Sea change begins with storm: Feds threatening action to stop infections
Sea change begins with storm: Feds threatening action to stop infections
Hospital funding could be withheld if infections persist
Warning that hospitals with poor infection control programs could have federal funds cut, a United States congressman recently lambasted the Centers for Disease Control and Prevention's longstanding effort to fight hospital infections.
"[The CDC has] been doing this for 35 years," Rep. Bart Stupak(D-MI) said at a recent Congressional hearing. "During that time, hospital stays have grown dramatically shorter yet infection rates continue to go up. It would appear that CDC efforts have not been very effective beyond the hospitals in their [sentinel hospital] network, which is less than 10% [of all hospitals] . . . What do we have to do to motivate CDC?"
Titled "Public Reporting of Hospital-Acquired Infection Rates: Empowering Consumers, Saving Lives," the hearing was held by the Committee on Energy and Commerce's Subcommittee on Oversight and Investigations on March 29. While ostensibly about the growing number of states adopting laws requiring infection rate disclosure, the hearing provided a snapshot of the increasing interest in — and dramatic changes within — infection control and prevention. A confluence of forces from within and without the field is dramatically elevating the infection control professional's mission. The old guard — bound by benchmarks and voluntary guidelines — is giving way to a new regulatory era with the expectation of zero infection rates. The pressure is not just from without, as epidemiologists and infection control professionals are showing that dramatic results are possible in programs with the audacious goal of flatlining infection rates. Congress has noticed, and all those lost patient lives are turning from numbers into human faces.
"Every year, nearly 2 million people pick up infections in hospitals and 90,000 of them die from those infections," said Rep. Joe Barton (R-TX), chairman of the subcommittee. "[That is] more than die of breast cancer and automobile accidents. Each and every hospital-acquired infection represents a real human being. I hope that fact will remain front and center as we discuss this important and complex issue."
Indeed, the patient has been too long obscured under a shroud of arcane terms and abstract epidemiological measurements, a clinician told the panel.
"We were reporting for years an average rate of 5.1 infections per 1,000 line days," testified Richard Shannon, MD, chair of the department of medicine at Allegheny General Hospital in Pittsburgh. "One day, I said, 'How many human beings is that? Five, 10, 50?' There was simply no way to know. When data are presented in an obscure fashion, it may be understandable to an epidemiologist but I submit that we — and I venture to say most health care professionals — were totally unaware of the tragic human consequences or of our primary involvement in them."
Threat of federal action
The threat of federal action was very much in the air, as the committee grasped the severity of the problem and heard descriptions of new aggressive programs such as Shannon's that are proving effective against hospital infections. Committee members threatened to bring fiscal pressure to bear on hospitals in the form of withheld hospital reimbursements from the Centers for Medicare & Medicaid Services (CMS). Indeed, Rep. Tim Murphy (R-PA) recently followed up on a threat he made at the meeting, releasing a report recommending in part that "health care providers should reduce health care acquired infections with pay-for-performance incentives through Medicare and Medicaid." Moreover, Stupak argued that the CDC could apply regulatory enforcement immediately under current law by partnering with CMS and withholding federal funds from hospitals that have poor infection control programs. Citing Section 1864 C of the Medicare & Medicaid Act, Stupak said the CDC and CMS could revoke accreditation of hospitals that have serious infection control deficiencies.
Drawing the brunt of Congressional ire — somewhat unfairly since much of the outrage drew from policies put in place long before her tenure —was Denise Cardo, MD, chief of the CDC division of healthcare quality promotion.
"Earlier, you said CMS was one of your partners, so wouldn't [revoking accreditation] be an enforcement mechanism if you really wanted to drive down infection rates?" Stupak said to Cardo during her testimony. "We have Mr. Murphy over here proposing that we withhold federal payments to hospitals that don't lower their infection rates. I believe you already have that authority under Section 1864 C, so the Murphy legislation may not be necessary. Thirty-five years that CDC has been looking at it, and I don't see any dramatic [effort] from CDC to tell hospitals to clean it up or we will enforce this."
