Article Discusses VA's "Big Data": Benefits and Challenges
Healthcare lags behind other industries in applying advanced tools and approaches to large, complex, rapidly changing data sets – a rapid evolution often termed "big data." VA has collected data on the 20 million Veterans it has served during the past three decades. This paper provides an overview of VA's evolving approach to "big data" and illustrates how advanced analytics support clinical activities, with particular emphasis on the Patient-Aligned Care Team (PACT) model of patient-centered primary care. It also shares some of the challenges, concerns, responses, and future plans that have emerged from these initiatives.
VA's information infrastructure – the Veterans Information Systems Technology Architecture (VistA) – became operational in 1985 and now incorporates the Computerized Patient Record System (CPRS), launched in 1997. Since 2004, CPRS/VistA has been used to document all routine clinical activities, retrieve results (e.g., test, procedures), as well as order tests, procedures, and consultations. More than 16 billion clinical entries have been captured since its inception. In 2006, VA began to build its Corporate Data Warehouse (CDW) – a repository for patient-level data aggregated from across VA's national healthcare system. The CDW does not include all local CPRS/VistA data; rather, it consolidates more than 60 domains of key clinical and operational data (e.g., demographics, laboratory results, medications dispensed from outpatient pharmacies). The CDW is still under construction, but is already one of the most formidable data aggregation efforts undertaken by any healthcare system. However, massive data storage alone does not define "big data," which also requires the ability to access and act upon data using a variety of advanced tools. Therefore, the CDW includes advanced data management, statistical, graphical, and business analysis software. More than 20,000 analysts, program managers, researchers, and others have accessed the CDW for a variety of initiatives. One common goal of these initiatives is to assist clinical and operational decision-makers by providing information and insights not feasible using local data alone. For example, the CDW is used to calculate risk for hospitalization and mortality for VA's entire primary care population, information that is provided regularly to primary care teams to help direct services to highest-risk patients. In another example, CDW was used to construct an index that includes measures of staffing, continuity, and access to assess the success of PACT implementation across hundreds of VA facilities.
In addition to many benefits, "big data" has also created organizational challenges. Managers and clinicians must help determine which reports and analyses are most critical to patient care, and which data elements need to be prioritized for standardization and validation at a national level. Another challenge for data governance is to appropriately balance the need to access data for purposes of clinical care, quality improvement, and research against VA's high standards for privacy and security.
VA has made substantial strides in creating an infrastructure to employ its immense data resources in advanced analytics to integrate those products into direct patient care and program evaluation. Continued success will require an understanding of both the strengths and limitations of systems that are created to utilize "big data."
Dr. Fihn is Director of VA's Office of Analytics and Business Intelligence (OABI); Dr. Francis is Director of OABI's Clinical Analytics and Reporting; and Dr. Clancy is Interim Under Secretary for Health.
Fihn S, Francis J, Clancy C, et al. Insights from Advanced Analytics at the Veterans Health Administration. Health Affairs. July 2014;33(7):1203-11.