Turning “big” data into “good data”: the foundation of actionable analytics

Understanding patients’ risk has become the first step in trying to find better, more cost efficient ways of providing care. But to get a full picture of patients’ health status, you need to analyze a lot of data.

But with all the talk about big data, it’s easy to lose sight of the need to make sure that “big data” is also “good data.” So, what makes data good?

Most health care organizations define good data in different ways. But there is a common thread among them: the need to gather and analyze information from all sources and from all sites of care. Assembling as much data as possible about patients is paramount to being able to know the quality and cost of care.

Once the data is gathered, however, it has to be turned into usable information. This means normalizing and validating it to make it suitable for analysis.Jeremy Orr, MD, MPH

Organizations have to pay attention to the sources of their data—one provider organization whose lab also served veterinarians inadvertently imported data for animals—and, they have to make sure it’s actionable. This could be clinical data that provides, for example, a pursuit list of high-risk patients who are likely to be admitted in the next six months. The real question to ask yourself is, how are you going to put it to best use once you have it?

One important use of actionable data is creating accurate registries for care management. Registries, which are a collection of health and demographic data, have historically been based on claims data. By combining claims data with clinical data, however, you begin to get a clearer picture of your patients.

Our research shows that nearly 20 percent of patients with clinical evidence of diabetes lack a coded diagnosis. That means one out of five patients aren’t showing up on EHR reports by diagnosis code, on problem lists or in registries. This is significant because, as our research also showed, patients whose conditions aren’t coded for reimbursement are relatively sicker, and use more acute care than ambulatory care.

In my next post, we’ll look at what comes next after gathering, normalizing and validating your data: using advanced analytics.

For more on using analytics to turn big data into good data, download “Getting from Big Data to Good Data: Creating a Foundation for Actionable Analytics.”

–Jeremy Orr, MD, MPH, Chief Medical Officer, Optum Analytics

Leave a Reply