About five years ago, Dr. Michael Hicks tells me, the UNT Health Science Center at Fort Worth invited a man named Christopher Hart to speak during an industry symposium regarding patient safety and medical errors. Hart had nothing to do with healthcare. Not with how it’s delivered, paid for, or scrutinized—none of that. In fact, Hart monitors the safety of the transportation industry. He’s now the chairman of the National Transportation Safety Board, the independent federal agency that digs into the details behind every single civil aviation accident in the U.S. 

Hart “made a very cogent point, I thought, in that he was asking us rhetorically why there isn’t some type of equivalent function in healthcare,” says Hicks, executive vice president for clinical affairs at UNTHSC. “If you look at the numbers, if you equate even the most conservative estimation of mortality and patient death from medical error, it’s equivalent to a couple of 747s crashing every day. If you have even a small private plane crash over here, the NTSB is going to come see what they can learn to help prevent that from happening again.” 

An awareness of medical errors is growing in this country, particularly because of a few high-profile studies and white papers that have attempted to wrap their arms around the problem. One, researched by surgeons at Baltimore’s Johns Hopkins School of Medicine, proclaims that medical errors kill more Americans than anything except heart disease and cancer. That study, published in May in The BMJ medical journal, argued that more than 250,000 Americans die from medical error each year. This goes for the more egregious ones—objects left in the patient, for example, the wrong organ operated on—as well as the more subtle miscues, like the ones caused by “variation” in practice or by a patient not taking the medicine they’ve been prescribed and no one catching it. 

More conservative estimates—mostly from a widely cited 1999 report by the National Academy of Medicine (then known as the Institute of Medicine) using data from a 1984 Harvard study—put deaths from errors at between 44,000 and 98,000 annually. 

Either way, Hicks says, “we are killing probably one or two people every day, maybe more,” in Dallas-Fort Worth.

No matter which estimate you believe—and most doctors I’ve asked think the true number falls somewhere in the middle—this is an epidemic that can affect all Americans, regardless of age, financial status, or political persuasion. There is a pure business argument to reduce errors, too. Just consider the huge financial incentive to prevent error, in both wages and lost productivity in the workplace for days missed unnecessarily (and that’s for those who live through the lost days). Medical errors are estimated to directly account for $50 billion in annual healthcare costs. 

“We are killing probably one or two people every day, maybe more.”

- Michael Hicks, UNT Health Science Center
One other problem, however, is that the numbers are difficult to pin down with specificity because of how deaths in this country are recorded. When a patient dies, doctors enter billing codes (known as ICD-10 codes) that list the supposed cause of death. The thing is, though, the medical error tends to cede to the patient’s underlying health problem. Dr. Martin Makary, one of the Johns Hopkins researchers responsible for The BMJ study, likes to tell of a young transplant patient who was given an unnecessary pericardiocentesis—a procedure where the doctor uses a long needle to drain fluid accumulation around the heart. It’s normal for a patient after an eight-plus-hour surgery to have some fluid accumulated there, likely from all the IV fluids pumped into the body, Makary says. In this case, the needle caused a “pseudo-aneurysm,” which ruptured, causing her death. 

“What do you put on that death certificate? You put what is the underlying cause of the patient’s death and the immediate cause,” Makary said. “Her death was counted toward the No. 1 cause of death in the United States, which is cardiovascular disease. But she didn’t die from that. She died from the care that she received, not the disease or illness for which she presented.”

Makary and research fellow Michael Daniel determined a mean rate of death from medical error by analyzing a series of studies that had been published on the topic dating back to 1999. They took that rate and applied it to the total number of American hospital admissions in 2013. Voila: 251,000. This research wasn’t finished when the 84th Texas Legislature was in session last year. But state Sen. Jane Nelson, a Republican from Flower Mound, prioritized the issue and secured support for a bill that sent $4 million to the UNT Health Science Center to launch the Institute for Patient Safety. 

Over the next 12 months, Hicks of UNTHSC and a team of stakeholders from Texas Christian University, John Peter Smith Health Network, and Cook Children’s Medical Center joined together to determine what the institute would actually look like. It was decided that it would be “the convener of all of these experts,” Hicks says. Currently, there are institutions that examine and research medical errors—including Johns Hopkins’ Armstrong Institute for Patient Safety and Quality—but they do so from an academic standpoint. Hicks says the program in Fort Worth will focus on actionable change, developed not just by schools but by caregivers and health systems.

It’s not that our prominent health systems aren’t addressing medical errors already. They are. Many are working to create an environment not of retaliation but of analysis and improvement, and to remove as many possibilities for human error as they can. But these are big tasks, and Hicks believes the best way forward is to work together—not only for a better understanding of how significant the problem is, but to share best practices about ways to cut down on what’s called variation.

Texas Health Resources, the Arlington-based system with 27 hospitals and more than 100 care sites across the region, began deeply analyzing its processes in the past five years. It’s released a public quality report, warts and all, and has made available the findings of an independent survey of its best-known medical error: The 2014 death of Thomas Eric Duncan, who was diagnosed with sinusitis and sent home only to return days later with symptoms of Ebola. 

That same year, THR began working with Press Ganey’s Healthcare Performance Improvement organization to develop a series of standards known as “reliable care blueprinting.” Dr. Dan Varga, THR’s chief clinical officer, says the system has been able to standardize some forms of care, cutting down on variation. As he puts it, “Don’t give us 57 ways to administer insulin to somebody that needs their blood sugar controlled. Tell us how they need to have their insulin managed, and we’ll do it.” Modules have been developed that do exactly that, and doctors are able to back up their success rate with data to convince any physicians who may be on the fence about the topic, Varga says. Best practices, after all, are proven. 

But, Texas Health isn’t the only game in town when it comes to medical errors. Parkland Health and Hospital System overhauled safeguards (499 of them), as mandated by the feds, after a patient died awaiting care in 2008. Baylor Scott & White has its own in-house initiatives as well. Hicks realizes the potential impact to be had if the institute brings all these institutions together, fulfilling that “convener” concept mentioned earlier. “You will see the health science center woven throughout the institute, but it’s going to be just as much TCU’s institute or JPS’ institute or, if they agree, Baylor or Texas Health Resource’s institute, because that’s how the problem is going to be solved,” Hicks says. “It’s not going to be solved because we have a few offices and a few people on this campus. It’s going to be solved because the community embraces it.”