All Samples > One Article

Coming soon: state reports on infection rates

Materials Management in Health Care

The New Mexico legislature, now closed for the year, failed to pass a bill that would require hospitals to report infections contracted in the hospital, but Eddie Benge, MD, fully expects it will happen in the future.

“I have no doubt that we'll have some kind of bill that will get through our legislature eventually,” said Dr. Benge, vice president of medical staff affairs at Presbyterian Healthcare Services , an eight-hospital system based in Albuquerque .

Since 2002, five states -- Illinois , Pennsylvania , Missouri , Florida and recently Virginia -- have adopted laws that require hospitals to report healthcare-associated infections (HAIs) to states, which would then be disclosed to the public. And 27 more states are now considering mandates in sessions that are still ongoing, in many cases.

Dr. Benge and many other infection control professionals across the country are not happy with the trend. They think hospitals might have to devote considerable resources to collecting information that, in the end, may be of little use.

Hospital officials stress that they do not oppose public reporting of clinical outcomes, which are widespread for a variety of other health care indices, but they are skeptical that HAI rates can be accurate.

“This sort of reporting has never been done before,” says Janet Hefel, RN, an epidemiologist at Mercy Medical Center in Dubuque , Iowa , another state with pending legislation. “To be accurate, the reports would have to take case mix, infection definitions and data collection systems into account.” She believes current collection methods are not sophisticated enough and, due to the nature of the data, may never be.

But these arguments have done little to dissuade state legislators, who have been reacting to campaigns by groups such as Consumers Union, a nonprofit consumer information group based in Yonkers , N.Y. , that publishes Consumer Reports .

“I've heard hospitals say that [HAIs] can't be measured, but I don't believe it,' says Lisa McGiffert, senior policy analyst on health issues at the Austin, Texas, office of Consumers Union . “What they're really saying is that we can't measure them with the system we're now using.”

Pennsylvania as poster child

To see if state reporting can work, all eyes are turning to Pennsylvania , the only state to actually collect data so far. Collection by the Pennsylvania Health Care Cost Containment Council began in early 2004 for data starting in January 2004, but it is still very much a work in progress. PHC4 gave facilities only a few months' notice and initially ordered hospital-wide reporting of all infections, something that is rarely, if ever, is done in hospitals' current voluntary collection programs.

“ Pennsylvania is a poster child for what not to do,” says Carlene Muto, MD, director of infectious diseases at Pittsburgh-based UPMC Health System, which runs eight hospitals.

Dr. Muto sits on an advisory committee that the state brought together to answer outcries about the start-up. On the committee's advice, the state agreed to limit reporting in the first two years to the most serious infections -- central line-associated bloodstream infections, ventilator-associated pneumonia, catheter-associated urinary tract infections, and surgical site infections for orthopedic, circulatory system and neurological surgery.

Pennsylvania hospitals must report numbers of infections by category on open fields on UB92 forms, which they have already been submitting to the state quarterly for other health care measurements. They have the option of submitting just the diagnosis of infection as an ICD9 code on the form. Dr. Muto says some hospitals, though not UPMC, have taken this option, but “it's not a good way of doing it.” Since HAIs are just a part of overall infection rates reported in ICD9 codes, these hospitals would have exaggerated numbers when public reporting starts. But, Dr. Muto said, “They have been so overwhelmed by the thought of having to report that they see ICD9 codes as an easy out.” Also, the state has not even decided yet when the data would be made public.

Dr. Muto says UPMC has not increased its infection control staff because it was already doing substantial collection and prevention work in voluntary efforts such as the Pittsburgh Regional Healthcare Initiative . Late last year, PRHI announced that member hospitals reduced central line infections by 55 percent over two years. But in January 2006, when hospitals will have to report all HAIs, Dr. Muto says her infection control staff, which now is at six practitioners, would have to add 4.2 FTEs.

Rick Shannon, MD, chair of the department of medicine at Allegheny General Hospital , says it's hard to pinpoint the cost of beefing up infection control, but it's worth it. He says Allegheny General, a 728-bed academic health center that is part of seven-hospital West Penn Allegheny Health System, loses $15,000 on each central line infection. “When you calculate that out for other infections,” Dr. Shannon says, “the amount is staggering.”

While Pennsylvania regulators are asking only for numbers of infections, the Hospital & Healthsystem Association of Pennsylvania argues that infection statistics should be expressed as rates per overall numbers of procedures – for example, total patient days on central IV lines. States like Missouri , which begins counting HAIs in July, plan to use rates. The hospital association says they allow for fair comparisons between facilities, even though they require more work because hospitals have to compile total numbers of procedures to create a denominator for the rate.

Dr. Shannon, however, argues that rate-reporting makes hospitals complacent. “If hospitals aim for a rate of six line days,” he asks, “when they get there, what's the incentive to go any further? Our real goal should be reducing infections to zero. ”

He says Allegheny General, in fact, reduced central line infections to zero this year by using a variety of strict precautions, such as prepping with the antiseptic chlorhexidine and creating a sterile barrier by using drapes, gowns, gloves and head cover .

Tightening infection control efforts

Hospitals across the country are already incorporating precautions to reduce infections. In addition, the Joint Commission on Accreditation of Healthcare Organizations is increasing efforts to make sure hospitals deal with infections. “You should try to fix it and come back to it and try to fix it again,” says Nancy Kupka, RN, project director of standards and survey methods at the JCAHO.

