Monday, April 7, 2008

Azyxxi/Amalga Site Visit and Review

The gracious folks from Washington Hospital Center and Microsoft recently hosted our Chief Medical Information Officer, Dr. David Liebovitz, and me for a deep dive session on Azyxxi, now known as Amalga.

One of the most impressive things about Amalga is the attitude and principles under which the product was originally developed by Drs. Mark Smith and Craig Feied, and Fidrik Iskandar. In terms of attitude, the team took control of their own destiny and, in an environment of very little funding and central IS support (that's my observation and I could be wrong), they simply made it happen-- Damn the torpedoes, they did the right thing for clinical care and for physician efficiency. I encouraged Mark Smith, our primary host at WHC, to publish an article on the culture, evolution, and leadership behind Amalga, not necessarily the technical capabilities, because I think there are as many valuable lessons to share with healthcare in the former as in the latter. In terms of principles, Mark and his colleagues established a set of guiding principles very early in the development of Amalga which held steady throughout the project, such as "More data is better than less data-- we can weed out the noise later." I won't go into the details of their other principles, but suffice to say that they were dead-on, 100% appropriate, but yet challenge what I would consider conventional wisdom in much of healthcare IT. Again, I hope Mark and his team publish an article or blog on those principles.

Is Amalga for real?
I didn't get a chance to crack into the code and actually look at the API or the object model, but we did review the data model and the overall system architecture, and I will stick my neck out and say that Amalga is for real. I assess systems and applications using the old PARMS acronym from my Air Force days-- Performance, Availability, Reliability, Maintainability, and Scalability. Lately I've added another “S” for Security and an E for Extensibility-- PARMSSE. :-) From what I saw, Amalga passes the test in each area quite well, though I'm still not sure how many engineers and servers it takes to keep the product running, which makes me a little hesitant to endorse its Maintainability.

Should I buy Amalga?
Right now, Amalga is best suited for hospital environments without an integrated Electronic Health Record (EHR), which makes sense, given the environment in which Amalga evolved and exists today-- i.e., WHC does not have a fully-integrated EHR with CPOE. Amalga's strength lies in its ability to aggregate and display data from disparate departmental systems, including imaging and document management. It is very strong in that role and if I were the CIO for a hospital without an integrated EHR, I would seriously consider Amalga as a first step in simply providing better and easier access to the data which already exists in the numerous departmental diagnostic systems in virtually every hospital. For our environment at Northwestern-- we have a hospital EHR, an ambulatory EHR, and an Enterprise Data Warehouse (EDW) -- the value and role of Amalga is not very clear, though we are still talking to Microsoft about the possibility of Amalga playing a role here, as well as assisting Microsoft in developing a strategy which addresses other healthcare organizations, like Northwestern, who are already far down the path of clinical data integration. In particular, I can see Amalga playing a very valuable role in smaller hospitals which lack an EHR and a robust IT team.

We did not dive into the details of the licensing and pricing model, though Microsoft assures me that they will make the product and the model flexible enough to achieve affordability. Given the vector of affordability in recent Microsoft product lines-- we use their Dynamics ERP system and their SQL Server product stack for our EDW-- I have faith that Microsoft understands the affordability issue in healthcare. (The notable exception to Microsoft's affordability trend is the Office product line, but I predict that open source and other competition will soon take care of that situation... that could be the topic for another blog.)

Closing Thoughts
Amalga is this strange and appealing hybrid between a transaction system supporting point of care clinical processes, and an analytic data warehouse system, capable of producing clinical reports as well as analytic reports and alerts for clinical management. It stretched my mind in its uniqueness and I sincerely appreciate both Microsoft and WHC, particularly Mark Smith, for being so open and hospitable with the site visit, knowing fully that I planned on writing a post-visit review.

Nuclear and Healthcare Decision Making

Nuclear warfare operations was my data-driven decision making environment before the healthcare phase of my career. It was all about recogni...