Jorge Grillo, who is CIO at Canton-Potsdam Hospital in Potsdam, NY, has been blogging on install Meditech 6.0. In his most excellent 11th installment (posted this week), he discussed two interesting points about integration:
One challenge for us right now is getting the 6.x interfaces up and tested. Those of you on Magic with existing interfaces will have to convert all those to 6.x. This is no small job, and we have a fair number of existing interfaces due to taking a Core Vendor approach over a Single Vendor model.
Thinking back to my best-of-breed environment days, interface development and testing was much more structured and controlled since it was all done in-house and had become a daily or common function. In the Meditech world, many small hospitals have to rely on business partners or external resources to get the job done. This can come with inherent challenges since you actually have little control over the time allocation by those resources, and they are most likely on numerous projects at one time. Also, internally to your organization, the user community may not be familiar with interface testing nor have a structured process for testing in place.
The first interesting point is the shift from treating Meditech as the “Single Vendor” to being the “Core Vendor.” Over the years many facilities and care systems have swung back and forth between best-of-breed and single-vendor. Jorge is dead on indicating that interfaces are more numerous and, I would add, more complex with a best-of-breed world.
The second interesting point was the concept of integration ownership — meaning: who will plan, build, test and deploy the interfaces? He’s swinging the right hammer at the right nail when reminding us that in-house testing staff is probably not used to deep interface testing if they are used to a single vendor philosophy.
There are some other juicy insights unrelated to integration proper, including discussion of training/staffing (“staff time costs were almost double all other costs combined, including hardware and software licenses.”) and a need to rebuild reports during the conversion (“custom reports will not convert into the 6.x environment and need to be rebuilt from scratch. There is no way your internal analysts will have time to rebuild each report prior to go-live.”)
Latest posts by Dave Shaver (see all)
- HL7 ADT Q&A with Dave Shaver - July 2, 2014
- Health Standards Community Membership Archetypes: Who uses HL7? - August 6, 2013
- Note from the Field: Meditech 6.0 HL7 Integration - September 6, 2011