NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Academy of Engineering (US) and Institute of Medicine (US) Committee on Engineering and the Health Care System; Reid PP, Compton WD, Grossman JH, et al., editors. Building a Better Delivery System: A New Engineering/Health Care Partnership. Washington (DC): National Academies Press (US); 2005.

Cover of Building a Better Delivery System

Building a Better Delivery System: A New Engineering/Health Care Partnership.

Show details

Bridging the Quality Chasm

David Lawrence

Kaiser Foundation Health Plan

This presentation focuses on the management of scientific and technological breakthroughs as they are made available to the health care delivery system—specifically on whether health care has kept pace with innovation by moving them into practice safely and responsibly. There is a substantial amount of overuse, misuse, and underuse of available science and technologies in the health care system—regardless of geography, type of payment, or when and where physicians were trained. To address this problem, the Institute of Medicine (IOM) undertook two studies, To Err Is Human, published in 2000, which focused on safety issues, and Crossing the Quality Chasm, published in 2001, which focused on quality issues. Both reports highlight the symptoms of a broken system. Both reports concluded that there is a mismatch between the rate and quantity of scientific and technological innovations and the ability of the health care system to use them safely and responsibly.

Wide variations in quality were documented as far back as 1975 in a small-area variation analysis by John Wennberg, M.D., and then in a variety of other studies across the country in the last 30 years (O'Connor et al., 1999; Wennberg, 1999). Recent safety studies, primarily but not exclusively studies by Lucien Leape and his colleagues at Harvard, identified a variety of medical errors that result in morbidity and mortality caused not because of physician malfeasance but because of system errors (Brennan et al., 1991; Leape et al., 2002; Thomas et al., 2000). The number of hospital deaths from these errors range from 30,000 to 80,000 per year. At this point, we have no understanding and little documentation of the number of errors in the ambulatory setting. Some early estimates in the United Kingdom and the United States have been published (Bubin, 1999; Fischer et al., 1997; Weingart et al., 2000). The total number of deaths attributable to errors in the health care system we think could be as high as 150,000 or even 200,000 per year.

Another measure was published by Barbara Starfield in an article in Journal of the American Medical Association in 2000. Dr. Starfield looked at the whole question of system-related deaths for all reasons, including errors. She concluded that 200,000 to 250,000 deaths per year were attributable to system-related causes, of which error is the most notable (Starfield, 2000). Starfield also made interesting comparisons between our system and others in terms of a variety of health outcomes. She concluded, as have many others, that although we spend an enormous amount on health care and lead the world in scientific innovation and technology, the results in terms of improved health do not match the level of investment.

There are also other symptoms of poor quality in the health care system. One of them has to do with responsiveness. In the Picker Institute studies of patient assessments of their health care experiences, about three-quarters of those surveyed indicated that their experiences with the health care system had led them to conclude that it was a “nightmare” to navigate (Picker Institute, 2000). They identified duplication, lack of communication, conflicting points of view about what should be done, and lack of understanding about what the science suggested. In short, the system is fragmented, fractured, and not patient-centric.

Finally, there is the cost of poor quality care, which has interesting implications for innovation. Between $.30 and $.40 of every dollar spent on health care is spent on the costs of poor quality. This extraordinary number represents slightly more than a half-trillion dollars a year. A vast amount of money is wasted on overuse, underuse, misuse, duplication, system failures, unnecessary repetition, poor communication, and inefficiency.

In this respect, the experience of General Electric Company and others in more tightly managed and highly organized manufacturing systems may be instructive. Companies often find substantial opportunities for improvement in the cost performance of the system by using quality-improvement tools. The health care system does not even come close to a well organized, systematically designed system like a production or manufacturing system.

Our experience in Kaiser Permanente suggests that these numbers are not exaggerations. When we look for ways to improve the organization and delivery of care, we often find that substantial improvements can be made in the underlying cost performance of the organization. In fact, the premise on which we compete is that we can drive costs down by improving quality.

There are five major problems with the health care system. The first is that most of the scientific and technological breakthroughs that have occurred since World War II have not simplified the task of taking care of patients. In fact, they have made it more complex. Here are some examples:

  • As we entered the 1950s, there were about 10 to 12 categories of health care professionals in the United States. Today, there are more than 220 categories of health care professionals.
  • Right after World War II, there were about six to eight—depending on how you counted them—specialties in medicine. Today, there are more than a hundred.
  • In 1970, there were approximately 100 published randomized control trials (RCTs) in the American medical literature. In 1999 alone, almost 10,000 RCTs were published. Half of RCTs published in the United States have appeared in the last five years.

