Improving Quality in a Changing Health Care Industry
The State of Health Care Quality: How Good Is Care?
Across the Nation, skilled practitioners and high-caliber institutions help to maintain or restore Americans' health and ability to function. While most Americans receive high- quality health care, too many patients receive substandard care. Quality problems have been identified, including (1) avoidable errors, (2) underuse of health care services, (3) overuse of services, and (4) variation in services. These shortcomings endanger the health and lives of all patients, add costs to the health care system, and reduce productivity. Efforts by public and private sector stakeholders have helped to identify problems and begin the process of finding solutions. What is needed now is a national commitment to measure, improve, and maintain quality care for all citizens.
The Best of Care, The Worst of Care
Every day, millions of Americans have access to the best quality health care available in the world. Skilled practitioners, high-caliber institutions, and top-flight researchers continue to advance the science and practice of health care. Yet many Americans increasingly are concerned about the quality of the care they receive and the ability of the health care industry to provide high-quality care on a consistent basis.
Exhaustive research documents the fact that today, in America, there is no guarantee that any individual will receive high-quality care for any particular health problem. The health care industry is plagued with overutilization of services, underutilization of services, and errors in health care practice.
These quality problems are not new, nor are they unique to any particular type of arrangements through which health care is financed and delivered. The term "managed care" denotes a wide spectrum of health care arrangements, including health maintenance organizations, preferred provider organizations, and point-of-service arrangements. Both the best and the worst health care our system has to offer can be found in managed care plans, as it can in traditional fee-for-service (or indemnity) arrangements (Miller and Luft, 1997).
Key indicators of the quality problems we face include:
- Avoidable errors. Many Americans are injured during the course of their treatment, and some die prematurely as a result. A study of 30,121 randomly selected records from patients treated in New York hospitals found that adverse events occurred in 3.7 percent of hospitalizations and adverse events attributable to negligence in 1 percent of hospitalizations (Brennan et al., 1991). About 70.5 percent of the adverse events gave rise to disability lasting less than 6 months, 2.6 percent caused permanently disabling injuries, and 13.6 percent led to death. A national study found that from 1983 to 1993, deaths due to medication errors rose more than twofold with 7,391 deaths attributed to medication errors in 1993 alone (Phillips et al., 1998).
- Underutilization of services. Millions of people do not receive care they need and suffer needless complications that add to health care costs and reduce productivity. For example, a study of Medicare patients with myocardial infarction found that only 21 percent of eligible patients received beta blockers, and that the mortality rate among recipients was 43 percent less than that for nonrecipients (Soumerai et al., 1997). An estimated 18,000 people die each year from heart attacks because they did not receive effective interventions (Chassin, 1997).
- Overuse of services. Millions of Americans receive health care services that are unnecessary, increase costs, and often endanger their health. For example, an analysis of hysterectomies performed by seven health plans estimated that one in six was inappropriate (Bernstein et al., 1993)
- Variation in services. There is a continuing pattern of wide variation in health care practice, including regional variations and small-area variations (Wennberg and McAndrew, 1998). This is a clear indicator that the practice of health care has not caught up with the science of health care to ensure evidence-based practice in the United States.
Despite this growing body of evidence, the American health care industry still does not have the ability to systematically measure and report on the quality of health care that is delivered to patients. What is known, however, indicates that there are substantial areas for improvement in the quality of care.
The lack of comprehensive information on the quality of American health care is unacceptable in an industry that accounts for more than $1 trillion in annual expenditures and comprises nearly one-seventh of the U.S. economy.
These problems with the quality of American health care present a clear and compelling argument for national action. Recent advances in the science of medicine, the measurement of quality, and the public and private commitment to accountability in the health care industry provide the Nation with a tremendous opportunity to improve the quality of health care in the United States. To achieve this necessary progress, we will need a national commitment to quality improvement, beginning with leadership from the highest levels of government, industry, labor, and the health care professions.
Evidence of Quality Problems
Quality of care has been defined by the Institute of Medicine as "the degree to which health services for individuals and populations increase the likelihood of desired health care outcomes and are consistent with current professional knowledge" (Lohr, 1990). Researchers have documented a tremendous amount of unevenness in the quality of health care services rendered in the United States. These problems fall into four distinct categories: avoidable errors, underuse of services, overuse of services, and variation in practice.
Evidence of Errors in Health Services
A review of the literature provides estimates of unacceptably high error rates in the United States (Leape, 1994). One study of 30,121 randomly selected records from 51 randomly selected hospitals in New York State found that adverse events occurred in 3.7 percent of hospitalizations and 27.6 percent of the adverse events were due to negligence (Brennan et al., 1991).
There are a variety of types of error that occur in the delivery of health services, including missed diagnoses, errors in the interpretation of laboratory or imaging studies, medication administration or prescribing errors, surgical errors, and errors in the care furnished by doctors, nurses, and other health care professionals.
- Missed or delayed diagnoses can significantly affect patient outcomes. A study in a rural trauma center found that about 3 percent of patients experienced a delay in diagnosis (Aaland and Smith, 1996).
- Errors in the interpretation of laboratory tests can contribute to missed or delayed diagnoses. One study reported that 10 percent to 30 percent of test results were inappropriately classified as normal based on rescreening reviews (Wilbur, 1997).
- Surgical pathology errors can contribute to missed diagnoses. One study found major errors (those likely to affect patient care management) in 1.2 percent of the pathology cases examined prospectively (Lind et al., 1995).
- Medication errors are a major example of misuse of health care services. One study conducted in a teaching hospital found about four errors per 1,000 medication orders; 70 percent of these errors were classified as those with the potential for serious adverse outcomes (Lesar, Briceland, and Stein, 1997). The most common type of error is related to knowledge and the application of that knowledge to the use of the medication.
Another study found that antibiotics, cardio-vascular agents, gastrointestinal agents, and narcotics were the most common medications involved in errors (Lesar, Lomaestro, and Polh, 1997). A study that examined the preventability of adverse reactions to medications found that about 28 percent were preventable overall and 42 percent of the life-threatening events were preventable (Bates et al., 1995). Studies of medication errors have evaluated the processes by which errors occur. One such study in pediatrics found lower rates of error associated with verbal orders (2.6 per 1,000) compared with handwritten or computer-entered orders (8.5 and 6.3 per 1,000, respectively) (West et al., 1994).
