C O N T E N T S

Advisory of February 27, 2002, announcing the hearing

WITNESSES

American Hospital Association, and Danbury Hospital, Matthew Miller, M.D.

American Nurses Association, Mary Foley

Institute for Healthcare Improvement, Donald M. Berwick, M.D.

Michigan Department of Veterans' Affairs, Veterans Health Administration, James P. Bagian, M.D.

Pittsburgh Regional Healthcare Initiative, and Jewish Healthcare Foundation of Pittsburgh, Karen Wolk Feinstein

American Academy of Family Physicians, Leawood, KS, statement

HEALTH QUALITY AND MEDICAL ERRORS

Thursday, March 7, 2002

House of Representatives,
Committee on Ways and Means,
Subcommittee on Health,
Washington, DC.

The Subcommittee met, pursuant to notice, at 11:10 a.m., in room 1100 Longworth House Office Building, Hon. Nancy L. Johnson (Chairman of the Subcommittee) presiding.

[The advisory announcing the hearing follows:]

Chairman JOHNSON. Good morning, everyone. The hearing will come to order, and while there are a number of other subcommittee hearings in progress, we are looking forward to a lot of our Members joining us over the course of the hearing.

Unfortunately, medical errors are an endemic problem that permeates our health system. According to the Institute of Medicine (IOM), preventable medical errors are the eighth leading cause of death in America, accounting for at least 44,000 and as many as 98,000 mortalities in hospitals each year. The number of injured is even greater. In addition, IOM also estimates that medical errors in hospitals cost between $17 and $29 billion each year.

This is shocking and unacceptable to caregivers and to patients alike. As medicine has become more complex, systems have not developed commensurate with the caregiving challenge. Patients rely on the system to improve their lives, not endanger them. And health professionals work long and hard hours after years of intensive, costly training, to help people, not to hurt them.

Not only do avoidable patient errors harm patients, they drive up health costs by requiring expensive medical interventions to correct subsequent problems. For example, adverse drug events and interactions in hospitals are prevalent and costly. According to estimates from Cardinal Health, Inc., there were more than 625,000 preventable adverse drug events in hospitals in the year 2000, at a cost of $2.9 billion. Reducing just half of those errors through innovations such as electronic prescribing could save billions and patient lives.

I am hopeful that this Committee can produce bipartisan legislation soon that will address this problem in a thoughtful, effective way, honestly recognizing the lack of malice that is causing such errors and the cost of the systems necessary to address them. Systemic approaches to reduce medical errors are endorsed by academicians and practitioners, and have the potential to dramatically improve health quality and patient safety while reducing costs.

The best way to reduce medical errors is to learn from our mistakes, and I would like to just share a short passage from the IOM study. One of the report's main conclusions is that the majority of medical errors do not result from individual recklessness or the actions of a particular group.

This is not a "bad apple" problem. More commonly, errors are caused by faulty systems, processes, and conditions that lead people to make mistakes or fail to prevent them. Thus, mistakes can best be prevented by designing the health system at all levels to make it safer, to make it harder for people do so something wrong and easier for them to do things right.

Reporting adverse events allows us to gain insight into those specific events and to identify patterns or common factors creating errors, changing practices, and preventing future errors. Individuals will not disclose their mistakes if they think they are going to be embarrassed or harmed. Appropriate legal and confidentiality protections are an essential element of a reporting system that works.

One common cause of errors is illegible prescriptions, because patients are often given the wrong prescription because the doctor's handwriting is misread. Electronic prescribing has the potential to dramatically improve patient compliance with drug regimens, reduce adverse drug interactions, and improve patient health and reduce costs.

Today we will hear from the Michigan Department of Veterans' Affairs (VA). They have adopted a successful reporting system that has improved patient safety. Also, the Pittsburgh Regional Health Care Initiative will discuss their success in implementing an ambitious zero tolerance policy for medical errors, and I might add that we do have testimony from Secretary O'Neill on this important system and issue, one that he has been very involved in in the past. Finally, we will hear from an academic and providers on their perspective to improve health outcomes and patient safety.

I look forward to the testimony. The solutions to this problem are not partisan. Working together, we have the opportunity to literally save lives. I look forward to working with my colleagues on the Committee to address this issue promptly.

With that, I would like to recognize Mr. Stark.

[The opening statement of Chairman Johnson follows:]

Mr. STARK. Madam Chair, thank you, and I want to direct this commentary not across the aisle but at the Congress in general and certainly at the medical care delivery system in particular.

Once again we are having a hearing. The IOM report came out in 1999. We have had hearings. We had a similar hearing to this identical hearing in 2000. Here we are, 2 years later, another hearing. I have introduced a bill. I am not sure it is any good, but there have been bills introduced to begin the process of setting in place a system for systematically reviewing medical errors and setting up procedures to prevent them.

Now, you are going to hear from some people that we should have a voluntary industry effort. That is just crap. The Joint Commission on Accreditation of Healthcare Organizations or JCAHO isn't worth a pound of salt. They have never punished a hospital that I know of in the existence of their so-called inspections of the hospitals. They are paid by the hospital.

So, unless we are willing to sit down--we went through this same thing with needle sticks. The hospitals fought us on safe needles like Billy-be-damned until some hospital got sued for $8 or $9 million and then they said, "Oh, maybe we should use safe needles." They won't move unless we move and make it a requirement.

Tort reform will come up. I think that is nonsense. If somebody cuts off the wrong leg, you don't need a system to tell you that you have been harmed, and you are going to sue. If you are too dumb and don't know it is the wrong leg, maybe you have got other problems.

I mean, there are tort suits in the medical system--it is very dangerous going to the hospital, and it is one of the few industries in this Nation that doesn't have a systematic, detailed requirement for keeping records, for protecting whistle-blowers, for doing all these things. It is going to take a major mind-set change that, if--in other words, you have got to get over, "You don't squeal on your buddies. That is not considered nice in the delivery of medical care."

Well, we have to make it very comfortable for people to do that. As you say, confidentiality is important. There is a whole host of things, but I just hope, Madam Chair, that you will get a bill out. We ought to be able to get a bill like this through on suspension if the hospitals don't fight us. All right? I mean, who would think, if we came under suspension with a bill that wanted to deal with quality of care, which is not mentioned much in Medicare, that we would have any resistance unless it comes from the very people who are committing errors?

Now, it is not pointing a finger and saying it is because they are not diligent. It is just, it is not required, and until we require it, I think this is one of the areas that we are going to have to say, I hope you will agree with me, that we can't just talk it into conformity. We have to mandate it, and that, I know it sounds like a regulation, but I hope we can move to begin to do it and put it in law, and we will have regular oversight perhaps.

So I thank you for getting the ball rolling, once again, and I hope that the next meeting we have will be a markup. Thank you.

Chairman JOHNSON. Thank you, Mr. Stark.

I do hope that this hearing will shed light on the two most difficult issues in this area, and particularly the first issue is what prevented legislative action following the hearing 2 years ago. There are two very difficult issues.

One is, how do you promote the kind of reporting that gives you a handle on all the little, tiny things that happen, that could have happened better? If you had known about them you would have seen a pattern or you would have seen an opportunity to put into place a system that would have prevented big errors. And how do you differentiate the need for protection in that system from the system that our malpractice laws serve? I think those are two different systems, and the rules in my estimation have to be different, but that is what we want you to talk about. Do they have to be different, and how different?

The second issue that wasn't as big an issue 2 years ago is, what is the cost of this? What do institutions have to be able to fund, to manage, and to invest permanently in? Not just the one-time cost. What are the ongoing, systemic costs of changing the system of health care delivery to enable us to put in place structural approaches that will minimize human error?

