Organisations today store huge amounts of information of many different kinds. But the more information one has, the harder it is to find and standardise what one needs.
This drives up software costs, as developers spend time locating the right sources of data. Trends in software architecture, particularly SOA, are reducing other cost factors but not this one.
However, the vendor-neutral Open Group's Universal Data Element Framework (UDEF) can help solve this.
Companies in all industries around the world, struggle non-stop with the problem of data quality, consistency and semantic interoperability. Integrating data between companies is an intolerable burden.
Harvard Business Press's "Enterprise Architecture as Strategy" reports: "In our visits to dozens of companies we have learned to recognise the warning signs of a company that doesn't have a foundation that supports its strategy." One of these: "A significant part of people's jobs is to take data from one set of systems, manipulate it, and enter it into other systems."
And: "When people key in data from one system to another, they are doing work computers do better. Worse, the process introduces opportunity for errors, and a company's scarcest resource - attention - is being wasted." And so the data challenge seems to be unsolvable.
Along comes the UDEF, a global initiative to tackle the data challenge. Driven by working groups within The Open Group:
* UDEF seeks to tackle the problem at its grassroots. The Open Group has thrown out a challenge to vendors: to deliver semantic interoperability, using UDEF, and to demonstrate it specifically for electronic health record data exchange.
Massive reduction
Healthcare provides a prime example of the problem of semantic interoperability. The RAND Institute has projected that electronic medical records could reduce US healthcare costs by at least $162 billion each year - about the same amount as businesses spend annually on research and development.
* GM says about $2 000 for every car goes towards retiree/healthcare costs.
* Intel estimates 20% of its R&D budget will be spent on healthcare costs by 2009.
* Wal-Mart's cost to provide healthcare to its employees will rise by $1 billion a year for the next five years.
* Medical mistakes lead to over 98 000 hospital patient deaths each year in the US alone.
The problem is that many data exchange standards exist for the medical field - such as those listed at www.ushik.organ/registry/x- but the medical systems used today within hospitals and doctors' offices do not all support the same set of standards, and therefore they cannot work together effectively.
Providing a framework
Integrating data between companies is an intolerable burden.
Stuart MacGregor is MD of Real IRM
The UDEF can help solve this problem, and similar problems of semantic interoperability in other fields, such as manufacturing and e-commerce. It is a standard way of indexing enterprise information that can produce big cost savings.
The aim of the UDEF project of The Open Group Semantic Interoperability Working Group is to develop and promote the UDEF, so it becomes established as the universally used classification system for enterprise metadata.
Vendors need to step forward with UDEF-based solutions that enable semantic interoperability between disparate systems. The UDEF Semantic Interoperability Challenge provides them with an opportunity to do this, and to give a high-visibility demonstration of those solutions, and show their effectiveness.
How it works
The UDEF Interested Parties Group will develop a set of detailed interoperability use-cases for medical records. The use-cases will be based on work done by Johnson & Johnson, the Stevens Institute of Technology in the US, and COEP in India, with the US National Health Information Network.
The use-cases will require integration of EMR/PHR standards such as HL7/CDSIC. For areas like diabetes/blood where HL7 has no standards, the US National Cancer Institute standards for blood/biomarkers will be used. The use-cases may also involve exposed Web services from the US Postal Service, for example, similar to those used in the award-winning Hurricane Katrina prototype.
The kinds of product featured in the use-cases could include:
* Metadata repositories that hold enterprise metadata and map it to the UDEF;
* Text processors that translate items "behind the scenes" using the UDEF;
* Browser plug-in tools to map terms to UDEF tags;
* Gap analysis tools to show differences and similarities between UDEF-annotated files; and
* Search tools to find exposed Web services that are tagged with UDEF.
The use-cases will be demonstrated at a number of Open Group conferences, and will continue to be demonstrated depending on the speed of response to the challenge. The use-cases will show how the UDEF-enabled products deliver the required semantic interoperability.
The Open Group will publish the results of the challenge.
* Stuart MacGregor is MD of Real IRM.
Share