Preventable deaths represent another view into the prevalence of errors in practice. An assessment of deaths from stroke, pneumonia, or heart attack in 12 hospitals indicated that 14 percent could have been prevented if adequate diagnosis and treatment had been provided (Dubois and Brook, 1988). An evaluation of fetal and maternal outcomes following emergency cesarean section found that 60 percent of the infant deaths (5 infants out of 20 that had fetal heart tones) could have been prevented with earlier diagnoses and by performing cesarean sections (Morris et al., 1996).
Evidence of Underuse of Services
Underuse of health care services occurs when there is evidence that a patient did not receive a service or procedure whose benefits exceeded its risks. Underuse may occur in the provision of preventive, acute, and chronic care. When underuse occurs in preventive care, opportunities to prevent the development of disease are forgone. In acute care, underuse may result in excess days in bed or the development of serious complications. And underuse in the provision of chronic care services may have serious consequences for functioning and quality of life and may result in premature death. Americans who have no health insurance or who are underinsured experience particularly high rates of underuse of all three types of services. Further, underuse of services plays a role in the large and widening gap in health status observed between racial minorities and other Americans (go to Chapter 8).
Preventive care. A classic example of underuse of services in the United States is childhood immunizations. Immunizations have virtually eliminated the incidence of polio, diphtheria, tetanus, and whooping cough from the set of infectious diseases of childhood. Vaccines for other conditions, such as haemophilus influenzae type B, measles, mumps, rubella, chicken pox, and hepatitis type B, offer the same potential. Yet a nationally representative survey of parents in 1995 estimated that only 76 percent of children had received the appropriate set of immunizations by 18 months of age (CDC, 1997). Recent data from the Health Plan Employer Data and Information Set (HEDIS) indicate that only 66 percent of children enrolled in 330 managed care plans had received appropriate immunizations by age 2 (NCQA, 1997). The range among individual health plans was from 20 percent to 95 percent.
Among adults, the two key immunizations are influenza vaccines, recommended annually for those over age 65 and for those with selected chronic health problems, and pneumococcal vaccines, recommended once for persons over age 65 and for those at increased risk of serious complications from pneumonia. A nationally representative survey of adults age 65 and older conducted in 1993 found that only 52 percent had received an annual influenza vaccine and only 28 percent had received a pneumococcal vaccine (CDC, 1995a).
Another type of preventive service is routine screening of persons without symptoms to detect the presence of disease at an early stage. Studies have found room for improvement in screening rates. For instance, a nationally representative sample of women age 50 and older found that 45 percent had a mammography in the previous year (CDC, 1993a). A survey of women age 18 and older found that 67 percent had received a pap smear in the previous 3 years (CDC, 1995b). Finally, a national study of persons 40 and older found that just 14 percent of men and 27 percent of women had received a fecal occult blood test to screen for colorectal cancer in the past year (CDC, 1995b). More recent data from National Committee for Quality Assurance show that managed care plan enrollees obtained screening services at higher rates than these national averages (NCQA, 1997).
Another type of preventive care is characterized by counseling or other interventions to prevent behaviors that are known risk factors for the development of serious chronic conditions. Smoking causes 19 percent of deaths annually (McGinnis and Foege, 1993), and advice from a practitioner to quit smoking has been demonstrated to be an effective intervention (USPSTF, 1996). Yet a survey of smokers age 18 and older indicated that 37 percent had been advised to quit smoking in the past year by a health professional (CDC, 1993b), again with significantly higher rates of such advice among smokers in managed care plans (NCQA, 1997).
Acute care. A leading indicator of the quality of acute care is the proportion of pregnant women initiating prenatal care in the first trimester. In 1992, about 78 percent of women nationally initiated prenatal care in the first trimester (PHS, 1995). The rate was 85 percent among women enrolled in managed care plans reporting HEDIS data (NCQA, 1997).
Care for heart attacks provides another important example of underuse of potentially effective care:
- In a study of Medicare patients in four States who had experienced heart attacks, 61 percent received aspirin within the first 2 days of hospitalization; 83 percent received aspirin at some point during the hospitalization; and 77 percent received aspirin at discharge (Krumholz et al., 1995; Ellerbeck et al., 1995).
- Thrombolytic therapy, which can reduce mortality among heart attack patients by as much as half if administered within a few hours of symptom onset, was given to 70 percent of ideal candidates (Soumerai et al., 1997).
- Among Medicare patients in New Jersey, only 21 percent of heart attack patients received beta blockers within 90 days of their heart attack; those receiving beta blockers had a 43 percent lower rate of mortality than those who did not (Soumerai et al., 1997).
Chronic care. Assessing quality of care for chronic conditions tends to be complex because multiple processes are required, stratification by severity may be necessary, the need for clinical detail from health records is greater, and the settings of care are varied. However, failure to receive high-quality care for chronic conditions may have the greatest potential consequences for the health and functioning of the population.
Chronic illness has replaced infectious diseases as the number one source of death in the world. Approaches for treating chronic conditions are growing, from high-tech treatments such as coronary artery bypass graft surgery or angioplasty to low-tech solutions such as home care and assistive devices. Yet knowledge about the quality of care for chronic illness or the efficacy of many treatments for producing desirable outcomes is in its infancy. For example, hip fractures and strokes -- the two most common disabling conditions in older people -- often result in profound functional losses. Among hip fracture patients, between 55 percent and 75 percent lose their ability to perform basic activities of daily living. Among stroke survivors, about 40 percent have moderate functional impairments 6 months later and 10 to 15 percent remain severely disabled. Despite such impacts, the effectiveness of different rehabilitation options for restoring function and preventing long-term nursing home placement is largely unknown (Kramer et al., 1997).
In other cases, knowledge of the most efficacious treatment of chronic conditions does exist, yet treatment approaches are unevenly applied. Care for persons with diabetes provides an example. In 1989, 49 percent of people with diabetes in the United States had received a dilated eye examination in the past year and 66 percent had received one in the past 2 years (Brechner et al., 1993). A study of Medicare patients in Alabama, Iowa, and Maryland reported that 46 percent had an eye examination and 16 percent had received a glycosylated hemoglobin test, a marker for glucose control in diabetes (Weiner et al., 1995).
Similar findings have been demonstrated for mental health conditions. One study found that only 20 percent to 30 percent of general medicine patients with depression were prescribed antidepressant medications; among those prescribed medication, 30 percent received a subtherapeutic dose (Wells et al., 1994). A recent consensus panel on the undertreatment of depression concluded that "there is overwhelming evidence that individuals with depression are being seriously undertreated" (Hirschfeld et al., 1997).