So those are the two issues. I hope we all approach this hearing with an open mind on them, because last time our inability to particularly resolve the issues around the reporters did prevent legislation. I hope that won't be the case this time, because there is simply too much opportunity for us to not only protect patients but also to better serve people who come out of medical training with enormous debts, who are in this business to provide care for people who desperately need it.

Mr. STARK. How about tying it to updates in the Medicare reimbursement, Madam Chair? That might get some action.

Chairman JOHNSON. You and I have long disagreed on the value of mechanical Federal formulas, and if what is happening to physicians isn't evidence that mechanical formulas cause problems, I don't know what is.

Okay, let's start. Dr. Bagian, Director of the National Center of Patient Safety, the Michigan Department of Veterans' Affairs, from Ann Arbor. Thank you for being with us and for sharing your experience in these areas.

STATEMENT OF JAMES P. BAGIAN, M.D., P.E., DIRECTOR, NATIONAL CENTER FOR PATIENT SAFETY, VETERANS HEALTH ADMINISTRATION, MICHIGAN DEPARTMENT OF VETERANS' AFFAIRS, ANN ARBOR, MICHIGAN

Dr. BAGIAN. Thank you, Madam Chairwoman, for the opportunity to come here, and Members of the Committee. I thought I would share the experience of a little bit of the Michigan VA and try to answer some of the questions you just posed in the initial remarks here.

We certainly agree that safety is the foundation upon which quality is built. You can't begin to say you have a quality system if you don't provide safety to the patients. The Michigan VA has been very interested and active in patient safety since 1997, almost 2 ½ years before the IOM report you mentioned in your initial comments. We have worked very hard on that, and I will share some of those lessons.

My history is slightly different. I began as an engineer, became an astronaut, and spent much time in aviation as well as medicine, and then came to this, which I felt was a very good chance to bring systems thinking and prevention to medicine, which we hadn't always done in very systematic ways. I think there are several points we need to recognize here.

Though often people in shorthand talk about errors, medical errors, and in fact if you look at the patient safety handbook in the Michigan VA, we don't use the term "errors" at all. It doesn't appear, because errors are just one subset of things you want to prevent. There are many things that occur that people would not view as an error but yet cause harm to the patient, and it is harm to the patient we want to prevent.

They are human beings. Whenever there are human beings in play, there will be errors made. No one is perfect. If we require them to be perfect in order not to harm a patient, that is a losing bet, guaranteed.

In aviation, for example, you have more than one engine on a plane that flies across the ocean. The reason is not that we try to build engines that are unreliable, but we recognize one might fail and yet we don't want the plane to crash. In medicine, on the other hand, we have single-engine planes, and we figure they have got to be perfect, and we know they always aren't. So we think we need to look at that.

We have to understand that it is systems solutions. As you pointed out, people don't come to work to hurt a patient. That is not the issue. The issue is when people that are well-meaning make mistakes or don't use things appropriately. There is a whole number of things. It is not just errors.

How do we figure out how to prevent those things and put systems in place to really study the causes? It is not so simple as the typical line that you hear, for instance, on a medication problem: "Tell the nurse to be more careful." That is not really a Nobel Prize winning strategy. We have to wonder why did this happen.

I think we need to understand that the systems, there are a number of accountability systems that have been in place and still are, and they are appropriate. There is appropriate use of the accountability systems, but you need a learning system, one by which we can learn when there is a problem. We share it very quickly with others, so they can learn. If we don't do that, we cause everyone to pay the same price for their own individual learning, and they don't share, so each one of us learns from the injury of a patient. That is a terrible way to do this.

You talked about cost. I can tell you in the Michigan VA we have looked at this as we put the system in place which is in place in all our facilities. We see the cost in aggregate, in full time equivalents (FTE), if you want to look at it that way, is about 1.1 FTE a year to do full root cause analysis and corrective actions in a large-size hospital.

That is peanuts. That is really nothing when you look--and I can show you a number of examples, I won't waste the short time I have here right now--but just small corrections often cause avoidable, just in costs, operating costs alone, often $100,000 a year, which more than pay for the cost of that. So to say it is a cost, it really isn't. It is a cost avoidance strategy, but you have to pay some money to make some money.

It is not blame and shame. It is you really want to learn from these things and set up a way to do it, and that is understanding what is blameworthy and what is not. We don't say our system is a blame-free system. Please understand what I mean by this.

What we did is, we say there are some actions that are blameworthy, that is what we call intentionally unsafe acts, those acts that are criminal acts, acts that involve substance or alcohol abuse on the part of the care provider, or acts that were intentionally unsafe. That is, the person knew it was unsafe and did it anyway.

I think we all would agree they should be in a system that is discoverable, that is available to a plaintiff's attorney or anybody else who wants to look at it. On the other hand, innocent mistakes, if you will, do not deserve that same treatment. They should be confidential so people can feel free to share them locally and globally so people can learn, and we have that ability within the Michigan VA system. We think that is important.

Since we put this system in place, we have seen a 30-fold increase in reports, we have seen a 900-fold increase in close call reporting, 900-fold. That is 90,000 percent, which means close calls are those bad things that could have happened but didn't. You want to learn by those. You don't want to wait until somebody is hurt before we decide to do something different. Learn from the close call. We do that, and I have numerous examples where concrete things that had global impact have been detected through that.

We also found in the old systems, 50 percent of the cases that would be looked at, people thought were not preventable because we didn't give them good systems tools to look at this, good human factors oriented tools. Now 100 percent come back with prevention, preventive strategies. That is huge. That means people are thinking differently. We think that is important.

To get there, though, we had to deal with certain barriers, and the barriers were, people were worried about punitive action, and that is from their perspective, not from the boss's perspective. That is not just, are you going to get fired, are you going to be suspended? That is are you going to be publicly humiliated, embarrassed?

All those things count. They are real roadblocks to sharing. We have to have protection so people understand that when they are not in the blameworthy category, that they can share, because the blameworthy ones won't tell you anyway. You know, if people did it deliberately, they are never going to tell you, so you have to have your other accountability mechanisms to deal with that.

Aviation learned this a long time ago with the Aviation Safety Reporting System (ASRS), because aviation had this problem. There was a crash not 40 miles from where we are sitting right now, where 92 souls perished, all because what was learned 6 weeks prior to that by another crew wasn't shared because of fear of punitive action. The Federal Aviation Administration, FAA, reacted decisively, started, had National Aeronautics and Space Administration (NASA) start the ASRS, and now they have a de-identified, that is not anonymous, reporting to NASA, which then shares those vulnerabilities with the community so they can be addressed.

We have an internal system at the Michigan VA, but we also have an external system we have set up that NASA actually runs for us, and it allows people to report. Now, I will say that that is something that can be used anywhere. We cannot right now, because of confidentiality issues, open it to outside members, but Kaiser Permanente for instance has approached us and wants to be part of it. The U.S. Department of Defense, DOD, does. University of Michigan does, Vanderbilt, and on and on and on.

Until there is legislation which allows people to be able to share these things between institutions within a State, and more importantly, across States, they can't do that, because to share that means it is all discoverable, which means people will not report. So in the safety realm we need protection if you want people to truly be able to share nationally. The Patient Safety Reporting System, PSRS, that has already been put in place with NASA, what we have done, could easily be opened up to others with the stroke of a pen, but without confidentiality protection it cannot happen. So we think this is extremely important and necessary and would really put it forward so people can share and learn.

I guess I would leave you with, until now the way we have worked in medicine is experience is the best teacher, but it is also the most costly teacher, and the people who pay our tuition are the patients. That is a terrible way to do business. We should really learn from each others' experience and not cause us to learn through harm done to others. Thank you.

[The prepared statement of Dr. Bagian follows:]

Chairman JOHNSON. Thanks very much, Dr. Bagian. We will be working closely with you as we use your experience, amongst others, to help guide us in making these difficult decisions. I thought the difference you drew between accountability and learning systems is really key here.

