14-14 Bonnie Lane
Bayside, NY 11360
Phone: 718.423.9167
Fax: 718.423.9169
Our goal is to meet and exceed all of the possible needs of your law firm in any health related area.

Long-term solution to malpractice crises: reduce harm to patients

Most people don't go into medicine or health care primarily for money or power. Indeed, most of us probably chose careers in health care because we want to help people.

But, unfortunately, there is ample evidence that we are harming people who put their trust in us to help them.

At a 400-bed hospital with an average length of stay of four days, about four people each day are suffering harm related to their care and one has been the victim of malpractice. Fortunately, only 2 percent of those whose medically related harm involves negligence or malpractice will go on to sue.

Physicians put the cost of malpractice insurance and malpractice suits at the top of their concerns about health care and medicine in the United States--above health care costs and insurance companies. Cost of liability insurance is also top of mind for hospital administrators--above the cost of medical devices, pharmaceuticals and equipment, and above Medicaid and private insurance reimbursement rates.

And what are physicians doing about this crisis?

We are protesting to our state legislatures and to members of Congress. We are demanding, and in many states have gotten, caps on non-economic damages.

What aren't we doing?

We are not addressing the major underlying problem leading to malpractice suits--harm to patients. When you look at the malpractice suits that are filed, even though only 2 percent of negligence leads to suits, people who have been victims of negligence are the ones most likely to sue, and people who have suffered non-negligent "adverse events"--in short, those who have been harmed by their care--are next most likely to sue.

Even though only a small percentage of those who could sue and would be likely to win actually do sue, it is out of this relatively small number that the malpractice crisis arose. So, it stands to reason that if we really want to help our patients and ourselves, we should be reducing the likelihood that we'll harm them.

Safety by the numbers

Health care today is not safe. When there are at least 44,000 deaths a year related to medical error, it is not possible to say that health care is safe.We believe that 44,000 deaths are the tip of the iceberg of harm to patients. It doesn't include deaths from care initiated outside the hospital; it also doesn't include most deaths due to delays related to poorly coordinated clinical decision making among multiple physicians or time delays in providing antibiotics or thrombolytics; and it doesn't include morbidity or disability related to harm.

Yet, over 60 percent of physicians when surveyed believe that there are only 5,000 or fewer preventable deaths each year in U.S. hospitals; and 70 percent believe that preventable medical errors are made "not too often" or "not very often." (1)

The public disagrees. Over 20 percent of adults who have been hospitalized or had an emergency visit in the prior two years, or have a chronic condition, believe that they have suffered from a medical or medication error and that most of these have been "serious." About half are reported from hospital care and half from ambulatory care. (2)

Furthermore, health care today is not as effective as it should be. When Beth McGlynn and her colleagues studied the degree to which ambulatory patients received "recommended care," only 55 percent was delivered. It varies by condition: Whereas a group of patients with breast cancer received three-fourths of recommended care, a group of patients with hip fracture received only one-fourth of recommended care. (3) Most of these failures to provide recommended care are errors of omission.

Reducing harm

We should do much better. The good news is that we could. There are examples of marked reduction of errors and intensive implementation of effective practices to suggest that if we put our minds to it, we can make an enormous difference.

A very old example still provides important lessons: Back in the mid-1980s, when Don Berwick was vice president for quality of care measurement at Harvard Community Health Plan (HCHP), his group examined the records of women who had had a positive Pap smear to determine whether they had a follow-up within a six-month period of time. He discovered much to the surprise and horror of his colleagues that only 70 percent met that requirement.

I (Stephen C. Schoenbaum) was the deputy medical director of HCHP and responsible for "fixing" the problem. HCHP consisted primarily of a large staff model group practice HMO. The staff model had put in place an automated ambulatory medical record back in 1971. Using that record, it was relatively easy to put in place a tracking and tickler system, so that women who had a positive Pap smear could be kept under surveillance by the management of the practice.

Each month following the initial positive smear result, if the woman hadn't had a follow-up, the clinician would be reminded. Over a short time, about a year, things had improved to the point where virtually every woman, over 99 percent, had had a follow-up. The quality of care measurement department documented this level of performance over a several year period.

