By CATHLEEN F. CROWLEY and ERIC NALDER HEARST NEWSPAPERS
Aug. 7, 2009, 5:22PM
Richard Flagg drowned in his own blood.
Stanley Stinnett choked on his own vomit.
Both were victims of the leading cause of accidental death in America — mistakes made in medical care.
Experts estimate that a staggering 98,000 people die from preventable medical errors each year. More Americans die each month of preventable medical injuries than died in the terrorist attacks of Sept. 11, 2001.
In addition, a federal Centers for Disease Control and Prevention study concluded that 99,000 patients a year succumb to hospital-acquired infections. Almost all of those deaths, experts say, also are preventable.
These numbers are not absolutes. There is no definitive study — which is part of the problem — but all of the available research indicates that the death toll from preventable medical injuries approaches 200,000 per year in the United States.
Ten years ago, a highly publicized federal report called the death toll shocking and challenged the medical community to cut it in half — within five years.
Instead, federal analysts believe the rate of medical error is actually increasing.
A national investigation by Hearst Newspapers found that the medical community, th federal government and most states have overwhelmingly failed to take the effective steps outlined in the report a decade ago.
Hearst also found that in states like California that have put some regulations in place, hospitals often ignore the rules without penalty.
Consequently, over that period, as many as 2 million Americans have died needlessly of preventable medical mistakes.
Secrecy built into the system has long kept both the scope of the crisis and the specific problem areas out of public view. Some of those lives could have been saved with innovations as simple as color-coding medical tubes to avoid confusion.
A Hearst data analysis lifted a corner of that veil of secrecy to show that in five states served by Hearst newspapers — New York, California, Texas, Washington and Connecticut — only 20 percent of some 1,434 hospitals surveyed are participating in two national safety campaigns begun in recent years.
Also, a detailed safety analysis prepared for Hearst Newspapers examined discharge records from 1,832 medical facilities in four of those states. It found major deficiencies in patient data states collect from hospitals, yet still found that a minimum of 16 percent of hospitals had at least one death from common procedures gone awry — and some had more than a dozen.
Now, as the Obama administration wrangles with Congress over access to health care, frustrated patient-safety leaders say another priority must finally be addressed — making hospitals safer.
Back in November 1999, the report titled “To Err Is Human” was issued with the highest of hopes. Its authors believed it promised the start of a revolution in patient safety.
The report certainly sparked awareness of the scale of the problem. But some of its authors say the revolution was doomed by a lack of political leadership and the health care lobby’s vested interest in maintaining business as usual, especially secrecy surrounding dangerous medical errors.
“We didn’t have any government efforts. We didn’t show leadership and take charge and do what needed to be done,” said Dr. Lucian Leape, one of the authors of “To Err is Human,” who is considered the father of the modern patient safety movement.
The report marked the first time that an authoritative voice — backed by the national Institute of Medicine — urged the industry and critics alike to stop blaming doctors and nurses. People make mistakes, the report said, so medicine must design systems that can reduce errors and prevent harm from reaching the patient when a mistake is made.
Like a car ignition that won’t release the keys until a driver shifts into park, the safe systems envisioned by the authors make it easy for health care workers to do the right thing. In the operating room, for instance, don’t allow two small containers, both filled with colorless liquid — one a dye and one a potentially lethal antiseptic — to be close at hand while a medical team performs a procedure. That mistake caused a patient at a Seattle hospital to die after the antiseptic was injected into her bloodstream.
No nationwide system
The report also called for a mandatory nationwide reporting system for medical errors. That never happened.
“The (American Medical Association) came out foaming at the mouth,” said Arthur Levin, president of the Center for Medical Consumers and an author of “To Err is Human.” “And I think the decision was made to let it fall off the table because the wisdom was that you needed the cooperation of the profession to make progress. I think the tacit compromise was, ‘We’ll let that go.’”
The AMA and the American Hospital Association vehemently opposed an attempt by President Bill Clinton to create a mandatory reporting system for serious errors. The leaders of both groups snubbed an invite to a White House press conference introducing the president’s patient safety agenda in 2000. The groups launched a multimillion-dollar advertising campaign that said mandatory reporting would drive medical errors underground. From 2000 to 2002, they spent $81 million on lobbying efforts, according to campaign statistics collected by the Center for Responsive Politics.
