Northwell Health Easton, Pennsylvania, United States
Background/Case Studies:
Background: Our current, multi-hospital blood bank Laboratory Information System (LIS), Cerner, does not support complex data integration and analysis. Until recently, blood bank data analyses were time intensive as they were performed manually. Integrating laboratory information system databases with a centralized quality analytics database (QADB) platform allows laboratory personnel and healthcare providers to gain insights faster and easier into key indicators, such as lab workload, blood utilization and wastage trends, and transfusion adverse events. Here we explain the workflow of developing an automated and efficient quality data analytics platform.
Study
Design/Methods: Methods: Experts in Transfusion Medicine and Informatics collaborated to develop a user-friendly tool to automate blood bank data analytics for a group of hospitals using the same LIS. Initially, a limited volume of detailed blood product and patient history was extracted from LIS and validated by two different parameters: comparing product counts defined in a certain period to utilization reports generated by manual tallying and comparing the extracted data parameters to the actual data within the LIS. The dataset was expanded by repetitive data extracts and validations.
The QADB was then designed, and the validated blood bank data was uploaded and re-validated to show that it accurately reflected the dataset. Once this was confirmed, an extract-transform-load (ETL) was created to allow for daily data flow from the LIS to the QADB. This daily extract was then imported into Microsoft Power BI to create various dashboards and visualizations for data review and analysis for end users.
Results/Findings:
Results: The capability to easily manipulate, analyze and share the data has made a great impact on our data collection and quality management programs. Blood bank personnel are now able to retrieve accurate and timely statistics by using Microsoft Power BI, Excel pivot tables and analytical tools available in any computer with access to Microsoft 365. Interactive dashboards allow filtering and color-coding to enhance visualization. The workflow and examples of dashboards can be seen in Figure 1. Conclusions:
Conclusion: Connecting a LIS to a quality analytics database has the potential to enhance the safety, efficiency, and effectiveness of blood transfusions, ultimately leading to better patient care. Our future goals include extracting data from the secondary LIS used at our tertiary hospitals, WellSky, to the QADB platform to obtain system-wide transfusion statistics. Long term goals include extending the use of this platform to improve patient blood management through the analysis of both blood transfusion & clinical laboratory data.
Importance of research: This abstract shows how an automated and efficient data analytics platform can be created by integrating existing relational databases. Automation allows accurate, precise, and validated statistics leading to a superior quality management.
