HL7 based Tree inventory systemPublication date: Jan 30, 2007
When one looks at Tree Inventory systems that are widely used in urban settings, and the data they exchange, the HL7 v3 "Patient" model, which defines a patient role as being played by either a "Person" or a "NonPersonLivingSubject" entity, can be used as a basis for trees that play the role of patients.
I was looking for a use-case to use in our HL7 version 3 implemention training. The use-case would have to show all of the issues an implementer would encounter when implementing an existing HL7 version 3 model. The trick is to pick a use-case that will be relevant to all attendees, whatever area of healthcare they are specialised in. We've had some issues with this: given that one and the same training is provided in multiple countries and contexts we get localization-based comments: "in this country we don't identify a patient in this way", "this is not a valid way to use a HL7 v3 address", or "this terminology isn't used in this country". They're fully right - but it's beside the point of an implementation oriented training. It deals with the generic mechanism of implementing "any" model.
The alternative was to use a seemingly slightly wacky use-case related to "Tree Surgeons". This is a classic joke we tend to use whilst presenting the HL7 v3 "Patient" model, which defines a patient role as being played by either a "Person" or a "NonPersonLivingSubject" entity. An example of the latter tends to be some kind of animal, because HL7 is also used in veterinary medicine. But we always also mention trees. A use-case with trees as patients would have the advantage that it would be outside of the experience of any of the attendees, allowing the training to focus on the principles behind localization of the models and the mapping of the model to application database structures and code tables. There won't be any distractions related to our choice of model nor about the vocabularies.
When looking for the background of the use-case it appears there are a significant number of applications related to the provision of care to trees. There are a significant number of Street and Park Tree Inventory Tools as well as non-invasive imaging techniques. Especially urban tree management seems to be an issue - no doubt because local governments are responsible for any damages caused by trees or falling branches. Comparison studies of the data models used by these applications have been published. There's more to this use-case than one would expect.
The training now shows how to localize the universal HL7 v3 Patient model to an implementable model used by one single application. The endresult is shown below (click to enlarge). We only need one extension to the existing model: "Place OfExistence". It isn't curently present in the model because trees have a unique characteristic: they don't have the autonomous ability to change location.
All of which means that if you attend our implementation training, you'll be looking at models related to trees and tree surgeons. If and when representatives from Tree Management Software systems show up in our training we'll have to move on to an even more exotic use-case. Probably aliens.
PermaLink to this page: http://www.ringholm.com/column/hl7_based_tree_inventory_system.htm
Index of columns:
About Ringholm bvRingholm bv is a group of European experts in the field of messaging standards and systems integration in healthcare IT. We provide the industry's most advanced training courses and consulting on healthcare information exchange standards.
Rene is the Tutor-in-chief of Ringholm.