The CDC has traditionally taken a voluntary, nonregulatory stance in issuing infection control guidelines, but Stupak charged that little progress has been made since the agency began focusing on the problem in 1970.
"The good news is that we are not the only ones now who think this is an important issue," Cardo said. "In the past it was [just] the CDC and the professional organizations directly involved in the prevention of infections. We see now that the partnerships have expanded [to] the public, the purchasers. We have CMS getting very involved in things and we are working very close together. We have the Joint Commission. I think the fact that we are here today with so many people talking about the problem and how to solve it shows that there is a commitment now at the federal level to make a difference."
Though the CDC has launched several major infection control initiatives in recent years and now has a former hospital epidemiologist as its director, Stupak lectured Cardo about both the CDC's historic approach and current results. Citing evidence presented at the hearing that aggressive, targeted infection prevention programs such as the Keystone initiative in Michigan are producing dramatic results, Stupak asked why the CDC and CMS have not demanded "that these preventive programs be instituted now in every hospital getting federal money?"
Reminding him that the Keystone project uses many guidelines and recommendations developed by the CDC, Cardo had to summon grace under pressure as Stupak hammered away.
"We have had several very strong recommendations that prevention should be a priority," she said. "I agree that there is a gap between what should be done and what exactly is happening. We are working with our partners [and have made it] a priority to show that if things are done the way they should be done we can prevent infections and save lives. . . . CMS is the enforcement authority with regard to hospitals. I would be more than happy to get an answer for you from CMS and provide it for the record."
Congress sees explosion of state laws
The general tone of the meeting certainly underscored that state and national interest in infection rate disclosure laws is increasing. Hospital infection rate reporting requirements have been adopted in eight states, with others preparing to follow suit even as discussion continues about creation of a uniform national standard.
"We need to figure out how to make public reporting as effective and fair as possible," Barton said. "A patchwork of state reporting laws presents a challenge in terms of differing standards and requirements, but also an opportunity. Uniform national standards might ultimately be appropriate, but in the meantime, these state efforts serve as an excellent laboratory to identify the best practices. On a personal note, I feel strongly that consumers should have the right to know how hospitals measure up on this important issue. Transparency has been a fundamental theme of this committee's health care work and hospital infections are no different."
Nevertheless, a veteran epidemiologist advised the committee not to enact national infection rate reporting requirements until the best lessons from the state efforts can be gleaned.
"At this point you see the disagreement of the nuts and bolts of how this should be done," said Robert Haley, epidemiologist at the University of Texas Southwestern Medical Center in Dallas. "[We have seen in] other federal areas that when you have states experimenting for a period of time you get all kinds of different models and then over time you learn about them. You have national scientific meetings where people debate them, and then two, three, four years from now, we might be in a position where you want to have a national standard — or one might just evolve as we all sort of agree on what needs to be done. I think to freeze it right now before there is experimentation might reduce innovation."
In addition — despite all the enthusiasm and clear political momentum — another epidemiologist pointed out that pesky but persistent problem with ensuring hospitals use risk-adjusted data to allow meaningful comparisons. Unfortunately, the risk adjustment methods currently available are limited in their ability to account for differences in patient population and acuity levels among facilities, warned Scott Hammer, MD, chief of infectious diseases at Presbyterian Hospital and Columbia University Medical Center in New York City.
"Unadjusted or poorly adjusted [infection] rates may lead to unintended and undesirable public health consequences," he told the panel. "For example, a patient misinterpreting [infection] data may avoid seeking treatment at a particular facility, despite its being more experienced and better equipped to treat the patient's condition."
While emphasizing that he was a "convert" on the issue of infection rate disclosures, Haley concurred that the process could backfire if states begin generating raw numbers unadjusted for factors like severity of illness.