Science and technology have certainly contributed to growing complexity in medicine—increasing the number of people involved, increasing the number of categories of people involved, raising expectations about what can be done to treat people, and increasing the amount of science and technology that must be managed. Largely as a result of advances in science and technology, the medical care system is far more complex today in terms of the number of institutions and types of health care practitioners than it was in 1950.

Second, the health care system, or nonsystem, has grown enormously over the last 50 years but has failed to keep the patient and the patient's family at the center of the enterprise. It is small wonder that people identify the system as a nightmare to navigate. It is not a patient-centered system.

How could the complexity of a system be significantly decreased? First, we could create a highly sophisticated production-design or manufacturing-design process to handle the complexity. Then an investment could be made in an information technology infrastructure. Next, we could create flow systems to manage the support activities required to carry out these processes, retain people, and set new standards of quality.

In medicine, we have done very little of this. Physicians are still trained on the principle of individual, professional autonomy, even though, in reality, they do not work in autonomous situations at all. Production design is a foreign word. In fact, it is considered almost sacrilegious to talk about production design in medicine. To many practitioners medicine is a religion, not a science. Therefore, the tools of production design have not been applied in the units where patients get care.

The third issue is that it is extraordinarily difficult to scale up medical care delivery. There are few examples of integrated care across ambulatory, inpatient, hospice, and home settings. Only a few systems enable us to capture capital and reinvest capital in the delivery system infrastructure. With 80 percent of physicians practicing in groups of fewer than 10, medicine remains largely a single interaction between a patient and a doctor. In reality, although the patient-doctor interaction remains absolutely essential, the enterprise itself now involves a much more complex set of interactions.

Except for the Veterans Health Administration, Kaiser Permanente, which has 10,000 physicians, is the largest health care delivery system. The next largest may be the Mayo Clinic. Most others are small, regional players on the delivery system side. Until there are more scaled-up enterprises, it will be difficult to collect and reinvest enough capital to build and support the production capability essential to the delivery of the science and technology that innovators are creating for us.

Fourth, our public policy environment is structured to inhibit the reshaping of the medical care delivery system. For example, in Wisconsin there are 27 licensed categories of health care professionals, each with its own board of practice. Medicine should be about removing boundaries so that people can flow seamlessly among a variety of practitioners, based on what the technology requires and what the patient needs. Yet regulatory and license-based silos create barriers between professionals. These barriers must be broken down to create teams and to deliver integrated care. This can be done, but only with great effort. The licensing system is designed to protect the interests of particular professional groups in medicine, not to further the delivery of integrated care.

On the reimbursement side, the fee-for-service system is designed to reward individual acts by individual clinicians. Our current reimbursement system does not support integrated delivery capabilities. Crossing the Quality Chasm called for experimenting with a variety of reimbursement approaches to determine which ones would stimulate the creation of integrated delivery capabilities—prepayment, perhaps, or capitation or other approaches. The fact remains that the classic fee-for-service system is a barrier to the development of collaborative medicine.

The final issue identified in Crossing the Quality Chasm is that information technology is not being used in the delivery system the way one would expect for such an information-rich industry. It is estimated that less than 2 percent of total revenues in health care is being invested in information technology infrastructure. Much more is being invested on the health insurance side, but investment on the delivery system side is much lower than in other industries or in the medical technologies industry. Finding capital, either by aggregating organizations to generate capital or by other means, is a major issue. As the system stands, we cannot make innovations in health care delivery that match the complexity of the science. A physician trying to keep up with 10,000 RCTs in a year cannot practice evidence-based medicine without an information technology decision-support system. It is simply beyond the capacity of the individual to keep up.

Now let's turn to opportunities for innovation, using the problem areas as the focal points. The first priority for innovation is to improve the ways patients can connect with the medical care delivery system. Innovations in monitoring, diagnosis, and treatment technologies will enable patients to self-manage, or at least communicate on a regular, ongoing basis, with the health care system. It makes no sense to continue to invest heavily in the bricks and mortar of classic delivery systems when there are other vehicles for taking care of patients in a far more responsive, patient-centric way. So giving patients the tools and creating bridges between the patient and the delivery system is one focus for innovation.

For example, one of the many promising innovations is the ability to test whether Coumadin is operating at therapeutic levels; this can be done by the patient using a hand-held testing device. A device for testing blood sugar is another. These and many other devices will substantially improve the connection between the patient and the system and put more capability in the hands of the patient. These innovations will also decrease our dependence on brick and mortar solutions for the delivery system.