I am going to go slightly out of order for a variety of reasons. I would like to recognize Dr. Miller from Danbury Hospital. First of all, his testimony will put in perspective the other half of this problem at the very beginning of the hearing, and secondly, I am very proud of his leadership, as I am of the Connecticut hospital system, and have worked very closely with the hospitals in my district because I have hospitals that are very small in rural areas, and hospitals that are superb teaching hospitals. Dr. Miller runs the Danbury Hospital, which is kind of a hybrid of both. It is a teaching hospital in a small city, surrounded by relatively small towns but on the border with New York and in the New York City environment.

So it is a pleasure to have you with us, Dr. Miller, and to have you share the experience of an institution in trying to grapple with these very problems.

STATEMENT OF MATTHEW MILLER, M.D., VICE PRESIDENT, MEDICAL AFFAIRS, DANBURY HOSPITAL, DANBURY, CONNECTICUT, ON BEHALF OF THE AMERICAN HOSPITAL ASSOCIATION

Dr. MILLER. Madam Chairman, I am the Vice President of Medical Affairs--

Chairman JOHNSON. You have to get very close to these microphones.

Dr. MILLER. I am the Vice President for Medical Affairs at Danbury Hospital, a community teaching hospital in Danbury, Connecticut. I have worked in health care for 30 years as a practicing physician, and for the last 10 years as a physician executive. I am here today on behalf of the American Hospital Association (AHA). We appreciate the opportunity to present our views on improving health care quality and safety.

We are extremely proud of the initiatives that hospitals and their staffs have adopted to improve patient safety. We are eager to work in furthering those efforts. I would like to talk about the importance of creating a culture of safety in health care, the role of medical error reporting, and the potential of technology to prevent adverse events.

We must create a system where medical errors are detected and analyzed so that we may learn how to reduce those errors in the future. To do this we must, one, provide a nonpunitive environment for reporting errors; recognize that most errors are failures of systems, not individuals; investigate errors thoroughly, using root cause analyses; and, four, collaborate with experts and peers on the lessons learned.

The AHA supports efforts underway in the Senate to create an improved confidential system for the voluntary reporting of patient safety information. We hope that this Subcommittee will take a similar approach.

We all agree that reducing medication errors is a critical goal for us improve health care safety. We must have systems in place to ensure that important clinical information is available to physicians at the time drugs are prescribed, so that orders can be complete and accurate. Effective medication management systems must ensure that the right patient is getting the right medication at the right time, the right dose.

New technology can be very helpful in reducing medication errors. Examples include computerized physician order entry, CPOE, and bar coding for drug and patient identification. We need to recognize, however, that these technologies do not provide a single silver bullet solution to these medication errors.

With CPOE, physician orders are entered directly into a computer. The computer provides real time patient data, including allergies and lab test results. It also provides pertinent formulary information, standard dosing protocols, and guidelines for care, all in one location. I do believe that CPOE systems, when fully implemented, have extraordinary potential to prevent errors.

We have begun implementing CPOE at Danbury Hospital, but the road will be long and arduous. We cannot discount a number of other very important considerations.

One, CPOE is expensive. We will need at least $2.5 million over the next 2 years, and we are getting off cheap because we had already purchased a system. Other comparably sized hospitals to mine will spend $5 to $10 million, and that doesn't include maintenance.

The CPOE is fairly new, not widely tested beyond large academic medical centers with home-grown systems.

Three, there is no off-the-shelf package ready to install. The CPOE requires substantial customization.

Four, CPOE systems must interface with other hospital information systems. We, for example, just discovered a problem with our pharmacy system. We will have to change that system entirely.

Five, staff support for implementation will be huge, involving physicians, pharmacy, nursing, information technology. Not all hospitals have those resources either for implementation or for maintenance, and ultimately, education of all the staff and major work redesign will be necessary to achieve all the planned goals.

We can't forget that there are multiple other medication management strategies that hospitals can and, in fact, must implement first, such as standardized orders, practice guidelines, formulary control, computerized access to clinical information. These strategies can accomplish a lot and they are actually much easier to implement.

Bar code technology is another tool to prevent medication errors. The AHA is very supportive of efforts underway at the Food and Drug Administration or FDA to promulgate bar code regulation that will enable information systems to verify correct patient medication and dose.

It is important to understand this issue against a backdrop of larger health care environment. Hospitals are facing unprecedented pressures, including a severe work force shortage, soaring pharmaceutical prices, and increased professional liability costs.

In conclusion, a third of all hospitals are operating in the red today and another third are teetering on the edge. It is vital that hospitals have adequate resources to meet the needs of their communities, and for this reason we are asking Congress to forego budget neutral provider payment adjustments; approve the full Medicare inpatient inflation update; protect Indirect Medical Education or IME payments; and help find a solution for the nearly 40 million uninsured. We look forward to working with Congress and others to help us cross what has been called a quality chasm.

I would be happy to answer any questions.

[The prepared statement of Dr. Miller follows:]

Chairman JOHNSON. Thank you very much, Dr. Miller.

I would like to recognize Dr. Berwick, the President and Chief Executive Officer (CEO) of the Institute for Health Care Improvements in Boston. Dr. Berwick.

STATEMENT OF DONALD M. BERWICK, M.D., PRESIDENT AND CHIEF EXECUTIVE OFFICER, INSTITUTE FOR HEALTHCARE IMPROVEMENT, BOSTON, MASSACHUSETTS

Dr. BERWICK. Thank you, Madam Chair. I have submitted my full remarks for the record.

Since the IOM report, there has been indeed a lot of progress in this country, and I want to acknowledge that. The issue is explicit and visible. Most consumers seem aware that there are safety challenges in the health care system. A lot of hospitals have begun to promise to improve safety, and a small but important minority of them are becoming much more open about their own injury rates and measuring their frequency.

Some of the highlights include the work at Pittsburgh you will be hearing about. I am aware of community efforts in Whatcom County, Washington and in Florence, South Carolina, to improve safety at the community level. There is a consortium of children's hospitals that is focused on this as their primary agenda. I am aware of a collective of 57 hospitals working on substantial safety improvements, and other important progress is being made nationally.

The Federal leadership on this has been dramatically important. The work at the VA, DOD, the Health Resources and Services Administration (HRSA), and most importantly, the Agency for Healthcare Research and Quality (AHRQ) has actually been, I think, the catalytic work in this country. I want to personally thank Congress on behalf of my colleagues for your commitment and bipartisan support for this work, but there are obstacles. We are not going fast enough. Let me tell you what some of the obstacles are.

First, there is still denial in the industry. There are a lot of doctors and hospitals that still think this isn't a problem. We need congressional voice, Federal leadership to say it is a problem and it must change.

The second obstacle is capital costs. Some of the technologies to get safety systems into place are beyond the reach of small hospitals especially. There is a threshold to cross here. Last week the Advisory Board published a report saying that CPOE, which I thoroughly endorse, costs something like $7.5 million per facility.

We know CPOE, for example, saves money. There is a good study in JAMA, the Journal of the American Medical Association, that said there is a 13-percent reduction in the total cost of hospitalization in hospitals with CPOE introduced, 13 percent on the bottom line of costs. I believe it is there, but it is downstream, and we have got to get over this activation issue.

The third problem is, we are still paying for defects. I don't think hospitals intend to hurt anyone, but it still is true that hospitals that injure patients get paid for treating those injuries.

Fourth, there is lack of standardization. I will come back to that in a minute. The Federal Government has a real opportunity to offer some standardization that would help the industry accelerate.

Finally, there is persistent fear. For a lot of reasons--lawsuits, embarrassment, loss of market share--organizations, individuals in health care are very frightened to discover and reveal and discuss openly their defects. We have got to get over this. Fear is incompatible with learning, and safety can't be achieved without learning. Safe systems are open, transparent systems. A frightened work force is not going to go there.

