In 1998, a doctor began operating on a patient’s right knee. After he cut into the knee, he realized there was nothing to do there — because it was the left knee that needed repair.

In 2001, surgical residents cut into the left side of a brain-surgery patient’s head, only to discover that they should have opened the right side.

In 2004, an anesthesiologist inserted a catheter on the wrong side of a patient’s neck.

And within the past year, on three separate occasions in two different hospitals, doctors seeking to drain blood that was pooling near patients’ brains drilled holes on the side that didn’t have excess blood.

All these wrong-site surgeries occurred at hospitals in Rhode Island, but this state is hardly unusual. Similar incidents have happened around the country, and they keep happening despite a decade of concerted efforts to prevent them.

Patients expect doctors to know where on the body to operate. Why can’t health-care professionals get such a simple thing right every single time?

The answer: it’s really not so simple. Surgery involves many people carrying out multiple actions, often amid distractions under tight time pressure. Innocent small mistakes can quickly cascade into disaster.

The persistence of wrong-site surgery errors around the country has prompted the Joint Commission, the national agency that accredits hospitals, to reassess its protocol for preventing them — which went into effect in 2004 but hasn’t reduced the incidence.

“All of us expected the rate of errors to be much lower” by now, said Dr. Mary Reich Cooper, the new vice president and chief quality officer of Lifespan, the parent company of Rhode Island Hospital. “We realize how much we have to think about every single step.”

Cooper, who has worked in quality improvement for years, had held her Lifespan job for scarcely a month when a wrong-site surgery occurred at Rhode Island Hospital, on July 30, the second such mistake at the hospital this year and the third in six years. This “pattern” prompted an unprecedented action by the Health Department — an order for immediate changes, investigations and monitoring. In each of the two previous cases, the hospital did an extensive analysis of what went wrong and put in place procedures to prevent a recurrence. And yet it recurred anyway.

As medical errors go, wrong-site surgery is considered rare, although dramatic. A study of malpractice claims in the journal Archives of Surgery last year estimated that wrong-site surgery occurs once in every 113,000 surgeries, about 10 times less often than surgeons leave foreign objects inside the body. Wrong-site surgery is also unlikely to cause severe, lasting harm, except in such cases as removal of the patient’s only healthy kidney. In all the Rhode Island wrong-site cases in recent years, the patients recovered.

Even so, government regulators and quality-improvement organizations consider wrong-site surgery a grave mistake: It shocks the public, shakes people’s faith in medicine, and can be a sign of deeper problems in a hospital’s quality-control systems. And everyone agrees it just shouldn’t happen.

A glance at the Rhode Island incidents shows the many different ways that things can go wrong. In the case of the 1998 knee surgery, at Kent Hospital, the staff mistakenly prepped and draped the wrong knee, and the surgeon just went to work on the knee that was ready.

Since then, hospitals have started marking the surgical site on the patient’s skin. And in the 2004 incident involving the catheter in the neck, which occurred at Miriam Hospital, the staff did mark the spot. But, the mark was obscured by the surgical drapes. (The procedure had to be canceled because the anesthesiologist had put the catheter, needed to monitor the heart, on the same side as where the surgeon planned to work, instead of the opposite side.)

In one of the brain-surgery incidents, at Roger Williams Medical Center last September, the surgeon misread the CT scan and planned surgery for the wrong side; from then on, every safety check was merely making sure that those plans, erroneous from the start, were carried out.

In another wrong-side brain surgery, in 2001 at Rhode Island Hospital, x-rays were placed backward on the viewing screen so that right looked like left. The staff failed to double-check that they were working on the correct side.

Since then, Rhode Island Hospital has switched to a digital x-ray system in which images appear on a computer monitor, making such reversals highly unlikely. But even so, wrong-side brain surgery occurred twice again at the hospital, one in a bedside procedure in January and the other in the operating room on July 30.

