eSource & Data Integrity

20th September 2016 (Last Updated July 18th, 2018 10:17)

Tom Haag, Novartis, explains how the adoption of eSource for clinical trials offers significant opportunities to improve data integrity

eSource & Data Integrity

Let’s start with what we mean by eSource. eSource is the concept of capturing clinical study data electronically at the time of data generation. eSource Platform/Applications reduce the use of paper to capture source data during clinical visits and transcription error. eSource Direct Data Entry (DDE) models combine source documents and case report forms (CRFs) into one application, allowing for the real time collection of clinical trial information to sponsors and other oversight authorities, for example.

The modalities of eSource models that allow for direct data capture from other electronic data sources include, but are not limited to:

  • Electronic Data Capture w/Direct Data Entry (EDC w/DDE)
  • Electronic Informed Consent Form (eICF)
  • Digital Medical Equipment
  • Interactive Response Technology
  • Imaging and Lab Data
  • Electronic Health/Medical Record Systems (EHR/EMR)
  • (Wearable) Sensors connected to Mobile devices (Connected Sensors)
  • Electronic Clinical Outcomes Assessments (eCOA)

Many of the above modalities have existed in health care for some time, but the clinical trial landscape is evolving to include more of this data, as we move away from paper records, and the subsequent transcription and associated Quality Control mechanisms.

eSource adoption presents meaningful improvements to Data Integrity in clinical trials. While cost saving initiatives, organizational flux, and therapeutic focus areas differ across the clinical trial landscape, Data Integrity is the air we breathe as practitioners. For Health Authorities, Sponsors, Clinical Institutions, and Patients - Data Integrity is the overarching goal. It is the key driver in our quest for truth.

So if Data Integrity is Driver, POLDAT is the map. The POLDAT Domain of Change Model is well known to business analysts and validation practitioners. It sometimes used during the development of User Requirements Specification (URS) documents.

From a Sponsor perspective, significant challenges are presented in the known domains of change. It is useful to consider key challenges in the areas of Process, Organization and Location need to be considered, along with more obvious (but less troublesome) challenges related to Data, Application and Technology.

Process

Governance Domain of Change

A sponsor’s present processes (especially if EDC is already in use) for protocol development, site management and study conduct may not prohibit the use of modern technology (such as eSource) in Clinical Trials. Nevertheless, there is operational change when adopting a new data collection model, and more detailed guidance and instructions are necessary for clinical teams and operations. New SOPs and Working Practices might be necessary to better equip clinical teams to leverage digital technologies. New Clinical Trial Protocol and Data Management/Handling plans might be created to include the use of eSource. Training materials for all data stewards need to be clear and distributed.

Organization

People Domain of Change

The relevant organizations responsible for the setup and conduct of Clinical Trials may have a varied level of understanding for how they will engage in the changing landscape. This variation in people’s understanding presents a key challenge to the adoption of any new technology model. It is important that the relevant Business, Information Technology and Governance groups (including QA, Legal, and Regulatory) are well informed and in sync.

A sponsor can have the best, most qualified systems, supported by strong, documented processes, but without well trained and motivated people, the change will not be effective.

It is recommended to evangelize the eSource concept (or any other modernization effort) to interested trial teams and leverage their enthusiasm to foster “change champions” across organizations. It is also advised to work closely with Quality Assurance and Regulatory Affairs to engage and solicit feedback about these technologies from Health Authorities around the world.

Location

Regulatory Landscape / Domain of Change

Global, regional outreach is necessary to better understand how local culture, politics, and regulations can impact the adoption of new technologies in clinical trials.

Local guidance on eSource tends to cover a wide variety of topics, such as:

  • eSignatures (Modality& Storage)
  • Source Document Management
  • Local Language Translations
  • Protection of PII (Personally Identifiable Information) / Data Privacy
  • Monitoring and Site Management Operations

It is in these areas that communication, across organizations internal to a company, must carry through to external partners, such as Institutional Review Boards/Ethics Committees and Health Authorities.

Data, Application & Technology

The more obvious Domains of Change

While the organizational makeup, company size, therapeutic goals, and overall spending ability differ from sponsor to sponsor, the increasing role of technology is a constant. The number of data sources and the use of data standards need to be understood as a more complex chain of data custody. It is unusual for a single application to collect and process the data, and the move from transcription to transmission of source data is bringing an undeniable tidal wave of change to the clinical landscape.

The technological wave is constant, and just as Electronic Data Capture (EDC) changed the landscape as fax machines were (partially) phased out, the notion of Direct Data Entry (DDE) will change the landscape as well. The anticipated change from paper transcription to real time entry and review is one that can certainly improve data integrity. When source data is substantiated with electronic audit trails, the data is more attributable and contemporaneous.

When data standards and transfer agreements are clear between the multiple parties (organizations), the overall management of the data is more straightforward, and frankly mechanical.

The diagram below compares a transcription-based EDC trial model against a data transmission model:

The validation of applications in the clinical landscape is complex, but it can be achieved with approaches that factor in both overall platform technology qualifications. The holistic balance is supported by project level validation. The validation process must be the right size for the right project. Part of such a process is to document the requirements, while considering all of the domains of change.

The documentation and realization of Data, Application, and Technology are arguably very clear and is a straightforward matter. While technology changes are usually the most disruptive trigger element to change, the domains of Process, Organization, and Locality need to be considered. These are more nuanced and fluid concerns that can hamper technology and data quality improvements.

eSource & Data Integrity (ALCOA+)

eSource in this context has been used as an example of a technical change. It is a stand-in for any technical innovation that has the potential to improve the process landscape. Consider paper as a technology for a moment; one might consider the change from monks transcribing the bible to the invention of the printing press as a technological change. It follows that, just as EDC changed the technology and process landscape, so too will eSource.

The analysis of this change through the POLDAT lens (the domains of change) helps to examine in what ways the change can impact or benefit a specific goal. It is helpful to carefully consider these facets in a holistic fashion to meet the intended goal.

If we consider Data Integrity as the defining goal for the adoption of eSource, we can endeavor towards the application and validation of these benefits:

  • Reduced Transcription combined with Reduced Transcription Errors leads to Increased Data Quality.
  • Enhanced focus on Audit Trails / Relational Databases in conjunction with reduction of handwritten notes results in Increased Data Attribution and Legibility.
  • Reduced use of flat files/paper files in tandem with increased use of relational databases leads to increased Data Endurance and Consistency.
  • Direct Data Entry increases Data Availability and allows for Contemporaneous Data Review.
  • Reduced Source Data Verification plus Centralized Data Monitoring creates a model of increased Data Availability.

 

 

*Tom Haag is the Process & Learning Lead, Global Development Operations, Digital Development at Novartis

 

Suggested Further Reading:

  • International Conference on Harmonization-Good Clinical Practice (ICH-GCP), Sections E6 (R2), 2015 and E9, 1998.
  • European Medicines Agency (EMA): Reflection Paper on Expectations for Electronic Source Data and Data Transcribed to Electronic Data Collection Tools in Clinical Trials, 2010
  • Food and Drug Administration (FDA): Guidance for Industry: Electronic Source Data in Clinical Investigations, 2013
  • eClinical Forum Risk-Based Monitoring Taskforce: Risk Based Approaches, 2013
  • Society for Clinical Data Management (SCDM): eSource Implementation in Clinical Research: A Data Management Perspective, 2014
  • Mitchel, et al: eSource Records in Clinical Research, 2015