Studies dash dogma of inevitable health care-associated infections
Findings may be met with 'skepticism, shock, incredulity'
Health care-associated infections (HAIs) have traditionally been viewed with a certain air of epidemiological inevitability, seen in many cases as the unpreventable result of keeping very sick patients alive via invasive devices and other medical interventions.
But this view has been falling out of favor in large part because consumer activists and patient safety advocates have taken the debate to the public square, where benchmark ranges and inevitable infections are difficult concepts to defend. The Centers for Disease Control and Prevention and infection control leaders have responded with a "zero tolerance" mentality, which implies that no infection is acceptable even if it is inevitable. Part of this seems to be less a matter of substance than style, a shift in mindset that puts the ICP on more proactive footing. But what if many of the infections that are occurring in the nation's hospitals actually are preventable? Has the field lingered too long in the shadow of the CDC's landmark Study of the Efficacy of Nosocomial Infection Control (SENIC). Now more than 20 years old, that study showed that hospitals with effective infection surveillance and control programs could decrease their infection rates by 32% — in other words, prevent about one-third of infections.1
Now comes the vanguard of a new body of research on this issue, three peer-reviewed studies recently published simultaneously in a special supplement to the American Journal of Medical Quality.2-4 If the findings of these papers are substantiated and expanded upon by subsequent researchers, infections may be increasingly viewed as preventable rather than inevitable. "This supplement will be met with skepticism by some, shock by others, and incredulity by the remainder. It will stir emotion across the health care spectrum," David B. Nash, MD, MBA, FACP, editor of the journal, notes in an accompanying editorial. "From this ferment, I sincerely hope we will make the necessary commitment to more deeply evaluate the contribution of processes of care to HAIs and take action to change those processes for the better."
Though very different in approach and methodology, the studies support an overriding view that "hospital factors" — not patient severity of illness — are the root of many infections. Such a finding suggests that if hospitals invest enough energy and innovation into modifying those factors, many infections may indeed be prevented. "Regrettably, many persons within the health care industry believe that HAIs are simply a risk of doing business — almost an expected outcome from the care of seriously ill patients, especially those in our high-technology settings such as the operating room, intensive care unit, or renal dialysis center," Nash notes.5
On the contrary, the studies suggest that it is the "the process of care" — not the underlying clinical condition of the patient — that is driving the current epidemic of HAIs, he emphasizes. The perception of infections as inevitable outcomes is an "anchoring heuristic," a tendency to stick to a firmly held opinion and ignore competing facts. "The work of the three teams highlighted in this supplement will do much to help us prove the fallacy of the anchoring heuristic that infections do occur and cannot be prevented," the editorial concludes.
"The point [the editorial] made was that severity of illness was not a powerful predictor of who got an infection," says Richard Shannon, MD, a professor of medicine at the University of Pennsylvania in Philadelphia and lead author of one the studies. "What that really indicated, in both our data and the other two studies was, that [severity of illness doesn't explain] the whole universe of these infections. There were many patients who came in for elective procedures who got [central] line infections. There were patients who came in with simple conditions like heart failure — which usually are three- to five-day admissions — that got line infections. So severity of illness per se, while it clearly increased risk, did not explain the entire experience with central line infections."
A cardiologist by specialty, Shannon has become something of an iconoclast in the infection control field with his bold application of industrial safety models, challenges to accepted dogma, and avoidance of arcane medical terms. Concerning the latter, he has called for more medical transparency in testimony before Congress and repeatedly used the term "human beings" instead of patients in an interview with Hospital Infection Control. Of particular interest, he has a vision of the empowered ICP that brings them into the patient care process and away from data collection and computer screens.
"It really calls into question whether ICPs and hospital epidemiologists need to think about forming new relationships," said Shannon, who conducted much of his published research at his previous post at Allegheny General Hospital in Pittsburgh. "Too many ICPs spend their days in front of a computer entering data onto spread sheets to aggregate and conduct surveillance. The deployment of the ICP back to the bedside is the powerful new model. It is not about pushing ICPs out of the picture. It is about re-engaging them at the point of care rather than in front of their computer. Information systems can be used to aggregate these infections. The ICP needs to have new partnerships with cardiologists, pulmonologists, and critical care doctors and anesthesiologists in an attempt to try and solve the problem at the bedside."
