Flexible yet Efficient Management of Electronic Health Records

This paper presents an approach for managing electronic health record (EHR) data in a flexible yet efficient way. The data model is based on the HL7 Reference Information Model. If clinical requirements change, the data model can be extended by editing a schema spreadsheet. A code generator and the use of state-of-the-art ORM tools allow consistency between EHR application and database. Due to the genericity of the approach, cross-cutting concerns like concurrency control and auditing can be implemented widely. The approach has been implemented in form of an EHR framework. As a proof of concept, a subset of a commercial melanoma care EHR application has successfully been implemented.

I. Introduction

Medical information technology has recently advanced in many countries, and enormous amounts of clinical data are already stored as electronic health records (EHRs) [1]. At the same time, the domain of clinical information has evolved and expanded rapidly and continues to do so [2]. For example U.S. healthcare data in 2011 is reported to have reached, 150 exabytes, while a Californian health network it is believed to have up to 44 petabytes of data from EHRs, including images and annotations [3], [4]. This leads to a dilemma. On the one hand side, efficient management (storage and retrieval) is necessary to deal with large amounts of EHR data. On the other hand, flexibility is needed to cope with the rapid evolution and expansion of EHR data schemas.