure is done. Is this overuse? The second scenario shows another potential type of overuse. The patient undergoing "the perfectionist procedure" is very healthy: there is high probability of some benefit but that benefit is very small because the patient is not very sick in the first place. The procedure is low risk in such a healthy patient. Do "perfectionist procedures" constitute overuse? A good outcome is likely, with or without the procedure. We can envisage, of course, even worse scenarios, such as when the patient is very sick and has a low probability of small benefit, so that a worse outcome is virtually certain from doing the procedure. When a new procedure is introduced—a common event these days—it is used first in the heroic mode. As providers get accustomed to the procedure, its riskiness declines. Providers then tend to recommend the procedure for less and less sick patients. As that happens, the probability of benefit to the patient fails; it may even fall below the probability of harm. This is likely to occur because new procedures are done at first in specialized settings: as the procedure moves out into less well-prepared settings, such as smaller hospitals with less experienced surgeons and staffs, the risk of harm rises. So then, appropriateness is not an invariable property of a procedure. It depends on the patient's characteristics and on where and by whom the procedure is done. Diffusion of new techniques to unsuitable patients and settings is a problem of overuse. Failure to diffuse techniques that are effective in improving patients' health is, however, a problem of underuse. Until recently, we commonly assumed that discouraging overuse could best be done by concentrating on areas where rates of performance of a procedure were high. Now we recognize that the task requires a more sophisticated approach because the rate of inappropriate use in areas with

OCR for page 96
Medicare: New Directions in Quality Assurance low usage rates is similar to that in areas with high usage rates. In other words, in both high- and low-use areas there is a mixture of over- and underuse occurring. A sophisticated approach to judging the appropriateness of a procedure is needed—one that adjusts for factors such as those in the scenarios shown in Table 13.2. It requires empirical knowledge of the benefits and risks of procedures in many different patient circumstances and many different care settings. Producing knowledge of this kind is the objective of the Agency for Health Care Policy and Research (AHCPR) Patient Outcome Research Teams (PORTs) and the Agency's outcomes, effectiveness, and practice guidelines program (MEDTEP). Knowledge of the net benefit of a procedure to the patient is not enough, however. The IOM report emphasizes the need to consult the patient's own preferences to incorporate judgments about the value of particular health states. This view is in keeping with the trend toward consumerism in our society. For health care financed by third parties, however, some way must be found to incorporate societal values into the decision to do a procedure or provide another type of service, because societal resources are being consumed. For instance, whatever the net benefit to the patient, should scarce resources be used for a heart transplant on a 90-year-old patient? If we say no, is that ageism and underuse, or is it prudent purchasing? How might the decision change if we knew the patient also had Alzheimer's disease? Judging appropriateness, then, is complex and is itself expensive. Given the observation made earlier that high-technology care is overused and low-technology care is underused, could we achieve more by changing the reimbursement methods that have encouraged this maldistribution of resources? The Hsiao resource-based relative value system for physician payment, for instance, is moving in this direction (Hsiao et al., 1988). The IOM report suggests that beneficiaries might not support this, since they did not complain about overuse. The shift of resources to low-technology care might, however, encourage doctors to take time to explain to their patients the dangers of overuse. The Medicare program may also be reluctant to encourage more use of any kind of service. To overcome this tendency to think only of short-term costs, we desperately need information on what the "net cost" impact would be if we dealt with underuse of low-technology care and of prevention and early intervention services. A shift to more of this kind of service could in the long run produce greater benefit at less cost than our current system of heavy use of high-technology procedures. Improving Technical Quality Third, since the committee report appeared, more evidence is available on the issue of poor technical quality. It comes from a study, called the

OCR for page 96
Medicare: New Directions in Quality Assurance Harvard Medical Practice Study (1990), of care in New York hospitals in 1984. Using a carefully standardized process of peer review, the investigators identified all adverse events, that is, unintended injuries to patients caused by medical management, and all negligent adverse events, that is, those that result from failure on the part of the physician to provide reasonably careful management or to reach the standard of care. Because these data came from representative sampling, the investigators could generalize their findings to the total population of hospital patients in New York State. The resulting estimates were startling. Of all hospital admissions, 3.7 percent had adverse events, and 1 percent had negligent adverse events. Adverse events were estimated to be primarily responsible for more than 13,000 deaths in 1984 in New York hospitals, and negligent adverse events for nearly 7,000 deaths. The authors also studied risk factors for adverse events and negligent adverse events. They found both to be statistically significantly more likely to occur in persons over the age of 65. The distribution of adverse events differed from that for negligent adverse events. For instance, simple adverse events occurred significantly more often in teaching hospitals where, of course, there are more complex procedures and sicker patients. The percentage of adverse events that were negligent was significantly higher, however, in nonteaching hospitals. This higher occurrence of negligence represents lower quality of care than in the teaching hospital environment. Technical quality in hospitals is clearly an important problem. Efforts must be directed to preventing each and every adverse event, irrespective of whether an individual provider's negligence was the primary cause. CONCLUSIONS There is sufficient information to conclude that deficiencies in quality of care impose a substantial burden of harm. This challenges providers, purchasers, and patients to commit themselves to quality improvement. How should this be done? Table 13.3 lists some recommendations for action. First, curbing overuse would also help with the problem of poor technical quality. We must recognize that new procedures are constantly being developed in centers of excellence and then diffusing to settings where the ratio of benefit and harm to patients is quite different. We should stop being surprised by that. We need a system for managing the introduction of new technologies and procedures. This would include determining the indications for performing the procedure and specifications about how and by whom the procedure can be safely performed. This would be a step toward ensuring that the probability of harm to the patient is not greater than the probability of benefit whenever and wherever the procedure is used. Second, it is time to pay attention to the potential long-term negative

