Sriram Swaminathan

BACKGROUND: Infections following durable left ventricular assist device (dLVAD) implantation are common and associated with increased morbidity and mortality. Despite documented interhospital variability, few studies have identified strategies to mitigate their occurrence. This national study uses a multistage mixed methods design to develop a customizable and deployable toolkit of expert-guided recommendations to reduce infections post-dLVAD. METHODS: Using purposeful sampling, participants (eg, clinical and operational ventricular assist device [VAD] team members) from low, medium, and high-performance hospitals (based on their risk-adjusted, 90-day post-implantation infection rates) across the United States were interviewed to assess factors contributing to postdLVAD infections. Draft toolkit recommendations were iteratively developed after integrating thematically analyzed qualitative and quantitative data from a merged national registry with Medicare and hospital survey data. A national advisory team of VAD subject matter experts provided mixed methods input to refine the toolkit’s content and structure. RESULTS: Seventy-three clinical and operational VAD team members across 8 US hospitals were interviewed, spanning low (n=4), medium (n=1) and high (n=3) performance groups. Fourteen subject matter experts provided stakeholder feedback to refine the toolkit. The resulting toolkit contains 39 infection prevention recommendations that address VAD program care processes (eg, real-time provider communication), clinicians (eg, multidisciplinary protocol development), patients and caregivers (eg, engaging patient advisors in patient education), and VAD leadership (eg, unit and service level data reporting). Accompanying resources (eg, team-based exercises, data collection worksheets) support implementing and evaluating site-specific strategies. CONCLUSIONS: Using mixed methods approaches, an infection prevention toolkit was developed to enhance care coordination among VAD team members and mitigate postdLVAD infections. Future work should evaluate the effectiveness of implementing this infection prevention toolkit within the dLVAD setting.

Background Infections following durable left ventricular assist device (dLVAD) implantation are common and associated with increased morbidity and mortality. Despite documented interhospital variability, few studies have identified strategies to mitigate their occurrence. This national study uses a multistage mixed methods design to develop a customizable and deployable toolkit of expert-guided recommendations to reduce infections post-dLVAD. Methods Representatives from low, medium, and high-performance hospitals across the U.S. were interviewed to assess factors contributing to post-dLVAD infections. Draft toolkit recommendations were iteratively developed after integrating thematically analyzed qualitative and quantitative data. A national advisory team of ventricular assist device (VAD) subject matter experts provided mixed methods input to refine the toolkit’s content and structure. Results Seventy-three clinical and operational VAD team members were interviewed, and 14 subject matter experts provided stakeholder feedback to refine the toolkit. The resulting toolkit contains 39 infection prevention recommendations that address VAD program care processes (e.g., real-time provider communication), clinicians (e.g., multidisciplinary protocol development), patients & caregivers (e.g., engaging patient advisors in patient education), and VAD leadership (e.g., unit and service level data reporting). Accompanying resources (e.g., team-based exercises, data collection worksheets) support implementing and evaluating site-specific strategies. Input from clinical and research experts provided preliminary evidence of the toolkit’s acceptability and considerations for enhancing the toolkit’s adoption and implementation. Conclusions Using mixed methods approaches, an infection prevention toolkit was developed to enhance care coordination among VAD team members and mitigate post-dLVAD infections. Future work should evaluate the effectiveness of implementing this infection prevention toolkit within the dLVAD setting. What is Known Six out of every 10 patients develop an infection within 2 years of durable left ventricular assist device (dLVAD) implantation. Infections contribute to an increased risk of major morbidity and mortality. Post-dLVAD infection rates vary across hospitals, even after risk adjustment. What the Study Adds A customizable toolkit of infection prevention recommendations was developed and pilot tested using mixed methods approaches. Recommendations focus on modifiable aspects of organizational structure, patient care, patient education, quality reporting.

While mixed methods research is increasingly used to examine determinants of unwarranted variability in healthcare delivery and outcomes, novel integrative approaches are required to meet the needs of mixed methods healthcare delivery research. This article describes novel refining strategies that enhance the linkage between qualitative and quantitative dimensions of a mixed methods healthcare delivery research study. Leveraging our study experiences, this paper demonstrates several refining strategies: (1) using mediated allocation concealment to facilitate qualitative sampling; (2) informing qualitative inquiry through quantitative analytics; and (3) training and immersing multidisciplinary researchers in qualitative data collection and analysis. Developing and implementing strategies in mixed methods healthcare delivery research could advance methodological rigor and strengthen multidisciplinary collaboration.