Kupka says the Joint Commission also has adopted strict hand-washing standards promulgated by the Centers for Disease Control and Prevention. In another initiative, the Surgical Infection Prevention Project, cosponsored by the CDC and the Centers for Medicare and Medicaid Services, hospitals working with Medicare quality improvement organizations have dramatically reduced their surgical infection rates.

Hospitals closely monitor certain infection rates. Kupka says the Joint Commission has long required hospitals to collect HAI data and organize them into a reportable form. It does not require hospitals to report the data to anyone and does not specify exactly what information has to be collected, but hospitals have to explain the reasons for their choices, she says.

Even in Pennsylvania , Dr. Muto says hospitals are not used to collecting all infection data. “The amount of information you get out of it [is not worth] the amount of time you put into it,” she says. Instead, hospitals tend to focus on ICUs and target problem areas like central line infections, which account for 40 percent of primary bacteremias in ICU patients and have substantial morbidity and mortality.

Data-collection can be a lot of work. In the absence of computerized medical records at many hospitals, infection control personnel have to spend a lot of time rooting out infections and then determining whether they are hospital-related. Personnel usually make rounds to find cases, which then must be verified through lab reports and the patient record. They must then make sure the problems are resolved.

Dr. Shannon says Allegheny has made collection more efficient by insisting that all units report suspected HAIs immediately to infection control. He says it would be easier to identify infections through ICD9 codes, but he does not use them because they are compiled after discharge, when the trail is cold. Even so, Dr. Shannon admits some cases are hard to decide and must be sent to him for review.

In search of national standards

Determining what cases are infections and then which of those are hospital-related is no easy matter. Hospitals rely on the CDC definitions that were created for hospitals that voluntarily report to a CDC database, the National Healthcare Safety Network (NHSN), which is still better known by its former name, the National Nosocomial Infections Surveillance (NNIS) System. The definitions classify 13 major site codes, including urinary tract, surgical site, pneumonia and bloodstream infections, and the CDC also provides rate-conversion formulas.

Gwen Davis, director of infection control at Memorial Hospital a 365-bed facility in Chattanooga , Tenn. , says the CDC definitions allow some “wiggle-room” in deciding what are reportable. She says that's acceptable when hospitals use data for their own in-house improvement efforts, but not when states such as Tennessee propose using the data to compare hospitals publicly. She says hospitals that more vigorously root out infections might be stuck with higher rates.

Dr. Shannon at Allegheny General says CDC definitions fail to define many cases. For instance, he says they do not identify half of all central line infections that he finds, using his own system for identifying infections.

Everyone in the infection control field seems to agree that a more precise measurement is necessary and that there should be a national standard for collection rather than leaving it up to each state. But Congress does not seem ready to take the ball. The House Energy and Commerce Committee has reportedly discussed the issue, but sources on Capitol Hill say no bills are imminent.

To fill the vacuum, the National Quality Forum, a healthcare quality improvement advocacy and standards-setting organization in Washington , D.C. , is poised to help create national standards, which each state could then adopt. NQF President and CEO Kenneth W. Kizer, MD, says the standards would be created through its consensus development process, in which it convenes stakeholders in private and government positions.

The forum first needs to raise about $500,000 but “that's not much in the health care field,” Dr. Kizer says. “I n a best-case scenario, we would have the financing by mid-summer and the consensus process would take eight to nine months to a year.”

There are signs that it may not be easy for NQF to forge a consensus among stakeholders. For instance, no definitions to replace CDC's have been promulgated by Washington, D.C.-based Association for Professionals in Infection Control and Epidemiology, which represents 10,000 professionals . And the only solid conclusion to come out of a consensus conference on the issue called by APIC in February was that “pursuit of national standards for reporting HAI embodies the overwhelming sentiment of participants.”

The CDC's Healthcare Infection Control and Prevention Advisory Committee grappled with the issue in a lengthy document released on February 28, titled “Guidance on Public Reporting of Healthcare-Associated Infections.” But despite its promising title, the document did not provide much guidance for state reporting.

Front-line infection control professionals were disappointed that the report did not address the state mandates. Pennsylvania regulators “would laugh at this,” Dr. Muto said. “There was not enough there.”

“I did not think they were helpful,” Dr. Benge said of HICPAC's recommendations. With New Mexico legislators poised to reintroduce a bill next year, “my expectation was a national standard. That's what we need.”

HICPAC avoided speaking for or against public reporting, arguing that there was not enough evidence on their effectiveness. It did, however. recommend collecting outcome measures in two areas: central line-associated bloodstream infections and surgical site infections, but it did not name specific operations to monitor.

HICPAC said it chose the two areas because they have high morbidity, mortality and costs and there are well-established prevention strategies for them. Also, there is strong evidence that hospitals need to improve prevention of surgical site infections. A study published in the Feb. 21 Archives of Surgery found that giving antibiotics within an hour before incision, a key prevention measure, was not done for 44.3 percent of Medicare beneficiaries undergoing major surgery.

One the other hand, HICPAC specifically advised against reporting catheter-associated urinary tract infections, saying their lower rates of morbidity and mortality do not justify the “the burden of data collection and reporting.” And it advised against reporting cases of ventilator-associated pneumonia (VAP), noting that they are difficult to detect accurately.

Eddie Hedrick, an emerging infections coordinator in the Missouri Department of Health and Senior Services, predicts it will be difficult to create accurate ratings.

Hedrick, who is helping to over see Missouri 's mandate, said, “We need to give the public something that informs them on how to look for healthcare. But coming up with a number that tells you anything about a hospital is a very, very difficult thing.”

Return to Samples Listing