Real world data collected in a functioning health care setting instead of a controlled clinical environment can provide opportunities for new and deeper insights across life science and health care organizations. However, managing, analyzing and extracting actionable information from the varied available sources can present unique challenges.

The sheer size of these observational and administrative sets of data alone can cause issues. Other factors include the variety of sources, data quality issues and the need to consider temporal relationships between the longitudinal data elements. Both human and computing platforms limit the available resources in many companies.  Clearly, a lack of internal resources limits the number of studies that can be performed.  To alleviate this issue, organizations can use an outsourcing model for research and analysis. However, this approach limits the ability to explore and analyze data freely. Beyond resource constraints, an appropriate computing environment can improve efficiency improvements for process standardization.

A number of regulatory issues also affect the collection and use of real world data. Of utmost importance is patient privacy.  However, unique patient identifiers are critical for the most accurate analysis of treatments in the real world. Over time, the regulations on the collection, use, sharing and sale of real world data will continue to evolve. With these changes, life will become easier for researchers while others will introduce additional hurdles.

Consider these three essential capabilities for any real world data project:

1. Data management capabilities are crucial for the utility of these sources. For instance, repeatable data import methods allow for the integration of periodic refreshes of data. Further, tools that provide monitoring of data transformations and other processes allow governance and resource management.
2. Modeling data from different vendors will lead to the standardization of a same or similar format. This standardization improves efficiency and makes it easier to reuse code and analytics across different vendors’ data.
3. Analytic methods and visualization tools designed to support these “big data” sources ease the process of analyzing and presenting results. The ability to quickly generate summaries across the full population or for selected subgroups of patients, and to interactively perform data exploration activities, provides insight to guide more sophisticated analyses and modeling.

Taken together, these capabilities can help organizations derive a variety of insights from their real-world data sources. These benefits decisions that improve patient’s lives and benefit the health professionals who care for them.

Join Laurie Rose and Robert Collins at PhUSE 2016 in Barcelona on October 10th  where SAS will discusses how organizations can address these issue in a presentation entitled, “The Playbook: From Real World Data to Real Word Intelligence” (Paper RW01).