Taking the Pain out of EDC Deployment
By John Russell and Paula McHale
Produced by Cambridge Healthtech
Media Group Custom Publishing
EDC has long promised to cut clinical trial costs, reduce errors, shorten trial duration, and enable emerging methodologies such as adaptive trials. Despite this promise, adoption has remained slow outside large CROs and sponsors. The truth is implementing EDC was expensive and challenging. Today, it still isn’t a trivial undertaking, but advances in the technology, the development of hybrid EDC systems able to handle paper-based processes, and greater use of software-as-a-service (SaaS) models are removing technical and financial barriers that prevented small and many mid-size organizations from reaping EDC’s substantial benefits. Read to learn more.
PAREXEL Informatics™ - Taking the Pain out of EDC Deployment
Worldwide, more than 100,000 clinical trials are ongoing. A single trial can last a few months or several years and may involve a handful of patients or many thousands. In the U.S. alone it’s conservatively estimated that there are about 1M patients per year participating in trials at roughly 7,000 sites for a total annual cost of around $35B, and the tab is growing 8-9% per year.1 The goal of biopharmaceutical companies is to successfully navigate the final leg of the decade long R&D marathon to deliver safe, efficacious drugs, diagnostics, and medical devices. Many technologies have been developed to improve trial effectiveness and efficiency. Among the most familiar are electronic data capture (EDC), electronic patient-reported outcomes (ePRO), clinical trial management systems (CTMS), medical imaging, and randomization and trial supply management (RTSM).
From early phase trials involving micro-dosing to pivotal Phase III studies upon which NDA approvals depend, the new technologies can help streamline trials, manage costs better, and allow investigators to pursue the complicated adaptive trial designs necessary for the personalized medicine approaches taking hold today.
EDC has received the most attention and perhaps deservedly because it interfaces with virtually every other trial system and promises the greatest rewards. Here are just a few of its benefits:
• Reduced Data Cleaning. By streamlining the design of the CRFs and utilizing built-in controls such as radio buttons and check boxes, the initial entry of data is cleaner which ultimately reduces the volume of work for data management.
• Early Data Review. It is possible to quickly view individual, site and cross-site patient data at any time throughout the trial as well as easily extract data for interim reviews such as Data Safety Monitoring Boards.
• Trial-wide Data Visibility. Anyone with the need – Sponsors, Program Directors, Clinical Research Associates (CRAs), Data Managers and Medical Monitors can be set up to easily view study data and site progress which assists in identifying high and low performing sites.
• Faster Database Lock. With careful planning and consistent cleaning of data through review and query management, it is possible to reduce the time from last-patient-in to database lock by freezing CRFs as they are cleaned
• Improved Trial Flexibility. Early data access and the ability to easily segment and track data make it much easier to modify trials (cohorts, protocols, etc.) such as in adaptive trials.
• Optimized Resource Planning. Critical information on patient recruitment, retention, and materials consumption is readily available for materials supply planning and logistics.
The potential gains from EDC are well understood by the clinical trial community and it’s estimated that just over 50 percent of trials today have an EDC component. But given how long EDC has been available, the real surprise is that the number is so low. Even the 50 percent statistic is likely skewed by the dozen or so very large sponsors who use EDC on virtually all of their trials, notes Paula McHale, Senior Director of Product Management, Data Management Solutions at clinical trial technology solutions and services provider, PAREXEL Informatics.
“At conferences you often hear huge percentages, 90 percent or higher, cited for studies using EDC, but it’s actually a common misconception. There are many companies who still aren’t using EDC at all because they just can’t afford to,” says McHale, an 18-year PAREXEL veteran and long-time EDC expert.
The practical reality is that implementing an EDC solution hasn’t been cheap or necessarily easy. While the technology has been maturing, EDC adoption has mostly occurred among large organizations with plentiful staff and financial resources. Consider some of the challenges to EDC adoption:
• IT Infrastructure & User Support. All studies require some base of computer equipment (servers, datacom, storage, etc.) and some measure of IT redundancy and IT support. A ‘help desk’ function for users with questions is also needed.
• Diverse Site Requirements. Sites in a single trial often vary greatly in assets and capabilities (IT resources, internet access, complexity of work being done, and personnel capability) and require different levels of IT provisioning and support.
