Hospital infections hard to gauge

Despite 5 years of data, the state still isn't sure about its progress in reducing them


Share with others:


Print Email Read Later

Next month, the Pennsylvania Department of Health will release its first full-year's worth of hospital-by-hospital data showing how many bloodstream, urinary tract and other infections patients contracted during hospital stays in 2009.

But even though the state has been collecting similar data since 2005, and anecdotal evidence indicates that at least some hospitals have made significant progress -- particularly in the Pittsburgh area -- it won't be possible to say how the state has done over the past half-decade in its nationally celebrated effort to reduce hospital-acquired infections.

That's because the way infection reporting has been collected and analyzed has changed so much year to year -- and agency to agency -- that drawing long-term conclusions using the state's data is impossible, state officials and experts say.

After all the work in the past decade that went into making Pennsylvania the first state in the country to publicly report hospital-acquired infection data, it's galling to those involved that the state is, in essence, starting all over again.

"There's been reporting out there that's bouncing all over the place and is just inconsistent -- and maybe some people like that," said Karen Wolk Feinstein, co-founder of the Pittsburgh Regional Health Initiative, which helped push for the state's first public reporting on infections.

While some hospitals with higher infection numbers might have a vested interest in obscuring the state's data, Ms. Feinstein said, "I know a lot of hospitals that are proud of their infection rate and the work they've done to get there. But it depends on the hospital."

This comes at a time when the national effort to reduce infections seems to have stagnated, even though 25 other states now have some form of hospital-acquired infection reporting -- some public, some not.

Although study after study has demonstrated how infection rates can be significantly reduced, a report issued last week by the U.S. Health and Human Services Department to Congress declared that "very little progress" had been made in eliminating hospital-acquired infections in recent years.

Nationally, the report found rates for three of five major types of serious hospital-related infections had increased in 2007 -- the last full year national data was available for analysis -- including an 8 percent increase in one of the most fatal infections, bloodstream infections following surgery.

Moreover, despite efforts to create a national, publicly available, hospital-by-hospital infection reporting system in the recently approved health care reform law, just a stripped-down version of it was enacted, using only a confidential reporting method. The law will impose financial penalties -- a 1 percent reduction in Medicare reimbursements -- on hospitals with the worst performance on a list of hospital-acquired conditions that include several types of infections.

Advocates say it's not nearly good enough.

"We need a national system that is standardized," said Lisa McGiffert, campaign director for the Safe Patient Project of Consumers Union, the advocacy arm of Consumer Reports. "We really need a system that shows every hospital's performance individually and not grouped together as a hospital system, like some reports allow. Individual hospitals in the same system can be very different."

Ms. McGiffert and others said a system like Pennsylvania's hospital-by-hospital public reporting is still the model, despite the problems that now hobble making long-term conclusions.

From 2005 to 2007, the Pennsylvania Health Care Cost Containment Council issued infection reports before the Health Department took over in 2008 with a vastly different system of reporting.

"It's just not possible to compare all the information that PHC4 was collecting to what the Department of Health is now getting," said Mike Doering, executive director of the Pennsylvania Patient Safety Authority, a state agency that uses infection reporting to educate hospitals on best practices and problem areas.

"It's now a different set of infections that are reported, a different set of patients, even a different set of hospitals. Plus the data is collected in a different way," he said.

PHC4 collected data directly from all of the state's 165 acute-care hospitals, first in four infection categories -- urinary tract, pneumonia, bloodstream and surgical site -- and then two more, gastrointestinal and an "other" category.

The agency also reported on other categories that hospitals took particular issue with: how many people died after contracting infections; the average length of stay for someone with an infection; and how much patients were charged.

"That's one of the things we liked about PHC4; they collected data that people wanted to see," said Ms. McGiffert.

The state was lauded nationally for its public reporting, even as the state's hospitals and their lobbyists roundly criticized PHC4 reports as an unfair, unscientific reflection of their work. They pushed for a new system handled by the Health Department, which was approved in 2007.