We have been distracted by the issue of public reporting. I have an opinion on this. I think we need to be committed to transparency, but we have a very frightened work force that needs some security to know that they can discuss safety issues safely, or we are not going to get further on this.

I think we have been stuck in a national debate about mandated reporting or not mandated reporting. I think we have to get over it. What we know is, safety has to be openly discussed, openly assessed, openly explored. The Centers for Medicare and Medicaid Services (CMS) should insist that health care organizations that it pays have to assess safety, they have to study it, they have to discuss it openly, they have to learn about their own injuries. That is for sure.

I have six suggestions for Congress about things you could do that would help us at this stage. It is not primarily a Federal problem, but let me tell you what I think would help.

First, please continue to support the investments that you are seeing in the VA, in HRSA, in DOD, and in AHRQ in this agenda. We are really getting some leadership out of these systems.

I want to editorialize. I am deeply concerned about proposed budget cuts this year in AHRQ's budget. If you proceed with the current budget, the Agency for Health Care Research and Quality will be unable to fund any new investigative research. It goes to zero, except in specific targeted areas.

Now, you have made safety a targeted area, but that doesn't solve the problem. Without investigator-initiated research, we are not going to get the learning we need in this field. If you do give AHRQ the funds it needs to support investigator-initiated research, it is a wise investment in how to manage a $1.5 trillion system, and I fear we are about to make a big mistake.

Second, ask CMS to sponsor some market area experiments to reward quality and safety. Right now a hospital or health system that reduces injuries to patients actually takes a financial loss often because it gets paid for treating the consequences of its own injuries. We have to figure out how to stop paying for defects in care and start putting the opposite incentives to work: grants, tax credits, low-interest loans. There is some current legislation on this, but right now we are paying for defects and we have got to stop doing it.

Third, I think we should, with all due respect to Mr. Stark, create a circumstance in which one, at least one State in this country can try a no-fault malpractice liability system. I think it will work. Here is the system we should try. It should have six properties:

Always let patients and families know when a patient is injured, extreme honesty, and you are not in the system if you don't adopt that.

Second, apologize. We have trouble saying we are sorry.

Third, compensate the victims of energy directly and fairly and promptly. Right now only 2 percent or 3 percent of the money that is being exchanged in the system is actually going to victims of injury.

Fourth, bear the responsibility at the entity level. Let the hospitals, the health systems, bear that responsibility. Don't go to the personal individual blame level, because that is going to increase the fear that is our problem in getting to safety cultures.

Fifth, learn from the events, continually reduce risks within and among organizations.

Sixth, deal differently with the criminal and grossly negligent. That has to be done promptly and severely, and that is 2 percent of the problem.

I think a test of this kind of system in one State or one region of the country would teach us, and we win either way. If it works, we learn that it worked; and if it doesn't work, we learn it doesn't work and can move the agenda on.

The fourth recommendation to Congress is, create some limited privilege for reporting on patient injuries for the individuals who make the reports. I think Jim has shown in the Michigan VA, we know from theory and practice in other industries, if there isn't some form of protection for the individual--I do not think it should extend to the entity, to the hospital or the health system, but at the individual level there has to be protection, or people are no fools, they won't talk about things when that talk is going to come back to hurt them.

Fifth, don't worry about constructing a large national database on errors or injuries. It will be a waste of time and money. We don't need it. We should fund AHRQ thoroughly, to have a research database on this, and you are about to make a mistake if you proceed with the current budget on that, but if we fund the research work, that is all the data we need. We don't need a massive national architecture on data on injuries.

Finally, ask CMS to adopt some simple information technology standards. We really are stuck. It is one of the reasons Danbury Hospital has to spend so much money. I can explain this in questions if you want, but there are a few decisions we could make about standard vocabularies for coding in this country.

If CMS said, "You must use these codes as a condition of participation," we are done. It would be behind us, and then we can have a national data backbone that would allow hospitals and health systems to talk to each other, that would have saved Danbury millions of dollars, and Congress is in a really important position here. The CMS could actually make the difference in getting us there.

We have made a great running start. I commend you for what has happened, and now we have a few more jobs to do and we can make it even better.

[The prepared statement of Dr. Berwick follows:]

Chairman JOHNSON. Thank you very much.

Ms. Foley, President of the American Nurses Association.

Ms. FOLEY. Thank you, and good morning. I really do appreciate the opportunity to be here and return to the table to talk about the importance of prevention of health care injuries.

As you know, there are more than 2.7 million registered nurses in the country right now, 2.2 million of us are still working in the field. We are very committed, and it has always been a cornerstone of our nursing profession to be worried about and try to attend to the issues of patient safety.

Of course, as you have mentioned, Congresswomen, the landmark report "To Err is Human" really brought this attention to the forefront. It was the most talked-about health care issue in the year 2000, and while I do also agree with our previous speakers, there has been some good work started, we need to continue to push on the issue of systems of care and systems of safety that protect our critically important patients.

I am here, however, to put a slightly different face on the discussion because the system of safety cannot overlook the people who provide the safety net and the monitors of safety, and for us nurses play that role in a very critically important manner.

We have long maintained, and we are beginning to get some significant research which validates the relationship between high quality, safe patient care and the amount and the skill of the nursing staff, and we applaud the research that has been supportive of those findings. The CMS and four U.S. Department of Health and Human Service agencies in the years 2000 and 2001 have reported correlations between safe staff and certainly the reduction of complications and injuries such as falls.

I support, as does nursing, the implementation of technology, for example, computer order entry, but the pursuit of technology should not be in a vacuum. it must consider the people who will then be part of that system. I think Dr. Miller mentioned the training, but importantly as well the people who provide the care and do the actual monitoring and assessment are just as critical to a good computer order entry system, so the pharmacists and the nurses have to be present in adequate numbers and with qualifications that are commensurate.

We also know, as I have mentioned, that--well, so on the issue of staffing, I just want to reiterate that we are recommending then efforts to support valid and reliable staffing measurements that really address what are the needs or the acuity of our patients, and that that concept be integrated into Medicare and Medicaid programs as a form of condition of participation.

We very strongly applaud the efforts of the Veterans Administration. They are demonstrating the best practices, particularly in the area of early reporting and near miss analysis, and I think that is a field that needs to be further promoted and supported.

Additionally, we are very concerned about skilled nursing facilities. I have been very alarmed by the reports that we have heard. Nurses are not surprised, however, by some of the shocking findings that we are hearing about care in skilled nursing facilities. We continue to support adjustable minimum nurse-patient ratios in those settings.

Let's talk about the issue of mandatory overtime. Again, the system of care and the human factors involved in the safe delivery of care, there is inadequate research. It is beginning to be performed. We have one particular study, an AHRQ-funded study, to look at nurses and fatigue.

Over two-thirds of nurses in a survey last September told us that they are being asked to work unplanned overtime, and of course the epidemic of mandatory overtime has been both a very negative image for nurses, it is affecting our ability to recruit and retain, and we know we have a shortage coming. More importantly, we are very concerned about the safety of nurses who feel that they are too fatigued to provide safe care, and there are ample studies that show that fatigue has a correlation to injury and safety issues. So we do support Congress's continued discussion and support for the Safe Nursing and Patient Care Act of 2001, which we think is a balanced approach.

Let me just quickly mention the reporting, the role of the whistle-blowing. Mr. Stark, thank you for mentioning that, very much. The ability to safely report, in fact the enhancement of safe reporting mechanisms is very important to nursing. If we cannot speak out and let people know when we see care that is inadequate or unsafe or lax quality and there are negative repercussions, it will be difficult to be the full advocate that we believe we must be and that we should be.