The cause of the July 30 error is still under investigation. Cooper, the new Lifespan quality officer, says she’s focusing on the transition from the emergency room to the operating room. She said two checks were done to make sure the correct side was cut, but “one relied on a person who perhaps did not have the correct information on hand.”

Health Department officials have said that at least part of the problem was the failure of the staff to follow proper procedures; in particular, pausing before surgery to make sure they’re about to do the right thing.

Hospitals are learning that establishing good error-prevention policies doesn’t necessarily solve the problem: sometimes the staff doesn’t follow protocols and sometimes the procedures themselves are subject to errors. A person running down a checklist, for example, could inadvertently skip an item.

“Having a case of wrong-site surgery is not an indication that the hospital is bad,” says Frances A. Griffin, a director at the Massachusetts-based Institute for Healthcare Improvement. “Every hospital in the country has errors occurring every single day. … The errors are always going to be there because we will always have processes designed by humans, being carried out by human beings.”

Errors can occur at almost any point, including even before the patient gets to the hospital. “There are many steps in the process and if one of those steps goes wrong, you can end up in a domino effect,” Griffin says. “Somebody switches right for left, and everyone references incorrect information.”

In one instance of wrong-site surgery, in another state, the surgeon relied on an MRI sent by an imaging center. It was the MRI of a different patient than the one he was operating on — who happened to have the same name.

If the wrong wrist band gets put on a patient in the emergency room, misinformation can follow the patient to the operating room.

Sometimes doctors are thrown off by the patient’s position. In a Pennsylvania case, a patient needed surgery on the right heel and left hand. The surgical sites were properly marked while the patient was lying face up. But the patient was flipped over for surgery, the markings were obscured, and the procedures were performed in reverse.

In 2004, the Joint Commission required hospitals to adopt its Universal Protocol, listing three measures to prevent surgical errors: before you start to operate, verify that you are doing the right procedure on the correct part of the right patient; mark the site of surgery ahead of time; and take a “time-out” right before surgery to make sure everything is in order.

Yet the Joint Commission continues to get reports of 70 to 85 wrong-site surgeries every year, and states that require reporting haven’t seen any drop-off. As a result, in February the commission held a summit with 60 organizations to reassess the Universal Protocol. The consensus, says Dr. Peter D. Angood, vice president and chief patient safety officer, was that the protocol is reasonable, although some changes might be needed.

The problem appears to lie with hospitals’ ability to carry it out. The health-care industry is not good at bringing about major changes in how it functions, Angood said. There’s a history, hard to undo, of each person focusing on his or her own “silo” instead of teamwork — nurses do nursing, surgeons do surgery, etc.

Oftentimes, an irritable or tyrannical surgeon creates an atmosphere that discourages people from speaking up when they suspect a problem, says H. John Keimig, the former president of St. Joseph Health Services who is now chief executive officer of Quality Partners of Rhode Island, an agency that works to improve care for Medicare recipients. The hospital leadership has to encourage a spirit of teamwork in which even the lowliest technician is not intimidated. “It really does come down to the culture of the organization,” Keimig says, “and culture in health care evolves slowly.”

Every hospital establishes its own procedures for implementing the three measures in the Joint Commission protocol, and some involve as many as 20 steps. The more steps involved, the more likely the staff will skip some when under time pressure.

“Many protocols involve considerable complexity without added benefit,” write the authors of the Archives of Surgery article on wrong-site surgery. “No published evidence offers guidance on the effectiveness of site-verification interventions.” Indeed, in 5 of the 13 cases studied for that article, which appeared in April 2006, the Joint Commission’s Universal Protocol would not have prevented the error.

“It’s easy for an academician in some ivory tower to write some best-practice protocol,” Keimig says. “When you get on the front lines of battle, so to speak, and you’ve got adrenaline running and nerves frayed and people tired, those perfect systems just break down because of human frailties. It’s the nature of health care — it’s human-being-intensive.”

ffreyer@projo.com