Are Hospitals Less Safe Than We Think?

Bad doctors, prescription errors, surgical slips, and medical mistakes injure or kill hundreds of thousands of Americans every year.

Why patients are kept in the dark?

When I was a medical student, modern medicine began to seem as dangerous and dishonest as it was miraculous and precise. The defining moment came when I saw a sweet old lady I cared about die after a procedure she didn’t need and didn’t want.

I had been assigned to follow Ms. Banks, whose scans revealed advanced ovarian cancer. Despite the poor prognosis, the conventional treatment is major surgery to remove the uterus, cervix, fallopian tubes, and ovaries. But I got to know Ms. Banks, and she told me that she just wanted to spend time with her family and do a few more things before she died. I explained to her that she could be passing up a potential, albeit unlikely, cure; then at the morning staff meeting I tried to communicate her wishes to forgo both a biopsy and treatment. I was shredded up, down, and sideways.

The drive for the doctors to do a biopsy was like a train no one could stop. Eventually, by overstating the benefits and understating the risks, the doctors convinced Ms. Banks to undergo the biopsy to confirm her diagnosis. Then, during the procedure, the biopsy needle accidentally punctured a major blood vessel, which resulted in an added six-week stay in the hospital, marked by blood transfusions, multiple CAT scans, and malnutrition, since most of the time she was not able to eat. Those six hellish weeks turned out to be six of her last nine on earth. Despite the apparent problems with her care, information about her preventable complication and prolonged hospitalization were never presented in our staff meeting or reviewed internally in the same way that other industries learn from their bad outcomes. I realized that hospitals did not have to disclose their outcomes to anyone, even when they were much worse than the national average. In fact, when I explained to the head attending surgeon what happened and recounted Ms. Banks’s objections to the biopsy, I was told that sometimes patients don’t know what they want and we need to decide for them.

A host of new studies examining the current state of health care indicates that approximately one in every five medications, tests, and procedures is likely unnecessary. What other industry misses the mark that often? Others put that number even higher. Harvey Fineberg, M.D., president of the Institute of Medicine and former dean of the Harvard School of Public Health, has said that between 30 percent and 40 percent of our entire health-care expenditure is paying for fraud and unnecessary treatment. While patients are encouraged to think that the health-care system is competent and wise, it’s actually more like the Wild West. The shocking truth is that some prestigious hospitals participating in a national collaborative to measure surgical complications have four to five times more complications as other hospitals. And even within good hospitals, there are pockets of poorly performing services.

A new generation of doctors has been developing fair and simple ways to measure how well patients do at individual hospitals. In hospital-speak, we call the information “sensitive data”—data that would tell you which hospitals have much worse outcomes than others.

It’s the kind of data that, if you had access to it, would help you know just where to find the best care. But you don’t. And that is precisely the problem with the entire system: because a hospital’s outcomes are hidden from the public, neither consumers nor payers have any way of measuring whether the medicine they provide is good, adequate, or even safe. Much as the financial crisis was incubated when bank executives turned a blind eye to the ugly details about their mortgage-backed securities, so too does medicine’s lack of accountability create an institutional culture that results in overtreatment, increased risk, and runaway costs.

Politicians debate different ways to pay for our broken system. But if we are going to get serious about reducing health-care costs—and improving health-care outcomes—we need to address the 20 percent of medical care that is unnecessary and dangerous. The public should demand disclosure of a hospital’s patient-outcome statistics. After all, we have information on a car’s safety record to inform our decision about which car to buy. But when it comes to choosing medical care, the consumer is left to walk in blind. While we currently have a free market for health care, the competition is at the wrong level. Many patients tell me they choose their medical care based on parking. For an industry that represents one sixth of the U.S. economy, we can do better than that.

Years ago, one of my favorite public-health professors, Harvard surgeon Dr. Lucian Leape, opened the keynote speech at a national surgeons’ conference by asking the thousands of doctors there to “raise your hand if you know of a physician you work with who should not be practicing because he or she is too dangerous.” Every hand went up. Doing the math, I figured that each one of these dangerous doctors probably sees hundreds of patients each year, which would put the total number of patients who encounter the dangerous doctors known to this audience alone in the hundreds of thousands. If, say, only 2 percent of the nation’s 1 million doctors are seriously impaired or fraudulent (and most experts agree that 2 percent is a low estimate), that would mean 20,000 impaired or fraudulent doctors are practicing medicine. If each one of these doctors typically sees 500 patients each year, then 10 million people are seeing impaired or fraudulent doctors annually. Incredulous at the numbers, I took to asking the same question whenever I spoke at conferences. And the response was always the same.

A 2010 New England Journal of Medicine study concluded that as many as 25 percent of all hospitalized patients will experience a preventable medical error of some kind, and 100,000 will die annually because of errors. If medical error were a disease, it would be the sixth-leading cause of death in the country. My research partner lost his father due to a medical error. My medical partner lost his younger sister due to a medication error. My best friend’s mom had her breast removed unnecessarily because she was mistakenly told she had stage-III breast cancer. My grandfather died at age 60 from a preventable infection following a surgery he didn’t need. Andy Warhol died prematurely of a mistreated gallstone at 54; Saturday Night Live’s Dana Carvey had open-heart bypass surgery on the wrong vessel; and the singer Kanye West’s mother recently went to a surgery center for a routine plastic surgery, developed a rare complication, and died.