Evidence of Overuse of Services
Overuse occurs when a health service is provided that poses substantially greater risks than potential benefits to the patient. Overuse of services is found in all areas of medicine and in all settings of care although the magnitude of overuse varies widely from procedure to procedure. Examples include:
- A study of appropriate use of coronary artery bypass graft (CABG) surgery among the Medicare population in the early 1980's found that about 14 percent of procedures were inappropriate and an additional 30 percent were of uncertain clinical value (Winslow et al., 1988). A more recent study in New York found that just 2 percent of CABG procedures were inappropriate and about 7 percent were of uncertain value (Leape et al., 1993).
- A study on the appropriateness of carotid endarterectomy found that 18 percent were inappropriate, 49 percent were of uncertain clinical value, and 33 percent were appropriate (Wong et al., 1997).
- A study of cataract surgery in 10 academic health centers found this procedure to be appropriate most of the time, with just 2 percent of procedures being judged clinically inappropriate (Tobacman et al., 1996).
- A study of appropriateness of hysterectomy use in different health plans found that about 16 percent of women underwent hysterectomy for reasons judged to be clinically inappropriate (Bernstein et al., 1993). Nationwide, over 510,000 hysterectomies were performed in 1994 (Agency for Health Care Policy and Research, 1997).
- One study found that 23 percent of requests to insert tympanostomy tubes in children with chronic ear infections were clinically inappropriate and an additional 36 percent were of questionable clinical utility (Kleinman et al., 1994).
Another significant area of overuse of health services lies in the use of medications. Treatment of respiratory infections (common cold, influenza, sore throat, sinus infections, bronchitis, and pneumonia) is the most common reason for Americans visiting a physician, representing about 10 percent of all visits among adults and 13 percent of visits for children (NCHS, 1994a; NCHS, 1994b). The cause of many upper respiratory infections is a virus, and antibiotics are not an effective treatment for viral illnesses. Yet several studies have reported high rates (30 to 70 percent) of inappropriate prescription of antibiotics for viral infections (Mainous et al., 1996; McCaig and Hughes, 1995; Dowell and Schwartz, 1997). Another study found that unnecessary prescriptions for nonsteroidal anti-inflammatory drugs were written in 42 percent of visits; this pattern of overprescribing can increase risk for hospitalization and death from gastrointestinal bleeding and perforation (Tamblyn et al., 1997).
The appropriateness of hospitalization has been examined in several different studies:
- One study of hospitalization rates in six communities found inappropriate admissions ranging from 10 percent to 35 percent (Siu et al., 1986).
- A study of internal medicine admissions in one hospital found 15 percent of admissions were inappropriate (Perneger et al., 1997).
- An examination of the appropriateness of admission for patients with community-acquired pneumonia found that 61 percent of admissions were medically inappropriate (Porath et al., 1996).
Evidence of Variation in Services
A clear sign of quality problems is the tremendous variation in the delivery of health care services, including sharp regional differences and significant variation even within small areas. Although some variation in services is expected due to differences in health status, available resources, and ambiguity in clinical knowledge, variation of the magnitude observed is not scientifically defensible (Chassin, 1997).
The first evidence of small-area variation in the rates of common surgical procedures was published in the early 1970's (Wennberg and Gittelsohn, 1973). A 1981 study of 123 medical and surgical procedures performed on Medicare beneficiaries in 13 metropolitan areas in the United States found threefold or greater differences for 54 percent of the procedures. A review of the literature in 1987 demonstrated variations at every level (e.g., regional, State, hospital area) and for a variety of utilization dimensions (e.g., length of stay, discharge rates, surgical procedure rates) (Paul-Shaheen et al., 1987).
Regional variation is quite apparent in the use of inpatient hospital services. In 1994, for example, overall hospital discharge rates were 49 percent higher in the Northeast than in the West and lengths of stay were 40 percent higher (Graves and Gillum, 1997). These variations also are apparent among specific conditions. Rates of hospital discharges for diabetes mellitus were 85 percent higher in the South than in the West and discharges for pneumonia were 76 percent higher in the South than in the West. Discharge rates for psychoses were 160 percent higher in the Northeast than in the West and 107 percent higher for asthma (Karuza et al., 1995).
Utilization rates also have been shown to vary within States and regions. A study of hysterectomy rates in Massachusetts between 1982 and 1989 found a fivefold variation (Haas et al., 1993). Hospitalization rates among children with asthma also varied, with children in Boston having a 3.8 percent risk of hospitalization compared with a 2.3 percent rate for children in New Haven, Connecticut. Researchers even found variations in utilization within Los Angeles County (Carlisle et al., 1995).
Variation in utilization is not limited to hospitals. A recent study of the use of hormone replacement therapy (HRT) found that HRT was prescribed in 13.7 percent of visits by women age 40 and older in the West compared to 3.9 percent of visits by women in the Northeast (Stafford et al., 1997). Women living in metropolitan areas were more likely to receive HRT than those residing in nonmetropolitan areas. A study of diabetes care among Medicare beneficiaries in three States found more than a twofold variation in care (Weiner et al., 1995).
Variations also have been demonstrated among individual practitioners. A study of the intensity of treatment provided by psychiatrists and psychologists found that the average time in face-to-face treatment per patient per month varied 244 percent among psychiatrists and 128 percent among psychologists nationally (Knesper et al., 1989). Cesarean section rates were shown to vary from 19.1 percent to 42.3 percent in a study of affluent women cared for by obstetricians practicing at a single community hospital (Goyert et al., 1989). A study of physician differences in the management of low back pain in a single managed care organization found variation in the frequency of prescribing pain medication and bed rest resulting in cost differentials of 79 percent between the high- and low-use groups (Von Korff et al., 1994).
Are Things Getting Worse?
In medicine, as in many other fields of study, as we get better at detecting problems, we find more of them. This is true both in clinical practice and in quality assessment. While we have made advances in our ability to detect problems, it also is quite likely that the rapidly expanding base of knowledge, coupled with changes in technology, is contributing to an ever-widening gap between the best care possible and the care that is actually delivered.