"When you are trying to put up a web site where you have data that consumers are going to look at — and presumably it's there so they can look at it and make a decision on which hospital to go to — if you don't risk adjust the data chances are the hospital with the lowest infection rate is the one that is going to be the one that takes care of the [healthiest] patients," he explained to the committee. "Whereas, if you risk adjust [you find that the hospital] that takes care of the more complicated patients actually does a better job. The patient could go to the hospital with the higher rate because that's where they actually do a better job and prevent infection better. If you don't risk adjust, you are in the paradox of sending the patients to the place where they have a higher risk of infection."
Risk adjustment can't justify inaction
On the other hand, it would be a mistake to delay action while trying to develop ideal reporting methods, argued Shannon. It would be better to remove data on high-risk patients — such as trauma and burn victims — and compare them in a different data set, he noted. Indeed, data being generated by reporting requirements under Pennsylvania's disclosure law suggest high-risk patients are not going to confound reporting efforts, he said.
"The experience in Pennsylvania [is that] of the 11,668 reported infections [in 2004], only 300 were in any of the high-risk groups that were mentioned today," he said. "What about the 11,300 other people that aren't in high-risk groups? I would be willing, in deference to public reports, to take out [high-risk patients] or compare them separately if that is what you want to do, but I think to hold up the process by virtue of waiting for an acceptable risk adjustment strategy is going to be a delay."
Taking an iconoclast's view of traditional health care epidemiology, Shannon dismissed the old dogma that hospital infections are the unavoidable result of providing invasive medical care to immune compromised patients.
"It is my goal today to convince you that hospital-acquired infections are not inevitable, but products of unreliable processes and maligned incentives that reward activity, not clinical outcomes," he testified. "A major challenge to the concept of public reporting has been the theory of inevitably — the notion that a hospital-acquired infection is an inevitable consequence of complex care and therefore an acceptable form of collateral damage in our daily battle against disease."
There only is one acceptable benchmark for infections — zero, Shannon argued. "The argument that the data need to be normalized in order to compare hospitals of different sizes and types simply, to my mind, focuses the attention on the wrong set of comparisons," he said. "Rather, I would submit the correct approach is for each hospital to benchmark against itself in its current condition and demonstrate rapid and consistent progress toward the theoretical limit."
Shannon has adapted business production principles to the clinical setting, focusing on standardized approaches to care that can eliminate some of the variables that lead to infections.
"This is about processes of care in which one places a catheter in an arm in a standard way and then guarantees that it is going to be maintained in a sterile way in the period that it is indwelling," he said. "This has to do with educating operators that variations in the way to put in a line are not helpful to a process. Can't we, in an intelligent way, agree that there is one way to place a subclavian line? Some of us gown, some of us don't. Some of us glove, some of us don't. Some wear caps; some don't. . . . What we have done is standardize those processes so that at any moment a variation can identify a potential harmful circumstance that might propagate into infection."
Such techniques over a 32-month period resulted in a reduction of central line infections from 49 to three and deaths from 19 to one, he told the panel. "Those results occurred at the same time as we have seen a doubling of the use of central catheters. And we have seen a steady increase in the acuity of illness in our patients," Shannon emphasized. "Said differently, using more catheters and caring for sicker patients is not justification for higher numbers of infections."
To underscore the cost of infection in human terms, Shannon detailed one case: a 39-year-old patient who was admitted to the hospital with pancreatitis. On the sixth hospital day, the man developed a central line associated bloodstream infection with methicillin-resistant Staphylococcus aureus. As result, he developed multiple complications, developing abscesses that had to be drained and renal failure that required dialysis. His hospitalization extended to 86 days before he was transferred to a long-term care facility.
"I don't share this with you because I am proud of it," Shannon said. "I share this with you because unless you understand the human consequences, there is no motive to change. Health care workers are not motivated by infections per 1,000 line days, but they renounce the current condition when they understand the magnitude of human harm expressed in such human terms. They begin to believe they can do something about it."
The economics of the case were staggering, with the patient incurring unreimbursed costs of $41,000 on a bill that approached a quarter-million dollars. "Notably, the care provided as a result of the complication costs $170,000 or nearly 71% of the entire costs of care," he said. "These costs do not include payment to physicians, his ongoing dialysis, or the loss of a productive worker."