Information Technology Reference
In-Depth Information
data that can be specifically linked to a patient. Sharing of PHI among providers
requires special safeguards and technologies specifically crafted to support these
requirements. Recently technology to support sharing of clinical information has
undergone significant transitions to approaches that take advantage of the Internet
and cloud computing.
Broadly speaking the technology models for Health Information Exchange (HIE)
are either “centralized” or “federated”. The centralized model is similar in many
respects to client-server computing. It is implemented in something similar to a hub
and spoke arrangement in which each EHR in the geographic area being served
sends and receives the information to be shared through a central system. In con-
trast, under a federated model there is a minimal central system.
A centralized HIE has many advantages, particularly if it can translate among
EHRs that use different terminologies to represent similar clinical concepts. The
premier example of this approach in the US is the Indiana Health Information
Exchange (IHIE) that connects over 90 hospitals, long-term care facilities, rehabili-
tation centers, community health clinics and other healthcare providers in and
around Indianapolis serving some ten million patients and 20,000 physicians.
A primary service of IHIE is the Indiana Network for Patient Care (INPC), a city-
wide clinical network that stores and delivers laboratory, radiology, dictation, and
other documents to a majority of Indianapolis office practices. It handles over 2.8
million secure health transactions daily and contains over 3 billion pieces of clinical
data including 80 million radiology images, 50 million text reports and 750,000 EKG
readings. Using INPC a physician can quickly see all the images a patient has had no
matter where in the area covered by IHIE they were performed. [
4
]
IHIE has some 70 employees devoted to developing and supporting its services
and the sophisticated data normalization and standardization that makes them pos-
sible. Given IHIE's success it is easy to imagine that we should have something like
this in every region of the country. However, in the complex adaptive healthcare
system of healthcare, it has proven formidably difficult to fund an operation like
IHIE. The economic challenge is familiar. How do you get the beneficiaries of the
information sharing to pay for it?
In fact, IHIE might well not exist without the support of the Regenstrief
Foundation, a 25-year old research organization dedicated to the study and improve-
ment of health and healthcare delivery. There are successful centralized HIEs in
Wisconsin and a few other communities but, in general, the model has been difficult
to launch and sustain. In our country which places a great premium on privacy and
independence there are also “political” issues with this model. People understand-
ably worry about who is looking at their data. A very specific concern has been that
health insurance companies might see health data and deny coverage for pre-existing
conditions. Assuming the new health reform bill remains in force, now that the
Supreme Court has upheld it, this concern might be mitigated, since it outlaws dis-
crimination in health insurance based on pre-existing conditions. Additionally, in
many communities, health systems view HIE as a way of creating close business
relationships with their referral sources and do not necessarily want to support an
more inclusive approach that essentially levels the field with respect to access to
clinical data.
