Patient self-reported data has become increasingly important in today's clinical trials. Trials in some disease indications rely upon patient recorded diary data as the primary endpoint to demonstrate drug efficacy - including, for example, indications such as insomnia, migraine and pain. In addition, improvements in quality of life measured using patient questionnaires can now be included as claims on drug labelling. Traditionally these data have been collected using paper questionnaires and diaries issued to subjects. Regulators and the industry have become increasingly aware of the limitations of recording patient reported outcomes data on paper including data quality and integrity issues. As a result there is a growing interest in collection of patient reported outcomes data using electronic means (ePRO). Solutions include handheld PDAs, Interactive Voice Response (IVR) systems, and other site-based hardware such as touchscreen PCs. Recently, there has been much open debate with the regulators around the use of ePRO in clinical drug submissions.
US and European agencies have approved new drugs that have included ePRO data in the submission dossier, but there are many questions around the adoption of the technology that concern the community. These include: How should instruments developed on paper be adapted for electronic use, and what degree of validation should be done between paper and electronic forms? How can researchers ensure they are complying with regulatory requirements including the PRO guidance published by FDA in 2009 when using ePRO solutions? Can fewer patients be exposed in a clinical trial as a result of improved data quality obtained using electronic diaries? What type of solution should be used for certain patient populations and protocols, and how can ePRO solutions be designed optimally to increase patient acceptability and compliance? Bill Byrom and Brian Tiplady's "ePro" addresses all these issues, reviews the new FDA guidance, and provides a very contemporary view on this important subject.