Perhaps you’ve been working on CLABSI prevention and are certain there is nothing more that can be done to improve outcomes. You’ve reviewed bundles, implemented everything with some evidence of benefit and none of harm. This is a topic very hard to research using the strongest method of research, that is to say, a randomized controlled trial.
It would be unethical to randomize a baby to a group where care of the PICC line was not meticulous. Much of the work of performance improvement focuses on process. In other words, the outcome is admirable but if we don’t comply with interventions we know to decrease infection, we just maintain the low rate of infection through chance. Eventually, the rate will return to previous levels if practice doesn’t change and become a new habit and standard. So, one might ask, what is it exactly that makes the difference? The concept of having a “bundle” of interventions is that there is not any one thing that makes the difference. It is a collection of interventions, the whole being greater than the sum of the parts. In addition, there is not a 1:1 relationship between bundle compliance and non-compliance and infection rates. Keeping your eye on the ball and monitoring compliance, whether as a leader in the organization, a co-worker or care provider is important. Not every breach in compliance will result in an infection but that doesn’t mean compliance is not important. The impact of bundle compliance has been documented in looking at large numbers of patients with large numbers of opportunities to use the bundle. The statistics show that better compliance leads to better outcomes but not 0 infections. There was a time we thought we would not be able to decrease infections, let alone maintain such low levels due to what we believed was a high risk of infection that could not be ameliorated. Research has shown that certain practices result in better outcomes. If there is a best way to do something, it should be done that way all the time. Standardization is a core principle of quality improvement. Unnecessary variation in process leads to two outcomes: poorer quality and higher costs. Unnecessary variation also leads to the impression that it is OK to not follow preferred practices and thus spread this ‘cutting corners’ approach to other processes such as hand hygiene or hub scrubs. We need to build and sustain a culture of adherence to best practices. It is not an option to pick which practices will be followed or not. Whenever cause and effect are not tightly linked e.g. not every break in technique leads to a poor outcome, it is hard to always appreciate the importance of 100% compliance with preferred practices. When breaks in technique/practice are looked at in the context of the unit’s culture, the spill-over to other practices, and the theory of harm in healthcare the need to audit and insist on adherence is pretty clear. Finally, it is important to appreciate the “Swiss cheese” theory of patient harm. This states that when an error reaches the patient, it is often the result of many small breaks in practice that all “line-up” and allow the error (infection in this case) to reach the patient. Not following best practices in any process allows small errors to occur which might eventually line up to cause harm to the patient. The system is only as good as those implementing it. It is also important to constantly monitor, not only the performance of known improvement practices but to question what other practices may make the difference and cause a significant drop in infections.
Several new papers have been published recently reporting on measures related to CLABSI prevention. One publication showed a relationship between leadership walk-rounds and improved or high patient safety climate.1 Leadership is important to support safe patient care. Leaders must be aware of what is going well and when additional or different personnel, supplies, or other resources are needed. Whether this is accomplished through formal leadership rounding or simply a leadership “presence”, the same result may be achieved. Unfortunately, while we all know these things are important, it is difficult to demonstrate an improvement in infection rates or other specific outcomes. The number of bad outcomes avoided by leadership presence cannot be measured. Most published reports of leadership rounding focus on outcomes such as surveys showing an improvement in safety climate, as this paper did or improvements in patient satisfaction surveys, often not influenced by the quality of care directly but more by good and personal customer service.
Two other recently published papers present two CLABSI prevention measures that may need consideration. Fenik, Celebi, Wagner et al present a simulation scenario in which the use of a prepackaged central line kit was used by some participants and not by others.2 These were novice residents placing central lines in adults with the assistance of novice nurses. Practicing in a simulation lab, one group placed the central line utilizing a manufactured prepackaged kit that contained most of the items needed for the line placement. Items not included in the kit were things with a short shelf life e.g. flush or lidocaine or items that needed to be personalized for the clinician e.g. gloves. The other group used a standard kit containing the central catheter and all other materials were available in a supply cart. Through video taped observation and scoring, fewer breaks in technique were found in the group utilizing the prepackaged kit. There were several weaknesses in this study but given a situation with a novice inserter, the use of a prepackaged kit may serve to avoid insertion related infections.
The last paper by Casner, Hoesli, Slaughter, Hill, Weitkamp discusses a phenomenon which has been observed in previous situations, that of an infection post PICC removal.3 This was a retrospective study and showed a 3.83-fold increase in culture-negative sepsis in infants in whom the PICC was not used for administration of antibiotics in the 72 hours after PICC removal and a 6-fold increase in very low birth weight infants. This phenomenon was also observed in some, but not all previous studies.4-6 However, the possibility of unnecessary exposure to antibiotics must be considered. By administering antibiotics as a prophylactic measure prior to discontinuation of a PICC, there is a possibility of resistant organism development, not only in an individual patient but in the environment. While there was an increase in infections post-removal in the paper by Casner, Hoesli, Slaughter, Hill, Weitkamp, this was found not to be statistically significant.3 The concept of post-removal infection is thought to be related to the “shedding” of bacteria colonized on the catheter. Basically, the removal of the catheter “frees” the bacteria into the patient’s blood stream.
The field of infection prevention is constantly evolving. Many measures are quite simple and have been a part of infection prevention since Florence Nightingale! Things like handwashing and general cleanliness just make sense. However, with these basic measures implemented, and infections still happening, we owe it to our patients to keep looking closer to find measures that can make a difference. Since the work of decreasing CLABSI has accomplished a significant drop in CLABSI related infections, the number of patients or line days needed to demonstrate a significant change in infection rates has increased. Even with thousands of line days in a large collaborative, only a few infections occur. Therefore, changes implemented may take quite some time to demonstrate an effect, either positive or negative, and even more to make this difference statistically significant. Rapid cycle testing and change can be helpful in figuring out what makes a difference in your setting and culture.
1. Schwendimann R, Milne J, Frush K et al. 2013. A closer look at associations between hospital leadership walkrounds and patient safety climate and risk reduction: A cross-sectional study. American Journal of Medical Quality. 28(5):414-421.
2. Fenik Y, Celebi N, Wagner R, et al. 2013. Prepackaged central line kits reduce procedural mistakes during central line insertion: a randomized controlled prospective trial. BMC Medical Education. 13:60-68.
3. Casner M, Hoesli SJ, Slaughter JC, Hill M, Weitkamp J-H, 2014. Incidence of catheter-related bloodstream infections in neonates following removal of peripherally inserted central venous catheters. Neonatal Intensive Care. 15(1):42-48.
4. Brooker RW, Keenan WJ, 2007. Catheter related bloodstream infection following PICC removal in preterm infants. Journal of Perinatology. 27:171-174.
5. Van den Hoogen A, Brouwer MJ, Gerards LJ, et al, 2008. Removal of percutaneously inserted central venous catheters in neonates is associated with the occurrence of sepsis. Acta Paediatr. 97:1250-1252.
6. Hemels MA, van den Hoogen A, Verboon-Maciolek MA et al. 2011. Prevention of neonatal late-onset sepsis associated with the removal of percutaneously inserted central venous catheters in preterm infants. Pediatric Critical Care Medicine. 12: 445-448.
Read more of Sandy Beauman’s insight into infection:
View her blog entry Infection Prevention, Part 1.
Click here to read the full blog entry.