Lessons Learned: System Selection and Implementation of a HIE

A Q/A with Kristin A. Myers, Bachelor Law/IT; PMP; PgMP, Vice President, IT, and Jason Martin, Senior Director, Integration/Interoperability of Mount Sinai Medical Center, New York, NY, a recipient of the 2012 Davies Enterprise Award.

Q. At HIMSS13 Online, your session, “Lessons Learned: System Selection and Implementation of a HIE,” you discuss  key challenges and lessons learned in the selection and implementation of a private health information exchange (HIE) in an academic medical center to respond to the challenges of coordinating care. What are some of the challenges you encountered while implementing a private HIE in an academic medical center?

A. Two of the biggest challenges we’ve encountered are patient matching/master patient index (MPI) synchronization issues and the vendors’ variable use of interoperability standards.

The biggest issues we’ve experienced, thus far, relate to the MPI synchronization of patients across all the entities. Public HIEs, as well as the private HIE vendors, should focus on this challenge.

Technically, interoperable platforms, EHR’s, etc. are often hindered by the inability to determine an exact patient match, because the eMPI solutions aren’t robust enough. This also potentially creates data integrity and patient safety issues, if the clinical data goes across multiple records.

We’ve also had some struggles with the variable use of the interoperability standards between the EHR vendors. The standards are left open to some interpretation and the variance by vendor could add to your development timeframe.  Consideration needs to be given to constraining the institution profiles too much and allowing for adjustments as needed.

Q. Can you describe the benefits of a private HIE versus a state HIE?

A. As you’ll see in our presentation, there are several key factors of differentiation between a public (regional, state, etc.) and a public, or enterprise HIE.

Public HIEs are typically funded by grants and/or membership/subscription fees. In NYS, there have been a series of mergers between RHIOs, etc. and little value returned from participation.  Sustainability is a definite concern, and the lack of strategic alignment between the members makes it difficult to accomplish individual institution goals and meet the aggressive timelines associated with them.

On the other hand, the private HIE is funded by a capital investment from the integrated delivery network (IDN) or individual hospital and the development is tailored to the specific goals of that organization.  The flexibility to address the analytic and data needs to support meaningful use, ACO and health home initiatives is also a significant benefit.  It is much easier to develop a standardized business model without competing with the other public HIE participants.

Q. What kind of challenges do you anticipate for private HIEs, with upcoming changes in meaningful use requirements?

A. The biggest challenge, as we see it, will be incorporating the meaningful use requirements into the existing development timelines and adjusting the priorities of the ongoing interoperability initiatives accordingly.  The normalization of the required data elements from disparate systems for the purposes of submitting syndromic surveillance data, etc. to public health agencies could also prove to be problematic.

Kristin Myers is a  health IT professional with a reputation for developing a strategic plan, collaborating with clinical and operational stakeholders, building a cohesive team, and implementing systems that are on-time, on-budget, and generate business value. She is responsible for implementing a $100 million EHR program on time/on specification and on budget at Mount Sinai Medical Center, which is a 1,100 bed teaching hospital with one million outpatient visits per year. Myers converted Mount Sinai to a Stage 6 hospital and received the 2011 HIMSS and PMI New York Chapter Project of the Year Award.

Jason Martin is the senior director of Integration and Interoperability at The Mount Sinai Medical Center.  He has been responsible for all interface development, integration/interoperability and data conversion for Mount Sinai’s Epic EMR implementation. Martin is also responsible for the Hospital Interface Engine, Internal HIE and all of the interoperability/HIE initiatives. He has 15 years’ experience in health IT, serving in various roles, but the main focus has been on integration and interface development.

This entry was posted in HIMSS13 Online, Interoperability & Standards, Public Policy and tagged , , , , , . Bookmark the permalink.

One Response to Lessons Learned: System Selection and Implementation of a HIE

  1. I believe that our master patient index (MPI) paradigm and the centralized control, architectural constructs it is based on may be a flawed concept. I believe third generation, legacy system constraints, e.g., administration (EHR/CPOE), health care delivery processes, and claims processing systems limitations undermine MPI paradigm success, and that the MPI architectural constructs do not take into account the human condition that is contingent, transient, mobile and variable, as well as, dependent on exponential changes in technology. For example, if the State of Delaware that boarders on Pennsylvania and New Jersey communities of referral, has an MPI for all its residents, and they import millions of visitors each year from all over the world their MPI by default will have duplicates and errors. Even the social security system has difficulty dealing with births, deaths, and duplicate entries. We unfortunately live in a time of identify theft, hence, the question raised: Can the “sick care” industry afford to duplicate the social security identification system? I do not think so, nor should it. It would seem that the attempt is akin to building the “Tower of Babel.” I believe acute care, ‘sick care,” is attempting to recreate what could be referred to as the “universal telephone directory” without establishing a framework that enables uniqueness. For example, an independent white pages directory structure, or a distributed clearinghouse that assigns and resolves duplicates to ensure uniqueness leveraging basic data structures, such as: name, address, telephone number, thumb print, genetic marker, or IP address. I believe each “sick care” silo is attempting to resolve the patient “biometric index” or MPI at the central control system level internal to each independent HIE Silo. The ubiquitous telephone network, internet network, banking system, USPS addressing and IATA addressing mechanisms work because they establish uniqueness by intelligently distributing the addressing mechanisms first and by requiring human intervention aided by technology to ensure duplication is manageable. The HISP structure could provide the missing piece. The current HIE Silo MPI approach, I believe, is missing the basic structures needed. Further the HIE Silo MPI may be completely ignoring the requirement for patient and clinician uniqueness through authentication, authorization and encryption; hence, it is almost impossible to ensure a patients identity, safety or security.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s