Quality data

By Scott Dixon

Understanding the final FDA guidance for patient reported outcomes.

“The experience of many providers has shown that the data collected by ePRO is more accurate, more timely and more comprehensive than traditional paper models”
-Scott Dixon

In December 2009 the US Food and Drug Administration (FDA) released its final guidance for industry on patient-reported outcome measures to help the biopharmaceutical industry understand what the agency will be looking for when it reviews new drug applications that include patient reported outcome (PRO) data. This guidance has been under review for the last three years and has become increasingly anticipated with the rise in post-approval drug surveillance and the growing need for quality data gathered directly from patients across all phases of drug development.

In the guidance, the FDA affirms that the federal code (21 CFR 312 Subpart D) outlining the general responsibilities of sponsors applies to PRO regardless of the technology (including paper) employed. Within these responsibilities are strict mandates for detailed record keeping and providing access to the FDA to conduct source verification on the data. This traceability requirement recognizes the limitations of paper-based data collection methods, particularly the lack of an audit trail. Further, the FDA wants to ensure that study participants are adhering to trial protocols, meaning if they are supposed to be completing a diary at specific intervals they are complying with that protocol rather than sporadically entering historical data. Again, this acknowledges the restrictions of paper, as there is no ability to track when and where a patient enters data. These assertions provide validity to electronic PRO (ePRO) as there is a time stamp, and hence an audit trail, when data is collected electronically. Many providers experience has shown that the data collected by ePRO is more accurate, more timely and more comprehensive than traditional paper models.

Along these lines, the guidance also provides direction for moving from various modalities and validating the modal change. An example would be moving from paper to electronic, be it web, personal digital assistant (PDA) or interactive voice response (IVR) system. When the technology moves from paper to electronic there is a substantial change to the structure of the instrument and the FDA wants to ensure this technology shift does not impact the validity of the data entered. The closer the new mode can approximate the previous structure the more easily the format's adequacy can be confirmed.

With web-based ePRO, an entire form can be displayed, and font size, color schema and so on can more closely match the paper form than what is possible with a PDA or an IVR. A PDA device only allows for review of one or two questions at a time, resulting in context being lost, which could impact how a subject completes a form. Additionally with IVR, the technology itself can limit the type of answers a patient could provide. For example, with multiple choice questions, paper, web-based ePRO and PDAs allow more than one answer to be selected, giving a range of data. An IVR, however, is often limited to only accepting a single answer, affecting the scope and accuracy of the data. The validation doesn't simply go away when shifting from a paper instrument to a web-based ePRO, but it can become more straightforward.

The focus of this guidance, in essence, is the need for sponsors to follow accepted development models, good documentation practices and ensure adequate validation is in place, as well as to consider various patient-reported outcome options. It requires sponsors to be able to clearly demonstrate the documentation, validation and change control, and have adequate records of all the development testing and validation. All of this is consistent with and further re-emphasizes previous FDA guidance and regulations; including 21 CFR Part 11 and other guidance for computerized systems and clinical investigations. This guidance extends earlier instructions laid out by the FDA and conforms to the general regulatory framework they have established for collecting clinical trial data.

Scott Dixon is Vice President of OutcomeLogix Group and Global CRO Partnerships. He has more than 18 years experience in the healthcare, life sciences and pharmaceutical industries. Prior to this role, he was chief operating and strategy officer at ePRO provider Maaguzi LLC, which was acquired by Phase Forward in July 2009.