The Rough Guide to Electronic Data Capture

Electronic Data Capture (EDC), three little words that can divide the clinical research community like few others. To some, EDC is a panacea for all the troubles and rising costs of the clinical development process, whilst to others it spells broken promises, and forms the butt of many less than flattering anecdotes.

But what is the reality, and where is the border between the promise of providers and the expectation of end users?

This article is not designed to provide definitive answers to the great questions of whether EDC is the future of clinical trials, quite how much can it reduce costs and query rates, or what is the best way to implement it, but to blow away some of the fog of misconception and misunderstanding that currently surrounds the EDC debate, and answer some of the more commonly asked questions.

What do we mean by EDC?
The results of a 2005 survey by CDISC (the Clinical Data Interchange Standards Consortium) suggest that EDC was used in over 40% of trials in 2004, and that it’s use is nearly doubling every two years.
Most people reading this will be scratching their heads at this point and thinking that there is no way that 40% of their trials use EDC. And they would be right; when the underlying data from the survey is examined it turns out for this survey, EDC included electronic patient diaries, IVRS and fax/scan systems as well as Electronic Data Capture of CRF pages. Not surprisingly the largest portion of the 40% is made up of eDiaries and IVRS, rather than CRF related EDC.

Whilst for the rest of this article the term EDC will only refer to non-scan/fax electronic methods for CRF data collection, it is worth noting that the viable scope of an eClinical Trial is growing fast. This involves integration of data from a large number of disparate systems with the resulting elimination of unnecessary duplications of effort. These systems include:
 Electronic Medical Records - direct import of demographic, medical history and concomitant medication information
 IVRS - screening / randomisation, drug shipments and reconciliation and patient reported outcomes
 CTMS - regulatory documentation, enrolment information, site visits and investigator details and payments
 CDMS - back end data cleaning and central repository for study data
 EDC - collection of data at site
 Electronic patient diaries - collection of patient reported outcomes
 External electronic data sources - e.g. ambulatory blood pressure monitor, central laboratory results
 SAE reporting/pharmacovigilance systems
 Electronic regulatory submission software - Clinical Trial Applications and Marketing Authorisation Applications
As a result of the early stage of computerisation of medical records, and the lack of common data standards between the pharmaceutical and medical fields, the vision of complete integration of all these systems is still at least 10 years away. However, integration of all systems other than electronic medical records is currently possible, and is already happening.
What is EDC All About?
Over the past few years EDC has been the source of plenty of buzz at conferences and in various publications. However, with still only 10 - 15% of studies actually using this technology, I thought it would be helpful to give a quick “EDC 101” crash course - please accept my apologies if you already know this bit.
In a traditional paper-based study, trial data is written on costly three-part NCR paper CRFs at investigator sites. The CRFs are periodically reviewed by CRAs in an attempt to make sure the data is valid and complete.
Once this has been done, two copies of the CRF are sent to the Sponsor or CRO’s data management department where the data is entered into a database not once, but twice (to try to eradicate transcription errors). Data managers then review the data in the database and identify any discrepancies in the data that are resolved using a paper Discrepancy Clarification Form (DCF) sent to the investigator site via the CRAs.

Eventually, after much back and forth, the data in the database is deemed to be “clean”, at which point the database is locked. This unfortunately is not the end of the story as both the investigator and the Sponsor company are required to keep copies of the CRFs for at least the next 15 years, creating a massive storage problem.

Whilst this has worked very successfully for many years, increasing pressure on Sponsor companies to get drugs to the market faster and cheaper have brought inefficiencies in paper trials to the fore - for example the cost of printing and distributing CRFs, the huge amount of time CRAs spend resolving simple data entry errors (such as incomplete dates), the lack of real time study metric information available to project managers, and the duplication of effort between departments with regards to SAE recording and reporting.
EDC is nothing more than the replacement of the entry of study data into paper CRFs, with entry into electronic ones that have some form of in-built real-time data validation checks.
Entered data is written directly to the central Clinical Database, and is then immediately available to all study staff.

There are numerous advantages of EDC to many different groups involved in the clinical trials process, however, they can all be summarised with another three work phrase  - “cleaner data faster”.