Mandatory reporting was dead on arrival.
By contrast, Americans know exactly how many people die from car accidents each year because lawmakers long ago decided that was a step toward preventing them. Motor vehicle deaths are the No. 1 cause of accidental death in the Unites States, with more than 43,600 deaths in 2006, according to the CDC. The next three causes — poisoning, firearms and falls — account for 90,000 deaths, combined.
But it is clear that if medical errors and infections were better tracked, they would easily top the list. In fact, a visit to your doctor or a hospital is twice as likely to result in your death as is a drive on America’s highways.
Now, AMA officials say they support voluntary reporting but still have the same concerns about mandatory reporting as they did a decade ago.
The American Hospital Association supports disclosing mistakes to the families involved, said Nancy Foster, vice president for quality and patient safety for the hospital group.
“The people who are injured and their loved ones need to know about what happened and why it happened,” Foster said.
That enthusiasm does not extend to the general public. Foster wouldn’t say whether the hospital association supported mandatory public reporting of medical mistakes, saying only that the association prefers voluntary reporting.
A sorry scorecard
It’s revealing to review other key recommendations from “To Err is Human” in the context of what has been done in the 10 years since they were made. The report:
• • Encouraged states to require medical error reporting. Only 20 states plus the District of Columbia have done so, and evidence shows that even in those mandatory-reporting states, hospitals report only a tiny percentage of their mistakes.
• • Said the public “has the right to be informed about unsafe conditions.” But 45 states plus the District of Columbia don’t provide hospital-specific information, either because they don’t allow access or because they don’t collect the data.
• • Recommended the creation of a national patient safety center. The center is underfunded and has fallen far short of expectations.
• • Urged that hospitals improve the level of safety within their walls. Hundreds of hospitals responded, a few of them comprehensively pursuing safer care. Thousands did much less.
• • Advocated a voluntary system for hospitals to report and learn from errors. Five years later, Congress approved legislation for “patient safety organizations” to serve this role, then took four more years to create rules to govern them. But the new organizations are devoid of meaningful oversight and further exclude the public.
Progress, but not enough
The positive steps are overshadowed by the continuing death toll.
Hearst Newspapers interviewed 20 of the 21 living authors of “To Err is Human,” and 16 believe that the U.S. hasn’t come close to reducing medical errors by half. Four did not know or declined to answer because they are removed from the world of patient safety, and one did not return numerous calls and e-mails.
In its 2008 annual report to Congress, the Agency for Healthcare Research and Quality, a part of the Department of Health and Human Services, reported that preventable medical injuries are growing each year by 1 percent, the first time it had reported such an increase.
AHRQ’s analysis showed that more people suffered accidental cuts and tears to their organs during surgery, more patients developed avoidable bloodstream infections from catheters, and one out of seven hospitalized Medicare patients experienced at least one adverse event during a hospital stay.
“Unfortunately, there hasn’t been a significant improvement in the level of medical errors, and what is getting worse is hospital-based infections that are preventable,” said Kathleen Sebelius, secretary of the Department of Health and Human Services, which oversees AHRQ, Medicare and the Food and Drug Administration.
Sebelius claimed the death toll is 100,000 annually, the same as 10 years ago. When asked whether that figure was based on the 1984 study used by “To Err is Human” to create the 98,000 estimate, she said, “No, that is from 2008 data.”
Sebelius’ office did not respond to requests to produce the 2008 data. Because it does not exist. Even AHRQ’s annual report to Congress cites the 1984 figure as the most current estimate of deaths from medical errors. Add infections, and the death toll nearly doubles.
Leape is frustrated by how few hospitals adopted 34 proven safe practices endorsed by the National Quality Forum, a coalition of medical groups. At a May conference of hospital leaders, Leape asked audience members to raise their hands if they had plans to implement all 34 practices. Only three hands went up in a crowd of 175.
“What you’ve seen has all been essentially voluntary,” Leape said. “It’s … all been done by people on the front line — doctors, nurses, pharmacists who don’t like to hurt people and they want to figure out a better way.”
But Leape added, “What we are hard-pressed to do is show that it’s made a difference.”
