Manage Learn to apply best practices and optimize your operations.

Clinical data analytics design strategies for healthcare providers

How next-generation clinical data analytics can and will improve care far beyond the capacity of current quality measures reporting systems.

This tip is part of a series exploring big data in healthcare. Each story in the series breaks down an aspect of analytics and where it fits into healthcare needs. This piece covers what questions a provider should ask themselves prior to implementing an analytics system.

Today, most analytics in the healthcare setting are driven by CMS Physician Quality Reporting System (PQRS) measures and value-based purchasing (VBP) initiatives. Though both PQRS and VBP are measures-based and not analytical in nature, healthcare providers can use more sophisticated analytics to improve their scores on these measures.

For example, predictive analytics can identify patients at risk for hospital readmission and provide foresight to providers to take additional steps in order to reduce those readmissions. Similarly, compliance analytics and clinical process analytics can help alleviate workflow bottlenecks en route to improving operational efficiency.

Not all analytics are the same, as care processes and clinical cognitive processes vary from specialty to specialty. However, when designing analytics in a care setting, you should consider the following 10 key questions:

  1. Purpose: What do you want to achieve?
  2. Usage: Where in the clinical care process (workflow) should data be analyzed?
  3. Who is the intended decision maker (the user of the analytics)?
  4. How will the analytics support clinical cognitive processes?
  5. What additional facts are needed to support analytics insight?
  6. Media: Who/how/which message will be conveyed/shared/sent to the subject?
  7. Register: How will you document analytics outcomes?
  8. Currency: How timely does the content fed into the analytics engine need to be?
  9. Decision time: How quickly must decisions be made on the analyzed data? (i.e. emergency rooms need immediate feedback, while chronic care plans evolve over time)
  10. Quality checks: What measurement framework on top of your clinical data warehouse will ensure the consistency and quality of analytics results?

To improve quality of care, each patient's candid input is very important. Social media is penetrating the healthcare domain, which has led to a new breed of analytics through comparative effectiveness research (CER). Use of big data from social networks and personal health content from smart devices, blended with actual clinical outcomes, is showing promise in understanding the geographical and socio-economic factors that can lead to improvements in patient care.

CER-based analytics provide a differential diagnosis that is similar to what IBM's Watson does, but as a cloud service availed to patients. The ability to integrate such cloud services into clinical processes depends on how open EHRs are to consume such services within their cognitive decision making processes.

Read the final tip in this series

Naeem Hashmi is chief research officer for Information Frameworks, as well as an expert in healthcare data analytics, information management and data exchange. He is an active member of HIMSS, CHIME and AMIA. Let us know what you think about the story; email [email protected] or contact @SearchHealthIT on Twitter.

Dig Deeper on Clinical data analytics software and systems