Every year brings advances in the interventions available to screen for diseases, prevent diseases from developing, make diagnoses, treat conditions, and monitor the progress of disease. For example:
- The reliability of fecal occult blood testing to screen for colorectal cancer has improved sufficiently for the U.S. Preventive Services Task Force to recommend regular screening for adults age 50 and older (PSTF, 1996).
- Immunizations for chicken pox and hepatitis B recently have been introduced.
- Thrombolytic therapy has had a major influence on the treatment of persons with heart attacks.
- Laparoscopic surgery has dramatically changed the functional impact of many procedures, such as cholescystectomy.
- Home glucose monitoring makes it possible for persons living with diabetes to more carefully monitor their treatment and make appropriate adjustments to maintain health.
While these innovations in medicine have significantly, and positively, affected the health of the population, they also pose a challenge for health care practitioners to keep their skills up-to-date. Just reading about advances (as compared to active training in or experience with new techniques) is a daunting task. For instance, according to an analysis of MEDLINE entries, the number of citations reporting on randomized controlled trials has increased from an average of 509 annually between 1975 and 1980 to 8,636 annually from 1993 through 1997 (McGlynn, 1998). Although no practitioner needs to follow advances across all areas of medicine and surgery, rapid expansion of knowledge is occurring even within specific areas. Randomized controlled trials published on coronary artery disease increased from about 6 per year between 1975 and 1980 to 149 per year between 1993 and 1997. For diabetes, the growth was from about 6 per year to about 176 per year in the same time periods. For asthma, citations increased from 13 to 200 annually. And, while there were no randomized trials reported in the literature on laser surgery between 1980 and 1985, about 30 were published per year between 1993 and 1997.
Randomized trials represent only one type of study report that practitioners are expected to track. Large observational studies also contribute new insights into the science of medicine. Considerable attention is being paid to methods for making information more readily available to practitioners, such as through the Internet and information systems within health plans. Meta-analysis, a technique that summarizes findings across multiple studies, is another mechanism that may assist practitioners in understanding the "bottom line" message from the literature. Only one citation classified as a meta-analysis appeared between 1985 and 1990, compared with 416 annually between 1993 and 1997.
There also has been an explosion in the availability of practice guidelines, which increasingly are viewed as mechanisms for communicating information about best practices. Between 1975 and 1980, about 1 publication annually was classified as a practice guideline, compared to 454 annually between 1993 and 1997. Initial resistance among practitioners to guidelines has given way to active participation in constructing such guidelines (Hoyt, 1997). The Directory of Clinical Practice Guidelines, published by the American Medical Association (AMA), currently lists 1,700 clinical practice guidelines. A survey of physician medical groups and independent practice associations found that 87 percent were developing or implementing guidelines to accomplish either cost containment or quality improvement goals (Fang et al., 1996).
The proliferation of guidelines and concerns about the scientific basis for guidelines in general led the AMA to create the Clinical Practice Guidelines Recognition Program, which will identify evidence-based guidelines meeting a set of criteria developed by a partnership among the AMA, national medical specialty societies, the Agency for Health Care Policy and Research (AHCPR), the American Hospital Association, and the Joint Commission on Accreditation of Healthcare Organizations. In addition, the AHCPR, the AMA, and the American Association of Health Plans are working jointly to develop a comprehensive, Internet-based source of clinical practice guidelines. The initiative, known as the National Guideline Clearinghouse (NGC), will make available a full range of current guidance on treatments for specific medical conditions. The NGC will make full-text clinical practice guidelines, abstracts, summaries, and comparisons available to every physician, health plan, provider, purchaser, and consumer with access to the Internet. It is anticipated that the NGC will be launched on the Internet in the fall of 1998.
While the "science" of health care advances, increasing attention is being paid to the "art" of health care as reflected in patient evaluations of their experiences with care. While there is a growing emphasis in health care on patient-provider partnership efforts, patients and their health care providers do not always agree on the importance of different elements of care. One study found differences between doctors and their patients on the importance of 73 of 125 elements of care (Laine et al., 1996). Patients ranked provision of information as the second most important aspect of care (behind clinical skill), whereas physicians ranked information sixth. Communication can affect health outcomes. A review of 21 studies found that in 16 a positive relationship between good doctor-patient communication and health outcomes was documented (Stewart, 1995). Outcomes that were positively influenced included emotional health, symptom resolution, function, clinical outcomes, and pain control.
Poor communication can lead to patient failure to adhere to recommended treatment regimens. A study in five urban teaching hospitals in the Northeast found that dissatisfaction with discharge instructions was a key factor that resulted in 12 percent of patients not filling their prescriptions (Thomas et al., 1996). A study conducted in Canada found that 20 percent of hospitalized patients reported that they were not told about important side effects of medications they were taking and 20 percent of patients undergoing tests indicated that results were not explained (Charles et al., 1994).
In summary, we know enough from the results of studies of variability, overutilization, underutilization, and errors to draw some conclusions about the level of quality. We know that there is a gap between best practices and actual practices. We know that even highly trained health professionals make mistakes from time to time. And we know that the likelihood that any one of us will get the best care possible varies considerably.
Quality Improvement Techniques
While a substantial amount is unknown about how to change behavior of patients and practitioners, we are learning more every day, and a wide range of effective techniques are already available to improve the delivery of health services. Much is being learned about effective techniques used in other industries. A key step in designing effective quality improvement interventions is identifying likely reasons for suboptimal performance and tailoring the interventions to address those deficiencies. Three main targets of intervention are the health professional, the patient, and the health service delivery system. Many interventions are directed at a combination of these targets.
One explanation for identified quality of care problems is that the health care practitioner has erroneous, outdated, or absent information or skills. Given the rapid increase in new information, this problem is hardly surprising. The traditional approach to keeping practitioners up-to-date is continuing professional education. A review of 99 education strategies designed to improve practitioner performance found that the traditional continuing medical education (CME) conference was not, by itself, an effective mechanism of change (Davis et al., 1995). More individualized outreach educational efforts involving leaders, often referred to as academic detailing, are more effective than traditional educational interventions (Soumerai and Avorn, 1990). A randomized trial in four hospitals in Massachusetts to improve the appropriateness of blood product utilization employed a combination of a departmental lecture, printed guidelines, and a 30-minute, one-on-one visit with each practitioner (Soumerai et al., 1993). The proportion of transfusions in compliance with guidelines 6 months after the intervention increased from 60 percent to 76 percent in the intervention group compared to a decrease from 60 percent to 56 percent in the control group. A trial designed to improve the likelihood that new practice guidelines for influenza immunizations would be adopted tested a small-group discussion and consensus process. Practitioners in the intervention group increased vaccination rates by an average of 34 percent compared to those in the control group (Karuza et al., 1995). Furthermore, all practitioners in the intervention group, compared to 54 percent in the control group, increased their individual vaccination rates.
