Skip to Navigation
University of Pittsburgh
Print This Page Print this pages

May 12, 2005

How to solve health system’s ills

“The U.S. health system is the poster child for underachievement. We know better. Why don’t we do better?” said a national health care system expert here last week.

Although the United States can boast about having the potential for the best health care available anywhere in the world, data on gaps and disparities in the quality of health care delivery in this country make the U.S. system a paradox, characterized by high variance, conflicting incentives and inherent fragmentation, according to Stephen M. Shortell, dean and Blue Cross of California Distinguished Professor of Health Policy and Management at the University of California-Berkeley School of Public Health.

Shortell delivered the annual Anne C. Sonis Memorial Lecture, which is sponsored by the Graduate School of Public Health’s Department of Health Policy and Management, the Center for Research on Health Care and the Sonis family.

Shortell, whose research focuses on evaluation of quality improvement initiatives and on implementation of evidence-based medicine practices in physician organizations, spoke on “What It Will Really Take to Improve Our Nation’s Health System.”

“I’d like to concentrate on three things,” Shortell said. “First, the problems and challenges we face; second, to suggest an approach that might work, which is not to say it’s the only approach or even the best one, and third, to present a little of the evidence for why this approach might work, that there might be a way out of our problems.”

Recently compiled data on health care delivery indicate the following widespread problems, Shortell said:

• Racial disparities are severe. Blacks received poorer quality of care than whites for some two-thirds of quality measures; Hispanics received lower quality care than non-Hispanic whites for 50 percent of quality measures.

• Physician organizations do not manage chronic illness well, as evidenced by data that say less than 50 percent of recommended care management processes — disease registries, guidelines, case management, feedback to physicians, patient self-management instruction— are utilized, and that only 1 percent of physician organizations nationally use all recommended processes for chronic illness patients with asthma, congestive heart failure or diabetes.

This failure of physician practices translates into an estimated 42,000-80,000 avoidable deaths per year; $1.8 billion in annual avoidable hospital costs, and 66.5 million avoidable sick days in the U.S. workforce annually, according to 2004 data compiled by the National Committee for Quality Assurance, Shortell said.

• There is great variance of treatment for patients in the last six months of life as measured by average hospital stays, average days in the intensive care unit and the number of physician visits.

“What we see is a widespread variance in the culture of health systems, the way medicine is practiced, the way things are organized in institutions, the way patents’ families are involved, and in different philosophies about end-of-life care, of death and dying,” Shortell said.

• There is inherent fragmentation in the ability to transfer knowledge into practice, he said. “Physicians comply with evidence-based guidelines for at least 80 percent of their patients in only eight of 306 U.S. hospital regions,” Shortell said, indicating a wide gap between knowledge and practice.

“In 1968 the clinical trial was completed for the flu vaccine, yet today there is only a 64 percent use rate,” he pointed out. Similar gaps exist in the application of other proven successful screenings and exams, such as thrombolytic therapy, approved in 1971, with a 20 percent usage rate, and diabetic eye exams, developed in 1981, with a 48 percent usage rate.

“These are things we know work and that are recommended processes, but are not being applied in practice 30-40 years later,” he said.

There are four common hypotheses about the root of the U.S. health system’s problems, Shortell said. “Is it lack of money? I sure hope not, since we’re spending $1.5 trillion, 15 percent of our gross domestic product, annually. Is it lack of people? We do have a shortage of nurses, for example, and in a few of the other health professions. But overall, I don’t think that’s the big problem,” he said.

“Is it lack of technology? I doubt it. We’re excellent at producing technology. Is it a lack of ideas? No, we have plenty of those. The real answer is that there is a lack of an organizing principle that links together the money, the people, the technology and the ideas,” Shortell maintained.

That principle should have agreed upon common aims for patient care: safe, effective, efficient, timely, personalized and equitable, Shortell said.

“The key to this organizing principle is alignment, on the one hand of external incentives and, on the other, of internal organizational capabilities, across four levels of change for improving quality — the individual, the group or team, the organization and the larger political and economic environment,” Shortell said.

