APIC Conference: Infection rate disclosure laws a ‘watershed’ moment for ICPs
APIC Conference
Infection rate disclosure laws a watershed’ moment for ICPs
Haley at APIC: Don’t let your hospital off the hook’
The nationwide push for infection rate disclosure laws is an historic opportunity to get full funding for clinically proven surveillance methods in the nation’s hospitals, one of the modern pioneers in the field said recently in Baltimore at the annual meeting of the Association for Professionals in Infection Control and Epidemiology (APIC).
"Mandatory reporting laws could provide the resources. Don’t let your hospital off the hook," said Robert Haley, MD, a professor in the department of internal medicine at the University of Texas Southwestern Medical Center in Dallas. "[A] watershed opportunity is at hand."
Haley directed the Centers for Disease Control and Prevention’s (CDC) landmark Study on the Efficacy of Nosocomial Infection Control (SENIC) project, which validated the efficacy of infection control programs and surveillance methods when the field was first taking shape in the 1970s. "SENIC remains the only controlled study of the effect of any quality improvement strategy on health care outcomes," he told APIC attendees. "What you do is proven."
Delivering an address that drew a standing ovation, Haley said ICPs must insist that the growing legal requirements are met with accurate data. If the effort really is about patient safety, then states and hospitals will have to provide the resources to do it right, he emphasized. For example, ICPs will need help gathering post-discharge surveillance data for surgical site infection rates.
"Do the right thing, and ask for the right resources," he said. "The time for the clinical surveillance model may be at hand. We are on the verge of mandatory public reporting. If we are mandated to measure infection rates with proven clinical surveillance methods in the context of a full infection control program it could greatly benefit patients. If not, it will be another make-work, major distraction."
Meanwhile, the U.S. map on infection rate disclosure activity keeps changing like it was election night. Infection rate disclosure laws are under some level of political discussion in more than 30 states and have been enacted in six others (Florida, Illinois, Missouri, Nevada, Pennsylvania, and Virginia). To head off a nightmare scenario of different rules of the road in every state, APIC is working with the National Quality Forum to create national model legislation.
The final result is expected to mirror the guidance issued by the CDC’s Healthcare Infection Control Practices Advisory Committee (HICPAC).
But with so much happening so fast at the state level, the question arises whether the national model standard is going to be too little too late.
"We are working with the state legislatures [to] let them know that there will be a national standard coming out," said Kathleen Arias, MS, MT, SM, CIC, of Arias Infection Control Consulting in Crownsville, MD. "So that when they write their bills the bills don’t specify exactly how the data are collected or what is collected. But they can still pass legislation requiring reporting with the clause in it recognizing that a national standard will be coming out. For instance, in Maryland, we asked our legislatures to hold off on passing our bill until the HICPAC guidance document came out. Maryland was the first state to reference the document. Our bill ended up being three sentences long basically, and it said that you will follow the HICPAC guidance document."
The matter is particularly critical for patients living on state borders who want to compare rates at their nearby hospitals.
"We cannot have 50 different reporting requirements," Arias said. "You [may] have a patient who lives on the border of Maryland and Pennsylvania, and they want to compare data for the two states. The states should be using the same kind of system for comparison, or patients won’t be able to use that data."
Despite the obstacles, Arias agrees with Haley that this could be one of the greatest opportunities infection control has had to expand as a field. "In the 20-some years I have been in infection control, I cannot think of a better opportunity for us to raise the visibility of what we do," she told APIC attendees.
Law of unintended consequences
But requiring hospitals to disclose infection rates could invoke the law of unintended consequences, warned Tammy Lundstrom, MD, JD, epidemiologist at Detroit Medical Center. In a nutshell, clinicians may be reluctant to treat high-risk patients because their subsequent infection or death may end up in a public report.
"There are a couple of recent articles suggesting that public reporting may actually backfire in some instances," she said. For example, a survey of practicing cardiologists found that 83% of respondents felt public reporting may deter surgeons from offering angioplasty to patients who might otherwise benefit.1 "That is a sobering thought," Lundstrom said.
Another study found that public reporting may have a "big chill" effect, with clinicians in states with reporting systems more reluctant to perform high-risk procedures. The study compared angioplasty cases in Michigan with those in New York, which has public reporting for certain health care outcomes. The authors cited "a propensity in New York toward not intervening on higher-risk patients because of fear of public reporting of high mortality rates."2
"The mortality rate in Michigan — not surprisingly since they seemed to be performing angioplasty on sicker patients — was higher," said Lundstrum. "This is suggestive although not conclusive that there may be some strategic behavior around public reporting. . . . It could potentially induce physicians to deny care or procedures to sicker patients. We need a lot of research in the states that do have mandatory public reporting to make sure we are not having negative consequences to patients," she added.
Other unintended consequences of public reporting laws could be administrative decisions to shut down high-risk programs such as neonatal intensive care units, Lundstrum explained. ICPs also fear that the effort will simply result in reduced funding for some infection control programs. "We all know that the harder you look, the more you find. We need to be sure that we don’t disincent hospitals to provide adequate resources to infection control programs."
Similarly, there may be more administrative oversight of infection control and pressure to remove borderline infections, she explained. "We now include these gray areas in our data that we use for improvement," Lundstrom said. "If it was publicly reported, there might be more pressure to take those gray cases out of the data. There also potentially could be surgeon decisions not to operate on high-risk cases like patients with diabetes, renal failure, or immunosuppression."
