Adapt EDC for Postmarket

While Electronic Data Capture (EDC) has been considered a technology of choice to run Phase II–III studies, the industry is now finding that EDC can be particularly effective for late-phase programs, improving the speed and even the quality of real-world postmarketing research. But just as late-phase programs are very different from Phase II–III clinical trials, the technologies to complete these studies are equally as different.

A flexible technology is essential because postmarketing research typically is conducted in the general medical community with physicians who do not have trained site coordinators and because patients often need to answer questionnaires electronically (e.g., via electronic Patient Reported Outcomes or ePROs). Thus, it is important to have one solution that permits ultimate flexibility and ease of use by both prescribers and patients.

Unlike clinical development sites, which are expected to adapt to the study requirements, sites involved in postapproval studies will only participate if the protocol meets their needs. In fact, the burden on site personnel can often be the biggest obstacle to recruitment and ongoing study follow-up, particularly if the protocol and data collection needs are perceived to be disruptive. These challenges are exacerbated by the fact that postapproval studies typically pay honoraria that are significantly lower than Phase II–III trials, and any perceived burden will cause investigators to decline to participate.

A flexible EDC system that's purpose-built for postmarketing studies can address these concerns for three reasons: it's highly configurable; it includes effective communication and reporting tools to fit the needs and constraints of all study participants; and it can be customized to meet unique study needs. Specifically, EDC can facilitate postmarketing studies by supporting:

• An increase in the volume of patients at minimal cost, since today's EDC systems are considerably more cost-effective than older data collection methodologies

• Streamlined enrollment, where sites are monitored in real time and participants can be stratified immediately

• The capture of data from various sources (labs, physician assessments, and patient-reported outcomes) that can then be seamlessly and cost-effectively integrated

• Better data quality, with less missing data and lower loss-to-follow-up rates, as well as enhanced regulatory compliance.

Case in point

The following case study illustrates how EDC supported the rapid start up of a rescue study. United BioSource Corporation (UBC) was asked to replace a vendor in running a large FDA-mandated postapproval observational safety registry, taking over project management, IT, EDC, central monitoring, and data management activities.

The primary reason cited for the move to UBC was that the previous vendor was unable to meet the requirements of the study. This was, in large part, because that vendor's EDC system was not flexible enough to allow the kind of tailoring required by the Sponsor. This resulted in below target enrollment and negative feedback from investigators who wanted a more simplified and streamlined look, feel, and function.

Stepping into this role at the six-month mark following first subject enrollment, in a large study that anticipated a 24-month enrollment phase, UBC was asked to:

• Develop a new EDC application, referred to as eRegistry, importing all data currently captured in the previous vendor's EDC—and complete this in three months

• Handle all data management activities, including development of new data collection tools (eCRFs), mapping of data variables in SAS from the previous vendor's EDC to UBC's eRegistry, and all legacy data entry activities, and make this transition immediately

• Launch a centralized Coordinating Center with dedicated points, one for subjects and one for sites, with a phased transition (participant line in two months and site line in three months) to coincide with the eRegistry release

• Coordinate all transition activities and manage legal and regulatory aspects of the program takeover, from day one.

The first steps, of course, involved defining expectations, outlining timelines, and maximizing the transfer of knowledge and study materials from the previous vendor. To accomplish this, various meetings were held to enable teams to work in the compressed time frame.

In addition, the UBC Coordinating Center began planning for the transition, including setting up designated telephone lines with the already established toll-free numbers, identifying dedicated program resources, and developing call schedules to cover the inflow of calls. The Coordinating Center became the primary resource for both investigator site and participant program communications.

At latest count, more than 37,000 subjects have been enrolled and it appears all enrollment targets will be met, albeit with a slightly expanded recruitment period (see Table 1).

Case Study Logistics

Lessons learned

The sponsor's project director noted that the sponsor wrote the protocol without getting outside input. What he discovered was that detailed consideration must be given to every data element and how it is going to be collected in order to meet FDA requirements in the most straightforward manner possible.

So the primary lesson learned was that the design phase of this kind of study is of critical importance to effective utilization of EDC. Before selecting a method of data collection—and even before choosing a particular vendor—study objectives and data collection needs have to be clearly understood.

In addition, the other vendor's EDC system (designed primarily for earlier phase clinical research) was insufficiently flexible to handle the flood of requests for study modifications—from site requests asking for changes in the flow of the CRF to sponsor requests for what seemed to them like minor changes in data ranges or field labels. However, any change proved time consuming, costly, and ultimately too cumbersome. Moreover, it lacked the look and feel requested by the investigators.

Further, the other system's business logic and edit-check engine were also designed for earlier phase clinical research. In postmarketing, with less experienced sites entering information, it is critical for the system to flag missing or incorrect information immediately, sometimes as the person clicks out of one field and moves to another. It is not adequate to wait until the entire form, or even the entire screen, is completed. This is even more critical when collecting data from subjects who will quickly lose interest or become frustrated and decide to drop out. Data validation must be immediate and user-friendly.

As important, the previous vendor did not field-test the interface with the physicians and nurses. Field-testing the application with only six representative physician offices provided the sponsor and UBC with invaluable feedback that was used to extensively customize the EDC system. This fostered faster enrollment yet required limited time.

Additional perks

Another benefit of this EDC system's flexibility was that we could deploy only the baseline questionnaire at go-live, then complete the system and add the follow-up questionnaires a few weeks later, providing significant time savings.

We also relied on the system's flexibility to solve one of the most critical problems in a long-term follow-up study. The study requires an annual patient questionnaire over a 10 year period. As a result, we had to develop an easy-to-use login system and a powerful reminder system for patients.

Finally, the system provides study-specific reports to the sponsor. As is typical with large postmarketing studies, there are multiple objectives unique to the therapeutic area and to the particular design questions. As such, standard EDC reports usually provided in clinical EDC systems prove of little value. For instance, 55 field CRAs are helping the sites with their enrollment activities. The CRAs have a dedicated portal that provides them with an environment where they can keep track of their site support and data review activities, while allowing the sponsor and CRO to track study progress on a real-time basis across the entire CRA team.

Conclusion

In postapproval research—where fast results, design study flexibility, and branding opportunities are far more important than in earlier stage research, and where dedicated trained site coordinators are rare—electronic data capture can be the customizable technology of choice. Moreover, the ease of data entry, enhanced by hide/show and enable/disable logic, is translatable across global boundaries with systems that are completely internationalized and localized to embrace local languages and cultural preferences. ?

Hugh P. Levaux,* PhD, is senior vice president, strategy and market development, and Chad Clark is general manager of safety, epidemiology, registries, and risk management at United BioSource Corporation, email: hugh.levaux@unitedbiosource.com

*To whom all correspondence should be addressed.