JMIR Publications

ABSTRACT

Background:

The Medical Informatics Initiative Germany (MII) established 38 Data Integration Centers (DIC) in university hospitals to improve healthcare and biomedical research through the use of electronic health record (EHR) data. To showcase the value of these DIC, the HELP Study was initiated as a use case. This study is a stepped-wedge cluster randomized controlled trial designed to assess the impact of a computerized decision support system for managing staphylococcal bacteremia.

Objective:

In this paper, we present the lessons learned during the use case from a technical perspective. Our goal is to outline the challenges encountered and the solutions implemented in our early-stage use of this new infrastructure. These insights are organized into three key areas: study-specific data definition and modeling, interoperable data integration and transformation, and distributed data extraction and analysis.

Methods:

An interdisciplinary team of clinicians, computer scientists, and statisticians created a Catalog of Items (COI) to identify data elements necessary for the study’s evaluation and developed a domain-specific information model. Data Integration Centers developed ETL pipelines to collect the disparate, site-specific EHR data and to transform it into a common data format. HL7® FHIR® and the Medical Informatics Initiative's Core Data Set profiles were adopted for consistent data representation across sites. Additionally, data not present in EHRs was gathered using electronic Case Report Forms. Analysis scripts were then distributed to the sites to preprocess the data locally, followed by a central analysis of the preprocessed data to generate the final overall results.

Results:

The use of FHIR to handle rich metadata and complex data structures proved critical to using EHR data at all. However, even with interoperable data across all sites, the presence of heterogeneous data sources and evolving project needs required a highly iterative and multidisciplinary approach. This applied to all phases of the project, starting with data definition and modelling followed by data integration and transformation and finally distributed data extraction and analysis.

Conclusions:

We identified six key lessons learned. Distributed analyses are essential for secondary usage of EHR data and so are interdisciplinary teams with diverse expertise. Successful EHR data analysis also requires reusable and sustainable systems for data processing, a strong focus on EHR data interoperability, and an overall improvement of the quality of clinical documentation. Finally, a high degree of iterativity direcly improves the adaptability needed to deal with the unpredictable and diverse nature of today's healthcare data sources. The experiences and findings from the study are helping to establish better data management practices for future multi-center, EHR-based research projects.