So, in conclusion, I would like to just continue to support this system approach, but the people as part of that system must be considered. They have to be able to be there. All nurses want to do, and I have said this many times, all nurses want to do is give good care. We must attend to the people and the equipment and to the technology that makes that care possible.

I would have to say, very importantly, though, Congressman, you asked us about the cost. The alternative to a safe system with adequate staff is the cost of untold numbers of injuries, errors, terrible patient satisfaction, and the cost of that type of system is unacceptable. I am a believer in prevention. I believe the investment in a safe environment with adequate staff who are well qualified and well cared for will indeed pay dividends over the many years to come.

Thank you.

[The prepared statement of Ms. Foley follows:]

Chairman JOHNSON. Thank you very much.

Dr. Feinstein, Chair of the Pittsburgh Regional Health Initiative from Pittsburgh. Welcome.

STATEMENT OF KAREN WOLK FEINSTEIN, PH.D., PRESIDENT, JEWISH HEALTHCARE FOUNDATION OF PITTSBURGH, PENNSYLVANIA, AND CHAIR, PITTSBURGH REGIONAL HEALTHCARE INITIATIVE, PITTSBURGH, PENNSYLVANIA

Dr. FEINSTEIN. Thank you, Madam Chair. I also want to express the regrets of Secretary O'Neill, who is in Kuwait, who would be here to testify. His testimony is on the table. Secretary O'Neill and I, before he was Secretary, Co-chaired the initiative that you referred to, the Pittsburgh Regional Healthcare Initiative, which we view as a vast learning network involving the total collaboration of all the hospitals in our region to reduce medical error, hospital-acquired infection, and most importantly, Madam Chairman, to follow through on what you suggested, which is the critical systems redesign on all levels, so that people within health care can do things right.

That is so critical to us because getting to the root cause of patient safety issues really isn't another layered-on activity with all the other things we layer on at the hospital level, but it is part of a total systems redesign, a new way of regarding the health care enterprise that is so critical to overall process improvement.

At the moment, we are the first region in the country where competing hospitals have come together in total collaboration, to count every medication error and infection, count them the same way, and share that information openly. Thirty hospitals, twenty corporations, four health plans, a small business purchasing coalition, hundreds of health professionals have all come together, many of whom have signed contracts pledging themselves to work toward zero perfection goal in medication errors and in hospital-acquired infections.

As I wanted to mention, even though we have come together, we have all installed the same databases and we are using and sharing information, I do want to assure you, because this issue has been raised, our hospitals are no less rancorous and competitive than any others. As I like to say, our health system is just as sick as any other community's. It is not because everybody gets along in Pittsburgh that we are able to collaborate.

The reason we have been able to come together is a collective awareness of the scale of the problem, of the errors that plague our hospitals and the overall system errors, not just medication and infection, that result in waste, overuse, inefficiency, and often a lack of applying the best practices even though there is good knowledge to suggest what we should be doing.

The core in our mind of what was wrong is that our systems have lost a collective focus on helping care teams deliver the right care every time for every patient, which leads to the inevitable imprecision, waste, and errors about which we have convened today. When we realized that the systems issues were the root of the problem, we wanted to look at a powerful model for process improvement, and we turned to Alcoa. We couldn't find anything as powerful.

We wanted to look at some organization, that had adopted a successful framework for quality and safety, so we went to Alcoa. Alcoa is headquartered in Pittsburgh, and it had the best safety record in the world. Think of this, it is 18 times safer to work in Alcoa with molten metal and sheaves of aluminum, all four edges are like knives, than it is to work in your average hospital.

So we were looking for something powerful in the way of process improvement that would move us to the kind of systems change that we had seen at Alcoa, engaging every professional at the point of care in becoming an improvement scientist, helping everyone look at each problem, each error, as an opportunity for learning, for the root cause analysis, experimentation, measurement, and shared learning you have heard here.

I would like to echo Don Berwick's comments. We really couldn't be doing what we are doing now in our collaboration, with our databases installed to capture our hospital acquired infection data and the shared learning that is going on, without the Centers for Disease Control Prevention (CDC)and their NNIS, the National Nosocomial Infectious Surveillance, system for measuring infection.

Obviously, good data entry, analysis, attempting solution and then sharing the learning, are essential. Believe me, the databases alone, without the support in how to use them and how to gather learnings, wouldn't get us to where we are.

Let me quickly say, the kind of support that we would love to have, that we would need from the Federal Government to get to this system of perfect patient care to which we aspire, is increased confidentiality protections. We can't deal with errors, we can't solve them to root cause, without confidentiality protection.

Secondly, we need support for an environment where these clinical data systems and other methods we are using to get better patient outcomes, where constraints are removed, rewards are introduced. Expanding a Federal partnership, working with CMS, AHRQ, CDC, and others to continue doing this is incredibly important to us.

One of our most powerful learning tools as well as our databases is observation. We want to engage the appropriate Federal agencies in direct learning and more collective observation at the point of care, to see where rules and regulations and other constraints impede our efforts to bring about the systems change.

We suggest Federal support for more demonstrations applying successful industrial models, such as the Toyota Production System model we are using, in local hospitals. We would also like to see training in safety, both worker and patient safety and systems improvement, a core component of the education of all health professionals.

We would also like to see more medical research and education funding dedicated to improving quality of care delivery at the point of care. We need new knowledge about how to deliver health care, and how to apply the knowledge we already have more perfectly. Along with that, of course, I would like to suggest we would benefit from experiments in methods of payment that provide incentives for doing the right thing at the right time. We have tried very hard to enlist the professionals at the point of care as the driving force in our initiative. I want to emphasize how important I think that has been to us, and the fact that they are allowed to experiment and come up with their own solutions is a critically important learning tool.

Thank you very much.

[The prepared statement of Dr. Feinstein follows:]

Chairman JOHNSON. Thank you. I thank the panel for their thoughtful remarks.

I want to ask two questions. The first one is, some of you have had very direct experience in creating what one of you referred to as a culture of safety. Many of you mentioned openness, transparency, that you have to be able to let people and encourage people to say whatever is on their mind, if they observed something they thought was odd, and so on and so forth, so that there is free and open communication. What is the relationship to creating that kind of system and also having whistle-blower protection? Dr. Bagian?

Dr. BAGIAN. This is Jim Bagian.

Chairman JOHNSON. Am I saying that right?

Dr. BAGIAN. Bagian. That's fine.

Let me tell you experience we have had, you know, because we have had to deal with this, certainly. The big issue is that the caregiver doesn't view it as punitive when they go to do it, and we don't have whistle-blower protection per se, though there is some that exists.

Within our system, if you submit data or there is data in the safety or quality system, you may not divulge that to anyone, and there actually is a criminal penalty of, I think it is like a $10,000 fine and 6 months in jail, for disclosure of anything in there. So people understand when they submit into the quality system that it is to be used for quality and safety improvement and nothing else. That has always been very clear, and from a leadership standpoint we have always been very aggressive to enforce that.

The other thing is, as I mentioned before, the whole definition of getting people to even want to trust the system, that they have to see it will do good. That is where we saw a ramp-up over the first 10 months of the new system, that they saw reports of the same type of incidents that had been reported in the past. There is like nothing new under the sun, pretty much. The problems that occur today have occurred for eons, certainly decades.

When they saw the results were more systems-oriented, that made a huge difference to the culture change, and that is why we saw the dramatic increase in reporting and as far as actual really meaningful solutions that actually prevented problems, rather than just document problems, which really doesn't do much at all. So I think it is people have to understand how it will be used, and then you must demonstrate it, and leadership, and I can't emphasize this enough.

One of the things we did is at every facility, the facility CEO has to actually concur or not concur with each line item of a recommendation that a team--this is a team of nurses, physicians, pharmacists--says this is what needs to be done. If they nonconcur, they have to put in writing, in the record, why they don't concur, and then it goes back for revision, but they can't just say, "I don't like to do this."

