This study will be a multicenter, cluster randomized, step-wedged design. The unit of randomization will be the inpatient units admitted critically ill patients, with the interventions being carried out over 4 months. The study plans to enroll 14 units at 2 sites. Four sequences will be enrolled into the intervention each month; each sequence will have 3-4 units (see figure 1).

Units eligible for the study will have patients that are critically ill admitted and have had at least 1 CLABSI events over the past 12 months (fiscal year). Given the pragmatic nature of the study design, there will be limited additional inclusion and exclusion criteria. Prior to the beginning the study, all nursing staff (RNs and nursing assistants [NAs]) will receive a survey link (RedCAP) to understand their perceptions of CHG bathing. After all units have been enrolled (approximately 4 months), nursing staff will be sent a post-survey to see if their perceptions have improved after the program. Further, the Context Assessment Index (used with permission; via RedCAP) will also be used to assess the context (i.e., culture) in which clinicians works and the effect this has on using evidence in practice. This will be provided to infection prevention champions on each unit one time, at the beginning of enrollment.

All hospitals in the trial will receive access to site-level quality reports on CLABSI data. The interventions will include a "direct engagement" at the site level; this strategy will build upon current quality improvement interventions developed from the Agency for Healthcare Research & Quality for optimization of care for critically ill patients at risk for CLABSIs. The multidisciplinary teams will include national key opinion leaders in quality improvement working with local infection prevention specialists and support staff to help healthcare systems and hospitals design or revise quality improvement plans. Units will receive feedback on quality improvement efforts, including audit and feedback reviewing their CHG bathing compliance and CLABSI rates. Duke will serve as the primary statistical center and analysis will be generated by Duke. The only risk in this study is the possibility of breach of confidentiality.

The primary objective of this study is to assess the effect of a customized, multifaceted quality improvement [QI] program on compliance with daily chlorhexidine gluconate (CHG) bathing per the AHRQ protocol in inpatient units that admit critically ill patients. Further, we will assess the effect of this program on nursing staffs' perception of the importance of CHG bathing. The Context Assessment Index (used with permission) will also be used to assess the context (i.e., culture) in which clinicians works and the effect this has on using evidence in practice. The secondary objective of this study is to examine the effect of the QI program on central line-associated bloodstream infection (CLABSI) rates. Primary outcomes and CLABSI rates will be measured at 12 months to assess sustainability.

Inclusion Criteria:

• Units eligible for the study will have patients that are critically ill admitted and have had at least 1 CLABSI events over the past 12 months (fiscal year).

Exclusion Criteria:

• Units with 0 CLABSIs over the past 12 months (fiscal year).

Individual participant data (IPD) will not be available to other researchers.