In addition to its implications on the inevitable/ preventable debate, Shannon's article addressed a different but equally troubling misperception. Despite increasing efforts to make the "business case" for infection control, there is a lingering perception that some infections may trigger enough additional hospital reimbursement that many may actually be cost-neutral or even profitable. Such a perception creates a disincentive to make infection prevention the prime objective, but is there any truth to it? Shannon and co-researchers note that "a tacit but potentially significant barrier to the eradication of health care associated infections in general, and central line-associated bloodstream infections (CLABs) in particular, rests in the complexities of the reimbursement system. There is a widespread but unsubstantiated belief that CLABs contribute to complexity of care, resulting in increases in the case mix index and increases in outlier payments. This is a product of a reimbursement system that pays for activity as opposed to outcome, such that complications, result in coding modifications that increase payments to hospitals."
Shannon examined both the hospital revenues and expenses in 54 cases of patients with CLABs over three years in two intensive care units and compared these financial data with patients who were matched for age, severity of illness on admission, and principal diagnosis. The work was conducted at Allegheny General Hospital while the ICUs were actively engaged in the process of eliminating CLABs by applying an industrial systems redesign to medical care delivery. A team composed of both clinicians and financial personnel examined the daily clinical care to determine how much of the total cost of care was attributable to the CLAB or complications related to the CLAB.
$27,000 per infection
The average payment for a case complicated by a CLAB was $64,894 and the average expense was $91,733. That means the average loss per case was $26,839, resulting in a total red-ink bottom-line loss of $1.4 million for the 54 cases. Most of the patients were insured through Medicare (43%), 22% were insured commercially, and 22% were insured through Medicaid programs, including Medicaid managed care. In 14% of cases, payment was made by another party or by the patients themselves. The increased payments that accompanied care complicated by CLABs and associated hospital losses were observed with all payers examined, Shannon found.
"We discovered that indeed our hospital got paid more, in some cases significantly more, when care was complicated by an infection," he says. "But we also discovered that it costs us more. It actually cost more than the increase payment we received so that in 50 of the 54 infections that we studied the hospital lost money. In only four cases did we break even or make money. And the average [overall] loss was $26,000. Yes, we did get paid more, but the cost of care exceeded that additional payment. This, to my mind, creates the alignment between doing what's right for the patient and recognizing that we can prevent a loss from operations by eliminating [infections]."
An incentive to invest in infection prevention?
Given such findings it would appear hospital chief executive officers would see the obvious benefit in investment in infection prevention, but Shannon says the project needs to be replicated on a national scale before being widely adopted.
"The first step is that we need to get more data like ours," he says. "We were one center in Pittsburgh with government and commercial payers, but it might be very different in California. We are really working to try to achieve that, to do a national demonstration at least 10 institutions in various states across the nation that repeat what we did. That would answer the question is this a phenomenon in Pittsburgh, or is this a generalized phenomenon across the United States? [Then] we do need to present these data to CEOs and [tell them] you could improve your operating margin at the same time you do an enormous service for human beings by investing in the prevention of these preventable infections."
While the study specifically dealt with central line infections, Shannon has conducted similar economic analyses on ventilator-associated pneumonia (VAP) and methicillin-resistant staphylococcus aureus (MRSA) infections. "A similar principle holds," he says. "We get paid more, but it costs more so we actually lose money on cases complicated by VAPs and MRSA. So we believe, looking at three different classes of infections in one institution that the principle holds. Now we need to make that demonstration across the country."
Good for business
Beyond the cost savings, a larger economic argument can be made that a prevented infection avoids prolonged lengths of stay that tie down ICU beds. "For three years, we increased the number of admissions by 388 in a 26-bed ICU," he said. "That is where CEOs and CFOs can really see the opportunity because you are making better use of a scarce and very valuable resource in an ICU bed that otherwise would require you to build more beds or potentially defer patients to other institutions, which is never good for business. So that is the other part of the story that is extremely important. Even if you didn't buy the argument that you are losing money on a central line infection you could certainly buy the argument that you are consuming valuable resources."