"This is a more robust and actionable reporting system, rather than trying to create a system based on an overall hospital visitation rate" like PHC4 was using, said Paula Bussard, senior vice president of policy and regulatory services for the Hospital & Healthsystem Association of Pennsylvania.

"It's not starting over. It's moving forward, and you always have to start somewhere," she said.

Pennsylvania hospitals no longer report directly to the state. Instead, hospitals report to the National Health and Safety Network, a surveillance system managed by the Centers for Disease Control and Prevention. The state pulls its data from that system.

The Health Department reports on nine infection categories, including four that PHC4 did not: eye, ear, nose and throat; lower respiratory tract; reproductive; and skin and soft tissue.

More institutions are included in the Health Department's reports -- 246 in all, 81 more than PHC4 studied -- adding mental health hospitals, inpatient rehabilitation facilities, specialty general acute-care hospitals and long-term acute-care hospitals. It also is reporting on more categories of patients, adding those with burn injuries, organ transplants, mental disorders, drug and alcohol disorders, and children younger than 2.

The change in reporting caused so many problems that the first half-year of data collected by the Health Department in 2008 was so replete with mistakes that the department decided not to use it and released information only from the second half of the year.

Although PHC4 has not issued a report on hospital-acquired infections since it released the 2007 data, it still has the authority to do so and may release future reports on infection-related deaths, length of stays and costs that the Health Department is not reporting. PHC4's board meets in July and will consider then whether to issue any future infection reports.

The Health Department report might not be much help in saying how the state has done in the past, but it is vitally important to hospitals' futures. That's because the report on 2009 data will become the base year that the state will use to set infection-reduction goals.

The law that gave the department reporting responsibilities set the first-year goal as a 10 percent reduction in infections for every hospital. The department will set the reduction goals for future years. If those goals are not met, hospitals could face financial penalties.

The Pittsburgh region long has been considered a leader in the infection battle after 33 hospitals here were involved in a successful joint project from 2001 to 2005 that significantly reduced central line bloodstream infections. Most local hospitals have already brought their rates down significantly, and they continue groundbreaking work on infections:

• Allegheny General Hospital has dropped its ventilator-associated infection rate in its trauma intensive care unit by 30 percent over the past two years by closely monitoring how many hours a day patients on ventilators were sitting up, rather than lying down;

• UPMC Montefiore recently began using a SwabCap, a device with a small, alcohol-doused sponge inside an orange cap that is placed on the unused hub of a central line that has been inserted inside a patient to keep the hub clean and prevent a bloodstream infection;

• Excela Health, which runs Latrobe, Frick and Westmoreland hospitals, has instituted a "storytelling" session for staff to try to figure out the source of an infection and reinforce daily hygiene practices.

"The thinking back in 2001 was, 'Well, if you have a more high-risk population of patients, you're allowed to have a higher rate,' " said Dr. Carlene Muto, UPMC's medical director for infection control. "We said the hell with that, our goal is zero."

Dr. Muto showed internal data that demonstrated UPMC's ongoing effort to drive the most lethal of infections -- central line bloodstream infections -- down to nearly zero, which is why it is trying novel approaches such as the SwabCap.

Dr. Stephen Ostroff, director of the bureau of epidemiology for the Health Department, said the work that began in the past decade appears to have had ongoing benefits.

"That is clearly seen in our report. Clearly, facilities in the western part of the state look better than the rest of the state," he said. "One of their challenges is that since they're so far out in front of others, they're going to have a hard time bringing it down even more, because ultimately we do want to bring everyone down to zero."


Sean D. Hamill: shamill@post-gazette.com or 412-263-2579.


Advertisement
Advertisement
Advertisement

You have 2 remaining free articles this month

Try unlimited digital access

If you are an existing subscriber,
link your account for free access. Start here

You’ve reached the limit of free articles this month.

To continue unlimited reading

If you are an existing subscriber,
link your account for free access. Start here