Cynthia Senerchia, vice president, quality and regulatory compliance, Optum® Analytics, helps address some of the questions around eSource data in clinical investigations.
1. What is the exact definition of an eSource?
When you conduct clinical trials, you are gathering information that the FDA requires to come from a source of truth: How did you find out this information about this patient?
This information has to come from a source that derives from some aspect of a patient’s care. If that information comes from an electronic system, then it is considered an “eSource.”
Historically, the source of truth was most often stored in a paper medical record or other paper-based document, such as radiology reports or laboratory test reports that were reported from a separate system.
With the advent of electronic systems “talking to” other electronic systems, the source is often an electronic system.
Examples of various eSources that may be used in a clinical trial include:
- Clinical investigator(s) and study staff, directly performing data entry into an electronic data capture (EDC) system
- Clinical investigation subjects (for example, electronic patient-reported outcomes (PROs) recorded on an electronic interface, such as a tablet or website)
- Consulting services, such as an electronic test report from a specialist directly loaded into the electronic health record
- Medical devices, such as blood pressure measurement data captured electronically and linked to the medical record
- Medical staff entering data directly into an electronic medical record (EMR)
- Automated laboratory reporting systems, such as central labs automatically loading data into the EMR
2. What is the FDA guidance on eSource and EHR utilization in clinical trials?
In 2013, systems were evolving as physicians and hospitals moved over to EMR, so the FDA developed guidance around capturing source data in electronic form. The guidance was meant to be clear on how these records would be accepted in clinical trials.
See Guidance for Industry Electronic Source Data in Clinical Investigations.
This was part of an effort to streamline and modernize clinical investigations. In an effort to help the industry, the FDA guidance was released to ease concerns regarding the use of data with no paper source.
This was intended to “assist in ensuring the reliability, quality, integrity and traceability of data from electronic source to electronic regulatory submission.”
Prior to this effort, research sites were often required to maintain a separate “extraneous” paper record that could be used to validate information entered into an electronic data capture system for data that was not normally included in a medical record.
Additionally, in July 2018, the FDA released an additional Guidance on the Use of Electronic Health Record Data in Clinical Investigations.
The focus of the FDA Guidance is to “assist sponsors, clinical investigators, contract research organizations, institutional review boards (IRBs) and other interested parties on the use of electronic health record data in FDA-regulated clinical investigations.
This guidance is hand-in-glove with the eSource guidance but provides additional information for use of the EHR as the eSource for transfer of information to an EDC system.
3. What do these FDA Guidance documents address?
The FDA Guidance documents address electronic source data in clinical investigations used to fill the predefined fields in an electronic case report form (eCRF), according to the study.
These documents also offer industry guidance on information to include in the protocol that clearly states the intention to use eSource systems.
The Guidance documents include information on:
- Identifying the authorized source data originators
- Creating data element identifiers to facilitate examination of an audit trail
- Capturing source data in the eCRF using either manual or electronic methods
- Understanding responsibilities of clinical investigator(s) with respect to reviewing and retaining electronic data
- Using and describing computerized systems in clinical investigations
- Ability to utilize EHR clinical data warehouses that store and integrate data in addition to the EHR system
- How to include both structured and unstructured data
- Promoting the interoperable systems that do not require transcription from EMR to EDC to limit errors and capture source data at the point of care
- How the systems should be validated
- How to manage multiple different EMR systems for one study
In planning clinical trials, sponsors should include information in the protocol, data management plan or investigational plan that specifies several details, including:
- The intended use of computerized systems used during a clinical investigation
- A description of the security measures employed to protect the data
- A description or diagram of the electronic data flow
The Guidance on the Use of Electronic Health Record Data also includes several best practices, such as “ensuring policies and processes for the use of EHRs at the investigative site are in place” and “ensuring that sponsors and study monitors will have access to all relevant subject information.”
Study sponsors will need to review the suitability of each EMR to be used as an eSource prior to using a study site.
4. How are eSources for clinical trials continuing to evolve?
The FDA eSource Guidance was written to support the moving of EMR data into the EDC for a clinical trial. But this approach is laborious and duplicative — and has the potential to result in errors due to the need for entering data twice.
Consider the fact that you have to create a list of key data points that must be captured to demonstrate the safety and efficacy of an investigational agent. This means literally hundreds of variables need to be captured at each time point by a data capture vendor, who creates a different system to do this.
That custom EDC trial system often takes a long time to build, is costly and is not generally able to be easily integrated with EMR. So, a nurse coordinator has to re-enter data from the EMR into the EDC.
The goal at the Optum Digital Research Network is to streamline this approach. It bypasses EDC data entry altogether and maps data directly from the EMR to a view-only EDC system that is used for data review and electronic signature compliance.
5. What are the benefits of capturing source data directly from the EMR and transmitting it to the eCRF?
There are numerous benefits that emerge from utilizing EMRs and other eSources in investigations.
- Eliminate unnecessary duplication of data entry and eliminate transcription errors.
- Encourage entering source data in real time during a subject’s visit, which opens up numerous opportunities for a wider range of physicians and patients to participate in clinical trials without having to bring in additional “transcription” staff to manage data entry.
- Enable remote monitoring of data, as the source is readily visible in the mapped data.
- Promote real-time access for data review, which reduces the time for analysis and provides early responsiveness to protocol concerns.
- Facilitate the collection of more accurate and complete data.
6. How is the use of eSource reshaping the future of clinical trials?
There are a variety of ways eSource and collecting data directly from the EMR are working to reshape the future.
- Using eSource mapped to the EDC system enables a reduction in study start-up, as no additional training on the EDC system will be required.
- Source data will be maintained in the EMR, thus archiving of paper documents used in the study will no longer be required.
- The amount of time it takes to obtain and review data from a study will be reduced, as data will flow in near real time from the EMR to the data tables (available for view in the EDC only).
- Innovative interventions can be offered to more patients when they are getting treated in community hospitals and physician’s offices, thus expanding the pool of patients eligible for research.
- Expanding trial access to the point of care enables faster enrollment for more efficient trials.
- Real-world evidence from direct patient care situations are not influenced by clinical trials.