Another method for improving care is structured encounter forms that serve to prompt practitioners to provide certain processes during single or multiple encounters. Prenatal care forms are one example. A randomized trial involving pediatric house staff found that the group using a structured form that included key developmental milestones had greater adherence to standards of developmental screening than the group without the form. Parent satisfaction also was higher (Zenni and Robinson, 1996).
Feedback to practitioners about their performance relative to expectations is another method that has been used to improve quality. New York State's Cardiac Reporting System is one of the best known examples of successful public feedback of results. Risk-adjusted mortality rates following coronary artery bypass graft surgery are published for individual hospitals and physicians in the State. Between 1989 and 1992, a 41 percent decrease in risk-adjusted mortality in the State was documented (Hannan et al., 1994). Private feedback to practitioners about performance is more common. For example, a combined strategies trial using an evidence-based protocol, an opinion leader, CME rounds, and feedback demonstrated a reduction in the use of superficial cultures to evaluate sepsis in neonates from 50.5 percent to 6.9 percent (Paes et al., 1994).
Patient refusal to participate in recommended interventions, such as routine screening, frequently is cited as a problem. Thus, some interventions have focused on patients, most commonly to enhance rates of preventive service use and to improve management of chronic conditions. In the review of the literature on changing practitioner performance, patient-mediated strategies were identified as one of the successful approaches. For example, a trial designed to improve the effectiveness of management for persons with depression included patient education materials (both video and written), along with increased intensity and frequency of visits and ongoing medication management. Patients with major depression in the intervention group were more compliant with medication dose and duration, rated the quality of care higher, believed that medication had helped their condition, and showed a significant reduction in symptoms and disease severity (Katon et al., 1995). Patients with minor depression also adhered more closely to medication regimens and believed that medication was helping them, but did not demonstrate significant improvement in symptoms, severity, or satisfaction relative to the control group.
Reminder systems have been used to help both patients and practitioners increase adherence to preventive service recommendations. A study of patient-directed methods for improving mammography rates compared three strategies: (1) mailed birthday card reminders; (2) a personalized letter from the medical director with educational materials enclosed; and (3) phone calls that included reminders, counseling, and appointment scheduling. Although overall rates of adherence were small, the most effective strategy was phone calls (28 percent), followed by birthday reminders (15 percent), and then mailed materials (9 percent) (Davis et al., 1997).
As with health care practitioners, consumer education strategies alone do not appear to be effective. For example, a randomized trial of a patient education program designed to reduce emergency room use for persons with asthma found no difference between the control and intervention groups in the first year following the intervention (Shields et al., 1990). A study of methods for improving influenza vaccination rates among the elderly combined community-based education with making the shots available for free. The educational intervention alone improved rates from about 41 percent to 54 percent; the addition of free shots increased rates from 41 percent to 64 percent (hospital-based) and 69 percent (physician-based) (Ives et al., 1994).
Perhaps the most important innovation in quality improvement has been the increased focus on system- related problems. Continuous quality improvement (CQI) and total quality management (TQM) programs have demonstrated the importance of examining multiple steps in the process of delivering care to identify opportunities for system redesign.
One starting point for quality improvement programs may be to examine the factors contributing to errors in processes or outcomes of care. For example, an assessment of factors contributing to errors in an intensive care unit found that human error was a factor in 55 percent of incidents and that violations of practice standards were a factor in 28 percent (Buckley et al., 1997). Both direct observation and monitoring systems were used to detect errors. A prospective observational study of the causes of adverse events in the hospital found that about 18 percent of patients experienced an adverse event and that 38 percent of these were caused by an individual, 16 percent were the result of systems factors, and 10 percent were due to administrative decisions (Andrews et al., 1997).
A related approach is to examine the processes by which certain interventions are provided, particularly those that require several steps to accomplish a goal. For example, staff in a health maintenance organization (HMO) examined the processes by which complete blood counts, mammograms, and surgical consultations were ordered and provided across nine clinics in the system. Variations in both the intent and successful implementation of steps in each of these interventions were noted; enhanced information systems were identified as one potential solution to some of the problems identified (Banks et al., 1995).
System redesign has been used frequently to improve the use of preventive services, particularly by reducing missed opportunities to provide the intervention. For example, to improve rates of mammography, medical assistants and licensed practical nurses were trained to determine if a woman visiting the clinic was due for a mammogram and, if so, were allowed to offer and order the test. The proportion of visits in which women were up-to-date on mammography screening increased in the intervention clinic from 68 percent to 77 percent compared with a 1 percentage point increase in one control clinic and a 2 percentage point decline in another control clinic (McCarthy et al., 1997).
Large-scale systems redesign has been shown to be effective in improving adherence to preventive service recommendations. For example, at Group Health of Puget Sound, a combination of strategies has demonstrated effectiveness. Eighty-nine percent of enrolled children have complete immunizations, the proportion of smokers has declined from 25 percent to 17 percent, bicycle helmet use among children has increased from 4 percent to 48 percent, and head injuries for this group decreased 67 percent (Thompson et al., 1995).
Quality improvement programs are designed around a team model, and the effective functioning of the team is likely to be critical to the success of the enterprise. A study of quality management teams in Veterans Administration hospitals found that commitment to total quality management by the facility and practitioners along with the number of teams involved in active quality improvement projects explained a substantial portion of the variation in quality improvement realized across the hospitals (Lammers et al., 1996). Inappropriate prescribing of medications for elderly outpatient veterans on multiple drug regimens was significantly reduced by adding a clinical pharmacist to the treatment team (28 percent decline in the intervention groups vs. 5 percent in the control group at 12 months) (Hanlon et al., 1996).
A National Commitment Is Needed
Quality of care in the United States varies considerably within and across communities, hospitals, practitioners, patients, delivery systems, geographic areas, and health problems. This variation is itself inconsistent -- meaning that the same groups do not always perform best or worst, making it difficult to develop programs to improve quality based on current knowledge. The real level of quality in most areas of the country is unknown, but what we do know suggests that we can no longer afford the luxury of ignorance.
