Understanding and Addressing Variations in Healthcare-associated Infections (HAIs) After Durable Ventricular Assist Device Therapy

Durable ventricular assist devices (VADs) provide long-term mechanical circulatory support as a bridge to cardiac transplantation or as permanent therapy, and have the potential to benefit more than 250K patients with advanced heart failure (HF) refractory to guideline-directed medical therapy. Upwards of six out of every 10 VAD patients develop a healthcare-associated infection (HAI) following implant. Broader adoption of durable VAD therapy is hindered by HAIs given their association with mortality (5.6-fold increased risk of 1-year mortality) and healthcare expenditures ($264K - $869K per patient).

Most studies evaluating HAIs in the setting of VAD therapy have been limited to post-hoc analyses of small trials, or single center experiences. While some argue that centers with lower HAI rates select healthier patients, others counter that improved practices (e.g., standardizing empirical and targeted antimicrobial therapy) or enhanced provider teamwork are more predictive of improved outcomes. While evidence-based HAI prevention guidelines and implementation tools (e.g., checklists) exist, these approaches are not comprehensive (e.g., neglecting device-specific determinants) or customizable to local context, thus limiting usability, adoption, and likelihood of significant effectiveness for preventing HAIs. Understanding the barriers and facilitators within individual centers for maximizing adoption of prevention measures would serve as the foundation for targeted strategies. Without this knowledge, evidence-based, action-oriented recommendations will have limited local adoption and ultimately effectiveness in preventing HAIs after VAD implantation. The investigators' long-term goal is to develop and subsequently promote wide-scale adoption of evidence-based HAI prevention practices following durable VAD implantation. The objective of this proposal is to identify prevention recommendations for the most significant HAIs after VAD implantation. To achieve this objective, the investigators will undertake a mixed methods study of adult patients receiving VADs in the U.S. from 2009 - 2017 and develop a modular toolkit of evidence-based recommendations. To determine best practices for preventing HAIs, the investigators will examine center-level differences in HAI rates to identify strategies used by centers with low rates, and barriers to lowering HAI rates among centers with high rates. The investigators will use a novel dataset including administrative claims and detailed clinical data, along with in-depth center surveys, to address the following aims: (i) identify determinants of center-level variability in HAI rates, (ii) develop a comprehensive understanding of barriers and facilitators for achieving low center HAI rates, and (iii) develop, iteratively enhance, and disseminate a best practices toolkit for preventing HAIs that accommodates various center contexts.

These studies will have a positive impact on clinical practice by identifying targets for an interventional study. Moreover, the mixed methods approach will serve as a model for evaluating and improving clinical care more broadly - especially in the setting of complex surgical procedures.