Emanate Blogs

EDC vs. eSource?

Abstract glowing polygonal hand using tablet with medical cells on dark background with DNA. Digital clinical research concept.

EDC systems reduced the use of paper, improved accuracy, and sped up the time-to-market for drugs and medical devices. EDC has grown over the years into a mature, well-understood system, and it is the industry standard today.

Now, with the advent of higher bandwidth, improved wifi and Bluetooth technology, and all manner of electronic data collection methods, comes eSource.

What is eSource?

At the most basic, eSource is about capturing data at its origin point, and avoiding the need to transcribe or move it from one repository/system to another. eSource is data initially collected in an electronic format, such as:

  • eConsent
  • EHR
  • ePRO / eCOA
  • ECG
  • Central Labs
  • Sensors

Transcelerate splits eSource into four categories, each with its own benefits and drawbacks:

  1. Direct Data Capture (DCC) - Entry of clinical data by site staff into a mobile application or EDC system; can be seen as an evolution of EDC.
    Benefits: Data available at the time of collection, zero transcription errors, improved data quality, remote monitoring, designed for clinical research.
    Challenges: Need for site training, site infrastructure, and logistics. Note that direct data capture is only possible when there isn’t an EHR solution (typically phase 1 units). Sites must enter data into their EHR systems to receive payments from insurance payors, therefore there is almost always an EHR. To do away with the EHR just for research purposes would be to leave sites with another major problem to solve.
  2. EHR (electronic health records) - Collection and reuse of data for use in clinical research from site/patient electronic health record systems. Although widely used across sites, these are not designed for clinical research.
    Benefits: Data only entered once, zero transcription errors, saves time, sites can use their own systems (no learning curve), supports Real World Evidence studies.
    Challenges: EHR were never designed with research in mind, so the challenges are significant, including privacy laws (HIPPA, GDPR), hospital sales and mergers which result in multiple EHR systems (much harder to gather the data), and lack of direct access to EHR data; researchers would need some sort of data integration to bring the data in-house to a sponsor.
  3. Devices and apps - Collection and management of data coming directly from third party vendors such as wearables, sensors, patient reported outcomes (ePRO), and clinical outcome assessments(eCOA).
    Benefits: Rich sources of data (vast amounts of data), automates time consuming methods, patient-centric.
    Challenges: Infrastructure, cost and budget concerns (software and hardware), regulatory challenges and uncertainties (are all devices calibrated the same?). Data set is so rich, we don’t have the right tools for analysis. For example, if someone wears an accelerometer (fitbit) it tracks every movement a person makes which is terabytes of data, and traditional tools like SAS are inadequate to review the data. Finally, a variety of data is being collected (e.g., questionnaire data is handled differently than the ECG data from a Holter monitor).
  4. Non-CRF - The collection and transfer of data from external vendors such as safety labs, imaging, and randomization without entering into a Case Report Form.
    Benefits: very rich source of data, no transcription, accurate, saves time.
    Challenges: time consuming reconciliation, require manual processes (uploading the data), standardization.

Is eSource the End of EDC?

EDC and eSource are data collection methods. All of it has to be compiled into SDTM (Study Data Tabulation Model) to prepare it for submission. In the end, it doesn’t really matter where the data is coming from, as long as it is a valid source.

Sponsor have built dedicated EDC systems that they have perfected over time. They've put a lot of work into their EDC systems and they are reluctant to abandon them.

There is also limited interoperability between healthcare (e.g., EHR) and clinical research systems and applications (e.g., EDC, data warehouse). How do you get all the systems to play nice and work with each other? Especially when research just isn’t part of the health care landscape.

Instead of making this an either/or issue, the most effective approach is for sites, sponsors, and data mangers to stop thinking of EDC as a standalone system, and start thinking of it as one of many data sources. As we’ve talked about before, there is an “explosion” of data in clinical research right now, coming from sources that didn’t even exist 10 years ago.

When you talk to the people designing clinical trials, and the data managers putting together the databases for each new trial, you realize that eSource is more of an ideal. In clinical data management, EDC and eSource are all part of the data that comes in. It’s the careful application of clear standards at the end of the collection process that bring the data together and allow it to be prepared for submission.

While issues like interoperability are discussed at a higher level, those of us on the front lines of clinical trial design need to start shifting our perspective to viewing both EDC and eSource as valid components in our Data Management planning.

Leave a Reply

Your email address will not be published. Required fields are marked *