• Training. Everyone using the EDC system needs role appropriate training; this varies by user type and can be time-consuming and often challenging to deliver effectively due to logistical issues and scheduling.
• Systems Integration. Input from systems such as external labs and any paper-based processes must be accommodated. Integration with other trial systems such as a CDMS (clinical data management system), CTMS, ePRO or an RTSM solution may also be necessary.
• Regulatory Compliance. The system must meet all relevant regulatory agency requirements.
Once outside the rarified atmosphere of big pharma and large CROs, the range of needs and capabilities among organizations conducting clinical trials is wide and varied. Many challenges to achieve EDC adoption can be intimidating. This particular community needs flexibility in technology and services, well-designed support for paper-based processes, and site friendliness. The latter point is often neglected, notes McHale. Sites bear most of the pain in carrying out trials, and making a concerted effort to identify and ease site pain points can yield enormous gains not only in trial performance but also in site receptivity towards EDC and other clinical trial technology tools.
STARTING POINT: EDC REQUIRES FLEXIBILITY
Given the diversity of needs among the user community, it’s important for vendors to be able to tailor their EDC solutions to the individual user organization’s capability. Some customers are able to simply license the software and do everything themselves, for example, the deployment, hosting and study design; others require some combination of software, hosting, and services from the vendor.
For users who have little knowledge about EDC, the system can be deployed as a full ASP solution in which the vendor works with the client from start to finish. A typical ASP model includes a study design team that creates and manages the study, and a dedicated project manager to provide support throughout the duration.
Another approach gaining traction in the industry is the software-as-a-service (SaaS) model, which simplifies many technology and financial management issues, can speed trial ramp-up, and substantially cut costs. Many companies are starting to offer such models with various adaptations of web-based study design tools. These tools begin to introduce a SaaS environment in which clients can independently activate a study design, build their own studies, and deploy online but leave the hosting and support of the trial to the vendor.
The SaaS approach will help drive the next wave of EDC adoption. It’s difficult for many CROs to take an ASP approach because they often have to commit to the service and licensing with the EDC vendor before receiving a signed contract, and more importantly funds, from the sponsor. Money may not be readily available within some smaller CROs, and the risk is high without a signed sponsor contract prior to EDC deployment.
With SaaS-based EDC models, companies can pay on a monthly basis without committing to a lump sum up front. From the technology provider perspective, it is much easier to maintain control over the technology and to release new features faster through a SaaS offering.
The other important SaaS advantage for CROs is that even the smallest CRO without any IT infrastructure can build their own EDC studies themselves because they can access the study design tool remotely. They can learn how to use the web based design tool and work with their clients to build their studies independently without the aid of the EDC vendor. Consulting is offered by the vendor as needed but the premise of the SaaS model is that the clients can be as independent as they want to be. This will open the door for small companies to be more in control of their EDC studies without needing to rely heavily on EDC vendors.
TAMING THE PAPER CHALLENGE
A large majority of trials have at least some paper based data collection activities. This persistence of paper in trials has many causes ranging from study design and complexity to site location capabilities and investigator preferences.
Most EDC systems are built specifically for single entry and geared towards sites so they are very much mouse-driven. In studies that have a paper component as well as EDC, it’s necessary to create a separate database to hold the paper data and then merge the two together at the end (or even mid-study if they have an interim analysis).
Straight EDC systems cannot be adapted to accommodate paper studies, therefore if the decision is made to enter paper components into the EDC system, the client will need to create specific procedures to ensure the entry is performed correctly and that queries can be generated outside of the system. These shortcomings discourage wider EDC use and can reduce the gains achieved on trials where EDC has been deployed but is hobbled by problems associated with the need to accommodate paper.
A few EDC vendors offer “hybrid” capabilities, enabling sponsors and CROs to conduct trials using their preferred method of data collection. Such systems can combine data collected on paper with data collected electronically into one electronic clinical data management platform.
