Preventing Hospital Errors by Howard C. Berkowitz

I have asked Howard Berkowitz to guest-blog on hospital errors.

Howard is in an unique position to write on this topic because he consults on medical information systems for hospitals and also has been a long-time patient. Over the years, he has taken an unusually detailed decision-making role in his own care for heart diseases and diabetes which, he says, “has kept me going, with bad heart genetics, at least 17 years more than my father.”

Howard also reports that “when no one else would coordinate my mother’s complex cancer care, I did so...and I know what it is to preserve the semblance of life, when only pain remains. Complex pain management is also one of my interests; too few doctors know that pain should always be controllable.”

As a result of his own health problems and his parents’ illnesses, he has spent more time in hospitals than anyone would ever want to endure. But unlike most of us, he understood what was going on. Originally trained in microbiology and biochemistry, Howard was doing independent research in antibiotic resistance and working in a clinical laboratory while in high school. He confesses that, for his 10th birthday, he asked his mother for a copy of the Merck Index of Chemicals and Drugs. Subsequently, he built the first clinical computer system for Georgetown University Hospital, developed virological systems for Electronucleonics’ “hot lab” and developed cardiac care simulators and for the George Washington University School of Medicine, Office of Computer-Assisted Instruction. He also developed the first automated blood bank laboratory tools for the Red Cross.

Full disclosure—he has two patents in process for hospital communications and staff management dedicated to keeping them informed, in real time, of patient needs. His post follows below.

IF YOU’D LIKE TO COMMENT ON THIS POST, PLEASE CLICK HERE TO E-MAIL MAGGIE WITH YOUR THOUGHTS.

In recent years, U.S. hospitals have been trying to reduce preventable errors that range from leaving the stray instrument in a patient’s stomach to letting bedsores become oozing ulcers.  But they haven’t been making as much progress as one might hope. According to the Fourth Annual Patient Safety in American Hospitals Study published last April, the data show that from 2003 to 2005 Medicare patients alone fell victim to 1.16 million preventable incidents, leading to 247,622 preventable deaths and $8.6 billion in preventable costs (click here for a short analysis of the data).

This has led some payers—including Medicare—to ask “Why should we be picking up the tab for the hospitals’ mistakes?”  Answering that question, the Centers for Medicare and Medicaid Services (CMS) is threatening to stop paying hospitals for costs incurred as a result of some of the most common and preventable medical errors—a list that CMS labels: “Things That Should Never Happen.”

On the face of it, this sounds reasonable. Today, there are no financial penalties for hospitals that report a large number of “adverse events.” The hospital simply recodes the diagnosis and receives reimbursement for the additional length of stay and treatment required to correct for the mishap. But let me suggest that, rather than penalizing hospitals for each errors, CMS might learn something from the U.S. aviation industry—as well as from Germany’s healthcare system.

Begin with our aviation industry. Modern aviation has a culture that formalizes the recognition and avoidance of errors. Pilots don’t try to cover up their mistakes because the Federal Aviation Administration has a mandatory program that requires reporting errors on a no-fault basis. In other words, under Crew Resource Management (CRM), pilots who barely avoid mid-air collisions are not penalized if they promptly disclose all the circumstances. Safety specialists will then examine what failed and spread the knowledge throughout the industry. (In most cases, aviation catastrophes, like many medical catastrophes, come, not from a single error, but from a sequence of them. If any of the errors had been prevented, the end result might have been avoided.) In the US medical system, by contrast, physicians are reluctant to admit to a slip-up, in part because they fear a malpractice suit.

Not long ago, when I was examining the design of some hospital information systems that track “workflow” (or “who did what when”), I realized that in cases where several patients were showing signs of a hospital-acquired infection, our system logs could be used to backtrack, and identify the places, staff, and equipment contacted by all of the patients. By looking at these common experiences, investigators could identify the source of the infection. Once known, infection control specialists could stop further infections. But when the marketing staff of the company I was working for suggested the idea to several hospitals, they learned that the hospitals were worried that the report would be discoverable by a malpractice attorney acting for any of the patients. So the hospitals turned down the idea. In other words, they would rather not know where their hospital’s infections originate—because they don’t want other people to know.

But fear of litigation is not the only obstacle to tracking and discussing errors. There also is the unfortunate fact that the culture of medicine still insists that doctors are supposed to be infallible. Put it this way: pilots use checklists every time they fly, but I encounter a lot of physicians who don’t want to check a reference book in my presence—even though I tell them that it reassures me to see them look something up.

The notion that the chief surgeon must be all-knowing intimidates other doctors and nurses, which explains why they don’t speak up, even as they witness a catastrophe unfold. Here again, the aviation industry’s Crew Resource Management (CRM) offers a useful counter-example. Aviation safety people encourage other crew member to speak out when they see a potential safety hazard. CRM failed when Air Florida 90 hit a highway bridge and then crashed into the Potomac. The voice recorder shows the copilot suspected they didn't have adequate takeoff speed. Had he been more assertive, the captain might have aborted the takeoff and a lot of people might not have died.

By contrast, United Flight 232 made a controlled crash landing in Sioux City, Iowa, after a total hydraulic failure. Other test pilots, knowing what the real crew faced, tried the situation in a simulator, and almost all crashed fatally in a very short time. The real crew essentially kept an unflyable DC-10 in the air for around 45 minutes. 111 people died and 185 lived, but it was amazing anyone lived. The crew distributed workload so while one pilot was physically flying and focused on that, others were planning the next step. Anyone with a good idea voiced it immediately.

Hospitals need to adopt CRM’s code. Just one example: in an operating room, any member of the team that has the slightest suspicion that the doctor is operating on the wrong limb should speak out loudly. But that doesn’t always happen.

Finally, hospitals hide their mistakes because, in our market-based system, they worry about losing market share if they begin admitting to botched operations. A five-country survey of hospitals in the U.S., the UK, Australia, Canada and New Zealand published in Health Affairs in 2004 revealed that hospitals in the U.S. and Australia were most opposed to public disclosure of medical error, infection, and mortality rates. It is probably not a coincidence that these were the two countries where loss of patients to competition was a major concern—along with malpractice costs.

In my next post, I’ll talk more about what is wrong with Medicare’s proposed solution—and what we might learn from Germany’s healthcare system.