HCHP was also concerned that it wasn't enough for there to be follow-up of a positive Pap smear result. Different types of positive results would ideally lead to different types of follow-up. Accordingly, HCHP's gynecologists developed a uniform set of recommendations for follow-up of different types of abnormal smears.

HCHP also had an electronic laboratory test results reporting system; and it proved possible to program that system so that when there was a positive smear, the responsible clinician received a prompt about the recommended type of follow-up for that type of positive result. So, not only was follow-up improved to near perfect levels, but the type of follow-up that was done was the most appropriate for the patient. (4)

As HCHP grew, it added a network of practices that did not have an automated medical record. By implementing simple manual test tracking and tickler systems in those practices, it was possible to achieve near perfect follow-up of abnormal Pap smears.

A major difference is that if you have an electronic system, it is pretty easy to program the system to track a variety of abnormal test results; whereas, it takes a separate manual tracking and tickler system for each type of abnormal test result.

Some points to consider from this example:

  • First, it is not unusual to find that in the absence of a specific test result tracking and tickler system, a significant percentage of positive test results, often 30 percent, will not lead to an appropriate follow up. Some of these are likely to lead to malpractice suits for delay in diagnosis.
  • Second, failure to follow up a positive test can occur for many reasons:
  • Sometimes the test result will end up getting filed in the record without being seen by the doctor.
  • Today, patients often have tests in one setting, such as a hospital, but show up next in another setting, such as a doctor's office, and the test results aren't available at the doctor's office.
  • Sometimes, the physician will seem not to recognize the positive result--we're all capable of human error and when we process hundreds of pieces of information each day, we are inevitably going to make some errors.
  • Sometimes, the physician, or the physician's office, will try to reach the patient, fail to, and the ball will get dropped.
  • Sometimes the patient will be reached, but will fail to follow-through on the physician's recommendation. Although the last is not negligence on the part of the physician, the others either are, or are preventable errors.

With an appropriate systems solution, the source of the error doesn't matter and the patient will get the needed follow-up.

Third, the costs of fixing this type of problem are small compared to the cost of an occasional malpractice suit for delay in diagnosis. Yet, remarkably, most physicians' offices today do not have test tracking and results tracking systems.

Another example of reducing harm is built into the Institute for Healthcare Improvement's 100,000 Lives Campaign.

Overall, using a set of standardized methods, now known as the central line bundle, there has been over a 60 percent reduction in central line associated blood stream infections (CLABS) across hospitals in the Pittsburgh Regional Healthcare Initiative--a collaborative effort of Pittsburgh hospitals, insurers, and others to make marked improvements in patient safety and quality of care.

Rick Shannon, chief of medicine at Allegheny General Hospital and a board member of the Initiative went one step further.

He felt it was possible to do better than 60 percent and set a goal of eliminating CLABS. In a 90-day period, processes were put in place that led to a 90 percent reduction. He also showed that it pays his hospital to do this. They were losing over $20,000 per patient with a CLAB, and the costs of prevention of CLABS are pretty small.

In short, there is an immediate business case for reducing CLABS. The patient wins and the hospital wins.

Next steps

Why aren't we seeing these types of examples everywhere?

It is hard to know for sure. It may be that we simply don't realize how much better we could do if we were committed to reducing error to near zero and increasing the application of effective practices to near 100 percent.

Paul O'Neill, who led Alcoa to become the safest organization in the world, says, "I don't know of an organization that has achieved better results than it set out to get."

Yet, we have behaved as if health care is inherently and unalterably dangerous. We talk about patients falling through the cracks as if an episode of medical care was a journey across thin ice. We need to understand that what stands between us and marked reduction in harm to patients and more effective care for patients is:

  1. Leadership commitment to ferreting out and addressing the problem
  2. Design and implementation of appropriate systems to reduce errors, provide elements of effective care, and reduce delivery of unnecessary care that might produce complications

Anyone who has been paying any attention to the quality improvement movement knows that the tools for doing #2 exist.

We cannot demand your commitment to reducing harm for patients, but, in the context of this malpractice crisis and likely future ones, we can try to provide a rationale for heading down this route and becoming passionate about ferreting out the problems that are harming patients and fixing them.

If many physician executives and leaders follow this route, there really will be a major reduction in harm to patients. Physicians will see fewer of their patients harmed, and many fewer subject to the delays, mix-ups and frustrations that are the iceberg below the waterline of injuries.