The costs of CLABs and the associated complications averaged 43% of the total cost of care. The elimination of these "preventable infections" constitutes not only an opportunity to improve patient outcomes but also a significant financial opportunity, Shannon and co-authors concluded.
As many as 80% of CLABs preventable
Indeed, Shannon argues the vast majority of CLABs are preventable because he was able to drastically reduce them by adopting an industrial model based on production principles created by the automobile manufacturing firm Toyota. The system uses intensive observations to detect variances from established practices and requires root cause analyses of all CLABs. A key component is that workers are empowered to implement countermeasures designed to eliminate the defects in the processes of central line placement and maintenance.
"When we looked at the root cause of central line infections we discovered in the overwhelming majority of the infections — and we looked at 49 of them — there was a process defect," he tells HIC. "There was some lack of specification — particularly in how the line was going to be maintained — that led to the infection. For example, we discovered that oftentimes if the catheter became kinked that a medical house officer would rewire the catheter and push it back in and straighten it out. That obviously dramatically increased the risk of infection. By identifying such variable or defective processes and by making sure that you eliminate those conditions, you can greatly reduce these infections."
As a result, Shannon concluded that CLABs are not an inevitable product of complex ICU care but the result of highly variable and therefore unreliable care delivery that predisposes to infection.6
"We actually went and looked at human beings — not just collected information — when they developed infections and were able to pinpoint things that went wrong that led to the infection," he says. "By doing that — not assigning blame but rather learning from it — we created new processes that prevented that same mistake from happening again. That's the value in using what we call the 'perfecting patient care' principles, the concepts we borrowed from industry. That's the value of using that real-time root-cause analysis. In industry when a product is defective people don't wait six months to figure out why it is defective. They try to determine that right away and try to fix it."
In general terms, using such process redesign systems could probably reduce certain types of infections by as much as 80%, he notes.
"The additional 20% are far more complex and probably require a higher level of thought and problem solving than is typically applied," he says. "But lets concede that 20% — I am not willing to personally, but for those who are fighting this issue — and [at least] accept the argument that 80% of the problem can be solved by redesigning the system for placing and maintaining catheters. That is my proposition."
Informing patients
Given such results, it is no wonder Shannon rejects the traditional dogma that many infections are inevitable because very sick patients are being kept alive via invasive devices and high-tech systems.
"One of the things that made this unique for me as a leader of this effort was that I had the opportunity when an infection occurred to tell the patient and their family that an infection had occurred and we were really sorry, but we had solved the process issue so that it would never happen again," he said. "When you do that, patients and families understand. Knowing why it happened and that I had fixed that problem gave me great confidence that I was to have a sustainable and very major effect on these infections. That's why I can say 80% and [the dogma is] 30% [can be prevented], because I am taking an approach that is very different on the conventional CDC approach."
Reference
- Haley RW, Culver DH, White J, et al. The efficacy of infection surveillance and control programs in preventing outbreaks of nosocomial infections in U.S. hospitals. Am J Epidemiol 1985; 121:182-205.
- Shannon RP, Patel B, Cummins D, et al. Economics of central line-associated bloodstream infections. Am J Med Qual 2006; 21(suppl):7S-16S.
- Peng MM, Kurtz S, Johannes RS. Adverse outcomes from hospital-acquired infection in Pennsylvania cannot be attributed to increased risk on admission. Am J Med Qual 21(suppl):17S-28S.
- Hollenbeak CS, Lave JR, Zeddies T, et al. Factors Associated with risk of surgical wound infections. Am J Med Qual 2006; 21(suppl):29S-34S.
- Nash D. Hospital-acquired infections: Raising the anchoring heuristic. Am J Med Qual 2006; 21: pp. 5S-6S.
- Shannon RP, Frndak D, Grunden N, et al. Using real-time problem solving to eliminate central line infections. Joint Commission Journal on Quality and Patient Safety 2006; 32(9):479-487.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content