Poor performance in safety areas
The detailed safety analysis conducted on behalf of Hearst Newspapers by Niagara Health Quality Coalition, a independent nonprofit in Buffalo, N.Y., used patient discharge data from 1,832 medical facilities in New York, Washington, California and Texas. It found that at least one in six of those facilities had preventable deaths from common procedures, including cases in which medical instruments were left inside patients and transfusions were done incorrectly. Analysts also discovered serious weaknesses and incomplete reporting in hospital patient data in California , Texas and Washington — many problems unknown to state regulators — that made measurement of many easily tabulated errors impossible.
Also, it found 399 of those hospitals had poor performance in at least one, and in some cases several, safety indicators developed in recent years by federal health researchers.
Deaths from medical injuries happen behind the doors of a hospital room. Unlike a national tragedy that takes hundreds of lives in an instant, these deaths are singular and often secret.
Doctors fudge death certificates, leaving out information that would point to medical error as a prime or contributing cause of death, according to court records and other documents examined by Hearst reporters and graduate students at the Toni Stabile Center for Investigative Journalism at Columbia University’s Graduate School of Journalism.
Norine Zazzara, 81, died of pneumonia at St. Joseph Hospital in Syracuse, N.Y., according to her initial death certificate. Hospital records show she initially went to the hospital for a shot of diuretic to treat leg swelling. She contracted a MRSA infection, or methicillin-resistant Staphylococcus aureus, and developed pneumonia. After several weeks on a ventilator, she died.
Her daughter, Betsy Zazzara, persuaded her doctor to change the death certificate to reflect the infection. She said the doctor asked her whether it really mattered what the death certificate said. “Yes,” she replied. “One of these days they may start counting these people who have died of MRSA, and I want my mother to be counted.”
The CDC — which is supposed to track the nation’s deaths and diseases — is aware of the inaccuracies in death certificates.
Medical error is “often not reported,” said Robert N. Anderson, chief of the CDC’s Mortality Statistics Branch.
He said doctors aren’t given enough motivation to report medical errors, and because of liability, “it would cause them problems down the road, so that there is a disincentive to report it.”
The 22 authors of “To Err is Human” debated public disclosure and fears that it would create more lawsuits, feed the blame game and drive errors underground.
The authors, in the end, decided patients deserved the information.
“These are the kind of things the American public has the right to know about and that patients should know about when they are selecting a particular hospital or surgeon,” said Janet Corrigan, who was the lead staff writer of “To Err is Human” and is now president of the National Quality Forum.
The authors recommended creating a patient safety center to conduct research and oversee a nationwide reporting system, and they said the AHRQ was the logical place to house it.
The agency excels at research. It sponsors more than 100 research projects and mails pamphlets explaining proven safe practices to every hospital CEO in the nation but has no authority to demand compliance.
As for nationwide reporting, AHRQ officials say it is too expensive and too difficult to implement, and they have no authority to mandate it, anyway.
Its patient safety budget is $49 million, half of what the “To Err is Human” report recommended.
The agency is “not positioned or equipped to be the social change agent that we need,” said Dr. Donald M. Berwick, a “To Err is Human” author and president of the Institute for Healthcare Improvement in Boston.
Dr. Carolyn M. Clancy, AHRQ’s director, said the agency has “returned a great deal of value to the American taxpayer for the resources we were given.”
Patient safety experts like Berwick and Leape argue that accurate measures of errors are not necessary for reducing them but concede progress can’t be proven until they exist.
The Obama administration does not support a nationwide, mandatory reporting system.
“The best thing to do is to create the incentives and the knowledge around best practices to prevent the errors from ever occurring,” Nancy-Ann DeParle told Hearst Television. DeParle is Obama’s health adviser and director of the White House Office of Health Reform.
“If we prevent the errors from occurring, then we don’t have to worry about … a massive reporting system,” she said.
States’ spotty records
Yet without a nationwide reporting system for medical accidents, states are left to collect the information. The result has been a chaotic, dysfunctional patchwork.