The public will recognize that health care is no longer a hazardous undertaking. As happened with the airlines, this is likely to be good for business. Once the public recognizes that health care is very safe and highly effective, there will be significantly fewer suits and even fewer victories than there are now.

The physicians we lead will feel supported by the systems around them. They will be happier about what they do and physician leaders will notice a benefit.

If you buy this line of reasoning, there are some logical next steps. One is working within your institution to get the next level of commitment. Another is using the energy of the current malpractice crisis to get help. (5)

As a physician leader, you and your colleagues can approach your malpractice insurer about beefing up its risk management activities to help you help your colleagues reduce harm to patients. This can take many different forms.

For example, CRICO, the malpractice insurer for the Harvard-affiliated hospitals and practices, not only pioneered in developing the anesthesia standards that helped anesthesiologists reduce their risk and relative malpractice insurance premiums, but in recent years has been running another innovative program for anesthesiologists.

CRICO offers discounts on malpractice premiums to those anesthesiologists it insures who go through a program of operating room team management in a simulator setting. Most anesthesiologists in the Harvard system have participated, and both potential claims and actual claims decreased since the program was put into effect a few years ago.

A similar program has now been instituted for obstetricians, and, in a few short years, CRICO believes it has more than recovered the value of the 10 percent discount it has been giving by seeing a reduced volume of potential obstetrical claims.

Back to basics

We understand that physicians have been lobbying for malpractice reforms, particularly caps on non-economic damages, and we favor malpractice reforms. But we do not believe that reforms such as caps on awards address the root cause of malpractice crises.

As physician executives, we have a long commitment to improving care. We need to help all physicians think beyond how we can get relief from the financial problem of rising and high malpractice premiums. We can do that by working on the root cause of the problem, which is not that our patients are wantonly litigious but that we, and the health care system, have been harming them and they don't fully trust us.

As physician executives and leaders, we need to engage our colleagues. Though most of the ways of reducing harm to patients involve "systems solutions," their design requires input from physicians as well as other members of the health care team. Most physicians have not been active and ongoing participants in quality improvement activities. (6) We need to change this. The physicians you work with can make an enormous difference once you've engaged them to participate.

If we do not address the root cause, we believe that current malpractice reforms will only delay slightly the next crisis. Conversely, by addressing the root cause in our organizations, with our colleagues, and also with external parties that could help (such as government and malpractice insurers), not only will we benefit, but so will our patients. And that is most likely why we went into medicine in the first place.

Stephen C. Schoenbaum, MD, MPH, is executive vice president for programs at The Commonwealth Fund in New York City. He can be reached at SCS@cmwf.org


Kenneth Segel is principal and partner of Value Capture / Value Capture Policy Institute in Pittsburgh, Pa. He can be reached at ktsegel@verizon.net


  1. Blendon, RJ and others. "Views of Practicing Physicians and the Public on Medical Errors." NEJM. 347(24): 1933-40 Dec. 12, 2002.
  2. The Commonwealth Fund survey
  3. McGlynn, EA and others. "The Quality of Health Care Delivered to Adults in the United States." NEJM. 348(26): 35-2645, Jun 26, 2003.
  4. Schoenbaum, SC and Gottlieb LG. "Algorithm based improvement a clinical quality improvement" British Medical Journal. 301: 1374-6, Dec 15, 1990.
  5. Schoenbaum SC, and Bovbjerg RR. "Malpractice reform must include steps to prevent medical injury." Ann Intern Med. 140:51-53, Jan 6, 2004.
  6. Audet A-MJ, Doty, MM, Shamasdin, J and Schoenbaum SC. "Measure, Learn, and Improve: Physicians' Involvement in Quality Improvement." Health Affairs. 24(3): 1843-53, May-June 2005.

By Stephen C. Schoenbaum, MD, MPH, and Kenneth Segel

RELATED ARTICLE: Case Study: Pittsburgh, Pennsylvania

Physician Leaders Work to Eliminate Illegible and Incomplete Orders

Over the summer of 2005, Cliff Orme, the CEO of LifeCare Hospitals of Pittsburgh teamed up with Cynthia Ayers, MD, chair of his patient safety committee, to champion the elimination of illegible and incomplete physician orders.


