St. Charles Health System’s four Central Oregon hospitals are among more than 20 in Oregon participating in the second round of a federal program designed to improve patient safety, but which has been criticized in the past for its poor design and lack of standardized reporting measures.

The initiative, called the Partnership for Patients, carries a goal of reducing preventable hospital-acquired conditions by 40 percent and 30-day readmissions by 20 percent. That’s the same goal it started with in the first round, launched in December 2011, but which was not met when that round ended in December 2014.

Centers for Medicare & Medicaid Services doled out $430 million for the first round and is committing a scaled back $110 million for the second, which began in September and will last one year.

Despite the apparent lack of nationwide rules from CMS around how hospitals should track and report their progress toward 10 safety measures, the campaign’s leader in Oregon said our state will adhere to strict standards that will also be used by hospitals in other states under the same reporting umbrella, called a Hospital Engagement Network.

There are 17 such HENs nationwide. Oregon’s hospitals, most of which are represented by the Oregon Association of Hospitals and Health Systems, are in a HEN led by the American Hospital Association.

Diane Waldo, OAHHS’ associate vice president of quality and clinical services, said her organization directed Oregon hospitals to use a set of reporting guidelines from AHA in the program’s first round, although the AHA did not explicitly require hospitals to use those standards.

“You’ve got to be able to be measuring the same thing to be able to know if you’re making a difference or if you have improvement,” she said. “If everyone is evaluating it differently, it makes no sense at all.”

This year, the AHA is asking hospitals to report on the measures using the same guidelines, she said.

At the national level, Waldo said CMS this year is directing all HENs to use reporting guidelines that were endorsed by the National Science Foundation, but she didn’t know whether the government was requiring that they use the same guidelines.

“I know they were not particularly happy with the data mess, really, from last time,” she said. “I’m speaking just really honestly with you that it was hard to evaluate improvement when there were measures all over the board.”

Raymond Thorn, a spokesman for CMS, said that although CMS has told hospitals which 10 areas of harm to report on, they can use “both national and local measurement strategies” to track and report their progress.

“Each contractor is required to have a local measurement system and to report the results of these measurements to CMS on a monthly basis,” he wrote.

In an August 2014 editorial in the New England Journal of Medicine, Dr. Peter Pronovost, director of the Armstrong Institute for Patient Safety and Quality at Johns Hopkins University, and Dr. Ashish Jha, a health policy professor in the Harvard School of Public Health, criticized the initiative’s lack of standardized performance measures.

Using the category of catheter-associated urinary tract infections as an example, the two wrote that HENs could choose to rely on either federal definitions or administrative data.

“But administrative data on these infections are widely considered insensitive and are subject to variation and changes in coding practices,” the editorial said. “Furthermore, since variation in measures rendered it impossible to compare all HEN hospitals, the ability to evaluate “improvement” was limited.”

‘Tight with their data’

Pronovost and Jha also criticized the Partnership for Patients’ lack of transparency, including its failure to make its results subject to peer review, which is when a study is examined by an outside expert.

Similarly, Kathy Luther, vice president of the Institute for Healthcare Improvement, said she hasn’t been able to find an analysis from CMS on the initiative’s first round.

“They seem to be very tight with their data,” she said.

The OAHHS declined to provide a statewide report of how all hospitals in Oregon fared on the 10 measures in the first round, nor would they provide hospital-specific data. Instead, the organization provided a list of bullet points and said whether or not each goal was met statewide: Hospitals met seven of the 10.

Leaders with St. Charles Health System also declined to provide either an aggregate report on all four hospitals’ performances or separate reports for each hospital.

Although Pam Steinke, the health system’s chief nurse executive and vice president of quality, said she had the data in a spreadsheet, a spokeswoman said it would be too difficult to extract the specific data relevant to Partnership for Patients.

Steinke shared some of St. Charles’ results in an interview, which included both improvements and areas where it lagged.

For example, when it came to catheter-acquired urinary tract infections, infections that occur when germs enter the urinary tract through a catheter, Steinke said the health system’s baseline was an average of 5.1 infections per 1,000 days of patients using catheters across all four hospitals. It’s since fallen to 1.9 infections per 1,000 catheter days, she said.

The Bend hospital sees the majority of catheter-acquired UTIs. It saw eight such infections last year, almost double the state average of 4.3 infections across 42 hospitals, according to Oregon Public Health Division data. In 2013, St. Charles Bend saw higher rates of such infections than 76 percent to 90 percent of hospitals nationwide.

St. Charles Redmond saw one catheter-acquired UTI in 2013 and one in 2014.

The initiative also examined falls with injuries. Back in December 2011, Steinke said the health system was at a baseline of 6.6 falls with injury per 1,000 patient days. Today, St. Charles is at 10 falls with injury per 1,000 patient days, she said.

To that end, the health system is in the midst of a fall prevention campaign that includes evaluating patients’ fall risk and placing special indicators on their doors if they’re identified as such. Those patients also wear yellow gowns, Steinke said.

Partnership for Patients will also look at preventable readmissions, which occur within 30 days of discharge from the hospital. St. Charles is currently seeing a 6 percent to 7 percent rate of readmissions for all causes. Oregon’s statewide readmission rate was 11.4 percent last year. In 2013, the national readmission rate was 17.5 percent among Medicare patients, according to CMS.

Luther, of the Institute for Healthcare Improvement, said it’s difficult to say whether hospitals will reach the goals for reducing hospital-acquired conditions and readmissions.

“It’s certainly a stretch goal,” she said. “It’s a lofty goal.”

Either way, she said, hospitals and hospital organizations will learn a lot of important lessons about quality and safety. They’ll learn new ways of thinking and improving safety and will carry those into the future, Luther said.

“That’s been fairly impressive, I think,” she said. “Those things are tough to measure.”

Whether or not the goals are met, Steinke said it’s just good to focus on patient safety.

“Regardless of how big of a goal or pie in the sky it is, the real mantra is zero harm,” she said. “Can we ever get there? I don’t know, but that’s where we want to be, is zero harm.”

— Reporter: 541-383-0304,

tbannow@bendbulletin.com