That has caused tremendous better communication within the hospital in general, and actions taken, because where the accountability, if there is any, resides, if you want to look at it that way, is with the CEO of the hospital, that they take responsibility that these improvements will be made. I think when people--and we have gotten reports from the field that people can't believe that problems that were dealt with in a punitive manner before actually are dealt with in a constructive manner, that they actually see improvement, and that kind of primes the pump and it builds on itself.

Chairman JOHNSON. Thank you. Anyone else wish to comment? Mary? Ms. Foley?

Ms. FOLEY. Yes, thank you. I think it is an excellent question. I think it is consistent with the environment of both the blame-free and the non-disciplinary approach. I mean, the ultimate, not the ultimate sacrifice but a severe sacrifice for someone who believes it is imperative for them to professionally report an incident or a practitioner is the loss of a job, and that indeed is what we are attempting to prevent by passing the whistle-blower protection.

So, it is an extreme type of discipline or intervention that really just puts a, casts a pall over people's belief that it is the right thing, it is the necessary thing to do. It is not the only approach. I think there are many levels of just that stimulating the reporting in an environment in which it is safe, that it is protected, that it is promoted. The whistle-blower is an extension of that same type of protection for that more severe action, and we see it connected because that whole attitude, that there are costs to doing the right thing, has to be eliminated or people will be repressed, and that is unfortunate.

Chairman JOHNSON. Dr. Miller?

Dr. MILLER. I certainly agree with what has been said already. I would just add that within an organization, often you want to promote the culture of reporting but you also want to make that easy to do, and it is not always so easy. It takes time to report. The reporting can be confidential or not confidential.

In terms of whistle-blower kinds of things, we certainly have those things in place in our organization, but we actually tie it to a little bit of what Dr. Bagian was talking about. If someone wants to report an error, they can do it confidentially, they can do it by phone, they can do it in writing, but we have a policy that within 5 days we get back to them.

Whoever, if they say who they were, we not only protect them for reporting it and encourage it--and my report card internally is to have more errors reported this year, not fewer--but that we have an obligation to get back to whoever blew that whistle, if you will, and tell them what our action was. Sometimes there is no action that can be taken, but they deserve a response, and I think that takes care a lot of the disgruntled employee, or nurse or pharmacist or doctor that says, you know, "I can't bother complaining anymore. Either they will go after me or they won't do anything about it anyway."

Dr. FEINSTEIN. Madam Chairman?

Chairman JOHNSON. Dr. Feinstein?

Dr. FEINSTEIN. Can I say one thing, too? I would de-emphasize issues of whistle-blower protection and focus on developing a positive reward system, with incentives that a CEO who is really committed to solving problems and having an error-free environment would endorse. Certainly that is something we learned when we observed Alcoa. If there is a passion for this, if the person at the top says, "This is a learning environment, we are here to learn, we are not here to hurt anybody," it creates an environment and culture that is so important to error reduction.

Anything that can be done to support that should be encouraged. We do have an example of a hospital among all of our hospitals who is doing the most reporting on medication error. I would say it is the culture at the hospital and the support of the CEO that is bringing forth so many errors. They report many times, multiple times the number of errors that any other hospital does. They have the same protections, and we have great belief in protections. They have no more, no less than anyone else, but they have a CEO who really supports this.

Chairman JOHNSON. Dr. Berwick?

Dr. BERWICK. One quick word on that. If a system needs whistle-blower protection, it isn't going to get safe, because it means there is fear in the system. I think you should have the protections, but don't expect--that is not culture change. That is just good police work.

Let me explain for a second what happens in a hospital where the culture change exists. You just get the other image in your mind, is what Karen said. Take Luther-Middleford Hospital in Eau Clare, Wisconsin, a great place. The CEO there did what Karen said, she said, "We can't be safe if we don't know what's going on. I will be behind you. Tell us what's going on." Nursing reports of injuries to patients went up 40-fold within 3 months.

The local newspaper got hold of it and ran a headline, said "Hospital Injuries Increase 40-Fold." They got it completely wrong. All that happened was, it became a transparent system and finally they could get to work on it. The courage it took on the part of that executive and the board to do it, and then to go to the community and say, "No, let us explain what happened," that is what culture change and leadership looks like. So do whistle-blowers, but we are after a different phenomenon in the industry, which is a different kind of courage at the executive and board level.

Chairman JOHNSON. Thank you. I do think that leadership is key, and nothing will happen without it, and no amount of law or whistle-blower protection will change the culture if there isn't leadership.

You might all think about, how can we hold the top executive more accountable for that leadership, as opposed to providing whistle-blower protection? Because in the end culture change can't be done negatively, it has to be done positively.

Now, I separate this entirely in my mind, although I know they're not entirely separate, from this issue of mandatory overtime which I think is a very serious development in our system, and will carry with it enormous potential for errors if we don't do something about it. So, you know, I want to recognize that, because I have had some terrible examples come to my attention of the abuse of mandatory overtime. I think nonetheless cultural change cannot be legislated from Washington, so the question is, how do we change the way we hold our systems accountable and look at you?

Then I want to just get back to this issue of cost. You have given us some very good information about the initial cost. It is clearly multi-million. Why is it that our capital payment system isn't sufficient, or is it?

We no longer make you prove that you made capital expenditures in order for us to give you capital payments. There is sort of an automatic capital payment system now that gives you money, assuming all of the kind of technology change and the various things you have to do to upgrade your operating rooms and so on. So, why isn't that sufficient to focus on this issue, or is it? Dr. Miller?

Dr. MILLER. Unfortunately, it is not. I can give you an example from Danbury. We do run in the black, and we have dollars that are available for capital purchases every year, and the price tag on those purchases, capital purchases or renovations, goes up annually. Pharmacy costs are going up double-digit annually. New technologies that we need to have in order to provide quality care to our patients, I don't mean something esoteric or to compete, I mean quality care for patients, those things cost more money.

Very specifically, we talked about the dollars for computerized physician order entry. A single example, one 371-bed hospital. This fiscal year, our clinical leaders for capital equipment purchase requests was $50 million. I had $20 million to spend. That means $30 million that was requested, not for frivolous things but for replacement items, renovations, fix this, the physical plants that weren't as sturdy as they once were, they were built 25 years ago, new technologies that a new surgeon wants in the operating room, all those things cost money. I don't have enough money for this year's requests. New technologies like physician order entry, and I had to get the budget to approve that, meant there were $2 million worth of things that I couldn't buy instead.

Chairman JOHNSON. Thank you very much. Dr. Berwick?

Dr. BERWICK. Thank you, Madam Chair. I want to divide the capital question into several categories because it is not one problem.

First problem, wiring physician offices. Five years from now there is no reason in this country we should have handwritten prescriptions. There is no reason any physician shouldn't have access to a hospital medical record for one of his patients. There is no reason we shouldn't have a master drug list.

For an average doctor, it is a small amount of capital, $10,000, $20,000 to get into that world, but a lot of physician practices are having a lot of trouble getting there. We need some solution. I think it should be a public default system available to anyone, out of the VA or somewhere, where we just say, "You can have this if you want it. If you want something better, you can invest in it."

Second, small hospitals, small and rural. I don't know how big Danbury is, but there is a whole--there are thousands of hospitals that can't spend the $3 million it will take them to get CPOE. We have got to help them out.

The third, the big systems I am frankly not worried about. We have lots of investment going on, and big capital investments in enormous multi-hospital systems, they are going to make those investments. There is a little problem here on return on investment, because once you get into the world of safety and quality improvement, money is saved by the system but it may well be lost to the hospital.