The second major area for innovation is translating the tools used in the manufacturing and production of goods and services into a language that applies to health care. I would argue that the delivery of medical care today is the most complex production challenge on the planet. Think about what is involved in running a hospital with about 250 beds—a wide array of diagnoses, a multitude of judgments being made by teams of professionals interacting with patients, and all of the support production that makes this happen hour after hour, 24 hours a day, seven days a week. This is an extraordinarily complex production challenge.

The third area of innovation involves organizational design or scaling. It has proven to be extremely difficult to create sufficient scale systemwide to produce the necessary capital, systems, and training capabilities. Scaling up has been done successfully in certain health care settings, for example, hospitals, nursing homes, laboratories, and pharmacies, but it has proven to be extremely difficult to create any kind of organizational scale for building integrated delivery capabilities.

Last, innovation could come through interventions at the national policy level in regulation, reimbursement, and, possibly, the financing of the information technology infrastructure in medicine. Given the current organization of the health care system, the financing of the information technology infrastructure may exceed the capacity of the private marketplace. Perhaps we will have to create the medical equivalent of the Superfund for environmental cleanup to build the information technology infrastructure for the health care delivery system. This infrastructure involves more than electronic medical records, which simply capture and move information to support decision making. A robust infrastructure would incorporate analytic tools that would enable epidemiological studies of disease. Without this infrastructure, it is extremely difficult to test whether or not microproduction units are working well and whether we are getting anywhere with the larger organizational challenges facing us.

In closing, the message of Crossing the Quality Chasm should be taken to heart. The mismatch between the pace and scope of innovation in medical science and technology and innovation in the delivery system has created a chasm, which is aggravated by shifting demographics and the shifting of the disease burden from acute to chronic care. The complexity that both add to the task of taking care of patients has not been matched by equivalent sophistication in the delivery system.


  1. Brennan TA, Leape LL, Laird NM, Hebert L, Localio AR, Lawthers AG, Newhouse JP, Weiler PC, Hiatt HH. Incidence of adverse events and negligence in hospitalized patients. Results of the Harvard Medical Practice Study I. New England Journal of Medicine. 1991;324(3):370–376. [PubMed: 1987460]
  2. Bubin C. Mistakes happen: improved processes mean fewer mistakes. Ambulatory Outreach. 1999 Fall;:23–26. [PubMed: 10621195]
  3. Fischer GF, Munro AP, Goldman EB. Adverse events in primary care identified from a risk-management database. Journal of Family Practice. 1997;45(1):40–46. [PubMed: 9228913]
  4. IOM (Institute of Medicine) To Err Is Human: Building a Safer Health System. In: Kohn LT, Corrigan JM, Donaldson MS, editors. Washington, D.C: National Academy Press; 2000.
  5. IOM. Washington, D.C: National Academy Press; 2001. Crossing the Quality Chasm: A New Health System for the 21st Century.
  6. Leape LL, Berwick DM, Bates DW. What practices will most improve safety?: evidence-based medicine meets patient safety. Journal of the American Medical Association. 2002;288(4):501–507. [PubMed: 12132984]
  7. O'Connor GT, Quinton HB, Traven ND, Ramunno LI, Dodds TA, Marciniak TA, Wennberg JE. Geographic variation in the treatment of acute myocardial infarction: the Cooperative Cardiovascular Project. Journal of the American Medical Association. 1999;281(7):627–633. [PubMed: 10029124]
  8. Picker Institute. A Report by the Picker Institute for the American Hospital Association. Boston, Mass: Picker Institute; 2000. Eye on Patients.
  9. Starfield B. Is U.S. health care really the best in the world? Journal of the American Medical Association. 2000;284(4):483–485. [PubMed: 10904513]
  10. Thomas EJ, Studdert DM, Burstin HR, Orav EJ, Zeena T, Williams EJ, Howard KM, Weiler PC, Brennan TA. Incidence and types of adverse events and negligent care in Utah and Colorado. Medical Care. 2000;38(3):261–271. [PubMed: 10718351]
  11. Weingart SN, Wilson RM, Gibberd RW, Harrison B. Epidemiology of medical error. British Medical Journal. 2000;320(7237):774–777. [PMC free article: PMC1117772] [PubMed: 10720365]
  12. Wennberg JE. Understanding geographic variations in health care delivery. New England Journal of Medicine. 1999;340(1):52–53. [PubMed: 9878647]
Copyright © 2005, National Academy of Sciences.
Bookshelf ID: NBK22849
PubReader format: click here to try


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (4.4M)

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to pubmed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...