What is needed to move forward? Significant advances have been made in the methods of quality measurement so that it is possible to assess the quality of care for diverse conditions in a variety of settings. We need to implement such measurement activities systematically, and on a broad national scale. This will require investment in people, methods, and data systems. We also need to find mechanisms for making the lessons of quality improvement available to all participants in the health care industry. The quality of services can be improved, but doing so requires an understanding of the extent of the gap between standards and actual practice, the likely causes of the gap, effective tools for closing the gap, and a cultural change that values engaging in ongoing quality improvement.
To accomplish these vitally important aims, the United States needs a national commitment to the measurement, improvement, and maintenance of high-quality health care for all its citizens. The body of this report lays out a series of steps that must be taken by all sectors -- public and private -- to achieve the goal of a quality health care system for all Americans.
Aaland, Mary O., and Ken Smith, "Delayed Diagnosis in a Rural Trauma Center," Surgery 120(4):774-778, October 1996.
Agency for Health Care Policy and Research, Healthcare Cost and Utilization Project (HCUP-3). Statistics from the HCUP-3 Nationwide Inpatient Sample for 1994: Principal Procedures, AHCPR Pub. No. 97-0057, Rockville, MD, 1997.
Andrews, Lori B., Carol Stocking, Thomas Kriezek, et al., "An Alternative Strategy for Studying Adverse Events in Medical Care," Lancet 349:309-313, 1997.
Banks, Naomi J., R. Heather Palmer, Donald M. Berwick, and Paul Plsek, "Variability in Clinical Systems: Applying Modern Quality Control Methods to Health Care," Joint Commission Journal on Quality Improvement 21:407-419, 21, 1995.
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.
Bernstein, S.J., E.A. McGlynn, A.L. Siu, et al., "The Appropriateness of Hysterectomy. A Comparison of Care in Seven Health Plans. Health Maintenance Organization Quality of Care Consortium," Journal of the American Medical Association 269(18):2398-402, May 12, 1993.
Brechner, Ross, Catherine Cowie, Jean Howie, et al., "Ophthalmic Examination Among Adults with Diagnosed Diabetes Mellitus," Journal of the American Medical Association 270(14):1714-1718, October 13, 1993.
Brennan, Troyen A., Lucien L. Leape, Nan M. Laird, et al., "Incidence of Adverse Events and Negligence in Hospitalized Patients," New England Journal of Medicine 324(6):370-376, February 7, 1991.
Buckley, T.A., T.G. Short, Y.M. Rowbottom, and T.E. Oh, "Critical Incident Reporting in the Intensive Care Unit," Anaesthesia 52(5):403-409, 1997.
Carlisle, David M., B. Robert Valdez, Martin F. Shapiro, and Robert H. Brook, "Geographic Variation in Rates of Selected Surgical Procedures Within Los Angeles County," Health Services Research 30(1):27-42, April 1, 1995.
Centers for Disease Control, "Mammography and Clinical Breast Examinations for Women Aged 50 Years and Older -- Behavioral Risk Factor Surveillance System, 1992," Morbidity and Mortality Weekly Reports 42:737-741, 1993a.
Centers for Disease Control, "Physician and Other Health-Care Professional Counseling of Smokers to Quit -- United States, 1991," Morbidity and Mortality Weekly Reports 42:854-857, 1993b.
Centers for Disease Control, "Influenza and Pneumococcal Vaccination Coverage Levels Among Persons Age > 65 Years -- United States, 1973-1993," Morbidity and Morality Weekly Reports 44:506-515, 1995a.
Centers for Disease Control, "Trends in Cancer Screening -- United States, 1987 and 1992," Morbidity and Mortality Weekly Reports 45:57-61, 1995b.
Centers for Disease Control, "National, State, and Urban Area Vaccination Coverage Levels Among Children 19-35 Months -- United States, January to December, 1995," Morbidity and Mortality Weekly Reports 46:176-183, 1997.
Charles C., M. Gauld, L. Chambers, et al., "How Was Your Hospital Stay? Patients' Reports About Their Care in Canadian Hospitals," Canadian Medical Association Journal 150:1813-1822, 1994.
Chassin, Mark R., "Assessing Strategies For Quality Improvement," Health Affairs 16(3):151-161, May/June 1997.
Chassin, Mark R., Robert H. Brook, Rolla Edward Park, et al., "Variations in the Use of Medical and Surgical Services by the Medicare Population," New England Journal of Medicine 314(5):285-290, January 30, 1986.
Davis, David A., Mary Ann Thomson, Andrew D. Oxman, and R. Brian Haynes, "Changing Physician Performance: A Systematic Review of the Effect of Continuing Medical Education Strategies," Journal of the American Medical Association 274(9):700-705, September 6, 1995.
Davis, Nancy A., Esther Nash, Catherine Bailey, et al., "Evaluation of Three Methods for Improving Mammography Rates in a Managed Care Plan," American Journal of Preventive Medicine 13(4):298-302, July-August 1997.
Dowell, Scott, and Benjamin Schwartz, "Resistant Pneumococci: Protecting Patients Through Judicious Use of Antibiotics," American Family Physician 55(5):1647-1654, April 1997.
Dubois, Robert, and Robert H. Brook, "Preventable Deaths: Who, How Often, and Why?" Annals of Internal Medicine 109(7):582-589, October 1, 1988.
Ellerbeck, Edward, Steven Jencks, Martha Radford, et al., "Quality of Care for Medicare Patients With Acute Myocardial Infarction: A Four-State Pilot Project From the Cooperative Cardiovascular Project," Journal of the American Medical Association 273(19):1509-1514, May 17, 1995.
Fang, Edward, Brian S. Mittman, and Scott Weingarten, "Use of Clinical Practice Guidelines in Managed Care Physician Groups," Archives of Family Medicine 5(9):528-531, October 1996.
Goyert, Gregory L., Sidney F. Bottoms, Marjorie C. Treadwell, and Paul C. Nehra, "The Physician Factor in Cesarean Birth Rates," New England Journal of Medicine 320(11):706-709, March 16, 1989.
Graves, Edmund J., and B.S. Gillum, "National Hospital Discharge Survey: Annual Summary, 1994." Vital and Health Statistics 13(128):1-146, November 1997.
Haas, Susan, David Acker, Charles Donahue, and Martha E. Katz, "Variation in Hysterectomy Rates Across Small Geographic Areas Within Massachusetts," American Journal of Obstetrics and Gynecology 169(1):1-146, July 1993.