It’s important for a hybrid system to have the standard CDMS features in order to accommodate paper CRF processing. Indeed, several small CROs use hybrid EDC systems for their paper studies which has resulted in proven efficiencies. Full-featured hybrid EDC systems are not considered a CDMS, but have all the required CDMS features needed to run a complete paper study. Many CROs are attracted to the idea not having to pay and support two separate systems. This is also an area where a SaaS-based solution can be beneficial. Since CDMS systems are not web-based, CROs need to implement something in-house to manage their paper-based studies. Using a SaaS offering for EDC and/or paper studies provides minimal investment and the most flexibility.
In a typical paper study, most CRFs (case report forms) are entered twice while others that contain a lot of text are usually entered just once with full QC against the paper CRF to ensure accuracy. Therefore, the ability to configure each study to require either single data entry or double entry is essential.2 Other useful features include: the ability to configure hot-keys for data entry personnel to minimize the need to use a mouse; conflict resolution tools to identify discrepancies between first and second entry; CRF tracking tools with reporting for missing CRFs; and configurable data clarification forms (DCF) that can be printed from the system and sent to the sites.
Dealing with PDE (paper data entry) and EDC can be a significant pain point for the study teams. A flexible, intuitive EDC platform, allowing users to conduct a hybrid study, can mitigate that pain and enables sites to choose their preferred data entry method.
Technology aside, there are occasions when sites become overwhelmed with entering high volumes of patient data. For example, a trial involving an elderly patient population had a perfect use for a hybrid system due to the data volume associated with extensive medical history information and concurrent medication use. Because of the large amount of data per patient, entering data for the baseline visit on these particular forms became labor intensive for sites. To minimize frustration and reduce the data entry time, a decision was made to collect data on paper just for these specific CRFs. Once completed, the paper was sent to the study team to be entered via double-data entry, while the sites maintained responsibility for updating any new data and resolving all queries electronically. In this real-life example, advanced hybrid features and vendor-supplied services to assist with data entry helped speed the progress and enhance the site experience.
TACKLING TRAINING REQUIREMENTS
EDC is a powerful tool, but like all such tools, adequate training is required to maximize the result. Also, pressure is high to complete training in a timely manner, a mandate that is generally complicated by the system’s diverse users with differing roles and training needs. Moreover, training is a nagging pain point for trial sites, whose resources vary and who often work on multiple trials using different EDC systems.
Commonly, two different approaches for training are typically used. One is face-to-face training at the investigator meeting and the other is some form of self-paced e-learning. Sponsors tend to prefer training at investigator meetings because they can train in large groups which is ultimately less expensive than individual face-to-face training sessions at each site. However, there are several obvious drawbacks to this approach.
“Typically, at these face-to-face training sessions the trainer walks users through step-by-step how to perform the actions they need to learn. But people are rarely focused because of the volume of information presented to them during these meetings.
It is a case of information overload and a significant amount of information to absorb and retain in a typical two-hour period or more training session. Often, the investigator meeting occurs long before patient enrollment, resulting in sites unable to recall how to use the systems when the time comes to enter the first patient,” McHale notes.
More importantly, many study coordinators are unable to attend investigator meetings, and sometimes if it’s a large site with many studies, the study coordinator doesn’t actually enter data but hires a data entry person. Therefore, it’s possible to end up with users entering data for studies they aren’t familiar with, only entering patient information purely to take the burden off the coordinator. Sponsors often favor that approach because it’s more expedient but the results can be problematic.
Most EDC vendors have a standard e-learning tool to perform training which users may be required to take during study ramp-up. The efficiency of these tools can be increased if time is taken to interview users about their experience level and if the e-training can be customized to meet their particular needs. “Experienced EDC users, for example, may not need in-depth system training, but rather specific training focused on the significant differences between the systems they already use and the new system being implemented. Fine-tuning the training in this way can save valuable time and money,” advises McHale.
Site users who interact frequently with EDC systems only need to know how to log into the system, add a patient, enter data, edit data if incorrectly entered, respond to queries and sign eCRFs, among other tasks. Their training should be specific to their tasks. EDC systems are very similar in nature and they typically have the same major components and functionality so these experienced sites often learn quickly and require less hand-holding.
Other types of training are also available such as one-on-one sessions at the study site. Travel costs make this an expensive option but are often worth the additional cost because it is typically customized to the specific user and as a result is a very successful method of training. Online training through platforms like WebEx is also available but it’s not a preferred approach because of the difficulty in proving meeting attendees have actually learned the material.