Who are the EDC stakeholders?
Ask this question to just about any group of clinical trial professionals and the answer comes back “Data Management and IT”. This answer is one of the key reasons behind the poor rate of EDC adoption to date, and the “failure” of many EDC pilots.
Whilst on the surface, EDC would appear to have the most impact on Data Management and IT, the reality is that almost all groups within the trials process can directly benefit from EDC, and therefore need to be involved in, and feel ownership of the selection and implementation of an EDC solution.
 Project Managers - Access to real time project metrics enables PMs to more accurately follow study progress, and identify, and where necessary address possible deviations from the project timelines before they become major issues e.g. poor subject recruitment, high data query rates, long query resolution times etc.
 Clinical Monitoring Staff - The use of real-time data validation checks in the eCRFs, combined with the ability to remotely review eCRFs and send data queries to sites significantly reduces the time spent during site visits cleaning up minor (and less minor) data errors e.g. incomplete dates or “pregnant men”. This enables CRAs to use their time at site more efficiently, and may enable a reduced number of site visits or a re-organisation of site visits to every 10 subjects having completed a visit rather than every 6 - 8 weeks as at present
 Investigator Sites - whilst not involved in the selection of an EDC solution, Investigator Sites benefit from EDC by the reduction in the number of queries from CRAs or Data Management they need to resolve during the study. In addition they only need to store/archive a single CD of their study data rather than the current half a rainforest of paper files
 Data Management - In addition to the complete eradication of double data entry, the real-time validation checks when data is entered at the site significantly reduce the number of queries for Data Management to resolve. The time to resolve each query may also be reduced as DMs will not have to decipher handwritten CRFs / DCFs, but will at worst have to read a few typos
 Drug Safety / Pharmacovigilance - The combination of automatic notification of SAEs by the EDC system, and the importation of SAE data from the EDC system into the pharmacovigilance system increases the efficiency of response to SAEs whilst at the same time eliminating unnecessary SAE data entry work
 IT / Application Support and Training - Whilst these groups do no benefit from the use of EDC, they are essential to the successful rollout of EDC both within a sponsor or CRO, and even more importantly, at the investigator sites - leave these groups out at your peril.

Why aren’t we all using EDC?
Given such an impressive list of efficiencies and benefits that EDC should be able to bring to a clinical trial, you might ask why EDC isn’t being used for significantly more trials. Well, there are a whole host of reasons, almost all of which relate to basic human traits:
 The need to please - Unrealistic promises by EDC vendors have created wildly unrealistic expectations amongst sponsors or CROs that have no chance of being fulfilled. Fortunately this is changing as vendors realise that overselling the capabilities of their EDC.
 Systems may have short terms benefits to them, but in the long term raises doubt and distrust in the minds of users, and hence a lack of adoption
 Resistance to change - Whilst the shortcomings of paper-based studies are well recognised, the “If it ain’t broke, don’t fix it” mentality means there is significant resistance to moving away from tried and tested systems and processes
 Avoidance - As outlined above, many different functions in a clinical trial are affected by EDC, however, unless all these groups are prepared to buy into, and take ownership for making EDC work, rather than saying “It is a Data Management and IT issue - nothing to do with me”, then unsatisfactory results are guaranteed
 Indecision - If a company is going to adopt, or even pilot EDC, they need to do so fully rather in a half-hearted manner. Several well publicised EDC pilot studies have been deemed failures because of a lack of commitment to implement process change alongside the introduction of the EDC technology, and as the old saying goes “new technology + old process = expensive old process”.
 Fear - With so many EDC vendors in the marketplace, and the inevitability of future company consolidations and closures, many sponsors and CROs are afraid of committing to a particular EDC vendor.

Where do we go from here?
There is no simple answer to this, however, what is certain is the use of EDC in clinical trials is on the rise - maybe not at the rate suggested by the CDISC survey mentioned earlier, but certainly on the rise. Therefore, we all need to accept and embrace its use and work with it to deliver the benefits it can bring.
Nobody should kid themselves that this we will all become EDC fanatics overnight, however, a good start, and the reason behind this article, is to sweep away our misconceptions and fears about EDC, and help us make decisions based on a fuller understanding of the reality.

Andrew Salomon B Pharm
Account Director - i3 Statprobe
www.andrew.salomon@i3statprobe.com
www.i3statprobe.com