Planning for eSource Integration, Today and for the Future
Q&A FROM OUR WEBINAR
Richard Murg, Global Vice President, Business Development, Bioclinica Software Solutions
Sina Adibi, CEO and President, Adaptive Clinical Systems
Gain the Benefits of eSource Without the Hassle of a Complex Integration
In our recent webinar, Gain the Benefits of eSource Without the Hassle of a Complex Integration, Richard Murg, Global Vice President of Business Development for the Software Solutions Division of Bioclinica, and Sina Adibi, CEO and President of Adaptive Clinical Systems, discussed the data and workflow benefits of integrating participant data (from site-based devices and electronic health records) directly into eClinical systems, such as EDC and CTMS. Here, they answer questions from the webinar attendees. The topics span considerations for integration of existing eSource as well as predictions and advice for the future.
When we refer to frictionless data transfer, how would existing eSources (especially devices) be optimized for FHIR?
First, a quick background: because electronic health-related data originate from and are stored in different sources (e.g., electronic medical records [EMRs], web-based forms, wearables, medical devices, mobile devices), compiling all the data in a single location is challenged by a lack of standardized formatting and communication. Therefore, interoperability is a key consideration when integrating from multiple sources.
The frictionless data transfer movement focuses on making it easier to acquire, share and reuse data by removing the “friction” caused by the large amount of data cleaning and mapping historically needed. Fast Healthcare Interoperability Resources (FHIR) is a standard developed by HL7 for exchanging healthcare information electronically between applications.
Back to the question: let’s look at it from the different sources.
First, a device as eSource: devices are, by nature, message-based, which makes them naturally compatible with FHIR. However, not all tool vendors offer FHIR, although most are trending in that direction. Furthermore, FHIR is updated continually (as with any standard), which creates some challenges in knowing which version of FHIR is supported by a specific device. It depends on the timing of that device release and how often it’s updated. So if the device specification states that it is FHIR- compliant, don’t be complacent. Expect you might need to do some work to avoid divergence in data flow.
Second, EMR as eSource: Almost all EMR vendors now support FHIR, but the same caveats exist as for devices as eSource.
Third, iPad-based data entry as eSource: data generated from iPads are very structured, and there are other prevailing formatting standards in place. Although these standards do not cover every type of data collection, they are very well established, which would make it disruptive to switch to FHIR.
If a source is added/corrected, will the eSource show any audit trail?
This is an important consideration to ensure continued compliance and avoid end-of-study surprises about incorrect data capture. Whether audit trails are captured is partially dependent on the electronic data source (EDC) system being used to collect data from the eSource. More advanced EDC systems will provide an audit trail, and you’ll want to make sure that there are meticulous logs about chain of custody (where data comes from, when it was sent, etc.). In addition, it is important to make sure that the data flow is validated.
Comprehensive planning plays a key role here, and consulting with companies that specialize in these systems will provide insight into what you should consider.
Will the sponsor provide eSource environment to sites or will sites develop their own environment? Which would be more common?
Sponsors have a vested interest in ensuring data quality and efficient workflows for faster, less expensive and more reliable studies. eSource achieves this by eliminating steps, so it is becoming more common for sponsors to require eSource in studies, as they do currently for EDC.
At the same time, some site management organizations are making their sites more attractive to sponsors by making significant headway in adopting technologies that will help data capture. However, sponsors need to make sure that data capture methods adopted by sites are compatible and interoperable with their EDC.
How do you see eSource changing in the next 1-3 years?
eSource will become more ubiquitous over the next few years, capturing reliable endpoint data at the individual and device levels, and becoming an integral component of clinical trials. Adoption has been accelerated with COVID-19 and will continue to be supported by the demographic shift. As technology natives enter the workforce as clinical research associates (CRAs) and data managers, their experiences have been defined by working with technology and will expect the same in their work environment. At the same time, standards will evolve, improving interoperability, which is going to be particularly important as we deal with more data fragmentation from the increasing number of sensors and data sources.
This increased eSource use will result in greater insights from more reliable, less expensive data, particularly as we remove the need for multiple people (clinical research associates, investigators, data management team) to be involved in manual data-related processes. It will reach a point where study teams do not have the need to chase data anomalies anymore; when there is an anomaly, it is truly an outlier. Our confidence in the data will significantly increase as we cut out the interim steps and get data direct from the source.
You know, 15 years ago, we were justifying the cost of EDC over paper, and now we’re doing the same with eSource. Incorporation of eSource has been expensive, but the initial cost is outweighed by the long-term benefits.
With so many devices and new data sources, how can my organization ensure that we can incorporate them as they come along?
Choosing vendors who have participated in a clinical trial will at least shorten the learning curve. However, because most device vendors likely have not been involved in a clinical trial before, you can make sure they are able to easily deliver the data, ideally through APIs in real-time. Or, if some intermediate evaluation must be performed (as in the case of cardiovascular studies), ensure you have access to both the raw data (e.g., wave file) as well as the reads (intermediate evaluation results).
Because interoperability is so critical, choose a partner with the technology, experience and capability to quickly interconnect new devices.
In addition, logistics are often overlooked. Make sure that the vendor is capable of fulfilling delivery of units you need at the various locations, including if you need to ship individual units to patient addresses. Arrangements will also need to be made to return, clean, and reset devices at trial end.
What does an organization need to do to get ahead of the curve?
Remember that old, traditional approaches to data capture are not scalable. Review and revise your data flow architecture. Similarly, evaluate your existing portfolio of eClinical tools and consider replacing those that have either fallen behind or are not interoperable with other third-party tools. Even small steps taken with each study are moving your organization forward.
Seriously investigate or otherwise adopt virtual trial practices. Even a Phase I study that requires bedside visits will benefit from being able to find patients in beds who are not co-located.
Consider your vendors as your partners. Bring your initiative to them and work collaboratively to find a solution before your study design is finalized. Vendors are working with multiple organizations and staying abreast of the latest developments in their respective area, making it likely that they will have a solution that you haven’t considered.
Considering incorporating or expanding eSource capabilities in your trials? Learn more about the considerations for the devices themselves as well as EDC systems to support them.
About Adaptive Clinical