CM: Nosocomial infections are those that were historically acquired in the hospital, but now it’s been broadened to include health care related infections, which would include infections from outpatient surgery. Many of these infections, unlike community infections, can be prevented. Many of the reasons you get these type of infections are because there have been lapses in protocol. There are clearly best practice guidelines published regarding how you can prevent these infection types, which are responsible for about two million infections and 100,000 deaths a year in acute care centers in the U.S. Clearly, there is concern and reason why we should work on reduction of these types of infections.

CM: To have zero preventable nosocomial infections. We’re not exactly sure what that will mean, as far as how low we can go, but a lot of experts agree that there’s a lot of prevention that could occur in hospital and health care settings. I don’t think anybody has continued to focus on sustained reductions of many of the different infections types. That’s what this group is hoping to do.

CM: The hospitals have representation from the nosocomial infection group and the medication error group, which are both part of the patient safety arm of PRHI. Here at our hospital I’m involved, as the director of Infection Control, as is one of our infection control practitioners. We attend monthly meetings at PRHI. More importantly, the CEOs at these hospitals have all agreed to be involved in this initiative. There are reports sent to the high-acting officials of all the hospitals, as well as all their top operational people. So, there is constant feedback to the hospitals as to what we’re measuring and, as the rates become available, they’re also fed back and distributed through the hospitals. What it allows for is comparison on a local level. That was never really available and isn’t available anywhere else. We not only can compare our rates of infection to the means that are nationally published by the CDC, we now have a more local group to see what the infection rates are looking like. Then, we can work together to figure out what aspects of prevention could be put into place to further reduce those rates.

CM: The first task involved getting all the hospitals to use the exact same terminology and definitions so that we could truly compare infections at the hospitals. We started with catheter-related bloodstream infections because they’re already being collected in most hospitals, so it wouldn’t involve extra work for the infection control professionals. Additionally, this type of infection is one that has a very high mortality associated with it: 10 to 40 percent of people who get these infections will die. And they’re very costly, about $33,000 per infection. We were all trained last year by the CDC, and those hospitals that historically couldn’t participate in the national database repository of these infection rates are now being able to, which was something that couldn’t have occurred otherwise.

The second part was to create an educational program which would address all of these targeted practices for placement and care of a central line. As a group we formulated a presentation which was then rolled out in all the hospitals by their infection control practitioners. Much of the time, health care people don’t try to not follow practice. Sometimes there are changes in practice that they may not be aware of, just because literature changes, and that sort of thing. But more importantly, we found that it is sometimes hard to do the right thing because all of the equipment isn’t easily assessable, and the health care person has to spend a lot of time collecting it for central line placement—a sterile barrier, a large drape, a sterile gown, gloves, a mask, things like this. If they’re not easy for the health care worker to get to, perhaps they may not follow those good practices each time.

Besides the educational component, we used the interaction with clinicians as a time to figure out what were the problems at each hospital and address those things to make it easy for all of the components necessary to place the line to be easily accessible. Another thing we were able to implement was the use of Chlorhexidene, which is a product used for disinfection but is new to this country for the purpose of disinfecting skin. We’ve always had it as a hand washing agent, but not as a disinfectant. We rallied to try to get this implemented in each hospital and many, if not all, of the hospitals have implemented it. It’s been shown to significantly reduce these types of infections. The feedback component of the initiative is an intervention in its own right. It’s been studied and shown that a feedback component is essential to create and support continual reduction of infections. So, it’s a much more formalized process now: all the hospitals receive their own data by individual intensive care units, as well as the mean data for the region.

CM: With the whole community watching and saying, "We want to have hospitals with better rates," it actually was not to your advantage to not join this initiative because no doubt you would be viewed as a hospital that wasn’t supporting patient safety. The other thing that happened is, Paul O’Neill, who’s now the Secretary of the Treasury, was here in Pittsburgh heading the company Alcoa, and they’re firm believers of safety issues. He greatly supported this program and, through association of other prominent business people in the community, it became much more easy to get the support of the hospitals.

CM: When you get your feedback, only your rates are identified by institution. The rest of the information that you receive is just the mean of the rest of the hospitals in the collaborative. The hospitals don’t get to see one specific facility’s data versus another. The Nosocomial Advisory Group, which includes infection control professionals from all of the hospitals that are represented, does get to see the information by specific facility and they sign a confidentiality agreement that says they can’t discuss any of the information and none of the information leaves the meeting room. They see facility-specific data mostly for the purpose of trying to figure out what hospital may have interventions that could be transported to other facilities. If there’s a facility that has a very low rate, we’re going to find out how they get their targeted practices in place. We can learn a great deal from places where there are low rates, as well high rates, what to do as well as what not to do.

CM: You can see where you fall, as far as the percentile distribution. If your hospital has a rate that’s much higher than the mean, that would be meaningful to most institutions. We would investigate why it is we’re having higher rates than the region and see if we can come up with a rationale and an intervention. Because the operational people and CEOs are seeing the data, I think it’s sort of forcing the investigation that may or may not have taken place before. Most folks were thinking that, if they were in line with the CDC’s numbers, that was okay and good enough. What we’re striving for is to no longer worship those means and to actually reduce infections below what had been considered acceptable, previously. We also present a utilization ratio and percentile distribution which tells you how often you’re using the central line device—based on the number of device days over the number of patient days in the specific ICUs. You can see if you’re utilizing more of these lines than the rest of the region. That’s very important because literature has shown that those devices are left in too long, sometimes. If your utilization ratio is very high, at least it would force you to look to see if you’re putting in too many lines, if you’re staying in too long, or what might account for that very high ratio. We’ve had four quarters of the bloodstream report sent out.

CM: Yes, the fourth quarter of our data is just back. Although these are preliminary numbers and still have to be completely confirmed, the aggregate reduction looks like it’s 22 percent from the second quarter of 2001 through the first quarter of 2002.

The final thing which we hope will be associated with even more significant reduction is the use of a procedure note that includes the things that are associated with increased infection risk. It will be used to collect information on the actual practices used for line placement. We went to the CDC guidelines for prevention of catheter-related bloodstream infections and summarized that document. We chose catheter insertion because it’s primarily done by physicians, as opposed to the catheter care, which is typically done by an IV team or the nursing unit personnel. The reason we wanted to target things that the physicians do is because they’re a much more hard-to-reach group. The nurses we can get together and ask them to use this dressing or this disinfectant. The physicians—there’s no forum in which you can gather them to do that sort of educational thing. You can go to their division meetings and there are grand rounds, but there’s nothing where 100 percent of them have to be there, so they’re a harder group to target.