The wide disparity in the quality of medical care is no secret among hospital staff. In a study I conducted in 2006, we asked hospital employees, “Would you feel comfortable receiving medical care in the unit in which you work?” While there were hospitals where 99 percent said yes, at more than half of the hospitals we surveyed, the majority of health-care workers said no. And to the question of whether their hospital gives priority to what’s best for the patient, again, in more than half of the hospitals surveyed, the majority of health-care workers said no.

In other words, everyone who works in medicine knows about this problem but few talk about it. A cardiovascular anesthesiologist once described to me a colleague who was one of four heart surgeons at his well-known heart hospital. This surgeon had six consecutive deaths during routine bypass surgery. Half the operations of his last 10 surviving patients took several hours longer than the norm, often requiring the patient to be put back on the heart-lung bypass machine after having come off it. I asked my friend if he ever thought about reporting this surgeon to someone. He laughed and asked, “Like who?” The hospital administration loved this young doctor and was making a mint off his work. The senior partners were very protective of him—he covered their holiday shifts and happily tended to whatever the senior surgeons did not like to do. Whenever one of his complications was discussed at a peer-review conference, they cut him tremendous slack, attributing the death to some extenuating patient circumstance. My friend listened to all these excuses offered up in conference. As an anesthesiologist who had to work every day with the surgeons sitting in the peer-review conference, he decided to keep his mouth shut.

Hospitals sometimes fire doctors and nurses who speak up, sending a powerful warning to the medical community at large to stay in line. Kiran Sagar is a prominent cardiologist who has trained hundreds of doctors in reading echocardiograms of the heart and who was one of the first female cardiologists in Wisconsin. At her own hospital, the Aurora St. Luke’s Medical Center in Milwaukee, she undertook a study of interpretation of heart-echo tests by doctors and found that the quality of the interpretation varied widely depending on the doctor. It’s a problem endemic in many U.S. hospitals, and at a national cardiology conference she presented her findings: 29 percent of heart-echo interpretations are incorrect, she said, and followed up with suggestions for quality control. After the release of the study, the 65-year-old doctor was fired. (In a public statement, the hospital said that Dr. Sagar’s “contract was terminated for a number of reasons, none related to her study.”)

Today we are seeing a revolution in the methods of collecting information on how well hospitals are doing. For the first time in history, computers can “talk” to each other and track patients who are readmitted. Previously, a patient who left hospital A and was readmitted to hospital B was considered to have no complications by hospital A; now all these patients are captured in databases that generate a readmission rate for each hospital by condition (called “bounce backs” in medical jargon).

Similarly, based on the notion that the best way to judge a hospital is by the attitudes of the people who work there, Dr. J. Bryan Sexton and I helped develop a study of “safety culture.” We asked nurses, doctors, and all other employees questions, including “Do you feel comfortable speaking up when you have a safety concern?” and “Does the teamwork here promote doing what’s right for the patient?” Sixty hospitals participated, and we got remarkable results: at one third of hospitals, most employees believed the teamwork was bad; hospitals also began to tell us that their bad teamwork correlated with infection rates and negative patient outcomes. And even when doctors knew what needed to be done to make things better, they felt disempowered because of what they perceived to be a growing divide between hospital management and frontline providers.

Perhaps the most important part of that survey, however, was this: all 60 hospitals agreed to have the data we collected published, but only on condition that the individual hospital names be kept anonymous. Viewing of the information is top secret. So while hospitals have plenty of data with which to assess their services, the public does not. If you were about to check into a hospital and wanted to look up outcomes by medical condition, the percentage of hospital-ized patients readmitted within 90 days, or the average length of stay for your medical condition, you’d have no way of doing it. Similarly, if you were considering surgery and wanted to review the various hospitals for their complication rates on that procedure, you’d quickly discover it’s not available.

Take, for instance, the National Practitioner Data Bank collected by the U.S. Department of Health and Human Services, which is also known as the national “blacklist” of doctors. The public has absolutely no access to it. When I requested the list, I was given a version with the doctors’ names deleted; the only groups that can query the list are state medical boards or human-resources departments doing background checks. Ironically, sex offenders’ names are broadcast to the community when they move into town, but doctors who lose their license in one state because of sexual misconduct with a patient are shielded by name in the database if their license is later restored or if they continue to practice medicine in another state.
But there is a solution: if any of this information—lists of sanctioned doctors, or employee-safety surveys, or hospital readmission rates—were made fully public, positive results would reverberate throughout the health-care system. The effect would likely be a global reduction in patient harm and a rise in customer satisfaction. We know that because it has been done—once.

In the early 1990s, New York state set out to address the horrific patterns of bad outcomes that health officials had heard about in some of the state’s heart hospitals. Mark Chassin, who became health commissioner in 1992, didn’t want to just slap wrists. Instead, he and his team did something radical: they made heart-surgery death rates public. Instantly, New York heart hospitals with high mortality rates scrambled to improve. Hospital executives held meetings with heart surgeons, nurses, and techs to find out what they had to do to improve quality and safety. At one hospital, the staff reported that a surgeon wasn’t fit to be operating; his mortality rate was so high it was skewing the hospital’s average. His hospital administrators ordered him, point-blank, to stop doing heart surgery.

The result of the release of this data? Big, broad improvements in mortality statewide. Despite some criticism of the program’s notable loopholes, with each passing year of public reporting the state’s average death rate went down. In addition, bad outliers, like the hospital with the 18 percent mortality rate, were reined in. Erie County Medical Center was the state’s worst-performing hospital, with an overall mortality rate higher than that of soldiers wounded in the Iraq War. Within three years the mortality rate was cut to 7 percent, and in the years since, it has fallen to 1.7 percent. Introducing transparency to New York’s heart centers brought something very novel and powerful to health care: public accountability.