Julius--a template based supplementary electronic health record system
These are the reviews for Chen's 2007 article, Julius--a template based supplementary electronic health record system. 
Different proprietary information models and inconsistent data quality from electronic health record solutions in Stockhom led the authors to develop a “template-EHR solution.” Clinicians enter there own predefined terminologies (data types and semantic links) in a template graphical user interface web service compatible with existing EHR solutions.
To resolve inconsistent clinical data acquisition and quality from local EHR solutions in Stockhom Sweden.
Medical terminologies are required to be entered (data types, names/labels, conceptual definitions) by clinicians interested in creating a template; this requires specialized knowledge of the reference terminologies. Design considerations also require that clinicians have expertise in the medical domain sufficient for re-use of templates.
The template based supplementary EHR system is comprised of three subsystems: 1) a system where terminologies are defined [“Concept Data Service”]; 2) a system where the user interface (layout) for new templates is determined (“Template Data Service”) - by grouping desired variables; and 3) a system for making web forms (for data entry) from a template created from the Template Data Service.
A Julius system template with 50 data variables was used in a pilot study. 250 patients’ data were recorded. The authors report positive feedback from clinicians. The reason cited for this feedback is that Julius “empowers them in deciding what and how the clinical information should be collected.”
This is certainly not a turn-key solution. Successful integration of Julius into the Swedestar EHR solution system during a pilot study may not translate to the other EHR solutions in other countries. Template construction and integration of a web based service into arbitrary EHR systems remains a primary obstacle for Julius. As clinical user groups are required to author complex variable definitions for each new template (each new *experiment*), a prohibitively high degree of technical expertise may be required to obtain new research data from new templates. Also, it remains unclear how the authors intend to avoid duplicating data already in local EHR systems.
How to facilitate the sharing ,integration and reuse of validated clinical information from different health care units and provided by different vendors ?
Purpose and Background
Electronic health record (EHR) systems are widely used in hospitals and primary care centres in the Stockholm County Council. However,within the region there are over twenty-five different EHR systems in use. Since different vendors are using both different proprietary information models and different terminologies when recording the (clinical) data, it's nearly impossible to collect consistent and validated data into centralized databases for further analysis. so the objective here was to provide a flexible solution enabling the clinicians to define which data to be recorded and shared for both routine documentation and clinical studies and the templates used for the data entry and presentation should be possible to use in combination with the existing EHR systems.
The authors stated some design considrations such as the new system should be able to record data both from clinical routine documentation and experimental studies and should be possible to integrate with the local EHR systems in clinical environment. the authors designed and developed a template based supplementary EHR system called Julius, which allows the clinicians to define data items they want to record and then design the layout of a template that guides the clinical user when recording the data.
The system conceptually consists of three subsystems:
- the Concept Data Service (CDS) where variables are defined
- the Template Data Service (TDS) for defining data entry templates
- Patient Data Service (PDS) where the patient data is stored and managed.
These three subsystems are communicating using web services. and can be deployed according to different operational strategies depending on the requirements of the organization. The system is designed according to modern system architecture with three principal layers .The presentation layer which allows seamless integration with existing EHR products,it is implemented in HTML and Java- Script, The domain layer consists of a template and term engine, transformation logic and validation logic. The data source layer utilizes an object relational mapping tool and a relational database. The system is implemented in the JAVA programming language with certain APIs from its enterprise extension J2EE.
The first release of Julius system has been deployed and integrated with Swedestar, an EHR system for the primary healthcare and during two years, over 250 patients have been recorded with the Julius system and reported to the National diabetes registry. The second release of the Julius system , with an enhanced template engine and an improved user interface has been deployed at Department of Infectious Diseases , at the Karolinska University Hospital. The feedback from clinicians has been positive since the Julius system empowers them in deciding what and how the clinical information should be collected. The web based solution of Julius allows data from different units to be collected from the same user interface regardless of the EHR software that the clinicians actually use.
The template system provides the clinical user groups with a set of tools to author variable definitions and templates, and to share the definitions of both , which alleviates the interoperability problem simply.another important aspect of the template system is that it is a highly flexible and adaptive system since its runtime behaviour is driven by templates and variable definitions created and maintained by the end users, the medical professionals themselves.There is some similar approaches to Julius system such as The OpenSDE system developed by Renske K. Los et al used row-modelling technique to allow generic structured data entry , Yamazaki etc reported using XML for sharing Electronic Patient Record screen templates in EPR platform independent way. and other authors have also reported template based EHR system and platform independent structured data entry, but none of these researches have the reached goal for sharing EHR data between different EHR sites.then the outhors describes some detailed comparison between the Julius system and the openEHR approach which is seems to be more comprehensive and well-thought .
The system has achieved its design goals. The system in combination with various local EHR systems can facilitate the sharing and reuse of validated clinical information from different health care units.
The article was so clear and understandable,the authors provided a detailed information about the issue and included some screen shots of the sytem , but there is two points I wants to state here :
- The article provided detailed description about the difference between the Julius system and the openEHR approach , I think it was more than enough ,as it is not the main issue , this comparison can be included but in less details .
- The authors included some UML class diagrams without discussing them ,so I think it may be provided as a reference for the readers that needs some detailed discussion about the implementation of the Julius system (may be with the link provided to the source code ) .
- Rong Chen, Gösta Enberg, and Gunnar O Klein. Julius--a template based supplementary electronic health record system. BMC Med Inform Decis Mak. 2007; 7: 10. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1868711/