“This is a mammothly complex challenge, as you know, and I’m oversimplifying. But I’m not trying to be glib. You can unpack those two words — incentives and capabilities — and it comes down to giving a focused set of incentives that will lead all of us in the same direction: the buyer community, the health plan community, the payers community, the patient community, the suppliers community.”

Shortell pointed to six health system “redesign imperatives”:

• Redesigned care processes.

• Effective use of information technologies.

• Knowledge and skills management.

• Development of effective teams.

• Coordination of care across patient conditions, services and settings over time.

• Use of performance and outcome measurement for continuous quality improvement and accountability.

“We need to make these changes at the organizational level. But how can we do all this with this perverse regulatory payment environment we have that pays me more money if I don’t do a particularly good job?” Shortell asked.

For example, if a patient has to return for another physician visit, or another range of tests is ordered, or a change in medication is required — things that might have been avoided in the first place — these have financial consequences throughout the health system, he said.

“We need a policy restraint, not a practice restraint, to think about interventions that take into account the interdependence of all levels simultaneously,” he said. “That does not mean you make one big massive change. But it does mean you think through the immediate and adjacent levels: What does the team intervention mean for the organization? What does it mean for the individual patient? We need to think across multi-levels simultaneously and align incentives accordingly.”

Shortell maintained that these changes are less likely to be consumer-driven; instead, they have to come from within the organizations themselves.

“I am less optimistic in the short run that consumers are going to buy into this. There is little evidence that patients take quality care data, which is difficult to understand, and ‘vote with their feet’ and go to high-quality providers,” Shortell said. “In addition to getting that data out there, it’s better to engage the profession and the organizations to make these changes and in the long run consumers will not only benefit but eventually will follow.”

In addition to ensuring professional competence and qualifications, organizations should focus on developing a patient-centered culture, building information technology, improving process initiatives, promoting team effectiveness and maximizing the ability to partner, he said.

“It’s amazing how much good work, best practices and so forth, is not spread around. We simply don’t share knowledge, and that’s true even within organizations.”

As for evidence that this alignment principle can work to improve overall health care delivery, Shortell cited research he’s done with colleagues at UC-Berkeley comparing medical groups against a chronic care management index that includes measures of self-management, linkages to community resources, delivery system re-design and decision support tools.

“We split off the data we have on about 1,000 medical groups into three chunks,” he said: A set of 12 groups of pre-paid multi-specialty delivery systems with 100 or more doctors that have aligned their systems across the measures in the index; a set of 468 medical groups, also with 100 or more physicians and with similar community resources, but without the same alignment across the index measures, and a set of all 1,028 medical groups in the study.

The 12 aligned pre-paid multi-specialty delivery system groups scored twice as high on the chronic care management index as both the 468 groups of 100 or more physicians and the set of all 1,028 groups studied.

A second study compared the same three groups against a clinical information technology index that measured laboratory findings, medications prescribed, radiology findings, progress notes, medication ordering reminders and drug interaction information.

The aligned group of 12 scored about four times as high as both the 468 other groups with 100 or more docs and the set of all medical groups in the study.

Implementation of the alignment principle is not a matter of snapping one’s fingers, Shortell said. “We need agreement on a standard measurement set for a range of clinical conditions. We need to focus on what to measure and how to instruct [practitioners] to do that. We need to understand incentives and how and when to apply them, because they can lead to unintended consequences,” he pointed out.

“At what level should incentives start? Should we be rewarding improvement over time? What if a group has no financial resources to use in order to improve, do we penalize them for not improving? We need to learn how to reach smaller, and rural, practices. We need to develop a truly patient-centered culture.”

But the U.S. health system will not improve without changes, Shortell concluded. “As John Kenneth Galbraith said, ‘When faced with a choice between changing and proving there’s no need to change, most people get busy on the proof.’

“Those of you who remember your history, when [Hernando] Cortez landed in the New World, to keep his crew from leaving, he burned the ships. There was no turning back.

So the question is: Where’s the ‘burning platform’ for changing the health care system? Who’s going to light that flame?”

—Peter Hart

Leave a Reply