Hanlon, Joseph T., Morris Weinberger, Gregory P. Samsa, et al., "A Randomized Controlled Trial of a Clinical Pharmacist Intervention to Improve Inappropriate Prescribing in Elderly Outpatients with Polypharmacy," American Journal of Medicine 100(4):428-437, April 1996.
Hannan, Edward L., H. Kilburn, Michael Racz, et al., "Improving the Outcomes of Coronary Artery Bypass Graft Surgery in New York State," Journal of the American Medical Association 271(10):761-766, March 9, 1994.
Hirschfeld, Robert M., Martin B. Keller, Susan Panico, et al., "The National Depressive and Manic-Depressive Association Consensus Statement on the Undertreatment of Depression," Journal of the American Medical Association 277(4):333-340, January 22-29, 1997.
Hoyt, David B., "Clinical Practice Guidelines," American Journal of Surgery 173:32-34, 1997.
Ives, Diane G., Judith R. Lave, Neal D.Traven, and Lewis H. Kuller, "Impact of Medicare Reimbursement on Influenza Vaccination Rates in the Elderly," Preventive Medicine 23(2):134-141, March 1994.
Karuza, Jurgis, Evan Calkins, John Feather, et al., "Enhancing Physician Adoption of Practice Guidelines: Dissemination of Influenza Vaccination Guideline Using a Small Group Consensus Process," Archives of Internal Medicine 155(6):625-632, March 27, 1995.
Katon, Wayne, Michael Von Korff, Elizabeth Lin, et al., "Collaborative Management to Achieve Treatment Guidelines: Impact on Depression in Primary Care," Journal of the American Medical Association 273(13):1026-1031, April 5, 1995.
Kleinman, Lawrence, Jacqueline Kosecoff, Robert W. Dubois, and Robert H. Brook, "The Medical Appropriateness of Tympanostomy Tubes Proposed for Children Younger Than 16 Years in the United States," Journal of the American Medical Association 271(16):1250-1255, April 27, 1994.
Knesper D.J., B.E. Belcher, J.G. Cross, "A Market Analysis Comparing the Practices of Psychiatrists and Psychologists," Archives of General Psychiatry 46(4):305-314, April 1989.
Kramer, A.M., J.F. Steiner, R.E. Schlenker, et al., "Outcomes and Costs After Hip Fracture and Stroke: A Comparison of Rehabilitation Settings," Journal of the American Medical Association 277(5):396-404, February 5, 1997.
Krumholz, Harlan, Martha Radford, Edward Ellerbeck, et al., "Aspirin in the Treatment of Acute Myocardial Infarction in Elderly Medicare Beneficiaries: Patterns of Use and Outcomes," Circulation 92(10):2841-2847, November 15, 1995.
Laine, Christine, Frank Davidoff, Charles E. Lewis, et al., "Important Elements of Outpatient Care: A Comparison of Patients' and Physicians' Opinions," Annals of Internal Medicine 125(8):640-645, October 15, 1996.
Lammers, John C., Shan Cretin, Stuart Gilman, and Emeloy Calingo, "Total Quality Management in Hospitals: The Contributions of Commitment, Quality Councils, Teams, Budgets, and Training to Perceived Improvement at Veterans Health Administration Hospitals," Medical Care 34(5):463-478, May 1996.
Leape, Lucian L., "Error in Medicine," Journal of the American Medical Association 272(23):1851-1857, December 21, 1994.
Leape, Lucian L., Lee H. Hilborne, Rolland E. Park, et al., "The Appropriateness of Use of Coronary Artery Bypass Graft Surgery in New York State," Journal of the American Medical Association 269:753-760, 1993.
Lesar, Timothy S., L. Briceland, and D.S. Stein, "Factors Related to Errors in Medication Prescribing," Journal of the American Medical Association 277(4):312-317, January 22-29, 1997.
Lesar, Timothy, Ben M. Lomaestro, and Henry Pohl, "Medication-Prescribing Errors in a Teaching Hospital: A 9-Year Experience," Archives of Internal Medicine 157(14):1569-1576, July 28, 1997.
Lind, Anne C., Chhanda Bewtra, James C. Healy, and Kenneth L. Sims, "Prospective Peer Review in Surgical Pathology," American Journal of Clinical Pathology 104(5):560-566, November 1995.
Lohr, Katherine, ed., Medicare: A Strategy for Quality Assurance (Washington, DC: National Academy Press, 1990).
Mainous, Arch, William Hueston, and Jonathan Clark, "Antibiotics and Upper Respiratory Infection: Do Some Folks Think There Is a Cure for the Common Cold?" Journal of Family Practice 42(4):357-361, April 1996.
McCaig, Linda, and James Hughes, "Trends in Antimicrobial Drug Prescribing Among Office-Based Physicians in the United States," Journal of the American Medical Association 273(3):214-219, January 18, 1995.
McCarthy, Bruce D., Marianne U. Yood, Mary Beth Bolton, et al., "Redesigning Primary Care Processes to Improve the Offering of Mammography. The Use of Clinic Protocols by Nonphysicians," Journal of General Internal Medicine 12(6):357-363, June 1997.
McGinnis, Michael J., and William H. Foege, "Actual Causes of Death in the United States," Journal of the American Medical Association 270(18):2207-2212, November 10, 1993.
McGlynn, Elizabeth A., written statement to the President's Advisory Commission on Consumer Protection and Quality in the Health Care Industry, January 28, 1998.
Miller, Robert H., and Harold S. Luft, "Does Managed Care Lead to Better or Worse Quality of Care," Health Affairs 16(5):7-25, September/October 1997.
Morris, John A. Jr., Todd J. Rosenbower, Gregory J. Jurkovich, et al., "Infant Survival After Cesarean Section for Trauma," Annals of Surgery 223(5):481-488, May 1996.
National Center for Health Statistics, Current Estimates from the National Health Interview Survey, 1993 (Hyattsville, MD: U.S. Department of Health and Human Services, 1994a).
National Center for Health Statistics, National Ambulatory Medical Care Survey, 1991 Summary (Hyattsville, MD: U.S. Department of Health and Human Services, 1994b).
National Committee for Quality Assurance, The State of Managed Care Quality (Washington, DC, 1997).
