Evidence-supported decisions key to big data success

With big data among the biggest buzz words in healthcare, a new report from the Institute for Health Technology Transformation (iHT2) offers guidance for leaders trying to harness all that information and gain insight from it.

The report highlights some success stories, including a partnership between the University of Ontario and IBM to develop a monitoring platform for newborns that allowed the hospital to predict the onset of nosocomial infections 24 hours before symptoms appeared.

What's more, the report points out, Johns Hopkins School of Medicine found that data from Google Flu Trends allowed it to predict surges in flu-related emergency room visits week before warnings came from the Centers for Disease Control and Prevention.

Yet, the report's authors warn: "It must be emphasized that the healthcare industry remains well within its infancy of leveraging big data for business and clinical use. Although there have been some successes, many are unproven at the outcome level and much work remains …"

The report focuses on five types of data that can be mined and analyzed: web and social media; machine-to-machine data, such as from sensors, meters, and other devices; transaction data, such as healthcare claims and other billing records; biometric data, such as fingerprints, genetics, retinal scans, and including X-rays, blood pressure readings and other data; and human-generated data, including unstructured and semi-structured data such as electronic medical records, physicians' notes, email, and paper documents.

All of this information can provide clues to improving quality and efficiency, earlier disease detection and fraud detection. The report's authors say, however, that technology is not the solution itself--and add that any technology must be a seamless part of work flow.

"The potential for benefits is predicated on the assumption that the organization/providers are committed to evidence-supported decisions using analytic tools with available information. If that commitment has not been made, analytic tools provide little value. … Key to achieving [these goals] is knowing specifically what metrics are necessary to measure progress," the report's authors say.

Among the strategies the authors suggest for leveraging big data:

  • Implement a carefully structured framework for enterprise-wide data governance.
  • Foster competition and transparencyHealth care organizations are attaching monetary incentives for improving patient satisfaction and quality metrics. Dashboards are among the ways peers and colleagues can chart their standings.
  • Make dashboards and other tools simple and easy to use.
  • Close the quality loop – Transformation requires follow-through to create sustainable changes in the structure and processes of health care.

Information silos are another big hindrance to quality improvements, Russ Richmond, M.D., CEO of McKinsey Healthcare's Objective Health consulting firm, wrote recently in a post on The Health Care Blog. He urged organizations to "connect the dots." A 2011 McKinsey report found potential for the healthcare industry to realize $300 billion in annual value by leveraging big data, the report points out.

To learn more:
- download the report

Suggested Articles

Healthcare software company Phreesia closed its first day of trading as a public company Thursday about 40% above its set price.

The announcement comes on the heels of the Trump administration's effort aimed at kidney care that includes expanding access to in-home dialysis.

Technology company Philips has acquired Boston-based startup Medumo, the developer of patient navigation and engagement solutions.