eSource in Clinical Trials: A brief overview
According to the FDA, “initial documentation of data in a clinical study is considered Source documentation or Source data*”. This is a fairly straight-forward concept. The source data and documentation is the original record of a data element, and integrity is maintained by using an audit trail to track any changes and modifications.
Now, introduce electronic source “eSource”. For all intents and purposes, the definition is the same, except the original record is captured electronically. This excludes source data that was captured on paper and transcribed into an electronic database. eSource is true to its name and the source data element itself must be electronic; and must be followed with a detailed audit trail as well.
On the face, the benefits are obvious. The FDA promotes eSource in clinical studies because it will help to:
- • “eliminate unnecessary duplication of data,
- • reduce the opportunity for transcription errors,
- • promote the real-time entry of electronic source data during subject visits, and
- • ensure the accuracy and completeness of data (e.g., through the use of electronic prompts for missing or inconsistent data).*”
Example 1: A patient receives an infusion treatment for a study drug. During her visit, all data (vital signs, concomitant medications, infusion data, etc) is captured electronically. That night the patient has a serious adverse event which results in a trip to the emergency room, and the study investigator is contacted. The beauty of having the patient’s most recent visit captured electronically is that the investigator has immediate access to the day’s treatment details (along with previous treatments) directly at his fingertips. If data relating to the infusion treatment was captured on paper, the investigator, at best, might have access to a chart at the infusion clinic. When time is critical, paper is not the answer, and can potentially jeopardize the patient’s safety. Using eSource allows the data to be accessed and assessed in real time. Any subsequent events relating to the emergency room visit can too be captured electronically and automatically disseminated to key team members.
Example 2: A patient reports symptoms on an electronic diary that are undesirable. The study staff is automatically notified via email. If this were collected on paper, the patient would report the symptom, and would likely not get any resolution until a site visit sometime in the future. Their recollection of the undesirable symptom or event will not be as clear as it was when originally reported. Using eSource allows the study staff to determine the severity in real-time and follow-up as needed. Any subsequent contact with the patient regarding the undesirable symptom can too be captured electronically and automatically shared with key team members.
From a data manager’s point of view, eSource can solve a multitude of problems, most notably consistency of data (pre-defined medication lists for example) and access to data as described above. The key is to avoid any potential gotchas relating to the integration of data with an existing study database (or EDC system). Every study team desires consistency and is not necessarily gung-ho about introducing another system/website/database into a study. This is one area where things can get complicated. If it is the study teams’ wish to use eSource, there must be a backend designed to integrate the eSource data into the existing database or EDC system. Audit trails must remain intact, and query processes must be clearly defined. What good is an eSource system if team members cannot use it seamlessly with study data reported into other system(s)?
The obvious benefit to any study using eSource is quality and access to data. The access to ongoing, real-time data can speed the overall process an average of 6 months. Since drug approval is what sponsors are after, this cannot be ignored. Equally as interesting is the substantial cost savings. If a true eSource solution is used, all transcription and source data verification costs are eliminated. This can cost upwards of $15 per page (CRF and non-CRF data). You do the math; the savings are significant. Just as noteworthy is the fact that data can be monitored remotely and queries fall, on average, by 70% (headaches and frustration decrease at the same rate!).
Although the FDA’s guidance on eSource is still in its “draft” stage, the implication is clear. The FDA appears to support the collection of source data electronically as a means to promote accuracy, integrity, traceability, and completeness of data. If your team is still tiptoeing around this powerful movement in clinical trials, consider starting with an eSource solution that collects primary endpoints electronically and integrates directly with a database that your team is confident using. Work out the kinks and transition to using eSource to capture all data elements once you’ve gotten your feet wet.
* U.S. Department of Health and Human Services, Food and Drug Administration (December 2010). Guidance for Industry: Electronic Source Documentation in Clinical Investigations. Available here.