Paes, Bosco A., Anjana Modi, and Ray Dunmore, "Changing Physicians' Behavior Using Combined Strategies and an Evidence-Based Protocol," Archives of Pediatric and Adolescent Medicine 148(12):1277-1280, December 1994.
Paul-Shaheen, Pamela, Jane D. Clark, and Daniel Williams, "Small Area Analysis: A Review and Analysis of the North American Literature," Journal of Health Politics, Policy, and Law 12:741-809, 1987.
Perneger, Thomas V., Pierre Chopard, Francois P. Sarasin, et al., "Risk Factors for a Medically Inappropriate Admission to a Department of Internal Medicine," Archives of Internal Medicine 157(13):1495-1500, July 14, 1997.
Phillips, David P., Nicholas Christenfeld, and Laura M. Glynn, "Increase in U.S. Medication-Error Deaths Between 1983 and 1993." Lancet 351(9103), February 1998.
Porath, Avi, Fransisco Schlaeffer, and David Lieberman, "Appropriateness of Hospitalization of Patients with Community-Acquired Pneumonia," Annals of Emergency Medicine 27(2):176-183, February 1996.
Public Health Service, Department of Health and Human Services, Healthy People 2000 Review 1994 (Hyattsville, MD: Public Health Service, 1995).
Shields, Marc C., Kenneth W. Griffin, and Wylie L. McNabb, "The Effect of a Patient Education Program on Emergency Room Use for Inner-City Children With Asthma," American Journal of Public Health 80(1):36-38, January 1990.
Siu, Albert L., Frank A. Sonnenberg, Willard G. Manning, et al., "Inappropriate Use of Hospitals in a Randomized Trial of Health Insurance Plans," New England Journal of Medicine 315(20):1259-1266, November 13, 1986.
Soumerai, Stephen B., and Jerome Avorn, "Principles of Educational Outreach ('Academic Detailing') to Improve Clinical Decision Making," Journal of the American Medical Association 263:549-556, 1990.
Soumerai, Stephen, Thomas McLaughlin, Ellen Hertzmark, et al., "Adverse Outcomes of Underuse of Beta Blockers in Elderly Survivors of Acute Myocardial Infarction," Journal of the American Medical Association 277(2):115-121, January 8, 1997.
Soumerai, Stephen, Suzanne Salem-Schatz, Jerome Avorn, et al., "A Controlled Trial of Educational Outreach to Improve Blood Transfusion Practice," Journal of the American Medical Association 270(8):961-966, August 25, 1993.
Stafford, Randall S., Demet Saglam, Nancyanne Causino, and David Blumenthal, "Low Rates of Hormone Replacement in Visits to United States Primary Care Physicians," American Journal of Obstetrics and Gynecology 177(2):381-387, August 1997.
Stewart, M. Anita, "Effective Physician-Patient Communication and Health Outcomes: A Review," Canadian Medical Association Journal 152(9):1423-1433, May 1, 1995.
Tamblyn, Robin, Leora Berkson, W. Dale Dauphinee, et al., "Unnecessary Prescribing of NSAIDs and the Management of NSAID-Related Gastropathy in Medical Practice," Annals of Internal Medicine 127(6):429-438, September 15, 1997.
Thomas, Eric J., Helen R. Burstin, Anne C. O'Neil, et al., "Patient Noncompliance With Medical Advice After the Emergency Department Visit," Annals of Emergency Medicine 27(1):49-55, January 1996.
Thompson, Robert S., Stephen H. Taplin, Timothy A. McAfee, et al., "Primary and Secondary Prevention Services in Clinical Practice. Twenty Years' Experience in Development, Implementation and Evaluation," Journal of the American Medical Association 273(14):1130-1135, April 12, 1995.
Tobacman, Joanne K., Paul Lee, Bridget Zimmerman, et al., "Assessment of Appropriateness of Cataract Surgery at Ten Academic Medical Centers in 1990," Ophthalmology 103(2):207-215, February 1996.
U.S. Preventive Services Task Force, Department of Health and Human Services, Guide to Clinical Preventive Services, 2nd Edition (Baltimore: Williams and Wilkins, 1996).
Von Korff, Michael, W. Barlow, Daniel Cherkin, and Richard A. Deyo, "Effects of Practice Style in Managing Back Pain," Annals of Internal Medicine 121(3):187-195, August 1, 1994.
Weiner, Jonathan P., Stephen T. Parente, Deborah W. Garnick, et al., "Variation in Office Based Quality: A Claims Based Profile of Care Provided to Medicare Patients With Diabetes," Journal of the American Medical Association 273(19):1503-1508, May 17, 1995.
Wells, Kenneth B., William Katon, William R. Rogers, and Patricia Camp, "Use of Minor Tranquilizers and Antidepressant Medications by Depressed Outpatients: Results from the Medical Outcomes Study," American Journal of Psychiatry 151(5):694-700, May 1994.
Wennberg, John E., and Megan McAndrew, eds., The Dartmouth Atlas of Health Care in the United States (Chicago: American Hospital Publishing, Inc., 1998).
Wennberg, John E., and Alan Gittelsohn, "Small Area Variations in Health Care Delivery," Science 142:1102-1108, 1973.
West, D.W., S. Levine, G. Magram, et al., "Pediatric Medication Order Error Rates Related to the Mode of Order Transmission," Archives of Pediatric and Adolescent Medicine 148(12):1322-1326, December 1994.
Wilbur, David C., "False Negatives in Focused Rescreening of Papincolaou Smears: How Frequently Are 'Abnormal' Cells Detected in Retrospective Review of Smears Preceding Cancer or High-Grade Intraepithelial Neoplasia?" Archives of Pathology and Laboratory Medicine 121(3):273-276, March 1997.
Winslow, Constance M., Jacqueline B. Kosecoff, Mark R. Chassin, et al., "The Appropriateness of Performing Coronary Artery Bypass Graft Surgery," Journal of the American Medical Association 260(4):505-509, July 22-29, 1988.
Wong, John H., Max J. Findlay, and Maria E. Suarez-Almazor, "Regional Performance of Carotid Endarterectomy: Appropriateness, Outcomes, and Risk Factors for Complications," Stroke 28:891-898, 1997.
Zenni, Elissa A., and Thomas N. Robinson, "Effects of Structured Encounter Forms on Pediatric House Staff Knowledge, Parent Satisfaction, and Quality of Care. A Randomized Controlled Trial," Archives of Pediatric and Adolescent Medicine 150(9):975-980, September 1996.