Hearst research shows that 20 states and the District of Columbia have mandatory reporting systems and five states are in the process of forming such systems. But only five of the 20 — Washington, Massachusetts, Minnesota, Colorado and Indiana — are transparent enough to be useful to consumers by revealing hospital names. California reveals some hospital-specific information and will provide more when its system is complete. Illinois, New Hampshire and North Dakota will also reveal hospital-specific information once their systems are running.
Meanwhile, 20 states have no reporting whatsoever. The remaining five states have, or are setting up, voluntary reporting.
In states that do report, standards vary wildly and enforcement is often nonexistent.
New York’s reporting system has run out of money and staff. The last public report it produced is 4 years old.
The law mandating reporting in Texas expired in 2007, and funding ran out. A new reporting law has been passed, but no funds have been allocated.
Washington state requires reporting but doesn’t enforce that requirement — and the Legislature didn’t provide funds to analyze the results.
Carol Wagner, vice president for patient safety at the Washington State Hospital Association, is a national leader in the hospital industry on patient safety. She favors nationwide reporting.
Wagner wouldn’t mind if it were mandatory or reasonably public, as long as the information is collected by a single, “competent” organization, focused only on patient safety, that analyzes reports and quickly disseminates its findings to the hospitals and the consumer.
If hospitals aren’t reporting medical errors in mandatory states like California and Washington, said Wagner, it is probably because many organizations, state and national, are repeatedly requesting various data. Too few solutions are coming back. A well-designed nationwide reporting system would provide “the ability to learn faster, ways to limit errors,” she said.
State Rep. Tom Campbell of Washington state, a longtime patient-safety advocate, isn’t convinced the hospitals will support such a move.
“We are going to play tug of war on this subject, like on everything we do in patient safety,” he said, adding that “the hospitals will do everything they can to keep the patients from getting close inspection” of error reports.
The inspector general of the Department of Health and Human Services sounded an alarm in December about the lack of uniformity between the state systems. The differences in regulation and standards make state data “unsuitable … to identify national incidence and trends,” the report said.
A new tool being tested around the country may evolve into a technique for measuring medical injuries nationwide. The Global Trigger Tool, being developed by Berwick’s Institute of Healthcare Improvement, reveals that many dangerous medical mistakes go unreported — even undiscovered.
Using the method, a nurse combs through a sample of medical records looking for hidden clues, called triggers. Among the tool’s findings at Intermountain Healthcare of Utah is that one out of 100 patients suffered serious injuries and three out of 1,000 died, with probably half of those resulting from medical errors, said Dr. Brent James, an Intermountain executive who led the study and is an author of “To Err is Human.”
None of the events was reported by staff, James said.
Little help for patients
So how do patients know which hospital is doing a good job? They don’t.
AHRQ created a software program that mines hospital billing and discharge data for adverse events and analyzes data 40 states volunteer to the agency. But AHRQ releases only aggregate information and promises the states it will keep hospital-specific information secret.
The Centers for Medicare and Medicaid Services, a division of HHS, collects and publishes hospital-specific information on mortality and “processes of care.” A care process involves actions like giving antibiotics to patients before surgery — practices that may influence infection but do not reveal a hospital’s true infection rate.
“Too often in health care we measure effort, and we need to concentrate on results,” said Bruce Boissonnault, president of Niagara Health Quality Coalition.
The Centers for Medicare and Medicaid Services’ “Hospital
Private groups do their best to fill the gaps.
The Institute of Healthcare Improvement recruited 2,000 hospitals to join its “5 Million Lives Campaign,” which meant implementing 12 of the 34 safe practices. The institute publishes the names of the participating hospitals.
The Leapfrog Group — a nonprofit created by a “To Err is Human” author, Charles Buck, and other business executives — attempts to measure hospital safety efforts. Many hospitals refuse to participate.
For-profit companies like HealthGrades and Thomson Reuters publish hospital ratings, but they also accept money from the hospitals and for consulting, and HealthGrades charges hospitals for the right to publicize their rankings
Boissonnault’s Niagara Health Quality Coalition, a nonprofit group based in Buffalo, N.Y., proves that hospital-specific safety measures can be published. Niagara posts mortality rates for 15 medical procedures and 14 safety measures for New York hospitals, including foreign objects left behind during surgery.
To get the information, Niagara analyzes hospital billing and discharge information — the same data AHRQ collects but keeps secret.
