Data quality is an integral part of EHR systems. Quality assurance for these systems not only identifies the current defects in the data but also aims for minimizing the risk of their future occurrence. Previous studies for secondary use of data in research projects presented several dimensions for such defects and proposed few methods for identifying them. Although those methods were successful in small scale research studies, their application to large scale day-to-day flow of information in EHR systems involves many challenges. In this paper, we highlighted those challenges for each method and each dimension and proposed a framework for using existing technologies to address those challenges.