Utah medical errors probed

A dramatic national report a few years ago predicted that 300 patients would die in Utah hospitals due to medical mistakes. From last October to this October, Utah hospitals reported a total of 34 adverse "sentinel" events, including deaths.

The true number lies somewhere in between what local hospitals are reporting and what the Institute of Medicine report predicted, according to Dr. Scott Williams, director of the Utah Department of Health.

Williams wants to see the number of acknowledged medical errors go way up in Utah, not because he likes errors but because only then will consumers know that errors are really being reported. And "until we find them, we can't fix them," Williams said. "We need to see how many could have been prevented and how many couldn't have been prevented."

He's convinced reporting by Utah hospitals is incomplete. "I think we got the events that came to the attention of the hospital through some risk-management program."

Utah is taking a more collaborative approach to medical errors than most states, inviting the hospitals to partner in the effort to report errors and fix root causes, rather than just demanding it through regulations, he said.

Two years ago, the health department created a rule requiring that serious medical errors, called sentinel events, be reported. Those events included death of a patient, loss of function, wrong-site (or wrong-patient) surgery, abductions or assaults on patients and patient suicide.

The health department also gave all hospitals until fall 2004 to have in place programs to reduce adverse drug events. Between July and December 2002, for instance, the most recent quarterly patient safety update shows that 3.5 percent of patients leaving the hospital had had at least one adverse drug event.

As for reported sentinel events, last year 15 patients died, compared to 19 the year before. (Years in the report run from Oct. 15 to Oct. 14.) Nine patients lost some function in the first year, six in the year that just ended. Both years, five wrong-site surgeries were performed. The number of patient suicides went from one to five. And three uncategorized events were reported the second year, as well. In 28 of the events, a root cause analysis was conducted. And in all but one case last year, the event was in an urban hospital.

The department is now working on a pilot project that relies on "case finding" rather than "case reporting" to help detect the sentinel events, using six hospitals that volunteered to participate.

The health department receives information on cases that go to the medical examiner. They also get death certificates and hospital discharge data, which indicates whether there were complications. Now they're beginning to compare all that data to find cases that weren't reported. If they had the resources, Williams said, they'd expand the pilot to include more hospitals.

Comments