Minnesota looks to evolve its patient-safety reporting system

The state of Minnesota is having a little trouble with data.

In its most recent report of statewide adverse health events, published in March, there were a total of 384 incidents from Oct. 7, 2017, to Oct. 6, 2018. That's the most events ever reported since the system began. The most common events were perennials; falls and pressure ulcers, accounting for 58% of all events. Of the 384 events, 118 resulted in serious injury and 11 deaths.

State health policy officials are alarmed but are also scratching their heads.

"We have been looking at the numbers and I would say struggling a little about how to talk about them," said Diane Rydrych, director of the health policy division at the Minnesota Health Department.

The state requires hospitals and ambulatory surgery centers to report the 29 adverse events recognized by the National Quality Forum, also known as "never events." These events, which include wrong-site surgery, patient suicide and retention of a foreign object after a procedure, only represent a sliver of harms that can happen to a patient in the healthcare setting. Furthermore, some events like pressure ulcers and falls consistently happen many times during a reporting period without much new knowledge gained from analyzing them, according to hospitals.

"On the one hand the numbers are going up and that is serious," Rydrych said. "But we also know that the set of 29 events only represents a really small percentage of situations. It's hard for us … to have a good sense of what the small numbers mean."

In 2003, Minnesota established one of the first state public-reporting programs in which hospitals and ambulatory surgery centers are required to report instances of serious harm and deaths at their organizations. That was in response to recommendations made three years earlier in To Err is Human: Building a Safer Health System. The report called for a mandatory and standard nationwide reporting system of adverse events.

Twenty years since the report, only 28 states have adverse-event reporting systems, according to the most recent report from the National Academy of State Health Policy, and there is no national database.

What that means is that there is no standard way to know how many people die in the U.S. because of medical errors. There is also no national database to glean knowledge from, and there are no numbers that can be used to hold any one organization accountable or in high regard.

Minnesota's legislators selected NQF's never events because they thought the IOM's recommendations would become a reality. To Err is Human called on NQF to develop a set of patient-harm events that states would all follow so comparisons could be made nationally. 

"That hasn’t really come to pass," Rydrych said.

Of the 28 states that have an adverse-event reporting system, just eight follow the NQF list. The other states either implemented a modified or partial NQF list of events or their own unique lists.  

Given the questions surrounding the figures in the most recent report, the state is now exploring how its reporting program can be modified. “Fifteen years ago we were just trying to prove we can report safety at all, and it was a huge lift to develop a system … but the (patient) risks are different now," she said. "How does the system need to evolve to meet where the needs are now?" 

Right now, Minnesota requires all hospitals and ambulatory surgery centers to conduct a root cause analysis for each event that occurs and how to prevent it from recurring. "It's a lengthy process," said Dr. Timothy Morgenthaler, vice chair of quality and affordability at Mayo Clinic.    

At Mayo Clinic Hospital, the system's flagship facility in Rochester, Minn., 41 never events occurred, according to the state's most recent report. Of those, 11 were pressure ulcers, 10 were wrong surgeries and six were falls. One fall killed a patient, while five caused serious harm. While Mayo Clinic takes every never event seriously, there are some events that aren't helped by conducting a root cause analysis, Morgenthaler said. 

For example, pressure ulcers typically happen now only in two circumstances: to patients in the intensive-care unit with medical devices or those with multisystem organ failure. 

"It requires a great amount of care and even with micro-turning, we aren't able to (prevent the patient) from developing a pressure injury," Morgenthaler said. 

Falls are also difficult to prevent. Patients in hospitals are much sicker and more frail than previous years, but mobility is still encouraged to speed recovery. 

"We can't really say to the patient, 'You need to be in bed,' " said Dr. Rahul Koranne, chief medical officer of the Minnesota Hospital Association. "If we did that we'd have zero falls, but that wouldn't be the right thing to do in serving our patients." 

Given the challenges with preventing some events and the resources it takes to report them, Morgenthaler wondered if the program should look at other areas of harm. 

"At some point doing the measurement and reporting can actually be helpful and at other times it can be burden," he said. "Are we really going to learn more and prevent more or should we take those same resources and put it into something else?"

The state department is asking similar questions. On the one hand, removing some frequent events would allow hospitals to focus on other areas of harm, but it's also likely not all hospitals do everything they can to prevent each event every time. "How do we know all the best practices are being done? And how do we know those are the right best practices? It's hard to throw up your hands and say, 'These could happen no matter what; let's cross them off the list,' " said Rachel Jokela, director of the state's adverse health events program.

The state is considering tracking diagnostic errors and expanding the program to other settings where patients now get care, like outpatient clinics. The department has also set up meetings with consumers throughout the state to understand how the system can better represent their needs. Most didn't know the reporting system exists, Jokela said.

Another consideration is releasing an aggregate report of events from across the state, rather than reporting them by facility.

Most states—16—publicly report the events with aggregate data, according to the National Academy of State Health Policy report.

Under those circumstances, hospitals may be more likely to report all the events that occur, Rydrych said.

Although hospitals and ambulatory surgery centers are legally mandated to report the events, the state does not conduct audits to ensure all events are reported. Jokela said the state will review death records, but not all never events cause death.

"It's kind of like speeding on the highway. Unless there is a police officer on every corner, you are going to miss some episodes of speeding. That's just the nature of the beast," said Dr. Julius Pham, associate professor of anesthesia critical-care medicine at the Armstrong Institute for Patient Safety and Quality at Johns Hopkins who co-authored a study on adverse reporting systems.

Rydrych said a concern with moving to aggregate reporting is that the accountability for facilities will be lost. Although the law allows the state to penalize a facility for failure to report an event by suspending or revoking its operating license, the state has never done so. Instead, if the state discovers an event occurred that a facility didn't report, it conducts an investigation and works with the facility to get it reported. On rare occasions, disagreements have come up with providers that don't want to report an event publicly, Rydrych said.

"If there is a death in their facility, they (the provider) care deeply it's going to be out there in a report," she added.

Morgenthaler said he recalls hospitals initially being nervous when Minnesota first implemented the reporting system. There was fear legal action could be taken against them because of some of the events.

Since then, buy-in has improved. Through the state's hospital association, the hospitals work together on ways to prevent events, Koranne said. For example, last year a state-wide initiative addressed wrong-site surgery. Now, hospitals across the state have agreed to take a timeout before surgery to make sure it's the right patient and the right surgery.

Morgenthaler said the types of events reported now show how much the industry has evolved. In the beginning years of the program, hospitals reported sponges or clamps retained in a patient after a procedure. Now such errors are rare and when a hospital reports a retained object in a patient, it's likely a small piece of a device that fell off while the physician was pulling it out.

"If we rolled the clock back 10 years ago, no one was reporting tiny slivers (of devices) left in," he said. "We have really upped the level of scrutiny."

Rydrych said the state expects to present recommended changes for the program in spring 2020, likely after releasing its 2020 report.