THE VALUE OF CONVERGED SYSTEMS
The proliferation of clinical trial technologies is a double-edge sword. Without doubt, EDC and other trial technologies can improve efficiencies, enhance trial flexibility, reduce error, and speed up trials. However, many sites are frustrated by the number of different systems (EDC, ePRO, RTSM, etc.) they have to work with. In addition, some sites are required to work with multiple vendor applications, depending on the sponsor they are working with for a particular trial.
Effective integration of these diverse clinical trial technologies is extremely important to achieve expected benefits. Sponsors can help mitigate these issues by choosing intuitive EDC systems that provide a great deal of flexibility for the site.
For example, having a workflow-based EDC system can make life easier for site users. Such systems identify specific activities for a user and clearly emphasize the prioritization of those activities. Often, there are many studies ongoing at a site at any given time, resulting in site users having difficulty identifying specific daily trial activities they need to perform. Workflow-based EDC systems have the capacity to identify specific tasks that are required for each patient based on the status in the trial. For example, if a patient is in the screening visit, the site user will only see the screening forms and tasks available for that particular patient such as enrolling or failing. Once the patient is randomized and enrolled in the study, the system displays the visit schedule and associated tasks related to enrolled patients (such as discontinue, complete, add unscheduled patient, etc.)
Converging systems to minimize the number of applications sites must deal with is another important trend. “One recent development in the industry has been the ability to modify EDC systems to work with RTSM solutions. It is important to understand that true convergence is much more than mapping RTSM data to the EDC system. With standard integration the site performs patient randomization by accessing the RTSM system either by phone (IVR), or more commonly, the web (IWR) and the randomization data is then mapped into the EDC system. The user then accesses the EDC system to begin entering patient data. The user is still required to access the RTSM system to perform all transactions such as requesting medication resupply,” McHale notes.
“With true convergence, access to the RTSM system is not necessary because users are able to perform all RTSM actions within the EDC system which minimizes the number of systems they need to access for their trial. In a system such as this the site coordinators would access the EDC system to enter subject screening information including all data required for patient randomization. In some sophisticated EDC solutions, an RTSM action pallet is available. For example, when screening data has been entered and the site is ready to randomize a patient, the user simply clicks on a randomization button, and the data is packed up and sent to the RTSM system. The patient is randomized, automatically enrolled in the EDC system and the randomization information is populated on the randomization CRF. This means the site enters their patient data only once and randomizes the patient without having to leave the EDC system. This advanced RTSM action pallet should also be able to accommodate most medication dispensing activities in a similar way.” says McHale.
Another important benefit of convergence is that there is no reconciliation needed between the RTSM and EDC systems by the data management group. Typically, the Data Manager reconciles the data entered into the RTSM system with the data that is entered into the EDC system to ensure they match and generate queries to eliminate any discrepancies. Since both systems share the same data in a convergence model, there is nothing to reconcile which saves the Data Manager valuable time at the end of the study. This benefit doesn’t directly impact the site although it will minimize the number of potential queries requiring a response.
Aside from convergence, integrating other electronic data whenever possible is worthwhile to enable sites to see everything in a single solution. It reduces the time spent performing redundant and duplicate data entry, ultimately improving efficiency and helping to increase site morale.
Implementing EDC and other clinical trial technologies still isn’t a trivial undertaking, but advances in technology and many years of experience working with EDC solutions are taking much of the pain out of the effort. SaaS models are removing many of the technical and financial barriers that prevented small and mid-size organizations from attempting implementation of an EDC system. Importantly, the gains in trial efficiency and flexibility from well-drawn and deployed EDC solutions have been demonstrated.
1 Sizing Up the Clinical Research Market, by Kenneth Getz, Applied Clinical Trials, Mar. 2010 (http://appliedclinicaltrialsonline.findpharma.com/appliedclinicaltrials/CRO%2FSponsor/Sizing-Up-the-Clinical-Research-Market/ArticleStandard/Article/detail/660749); Cost of Clinical Development, Hove Associates, personal communication (www.earthinstitute.columbia.edu/cgsd/events/documents/hovde.ppt)
2 Hybrid Hopes, International Clinical Trials, May 2010