Patient safety is not improving: studies

Well, this is discouraging. Two recent studies indicate that, after a decade-long, nationwide campaign to make hospitals safer for patients, essentially no progress has been made.

A patient checking into a hospital today appears to face at least a one-in-four chance of coming to some degree of harm there.

A study published this week in the New England Journal of Medicine looked at the records of 2,341 patients discharged from 10 randomly selected hospitals in North Carolina, which was chosen because of that state's "high level of engagement in efforts to improve patient safety."

The study took place between January 2002 and December 2007. What it found was, in short, that "harm to patients resulting from medical care was common in North Carolina, and the rate of harm did not appear to decrease significantly during a 6-year period ending in December 2007, despite substantial national attention and allocation of resources to improve the safety of care," the report stated.

A total of 588 patients were injured — 25.1 percent of study subjects. Harm was caused by, in declining numbers, procedures, drugs, hospital-based infections, other therapies, tests, falls and other causes, the study found. Sixty-three percent of these injuries were deemed to have been preventable. Nine preventable errors resulted in death, and 13 in permanent damage.

In addition, a report from the U.S. Dept. of Health and Human Services released earlier this month documented the experiences of 780 randomly selected Medicare patients discharged from various hospitals in October of 2008.

About one in seven of these patients experienced "adverse events" — serious harm that comes to a patient as a result of medical care.

A second group of about the same size in the HHS study suffered "temporary harm," a transient injury like bedsores (here called "pressure ulcers") for example, or hypoglycemia. Twenty-seven percent of temporary harm events were caused by drugs.

Twenty-eight percent of patients who experienced more serious "adverse events" also suffered some temporary harm during the same hospital stay.

About 44 percent of all these events — adverse events and temporary harm — in the HHS study were deemed preventable — the result of errors, substandard care, or insufficient monitoring.

In 1999, the independent, not-for-profit Institute of Medicine published a report on hospital safety, "To Err is Human," which caused a sensation and produced a massive effort to improve protocols at hospitals across the country. The goal was to decrease errors by 50 percent over a five-year period.

"To Err is Human" asserted that as many as 98,000 patients die in hospitals each year because of medical error.

Commenting on the two discouraging new studies, the authors of the NEJM report on patient safety in North Carolina write, "All the findings about extent of harm should increase our commitment to prevent it."

A cascade of errors

At the ACOG meeting last May, Dr. Robert Wachter talked about a case in a teaching hospital in which two women with similar names -- Jane Morrison and Joan Morris are the pseudonyms assigned to them -- were confused with one another, resulting in one of them receiving an invasive procedure intended for the other one. This case did not involve a birth story, but every medical specialty can take lessons away from it.

This case, which Dr. Wachter and colleagues published as "The Wrong Patient," in June, 2002, in the Annals of Internal Medicine, demonstrates how easily a series of oversights can cascade into a shocking medical error. (It could have been worse. The procedure was a cardiac electrophysiology study, not, say, a leg amputation.)

In fact, the team that analyzed the mishap identified 17 junctures at which the process could have been stopped but instead moved forward. No single mistake would have been enough to keep this juggernaut moving. Human error fed into institutional weaknesses, including "frighteningly poor communication," a lack of standardized protocols and a culture that had become sufficiently dysfunctional that more than one person thought, incredibly, Gee, this doesn't seem right, but I'm going to do it anyway.

"Human performance can beimproved but not perfected," the team concluded. Protocols must be in place to head off the inevitable errors before they converge into tragedy.