I was just in Bellingham, Washington, where the hospital is supporting a community effort to improve diabetes care. They know it is going to end up reducing their revenues by $2.5 million a year because diabetics aren't going to be in the hospital, and it is a system saving that doesn't go to the hospital. We have got to solve that problem. We have to get--that is what payment for quality would look like, that creative circumstance in which, when the money is harvested out of the system, it goes back into the system in a more creative way.

The last issue is a cost-reduction thing. It is kind of how could you reduce the cost of capital? The standardization problem is very serious. Right now, investing capital for some facilities is a very risky game, because they could capitalize a system with one language and tomorrow we could end up with a different language structure and they will have wasted a lot of money or have a lot of adaptation to do.

That is why I think the safety issue is related to a national move to say, "Here are the standard languages. We promise you this is going to be here. For laboratories it will be Loink. For diagnoses it will be Snomed." Whatever we decide, let's just make a decision and say to the Nation, "Now you can be safe in investing."

Chairman JOHNSON. Just briefly, do we know enough to establish a single standard language?

Dr. BERWICK. There are six standards we need, in my opinion; four we do, two we don't.

Chairman JOHNSON. Okay.

Dr. BERWICK. Laboratories, everyone agrees that Loink is the right system. If CMS said tomorrow, "That's the system, everyone code now or we won't pay your bills in 3 years," that will get that solved.

For diagnoses I think Snomed is the right answer. It is currently owned in a proprietary way, but negotiations are underway for the government to make those public domain. We should get that done. That has been going on 18 months. It doesn't make any sense.

There is a system for dialogue called HL7 which you have invested in. It could be the national standard. Everyone kind of agrees it is better than anything else.

There is the DICOM, the Digital Imaging and Communications in Medicine system.

We don't know quite yet nationally how to code drugs, medications, and we really need to. We need to have a standard medications coding system in the country.

We don't know how to code procedures yet, although there are several options. Within 6 months, if you told CMS, "Let's establish, let's have the backbone, let's say these are the systems, they're not going to change," you will save millions of capital for hospitals like Danbury, and we will get on with the job. I think it will require Federal leadership. We don't have another structure to get that job done.

Chairman JOHNSON. Thank you very much. I am going to turn to Mr. Stark, and yield the Chair to Mr. Camp of Michigan. Thank you.

Mr. STARK. Thank you, Madam Chair.

Let me just take one more crack here at EMTALA, the Emergency Medical Treatment and Active Labor Act, and the Patients' Bill of Rights bill, which has been passed in both Houses, and a section which I think has no quarrel from either side of the aisle or in either House. Is there anybody who would object to either of those standards being used?

Now, the American Hospital Association has, but tell them to get with it, Dr. Miller. I mean, I think we can find and take care of that, and I think everybody understands that it is the issue of the subordinate who reports to their supervisor, who ignores them, and then the subordinate goes elsewhere. Arguably that is not desirable, but it probably is more desirable than having a subordinate who will suborn those issues. I think we could get that one taken care of pretty quickly.

The issue of capital and getting the system working just may very well be a problem in the system. When you allow or encourage, depending on what State you are in, a lot of competition and oversupply, when you are running less than 60 percent of staff-bed occupancy, then you have got to say, "Well, wait a minute." Or when you have got to make the case that you want to buy a computer system for a 20-bed rural hospital. It would probably cost almost as much as it would for your hospital, Dr. Miller, but you have got 20 times more people that you are serving.

That politically, you know, you can't get any of my colleagues who want to close the hospital, as small as it may be, on their watch. Nobody builds statues to us in our home district for closing a hospital or a post office. So we have got to find some way for the communities to get off stage, to perhaps consolidate, to share equipment. I don't know, we aren't going to be able to do that, and that is part, I think has got to be part of this.

In the costing, one of the things that occurs to me is that you all in the hospital business don't take your recalls. I take my car in to get it tuned up and they do something, the dealer has to fix it for free if they screwed up, right? I go to the hospital and something happens subsequently, poor Blue Cross has got to pay again to send me back.

Now, if we changed the system and you had to do your own recalls, that is either a loss of revenue or an increase of cost, you slice it either way you want, and I suppose you ought to be compensated for that. I mean, I think if that is part of what is having people drag their heels on this, and I don't know whether they are so harsh that they are willing to say, "We won't do it because it's going to cost us more to correct our mistakes," I hope that people aren't thinking that way, but they may. I think that would be a valid issue for us to say, "Okay, if we are creating more procedures, then we're going to pay for it."

We are going to have to come to that same issue, I suppose, with pharmaceuticals. They are costing more. They are costing you more on the one hand, and they are probably, if they do what they say they are going to do, my Zocor, which is very expensive, means you are less likely to get me into some heart program where you make a lot of money on me.

The next thing is, what are we going to do about boutiques? As we balkanize your hospital, so all the cardiologists in Danbury say, "Uh-uh, we're going to create our own heart hospital," they are going to thumb their nose at you and pull a lot of your good revenue, high-margin revenue, out.

All of these things face you. It doesn't happen in the veterans' hospitals, I don't suppose, now, but I hope that you will work with us to address that. I don't think it does us any good. I mean, I think the balkanization thing hurts teaching hospitals, in which I have a lot of faith, and managed care plans, they aren't going to teaching hospitals if they can avoid it because it costs them more.

I think that in this overall review of safety we have to help you, but you have got to be willing to work with us. I mean, we can't just say no, we are not going to have regulation, because then we are not going to get a universal computer language.

I mean, you have to be willing to trust us that we won't impact too much, and we have to trust you that you are not going to just come and hit us all the time for more capital. When I will tell you, you ought to be thankful that Danbury is not in California, because retrofitting California hospitals for earthquakes, we are looking at $8 or $9 billion. Our hospitals, they are doing pretty well but not that well.

So, I mean, there is a shared responsibility here, and I assure you that while some of us may be more skeptical than others about absolute tort reform, I have no quarrel with the idea of no-fault. I mean, I wrote the original bill for the District of Columbia, and driving in this city with no-fault auto insurance, and you would wonder about what we were doing those many years ago. It can work, as long as you leave the outlier for the gross negligence, because I think that threat has a salutary effect on those chief executive officers who may not just completely want to do this out of the goodness of their heart.

So, thank you for being here. Please push us, because this, we can talk this to death. There are some bills. Tell the Hospital Association to give us their bill. I mean, the Secretary can get to the drafters of legislation more quickly than I can. Let Secretary O'Neill draft a bill for us and send it over. I will introduce it for him, but we have got to get going.

I mean, this process, we won't please everybody, so I will just shut up and say that the secret is when all of you are frowning, Ms. Foley, Dr. Miller, Dr. Berwick. Then we have got the right bill. If anybody is smiling, somebody got away with something. So let's get you all frowning, drop the gavel and say, "Go forward with a bill." I really hope we could do it. Thank you very much.

Thank you.

Mr. CAMP. [Presiding.] Thank you. I just have a couple of questions.

Dr. Berwick, you were the author of this groundbreaking study on this issue which said that preventable medical errors are the eighth leading cause of death in America, accounting for as many as 98,000 mortalities in hospitals each year, and I think this only gives us a window on hospital deaths and does not really include the number of patients injured. Do you have an estimate of the number of patients that are harmed or killed as a result of medical errors in America each year?

Dr. BERWICK. I don't have it, nor does any such estimate exist that I know of. The eighth leading cause of death figure is attached to the estimate of 44,000. If it is 98,000, it is the fourth leading cause of death, just hospital injuries. What we do know is that certain forms of ambulatory surgery centers are quite unsafe. We know there are injuries in nursing homes.

The other calculation that would be great to see would be deaths and injuries due to quality failures beyond safety hazards, for example, the failure to use the best-known medication or the failure to use the proper diagnostic procedure. That is not called an error in the errors report. That is a different kind of failure, and my own estimate is that there are many times that number of people who are suffering unnecessarily because of quality failures as there are of the more confined area of just errors.

