The Real Impact of Real-World Data on Clinical Research

The line between real-world data (RWD) and real-world evidence (RWE) is often blurred. RWE helps draw the clinical insights and conclusions from the RWD.

Real-world data (RWD) points are constantly being generated in our day-to-day lives, particularly as related to healthcare. A trio of RWD subject matter experts at Norstella, Citeline’s parent company, recently sat down to discuss RWD and the role it plays in clinical research.

The line between real-world data and real-world evidence (RWE) is often blurred. Dr. Shrinal Patel, VP of Medical Science, acknowledged that the two “go hand in hand with each other. So it is hard for us to differentiate what makes one different than the other. When I think of real-world data … it's the raw data that you're looking at. It's the lab values that you're pulling in.”

She explained that unstructured data — such as health records, claims data, and lab data — are in their rawest format. “I think a common misconception is that real-world data is clean, that it's complete” and can be used right away, Patel said. “Real-world data is in its raw form.” She said Norstella works to eliminate much of the noise and clean up the data via mapping. It's the RWE, she said, that follows through on the RWD, bringing the clinical insights or conclusions that we can take from that data itself.

“I like to think of real-world data as a series of data points that are being generated and collected throughout healthcare interactions that are happening in our day-to-day life,” said Ilan Behm, VP of RWD Engagement. “The real question is what are these interactions and how is the data being collected?”

Behm said these interactions could occur at the doctor’s office, at an urgent care facility, in the emergency room, during a hospital stay, at the pharmacy, at a lab for routine bloodwork. And with the advent of wearables, your data is likely being collected while exercising at home or the gym.

The next question, Behm posed, is how are the data being captured and what is collected? At the doctor's office, for instance, certain information will be put in your medical record or your chart. This could include your vitals: height, weight, blood pressure. Perhaps your gender, your date of birth, will be part of that as well, he added.

“If you wear a device like an Apple Watch,” Behm said, “it could be monitoring your heart rate, your oxygen, recording these every few minutes. If you wear a continuous glucose monitor, it could be measuring your blood sugar every few minutes.”

These data points tell us very specifically about that patient, “but it's not telling us too much about the broader spectrum of care. ... But when we start to aggregate this information … we can start to piece together different types of trends that can be used to further our mission of increasing access for patients.

“When we're using real-world data, it's really about getting the right patient on the right therapy and the right care at the right point in their healthcare journey.” One way that this data can be used, Behm said, is making sure the healthcare practitioners know they have patients eligible for a clinical trial or for a therapy that could help them.

Citeline’s Sitetrove and Trialtrove data give visibility into clinical trial performance. Alison Perry, Senior Director of RWD Engagement, explained that this data can be linked with patient-level data to understand patients moving across clinical trials. A good example, she said, is electronic medical records (EMR), which provide clinical trial information for treatments for specific conditions seen in the data. This enables longitudinal understanding of where patients exist within a larger clinical trial or competitor trials.

Behm noted that Citeline’s Trialtrove datasets easily and readily link to Norstella’s RWD dataset. NorstellaLinQ combines real-world data and Norstella’s proprietary intelligence from Citeline and its sister brands: Evaluate, MMIT, Panalgo, and The Dedham Group. He added that the differentiated insights and data points available in Trialtrove are “light years” beyond what's available in ClinicalTrials.gov. “We have been working with the industry,” he said, “to ensure that eligible patients are making their way to trials.”

Perry expanded on Norstella’s dataset, which brings together data from different health systems, hospitals, payers, health plans, and payment clearinghouses. She continued, saying that all the datasets are in-house and tokenized, enabling Norstella to link the same patient across different sources and different systems over time.

Perry was quick to add: “We do this in a way that allows privacy preservation, but then also allows us to see the patient's diagnosis, their treatments, their outcomes over time and across healthcare systems.” The way Norstella data are de-identified, means “we can't ever re-identify who the patient is. We don't have the true information about patient information that would allow you to figure out this is a patient with their name and identifying information. .... Patients are not able to be contacted, and I think this is something that is really important.” She reiterated that the data are for population-, HCP- or provider-level insights, not for patient identification purposes.

She said the data can be sliced by disease stage treatment pathway, lab values, tests, outcomes, and more. “So it goes beyond just surface-level analytics and allows for much deeper, richer information.”

Structured datasets, like a standardized diagnosis, are one thing. Referring to unstructured data as “super rich,” Perry said that because of this the data are difficult to analyze. That’s where artificial intelligence (AI) comes in. She said AI can uncover the why and how patients are being treated and doctors diagnosing.

Of course, RWD is not without its challenges. One challenge, Patel noted, is longitudinally tracking a patient across various care settings. While multiple data sources are a plus, she said, “It's hard to take all of that information and still say that it's the complete history or the complete journey that the patient has.”

Perry agreed, emphasizing the importance of connecting the dots in terms of what RWD is available, where it's going, and how it can be used. ”We all have this shared mission of smoothing the path to treatments for patients. … I personally think there's a lot of opportunity for our real-world data to add value in ways that might not be obvious and that might not be on the radar right now. I think we've really only been scratching the surface.

“I really think we're at the forefront of the industry,” said Perry. “We never just throw data over the fence.”

For additional resources, visit the RWD Hub on Citeline.com