OCR for page 96
Medicare: New Directions in Quality Assurance TABLE 13.3 Recommendations for Action to Improve Quality of Care 1. Curb overuse, which will help also to limit harm from poor technical quality. 2. Adopt policies for coverage and reimbursement that promote cost-effective prevention and early intervention services. 3. Manage health programs and the introduction of new health care technologies to improve technical quality. 4. Stimulate continuous quality improvement by health care providers and organizations. 5. Intervene for physicians and hospitals with serious quality problems. consequences of underuse. There is great danger that the Medicare program will make bad decisions about allocation of resources because it lacks information about whether better coverage might in the end reduce costs. Shortsighted decisionmaking is likely to produce worse quality at higher cost in the long run. Third, we need a direct campaign against poor quality. This will require a commitment to quality management as a personal responsibility of every health care provider, manager, and worker. Senior clinicians and managers must provide the leadership to harness these individual efforts to the common goal of continuous improvement. What should the federal government do about quality assurance in the Medicare program? Two major thrusts of the IOM committee's recommendations are reflected in the fourth and fifth points in Table 13.3. Fourth, the Health Care Financing Administration (HCFA) should become an active partner in promoting and facilitating continuous quality improvement among all physicians, nonphysician health professionals, and health care organizations. This is a change from an earlier mode of operation in which HCFA adopted a punitive and adversarial approach toward health care providers. It is gratifying to hear HCFA announce that it had already and independently adopted this positive and facilitative role (Morford, 1991). This new approach requires that HCFA hold providers responsible for their own quality improvement. It is important to emphasize this because some advocates interpret any external pressure to hold providers accountable as counterproductive to internal quality improvement. It is doubtful, however, whether American companies would have embarked upon continuous quality improvement if their customers had not started buying Japanese products. There is an external stimulus in the free market—customer preference—and that stimulus has encouraged American industries to change their ways. In health care, since the end users, that is, patients, cannot fully judge the quality of the care they receive, the proposed new Medicare Quality Review Organizations are needed to fill that role.

OCR for page 96
Medicare: New Directions in Quality Assurance Fifth, the IOM committee recommends that HCFA retain its role of preventing harm to patients from providers who are not maintaining acceptable standards of quality. The PRO evidence suggests that a relatively small number of physicians contribute disproportionately to problems in overuse, underuse, and poor technical quality. The Harvard Medical Practice Study suggests that the same is true for hospitals. Incompetent providers should receive special attention. The challenge for the federal government is for all its agencies that become involved with seriously substandard providers to cooperate with one another. The evidence from the IOM report is clear: there is a need to ''manage for quality'' in the Medicare program. It justifies a major increase in effort in the next decade. REFERENCES Chassin, M.R., Kosecoff, J., Winslow, C.M., et al. Does Inappropriate Use Explain Geographic Variations in the Use of Health Care Services? A Study of Three Procedures. Journal of the American Medical Association 258:2533-2537, 1987. Harvard Medical Practice Study. 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, 1990. Hsiao, W.C., Brown, P., Yntema, D., et al. Estimating Physicians' Work for a Resource-Based Relative Value Scale. New England Journal of Medicine 319:835-841, 1988. Institute of Medicine. Medicare: A Strategy for Quality Assurance. Lohr, K.N., ed. Washington, D.C.: National Academy Press, 1990. (See especially Volume I, Chapter 7.) Leape, L.L., Park, R.E., Solomon, D.H., et al. Does Inappropriate Use Explain Small-Area Variations in the Use of Health Care Services? Journal of the American Medical Association 263:669-672, 1990. Mills, D.H., ed. California Medical Association and California Hospital Association. Report on the Medical Insurance Feasibility Study. San Francisco, Calif.: Sutter, 1977. Morford, T.G. An Administration Response to the Institute of Medicine Report From the Health Care Financing Administration. Pp. 179-185 in Medicare: New Directions in Quality Assurance. Donaldson, M.S., Harris-Wehling, J., and Lohr, K.N., eds. Washington, D.C.: National Academy Press, 1991.