Written By: Eberechukwu Onukwugha, MS, PhD, Associate Professor in PHSR

Big data is a hot topic, but not a new one. In 2001, Douglas Laney characterized big data using the famous 3V’s:

  • Volume: Big data can include millions of patient records.
  • Variety: Big data can come from many different sources, such as administrative claims, electronic medical records (EMRs), smartphones, Fitbits, and more.
  • Velocity: Big data can be subject to frequent and not necessarily consistent updates.

More recently, an additional V was added to this characterization for value, recognizing the unique insights that are only possible with large volume, varied, and updated data.

Getting the Complete Picture

Big data is important because it provides researchers with information about more people, allowing us to develop findings and recommendations that can be generalized and applied to large populations. In addition, large volume data also can help researchers better investigate rare events.

Big data displays great variety, which allows researchers to use more measures to tell a richer story. For example, researchers can better describe patients’ health behavior by examining administrative claims data from an insurer that is supplemented with data from patients’ EMRs, Fitbits, and text messages about diet and exercise. This combined information provides a more complete picture of health behavior than researchers might have if they only examine the information available in the claims data. In addition, when researchers bring velocity into the mix, we get longitudinal health behavior data – data that tracks patients’ health behavior over a predetermined period of time – which allows us to better identify patients who may require additional support to achieve their long-term disease management goals.

Sharing Our Research and Expertise

Faculty members in the Department of Pharmaceutical Health Services Research (PHSR) at the School of Pharmacy are sharing our expertise in the analysis and implementation of big data through leadership service and research.

  • Peter Doshi, PhD, assistant professor in PHSR, joined the Reagan-Udall Foundation for the FDA Innovation and Medical Evidence Development and Surveillance Program Steering Committee, which uses data available through the FDA’s Sentinel Initiative to help improve drug safety.
  • Susan dosReis, BSPharm, PhD, associate professor in PHSR, used state-wide data provided by the Child Welfare and Behavioral Health Administrations to examine psychiatric medication treatment data and assess whether quality of care and practice patterns were consistent with federal mandates for oversight.

Increasing Our Impact in the Field

And lastly, I recently served as guest editor on a big data-themed issue for PharmacoEconomics — the benchmark journal for peer-reviewed, authoritative and practical articles on the application of pharmacoeconomics and quality-of-life assessment to optimize drug therapy and health outcomes — that was published in February 2016. This issue highlighted data, analytics, and perspectives on big data, particularly as related to health economics and outcomes research. I contributed an editorial as well as two peer-reviewed manuscripts to the issue:

  • A paper co-authored with Margret Bjarnadottir, PhD, assistant professor of management science and statistics at the University of Maryland Robert H. Smith School of Business, and Sana Malik, MS, graduate student at the University of Maryland, College Park, on the use of a visual analytics tool – EventFlow — to investigate, describe, and analyze prescription claims data.

Big data presents exciting, evolving opportunities for researchers, and we are just scratching the surface of its transformative potential. Ongoing big data initiatives across the public and private sector that are harnessing this not-so-new concept offer unprecedented opportunities to study health care delivery and improve patient care.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Copyright © 2016 University of Maryland School of Pharmacy