Mr. CAMP. So, then how serious would you say this problem is?

Dr. BERWICK. The biggest opportunity for improving the health status of Americans, beyond prevention of disease, is to improve the quality of health care.

Mr. CAMP. Dr. Feinstein, you mentioned that obviously if Federal policy doesn't have quick fixes, in your testimony, and relies less on mandates and punishment and more on what you called learning networks, what would be the ideal components, just to summarize, of an error reporting system?

Dr. FEINSTEIN. Well, error reporting systems and new technologies are tools that become part of a general quality and process improvement framework. I don't want to in any way miss the opportunity to say that databases are very important. Protections so that someone can enter data on problems, all problems is critical. It is how Alcoa became the best in the world. Even more important is to embed all of these into a process improvement framework. This is what I would encourage the Federal Government to do. We need to create in health care the same total safety environment that you have in aviation and the nuclear power industry.

That involves research and education efforts directed at understanding, for everybody involved in a health enterprise, how to create a quality and a safety-focused environment. Systems need to keep learning how to move continuously toward improved safety, and more application of what we know to be good practice. The partnerships we have had with CMS, AHRQ, CDC, to date have helped us to build this improvement framework, to understand what works, and also how you are continually, as you introduce new technologies and as you gather more data, you are continually coming up with new problems.

Mr. CAMP. All right. You have mentioned that you believe State peer review statutes are inadequate, and why do you think that is so, and is that why you think we need a single national standard?

Dr. FEINSTEIN. I will speak from our own example--our Pennsylvania State peer review statutes are good. The Medicare protections through our Quality Improvement Organization, QIO, are very good, but involves only the Medicare population. Hospitals are treating a very broad base of patients. We want to learn from all our errors, all our problems. Every one is significant, even the ones that are minor, even the ones that don't hurt people.

Mr. CAMP. Thank you. The gentlewoman from Florida.

Ms. THURMAN. Thank you, Mr. Camp. I apologize for not being here for all the testimony, but we have this little stimulus package on the Floor today that we seem to all be concerned about and want to see something done with.

Beyond that, I probably just want to get some ideas from all of you. I don't know if you are familiar with a piece of legislation that Mr. Houghton and I have introduced on medication errors. It is actually--and it is being done by Senator Graham from Florida and Olympia Snowe in Maine, and there are a few differences but not much of a difference.

I am curious because we think, at least looking at the numbers, medication errors seems to be a very high part of any of this system that we are concerned about. Some of the numbers we have seen, that there have been approximately 7,000 deaths, some 250,000 nonfatal injuries. What I also find interesting about it is that by doing some of this, and I think you all have alluded to this, that because of this that it costs about $4,700 per patient admission because of these issues.

So, we put a piece of legislation together specifically to look at technology, the kinds of things we can do with technology. Jim, I know that you all have done much of this at the Michigan VA, and DOD has similar technology. I happen to have had the experience of my mother being at Walter Reed. They have a wonderful system going on out there. It is all computerized. Doctors know what is going on. Everything goes into the computer, the patient's information, what medicines and all those kinds of things. I was fascinated by what had happened out there, and certainly the VA hospital in Gainesville and others have implemented some of that.

What we are trying to do is actually give about $1 billion over a 10-year period of time to help hospitals and nursing homes, because we think skilled nursing facilities have got to be in this mix, and we also have taken a part of these dollars and we have carved out about 20 percent of those dollars going to rural health areas, because that too is something we think is absolutely necessary.

Then we have actually tried to work on a board that would be called the Medical Information Technology Advisory Board, specifically because we also think that as we get this information, as we are using with the patients, as it is working, we also know that we want to be able to transmit this information so in case somebody needs the information for follow-up on somebody. We also recognize there is information, privacy issues come up, privacy issues that become a concern, and so the difference between Senator Graham's and our bill basically is that issue.

There will be some criticism of this bill because funding comes out of the Medicare Trust Fund. I don't want that to be where it comes. That just seems right now the best place that we can look for the dollars. We think it is a good investment, potentially saves money, and certainly would like to work with any of you on this issue. Certainly, based on your comments, there is not money in these hospitals. There is not anything to be able to be done with capital improvements. That is going to create a problem.

Just based on my explanation of what we have done here, can you give me a little bit of feedback on whether you think that this might be something that could work, would be helpful, and what we can do to get this moving?

Dr. BERWICK. I will start. I am sorry, I was aware of it as the Graham-Snowe bill. I think it is a good idea. I think it would at least get some acceleration into the system.

Here are two ideas. One is, continue as you are to make sure the money goes where the need is. Industry ought to be obligated to obligate capital to this when it can afford it. So as long as you are targeting rural, small areas, hospitals that are in underserved communities, I am happier that we are not offering money where it won't make a difference.

The second is an idea I have got. I have no way to know how to put it in the legislation, but I think it is a good idea. Why reinvent the wheel over and over and over again for a thousand rural hospitals, or if you want to go to physician offices, for 20,000 physician offices?

Could it be a government task to fund and create a single national default option which is available at extremely low cost as the basic system, so we could say to rural hospitals or to physicians' offices, "You can have this one. The VA built it. They put the money in. It's your money. It's your tax dollars." Or if not VA, someone else. "It's yours. You can have it for free. If you want to buy something with higher end or higher functionality, of course feel free to invest money further."

No one is going to do that, and it would really accelerate the action a lot. We have spent public dollars on wonderful systems that, with some adjustment and some investment, perhaps under your legislation, could become a national gift, I guess, to the system that is having trouble getting over this capital threshold. It would save in aggregate nationally lots of money, because then we won't have a thousand places doing a thousand different things. We will have one system that everyone can use.

Mr. CAMP. Just quickly, we do have a recorded vote with 8 minutes left, so if you could just quickly sum up.

Dr. MILLER. A couple of quick points. I am interested in the legislation. I don't know all of its details. I am a little anxious about designation of dollars more to rural hospitals. I would want to make sure we are putting the dollars where they need to be.

Ms. THURMAN. It is 20 percent. It just carves it out of 20 percent, so they don't get left out of the system.

Dr. MILLER. They shouldn't be, but the middle-size community hospital is often the one that should have more complicated systems in place, because it is big enough and may not have the resources.

The source of dollars is always problematic. We talked earlier about sources. Taking it out of Medicare funds is just going to take it out of one pocket and put it in the other. It is not going to solve my problem.

The third, I will be brief, but it is an issue. The whole business of cost savings with new technologies, I believe in it, I believe in that $4,000 figure. I have read the studies. I think those savings are way down the line, and they are illusive. The Advisory Board has published four stars for quality for physician order entry, and that is why we should do it; one star for cost savings. That will be hard to achieve.

Ms. FOLEY. I will sum very quickly, as well. Thank you. I haven't looked at the bill, but it sounds good, because the most difficult position that good administrators are put in is to choose where to spend their money. If they have to choose between technology and adequate staff, that is a terrible position to put good people in, so additional incentives for the pursuit of the technology while we also provide the adequate funding for the people.

The U.S. Pharcacopeia, USP, study showed that primary contributing factors to medication errors were distractions and work load increases. So, though we know there are some savings and error prevention to be achieved, we don't want to just go off in one direction. That is why that system approach, which is what we are all about, is very good. It sounds very exciting. Thank you.

Ms. THURMAN. Thank you all very much.

Mr. CAMP. All right. I want to thank you all for your very helpful testimony, and this Health Subcommittee hearing on health quality and medical errors is now adjourned.

[Whereupon, at 12:31 p.m., the hearing was adjourned.]
[Submissions for the record follow:]

American Academy of Family Physicians, Leawood, KS, statement