Sorry I haven't written...

It's been awhile since I started writing this so just to get you up to speed on where I am in building shangri-la...I'm now at a healthcare technology company and I've joined a group of truly innovative, cutting edge developers who understand and practice systems and design thinking.  They have built this platform to manage clinical data flow between the various healthcare delivery system participants, using a social (read: community) networking model.  What is fundamentally different is how they have structured the underlying data model.  The base model is the healthcare community consisting of multiple healthcare enterprises (rather than the base being an individual provider, hospital or payor).  From there, they have thought about how the end-user (consumer, if you will) of the data would access the data, where they would access the data and why they would access the data.  As a result of taking into account the community orientation and the end-user perspective, the platform consists of three major components, which I believe could apply to any HIT design:  a data integration bus, the longitudinal patient record and the communication tools.

First, you need something that can (1) take in data from a variety of sources, (2) be flexible but systematic in consuming the data, (3) have probabilistic patient matching, (4) normalize the data to fit a defined nomenclature standard and (5) create a longitudinal patient record.  All 5 of these functionalities are what is incorporated in the "integration bus."  The platform is able to collect administrative and financial claims data from payers, clinical data from laboratories, pharmacy claims data from PBMs, radiology images and reports from radiology providers, CCDs from EMRs, and patient-derived data from a PHR.

The data is then organized into a longitudinal patient record, which requires the platform to run the data through a probabilistic patient matching algorithm and a data normalization program.  Once the data has been attributed to the correct patient and the data normalized, the data can be subjected to a variety of computable, evidence-based, clinical guidelines, in order to make sense of the data - "transform data into knowledge."  The platform is able to risk stratify the population by running the data through the Johns Hopkins ACG risk model and assign disease condition markers to patients, which in turn can generate evidence-based care plans, based on the disease condition.

Finally, through these various applications, working on the longitudinal patient record and creating care plans, the platform can generate workflows to distribute the tasks from each of the care plans and involve and inform all of the members of the care team.  The data and tasks are available to access by the end-users securely, via the web, using any browser, smartphone or tablet.

Also, unlike most other platforms out there in this care coordination/medical management space, they have included the patient as part of the care team - so he/she can also get assigned tasks based on the care plans and the patient can track his/her progress with the plan as well.  The patient is also able to send secure email messages to his/her providers, order refills, make appointments, track clinical results (e.g., lab data), obtain health educational information, and communicate his/her care plan to other family caregivers.

So far, it's been a blast to be working with this group of technologists, who not only get the technology, but are also keenly aware of the people that will be using and benefiting from their work.

We are currently implementing this platform in a Medicare Advantage plan, to go live in January, 2013.  This project will also take advantage of the medical management capabilities of the platform - including all of the regulatory communications, the quality measurement and reporting and the tracking of operational measures associated with medical management, including concurrent review.  Will let you know what happens on go live.