Joint Commission Infection Control Conference

Between the unknown and the uninformed

ICPs must craft message,’ provide answers

Amid increasing sensational press exposés and consumer advocates demanding release of hospital infection rates, comes this cold truth from a leading public health official: Health care-associated infections are fraught with so many variables that epidemiologists don’t really know how many occur and how many can be prevented.

For example, it has been estimated 2 million nosocomial infections occur annually, but
projections are as high as 4 million, said Steve Solomon, MD, acting director of the division of health care quality promotion at the Centers for Disease Control and Prevention (CDC).

Solomon spoke at an infection control conference held recently in Chicago by the Joint Com-mission, which has taken an increasing interest in collecting data about serious nosocomial infections. While much is unknown, Solomon stated flatly that one thing is clear — there is an inherent risk in entering the health care system.

"Who is at risk? Everybody is at risk," he said. "Every time you come into contact with the health care system there is some potential risk."

He also reminded conference attendees that the traditional wisdom that some one-third of infections are preventable is essentially a baseline estimate. The CDC data from the study on the efficacy of nosocomial infection control (SENIC) was collected decades ago by researchers who were not trying to answer the question of preventability of any given infection.1

"How preventable are nosocomial infections? That is the $64,000 question," Solomon said. "It is uncertain. For many years at CDC, we used the figure 28% to 32%. We got that from the SENIC study. It certainly is a good baseline number. What they [really] said was in 1975 if all hospitals had infection control programs — according to certain criteria of infection surveillance and control — the number of infections would be 28% to 32% lower than hospitals that don’t have such programs."

Yet with scathing press coverage and public demands for medical data increasing, there is a growing perception that health care infections are medical errors that can be prevented. Given that, consumers want to know where and how many infections are occurring before they seek medical care.

"Health care is a commodity," he said. "It is bought and sold, and that’s why consumers want information. They want information about health care to be transparent, and they want it to be accountable."

Time for a public education campaign?

Epidemiologists have long stressed that data that are not risk adjusted for patient severity of illness — or otherwise taken out of epidemiologic context — actually could make good hospitals look worse than poor ones. For example, hospitals with lax surveillance programs detect fewer infections — and thus may have lower infection rates — than those that strive to identify and prevent every one. In light of such complex issues, Solomon was asked if the CDC would consider a public education campaign to explain the inherent risks of infection in the health care environment and the ongoing efforts of ICPs to prevent such outcomes.

"What is the [campaign] message?" Solomon said, pondering the question. "The message can’t be that there is a risk that you can’t control. The message can’t be that there are things out there that are preventable that we are not preventing. We’ve got to figure out what the message is [for] our industry. Is it risk-free care? Is it the best care available? I’m not sure that we know. I agree with you [that there] is a breakdown in communication. I think we need to have a frank and honest dialogue within the industry about what is it we really want to communicate."

Barbarians at the gates?

That dialogue might need to come sooner than later, because consumer groups and attorneys are starting to demand data that hospitals traditionally have fiercely protected. For example, the Consumers Union — publishers of Consumer Reports — have set up a web site:

"We’re going to have little [ratings] circles eventually, black and red, like they do for the cars," Solomon told conference attendees.

"It sounds silly but it may not be that far away. There is tremendous press and public scrutiny, a lot of discussion in Congress. This is a very real issue. In Illinois and Pennsylvania at the state level, [there are demands] for reporting infections. . . . We have to address the public perception. We have got to do something about those screaming headlines [in] paper after paper," he continued.

Yet while there are, so to speak, barbarians at the gates, there is little comfort on what is going on inside the castle. Solomon described a health care system overwhelmed by information, hampered by antiquated systems, and thoroughly undercut by chaos.

For example, there currently is an unprecedented exchange of medical information, with more than 20,000 articles published on infectious diseases in 2003, he said. Roughly 1,000 of those articles deal with severe acute respiratory syndrome (SARS).

"We have added close to a thousand new articles on a disease that didn’t exist last New Year’s Eve," he said. "We have today in health care reached and exceeded the ability of the system to adapt to the input that is coming in to it."

Chaos within, pressure without

As a result of all the chaos within and pressure without, the CDC is striving to become more "responsive and accountable," he said. New fast-track approaches under study for infection control in clinical settings include more practical surveillance methods such as random chart review, charts screened based on certain thresholds like antibiotic use, and syndromic surveillance to detect clusters. While many private companies feature elaborate tracking systems that can tell where your order is at any time, the health care system is in need of a massive overhauling of computer and information systems to meet current demands for information, he said.

"We need to free your time to do hands-on prevention," he told the ICPs in the audience.

Better statistical models are needed to delineate patients with "suspect" infections, who may or may not be epidemiologically significant but remain outside the grasp of surveillance definition. "There is a group of patients who have
suspect infections who live in a nether world between confirmed infections and no infection," Solomon said.

Given such gray areas, the demands by some for hospital infection data — like a reading off a meter — seem all the more counterproductive. But the demands will not stop, so the challenge
is to make the data more meaningful and understandable, he said.

"I think the thing for us to do as infection control professionals and epidemiologists is to figure out how we are going to make the data more meaningful, and convince the people that want that data that a lot of [that information] isn’t going to help them," Solomon said. "We have got to figure out which data are going to help them and show them why they can help them. We have got to turn that sow’s ear into a silk purse, and it’s not going to be easy."


1. 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.