Orme publicly committed the hospital to take any action necessary to eliminate all unsafe conditions for workers and patients.

The leaders first led detailed observations of the entire medication process, from the physician order to administration. These observations revealed that at least 40 percent of the orders that reached the pharmacy were illegible (requiring clarification or "educated" guessing), incomplete or both.

Clarifying just the most blatant of these problems required 17 percent of an order entry pharmacist's time every day.

These problems were generated by a poorly functioning system that did not use three important design principles for managing complexity:

  1. There was no signal to a physician or nurse writing an order that the order they produced was perfect. For example, the forms did not have columns for each component of necessary information (date, medication, dose, concentration, route, and frequency).
  2. There was no direct connection between the customer (the physician, acting on behalf of the patient) and the supplier (pharmacist).
  3. There was no method in place to call out any problem in the medication process and solve it to root cause quickly.

Ayers and other physicians then participated in a rapid series of experiments that redesigned the medication order form. Lifecare leaders enlisted physicians who produced illegible or incomplete orders to design personal experiments to eliminate the problem using the three principles for more effective connections between customers and suppliers.

Orme also set a clear expectation that any problem in the medication process, including problems within the experiments to improve the process, would be called out in real time.

After calling out the problems, the people who do the affected work attempt to quickly solve the problem to root cause, with the aid of a designated "help chain"--designated individuals assigned to help solve problems or seek assistance from the next level of the help chain while the line staff continues to do their work for patients.

Despite a tendency to drift back to traditional committee-based problem solving and a modest number of physician experiments under way, Ayers' and Orme's efforts produced a 43 percent reduction in illegible and incomplete orders between March 24 (baseline) and July 27, 2005.

RELATED ARTICLE: Case Study: Eastern Tennessee

Physician Drives Safety Learning at Wellmont Health System

Dale Sargent, MD, vice president of medical affairs at Wellmont Health System in eastern Tennessee in 2001 confronted a low (<15 percent) compliance with pneumovax administration in patients admitted with pneumonia at the Wellmont Bristol Regional Medical Center. This was despite the fact that pneumovax vaccination is a proven intervention to decrease recurrent pneumonia rates.


Sargent's investigation revealed some typical work flow issues, including that the vaccine was a low priority to nursing staff on day of discharge and that the clinical pathway wording was vague.

If a patient couldn't remember whether they had been vaccinated within five years (and many patients could not remember), this was considered a "yes" and no vaccine was administered.

Sargent also went to the admitting physicians and hospitalists and learned from them that two major concerns were limiting the rate at which they ordered vaccination.

  1. There was a perception among the physician group that vaccinating a patient who is still acutely ill (febrile) with the current episode might carry significant risk.
  2. And there was the thought that revaccination in a patient who has been recently vaccinated is costly and may carry a risk of adverse reaction to the vaccine.

Sargent asked a respected infectious disease specialist to research the physicians' concerns. She reported there was no risk to revaccination except for the slight risk of increased local reaction, and that the cost is minimal. The report also showed that there are no data to support that vaccination early in the course of acute pneumonia is risky or less effective.

After examining this research, the physicians then helped design a solution experiment. The pathway was rewritten to indicate that the vaccination is given on the third hospital day (ALOS is five).

This gives the vaccine a formal place on the medication administration record and the pharmacy will routinely dispense it without nursing making a special call. The protocol also now indicates that the vaccine will be given unless the patient or family specifically remembers being vaccinated. Those who don't remember are vaccinated.

As a result, vaccination rates went to 65 percent, which placed Wellmont among the top 10 percent of hospitals reporting in 2001. No adverse events were reported.

Physician leadership for safety has continued to accelerate at Wellmont. In the fall of 2004, Richard Salluzzo, MD, was named president and CEO of Wellmont. Salluzzo's first act was to publicly declare that Wellmont's most important objective was to become America's first safe health system.

Wellmont's publicity materials--including billboards--proudly proclaim that Wellmont facilities are "Learning to Become America's Safest Hospital." Salluzzo invested a total of 10 additional staff in two nursing units that are experimenting with safer approaches to care, including a transparent learning system that publishes a daily log of unsafe conditions and for many of those problems a root cause investigation and experiment to prevent recurrence.

© 2006 Anerican College of Physician Excutives
© 2006 Gale Group