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

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.

Predicting Pressure Injury Risk in Pediatric Patients: The Braden QD Scale.

OBJECTIVE: To describe the development and initial testing of the Braden QD Scale to predict both immobility-related and medical device-related pressure injury risk in pediatric patients. STUDY DESIGN: This was a multicenter, prospective cohort study enrolling hospitalized patients, preterm to 21 years of age, on bedrest for at least 24 hours with a medical device in place. Receiver operating characteristic curves using scores from the first observation day were used to characterize Braden QD Scale performance, including areas under the curve (AUC) with 95% CIs. RESULTS: Eight centers enrolled 625 patients. A total of 86 hospital-acquired pressure injures were observed in 49 (8%) patients: 22 immobility-related pressure injuries in 14 (2%) patients and 64 medical device-related pressure injuries in 42 (7%) patients. The Braden QD Scale performed well in predicting immobility-related and medical device-related pressure injuries in the overall sample, with an AUC of 0.78 (95% CI 0.73-0.84). At a cutoff score of 13, the AUC was 0.72 (95% CI 0.67-0.78), providing a sensitivity of 0.86 (95% CI 0.76-0.92), specificity of 0.59 (95% CI 0.55-0.63), positive predictive value of 0.15 (95% CI 0.11-0.19), negative predictive value of 0.98 (95% CI 0.97-0.99), and a positive likelihood ratio of 2.09 (95% CI 0.95-4.58). CONCLUSIONS: The Braden QD Scale reliably predicts both immobility-related and device-related pressure injuries in the pediatric acute care environment and will be helpful in monitoring care and in guiding resource use in the prevention of hospital-acquired pressure injuries.