The Arcus Experience: Bridging the Data Science Gap for Nurse Researchers

Abstract

Background: For years, nurse researchers have been called upon to engage with "big data" in the electronic health record (EHR) by leading studies focusing on nurse-centric patient outcomes and providing clinical analysis of potential outcome indicators. However, the current gap in nurses' data science education and training poses a significant barrier.

Objectives: We aimed to evaluate the viability of conducting nurse-led, big-data research projects within a custom-designed computational laboratory and examine the support required by a team of researchers with little to no big-data experience.

Methods: Four nurse-led research teams developed a research question reliant on existing EHR data. Each team was given its own virtual computational laboratory populated with raw data. A data science education team provided instruction in coding languages-primarily structured query language and R-and data science techniques to organize and analyze the data.

Results: Three research teams have completed studies, resulting in one manuscript currently undergoing peer review and two manuscripts in progress. The final team is performing data analysis. Five barriers and five facilitators to big-data projects were identified.

Discussion: As the data science learning curve is steep, organizations need to help bridge the gap between what is currently taught in doctoral nursing programs and what is required of clinical nurse researchers to successfully engage in big-data methods. In addition, clinical nurse researchers require protected research time and a data science infrastructure that supports novice efforts with education, mentorship, and computational laboratory resources.