Forces beyond hospitals’ control make the task of reducing errors difficult. The complexity of medicine is growing so fast doctors can’t keep up with the knowledge — and, perversely, hospitals are paid more when they err because they can charge for follow-up care to treat the mistake.
“It’s not because physicians are bad people or stubborn. It’s because the amount of knowledge and the amount of technology to incorporate continues to grow beyond the ability of either individual physicians or medical system to incorporate them in a systematic way,” said Dr. David Lawrence, a “To Err is Human” author and the retired chairman of the Kaiser Foundation Health Plan.
On top of that, a payment system gives hospitals no incentive to provide safer care.
When Intermountain improved its system for prescribing heart patients the proper medications on discharge, re-hospitalizations were reduced by 900 beds a year. As a result, the hospital lost $3.5 million in revenue.
“To my hospital administrators, there was actually a certain amount of whining about this, but they know better than to whine too much,” Intermountain’s James said.
Last year, Medicare stopped paying for eight types of medical errors, and some states and private insurers have followed suit.
Experts like Lawrence, Leape and James believe health care reform must include changing the payment system to reward hospitals for the quality of their care, not just the quantity.
More secrecy ahead
Under federal regulations issued in January on the last day of the Bush administration — a result of a 2005 federal law — medical practitioners can be fined $10,000 for publicly revealing information about errors.
Candid information about dangerous events in hospitals will be funneled into top-secret “black boxes” — dozens of privately run and sponsored patient safety organizations around the country that will presumably analyze the information to improve medical operations, with little or no oversight.
The HHS stated that there will be “little direct federal involvement” in the patient safety organizations and no funding.
The organizations, which now number 66, are expected to reduce adverse events by 1 percent this year, rising to 3 percent by 2013, according to the Federal Register.
Ironically, the only enforcement provisions with teeth involve wrongful release of information about patient safety cases, which will be investigated by the HHS Office for Civil Rights.
“What’s the public benefit? Supposedly, we are counting on the goodwill of hospitals,” said Levin, the Center for Medical Consumers’ president. “I’m not a believer. Show me the beef.”
If patient safety improves, it will happen too late for too many, including Richard Flagg and Stanley Stinnett.
Surgeons at Meadowlands Hospital in Secaucus, N.J., accidentally removed Flagg’s healthy lung, leaving behind a tumor in the 60-year-old barge captain’s diseased lung, according to the state Board of Medical Examiners. The tumor bled and made him cough. Flagg survived three years, attached to oxygen, until the tumor ruptured and he drowned in his blood.
Stinnett, 49, entered the emergency room at Memorial Medical Center in Modesto, Calif., with broken ribs from a motorcycle accident and left in a body bag.
A series of errors killed him — starting with improper treatment of an intestinal obstruction, according to testimony from his family’s medical experts. The drug oxyodone suppressed his gag reflex, and his own vomit shut off his airway, one expert concluded. The doctor denied responsibility, but the jury awarded the family $8.5 million. The doctor is appealing.
Stinnett was, like many Americans, at one of his life’s most vulnerable moments when he entered the hospital, and he didn’t have much choice about his care. Even when people have choices, they usually have no information about a hospital’s safety record.
“They did nothing for him but fill him with medication to let him die peacefully,” said his widow, Holly Stinnett. “There was nothing wrong with him to begin with but four fractured ribs.”
Is meaningful reform ever going to happen?
“There’s a point at which you have to say, ‘Is it ethical to allow preventable harm to continue to occur when you know how to prevent it?’” Levin said. “When do you say enough is enough?”
Contributing to this story were Olivia Victoria Andrzejczak, Kyla Calvert, Ana Azpurua, Andrew Schmid and Emily Witt of the Toni Stabile Center for Investigative Journalism at Columbia University; John Martin, adjunct professor, Columbia University Graduate School of Journalism; Sarah Hinman, director of news research, Albany Times Union; Laurie Kinney, reporter, Hearst Television; Terri Langford, enterprise reporter, Houston Chronicle; Lance Williams, investigations editor and reporter, San Francisco Chronicle; and Don Finley, medical writer, and Kelly Guckian, database editor, San Antonio Express-News.