Reporting a by-product’ of prevention
The bottom line is to design mandatory reporting systems in which the data collection is aimed primarily at reducing the infection risks, Haley emphasized. "Reporting to the state [is just] a by-product of prevention. If we don’t stress that, I guarantee you within two years, it is going to be a make-work medical records job and you’re going to be left out in the cold. You better grab it right now and say we are going to make state reporting just an extension of our clinical surveillance," he noted. "That helps patients, and we are going to report that to the state as a by-product. We are not going to create something new."
Indeed, while some have lamented the national push to disclose hospital infection rates, Haley argued that ICPs must use it as an opportunity to make sure their surveillance data are accurate and their programs fully funded.
"If ICPs don’t measure infection rates, others will," he told APIC attendees. "That’s been the history. So you need to get out there and do it. To measure is to control. It’s a play on the old [surgical saying] to cut is to cure.’ In our business, to measure is to control."
The approach validated in SENIC calls for the ICP to draw data from multiple sources and conduct post-discharge surveillance of surgical site infections. There is no easy one-stop shop to find the data, which must be gleaned from a clinical surveillance plan that looks at a variety of key factors.
"Clearly, nosocomial infection rates cannot be measured by ICD-9 codes and medical records," Haley said. "We need to say that over and over until people get it. Nosocomial infection rates cannot be measured by data mining of microbiology laboratory data. It’s just as simple as that. Nobody has ever believed it could be done. It’s like turning lead into gold. It’s alchemy."
Data mining and collection of clinical lab data certainly can be part of an overall clinical surveillance program, he noted. "[But] that is one of the six or eight things you’ve got to be doing to do a complete clinical surveillance model," he said. "Don’t focus just on outbreaks. What we are about is reducing the baseline."
In that regard, the infection rate is not some number in a black box. It must be reported back to the caregivers to have a preventive effect. It is not enough to simply "change the culture" by stressing infection control or introducing new tactics such as alcohol handrubs.
"SENIC validated the clinical surveillance model: measuring infection rates and using those to feedback and also to direct the control program," Haley said. "You have to be aggressively controlling, and you have to be measuring. I would submit to you, from time immemorial, the only time we are really successful is when we change the culture and measure it to hold people accountable. Otherwise, changing the culture becomes lip service."
Risk index models are improving steadily, and technological advances also will ease the process of recording and comparing accurate infection rates. "Make this meaningful," Haley urged. "What number does the consumer want to see? Do they want to see a big table with all kinds of rates on it? No, they want a [surgical infection rate] or something comparable to that. They want one that is already adjusted for patient risk — a level playing field."
To measure rates, for example, monitor device day denominators (e.g., number of central line days) in intensive care units. "[Use] the number of central-line-associated bloodstream infections as the numerator and the number of central-line days as the denominator," Haley said. "We all know this, but how many hospitals are doing it? If you are not doing this, feel guilty. But you will be doing it soon, because your state is going to ask you kindly to."
Codify the SENIC model
The SENIC model essentially documents the efficacy of infection control against the staphylococcal epidemics that spurred the creation of the field in the 1960s and 1970s. "Let’s codify this into state laws because this will reduce infection rates," he said. "If we tell the state to make us do it, then we will do it and all of our negligent colleagues who aren’t doing it [also] will do it. The full potential of the clinical surveillance model unfortunately has not been realized. We have not capitalized on this fully."
There are some favorable signs in the way the mandatory disclosure movement is taking shape, Haley noted. He praised the selection of high-volume, high-risk surgical site infections, saying it is time to capture them correctly rather then argue that post-discharge tracking will be too difficult.
"When we get mandatory reporting, you’ve got to tell [administration] how hard it is and what resources you need to do post-discharge surveillance," he said. "You may want to hire an MPH to help figure out how to measure these things."
Rate disclosure laws should be used to remove the impediments to clinical surveillance and enlist other departments in the gathering of the data, Haley explained. "[For example], automated downloading of surgical site infection denominator data from the OR computer to the infection control department," he said as the ICPs in the audience began to applaud. "ICU nurses collecting device day denominator [data] for you. That ought to be part of the SOP [standard operating procedure] of the ICU."
The key will be getting accurate definitions for all of the infections measured. "For all of these, there has got to be CDC clinical definitions," he stressed.
"In some cases, if they are not sufficient, CDC needs to fix them. For example, they have got to fix the ventilator-associated pneumonia [VAP] definition. It doesn’t work. Everybody who gets anything in their lungs has a VAP according to [the definition]. We have to fix that. The rest of them are pretty good, but they require data collection from multiple sources," Haley noted.
Fortunately, demands for hospitalwide surveillance seem to have fallen by the wayside, with the "targeted" or "surveillance by objective" model that Haley has long advocated deemed sufficient. "I think Consumers Union got the message early," he said with relief. "We don’t want [hospitalwide surveillance] to happen. This would be a holy disaster. We all know that."
References
- Narins CR, Dozier Am, Ling FS, et al. The influence of public reporting of outcome data on medical decision making by physicians. Arch Intern Med 2005; 165:83-87.
- Eagle KA, Share D, Smith D, et al. Public reporting and case selection for percutaneous coronary interventions. J Am Coll Cardiol 2005; 45:1,759-1,765.