Guest Column | November 4, 2015

Unlocking The Value Of Unstructured Patient Data

Unlocking The Value Of Unstructured Patient Data

By John Smithwick, CEO, RoundingWell

Twelve years and $5.6 billion dollars - these are the average benchmarks for a drug moving from development to market. Diagnosing patients correctly and documenting their health events appropriately depends on sponsors and investigators, gathering both quantitative data, which is typically structured, and qualitative data, which is typically unstructured. When comparing both types of data, it’s more challenging to manage and derive value from unstructured data.

Structured data – quantifiable, measureable data such as lab results, blood sugar levels, and cholesterol – can be put to use by clinicians at the point-of- care to aid their decision-making. Qualitative data – symptoms like pain, discomfort, and fatigue – are a different kind of challenge. Because this data is subjective to the patient, the doctor’s knowledge of this data depends on what the clinician asks and what the patient discloses.

If a patient’s encounter were like an academic exam, gathering structured data like vitals would be the fill-in-the-blank portion of the test. Gathering unstructured data would be like the essay portion. While it might seem like a simple exchange of niceties, these communications provide a lot of information to those running a clinical trial, like whether a patient is experiencing depression, or that s/he’s experiencing shortness of breath upon standing.

Correct diagnoses are dependent on managing both structured and unstructured data. Managing structured and unstructured data also greatly influences the successes a CRO, pharmaceutical organization, or sponsor is able to achieve during a drug approval process. As the costs to conduct clinical trials increases, ensuring clinical trials are efficient and effective becomes all the more important.

Managing structured data is usually handled well. It’s in the management of unstructured data where the problems arise. There are two primary problems with unstructured data:

  1. The first problem is process-related. Unstructured data is not gathered consistently or systematically. An investigator only knows about drug side effects if he or she asks the patient, or if the patient discloses the information. When symptoms are overlooked or patients withhold information, pharmaceutical companies are hindered in their ability to gauge the impact of the therapy being studied.
  2. The second problem is technical. Unstructured data is most often recorded on paper in free text or check-the-box fields. Data stored this way makes it very difficult for software systems to interpret, understand, and analyze.

The ballooning amount of data available is its own issue. In 2012, worldwide digital healthcare data was estimated to be equal to 500 petabytes. That’s an astounding number, and it’s only growing: the data is expected to reach 25,000 petabytes in 2020. It can be a daunting challenge for pharmaceutical organizations to unlock the value of unstructured data. And guess what? Industry consensus is that approximately 80 percent of all health care data is unstructured data.

So, what if technology could not only ensure patients were being tested correctly, but also automate the process? Cloud-based care management and patient engagement software are providing a new way for pharmaceutical organizations to unlock the value of unstructured data. How? In essence, by creating “structured symptoms” – gathering patient-reported symptoms and discretely capturing them in a way the data can be analyzed.

These platforms systematically assess patients for symptoms (and signs) that they might not get asked about directly by an investigator or doctor participating in a clinical trial, and that they might not self-disclose because they don't think it's important (or because they’ve simply forgotten). Care management software then stores patient symptom info in a structured way, allowing this previously unstructured data to be analyzed and made actionable.

For example, let’s consider the use of alerts which signal clinicians that a patient has experienced a change in his or her health status. Instead of discovering issues at a late stage, investigators are alerted early to leading indicators of a decline in a patient’s health status or an adverse effect of a drug.

The bottom line? Whether it’s gathered via care management software or patient-specific insights, all data should be structured and be ready for interpretation and analysis. Care management platforms can enhance test data provided from clinical trials and greatly improve the feedback loop between a pharmaceutical organization and patients.

About John Smithwick

John Smithwick is the CEO of RoundingWell.  He co-founded RoundingWell in 2011 following four years at Nashville's Healthways, where he led the design effort for its Web-based disease and lifestyle management product offerings. Prior to his work at Healthways, he worked in product management at Microsoft in Redmond, Wash. and in technology strategy consulting with Accenture in Boston, Mass. A graduate of the University of Richmond, he holds a master's of business administration from the University of Pennsylvania’s Wharton School of Business.