E le ctro n ic

H e a lth R e co rd s

Health Information
Exchanges

SPH 210 - 2010

 Overview

• What is an HIE?
• Historical perspectives
• HIEs today
• The value of an HIE
• Common challenges
• Building a basic HIE
• HIEs in California
•
 What is an HIE?

• HIE = Health Information

Exchange
 “HIE is defined as the mobilization of
healthcare information electronically across
organizations within a region, community, or
hospital system” (Wikipedia)

• What is it though?
– Is it an organization?
– Is it a statewide health IT
system?
– Is it a process?
 Historical Perspectives
• 1990’s – Community Heath Information
Networks (CHIN)
• 2001 – National Committee on Vital and
Health Statistics called for creation of a
National health Information Infrastructure
(NHII)
• 2004 – DHHS calls for the creation of a
“national health information network”
• Early HIEs
– Santa Barbara Data Exchange (closed 2002)
– Indiana Network for Patient Care (INPC)
– Indiana Health Information Exchange
Ovehage, Evans and Marchibroda. Communities’ readiness for health information exchange: The national landscape in 2004.
J Am Med Inform Assoc. 2005;12:107-112
 HIE’s today
• As of 2009 there were 193 active health information
exchanges in the U.S.
• A survey of exchanges in 2004 found that one in four
were no longer functioning (25% failure rate)
• 2008  2009 here was an increase of 40% in
operational HIEs
• 70% of operational HIEs reported cost savings
• Most common services
– results delivery
– connectivity with EHRs
– alerts for providers

/ m e d sp h e re . o rg / se rvle t/ Jive S e rvle t/ d o w n lo a d / 1 2 8 6 -1 3 0 2 / E co syste m % 2 0 C o m m u n ity % 2 0 C a ll% 2 0 -% 2 0 2 0 0 9 0 1 1 5 . p d f

 What providers value in an HIE
• Ross and colleagues studied small-to-medium sized family
medicine practices
• Goal to identify what small practices value as HIE functionality
• 9 practices agreed to participate
• Methods: extensive interviews with clinicians, administrators
• Existing valued processes
– ordering tests/studies and receiving the results
– communicating with hospitals and specialty practices
• Desired HIE functionality in order of priority
– #1 – Ability to lookup patient-specific information (test results,
clinic notes, discharge summaries)
– #2 – Automated results delivery to the practice
– #3 – Electronic prescribing
– #4 – Placing non-prescription orders (low priority)
– #5 – Creating reports (lowest priority for the group)
–
e t a l. H e a lth in fo rm a tio n exch a n g e in sm a ll-to - m e d iu m size d fa m ily m e d icin e p ra ctice s: M o tiva to rs, b a rrie rs, a n d
tia lfa cilita to rs o f a d o p tio n . In t. J. M e d In f. 2 0 1 0 . 7 9 : 1 2 3 -1 2 9
 HIEs and Physicians
• Wright surveyed physicians regarding their
views on HIEs
• Surveyed 1,296 physicians in Massachusetts
(2007), with 77% response rate
• Results
– 70% felt HIEs would reduce costs
– 86% felt HIEs would improve quality
– 76% said HIEs would save time
– only 54% were willing to pay for such a service
– up to 32% were willing to pay $150/mo
§ $150/mo was based on an amount planned by
one of the HIE organizations as a charge for
providers
Wright, et al. Physician attitudes toward health information exchange: results of a statewide survey. J Am Med Inform Assoc.
 Challenges for an HIE
• often must bring together competing
stakeholder groups to collaborate on a
common set of goals
• must manage stakeholders with different
HIT needs, agendas, and schedules
• must develop data exchange/trust
agreements
• must manage competing vendors
• must have a viable long-term funding
model that is acceptable and equitable
to stakeholders
• and Electronic Health Information Exchange: A Guide to Local Agency Leadership
blic Health
stitute for Public Health Informatics and Research. 2009
 Important Perspectives on
Adoption
• “Information technology is a tool, not
a goal”
• “you can’t ‘make’ standards any
more than you can ‘make’ friends”
• “Information won’t be shared until
there is a compelling reason to
share it and until parties that need
to share it trust each other”
• “People adopt standards after they
have a reason to share”
o n d a n d S h irky. H e a lth in fo rm a tio n te ch n o lo g y: A fe w ye a rs o f m a g ica lth in kin g . H e a lth A ffa irs 2 7 ( 5 ): w 3 8 3 -w 3 9 0 2 0 0 8
 A cautionary note on the effects
of misaligned incentives in
healthcare
• The Portland Metropolitan HIE was shelved when
a model forecasted a $10 million drop in
revenues for regional hospitals due to
elimination of duplicate testing – the hospitals
were also being asked to pay $2.5 million/year
to support the HIE!


• “Labs may decide it does not make business

sense for them to send electronic results to
physicians who do not constitute enough
business volume”
 Jonah Frolich, Oct 2009, Testimony to HIT Standards
Implementation Workgroup

l Ross. Facilitating Network Agility of Health Data. Invited Lecture. UC Davis. Nov 4, 2009.
 Typical HIE Stakeholders
• Physician practices
• Payers
• Hospitals
• Pharmacies
• Clinical laboratories (regional,
independent)
• Radiology practices
• HIT vendors
• Public health department
 Physician Practices

• A challenge because of their

relatively slow rate of technology
adoption
• The bar is fairly high before practices
gain substantially from the
efficiencies of computerization
– the smaller the practice, the more
difficult the argument for
computerization
§ lower volume means longer to recoup
investment
§ smaller practice does not always
 Payers

• Have a focus on electronic

transactions
• Can gain significantly from an HIE
but it must provide efficiency in
transactions that matter to the
payers
• Payers can gain from HIEs by seeing
population based data
– this can be counter productive if it
leads the insurance plan to leave
that region
 Hospitals
• Have a major role to play in an HIE
– improve quality
– compare with other hospitals
• Hospitals in the same region are competitors
– anxiety about making available census and
demographic data
– anxiety about ‘report cards’ on quality
• Often are ‘competing’ for physician practices as
well
• Need to exchange information with physician
practices
– follow up
– improved communication
– allow physicians to have all clinical information
relevant to caring for a patient, whether the
physician has privileges in the hospital or not.
 Pharmacies
• E-prescribing improves efficiency for
physicians and pharmacists
– improved prescription accuracy
– reduced number of calls
• E-prescribing improves safety
– improved legibility
– ability to introduce some form of decision
support in the e-prescribing modules for
EHRs
• HIE could send a pharmacist relevant and
important segments of the medical
record
– drug allergies, food allergies, co-morbidities
 HIT Vendors

• There is a large number of HIT

vendors
• Increasing number of HIE ‘services’
• Ability to consume HIE services and
furnish information through an HIE
will be critical
• HIT vendors are key in enabling that
functionality for their customer
base
• But lack of well-established
standards for HIE exchange makes
 Building an HIE – building
blocks
• EHR/EMR systems in the community
– how ready is the community?
§ only 15-20% of providers have EHRs in
some form
• A Health Information “Exchange” system
– what interfaces?
– what connectivity?
– repository for result viewing vs sending results
§ A survey of HIEs by Overhage in 2004
showed 3% were federated, 54%
centralized databases, 20% used peer-
to-peer connections, 18% had not yet
selected an standard architecture
• NHIN Gateway
– allows exchange across HIEs (across regional
/ m e d sp h e re . o rg / se rvle t/ Jiboundaries)
ve S e rvle t/ d o w n lo a d / 1 2 8 6 -1 3 0 2 / E co syste m % 2 0 C o m m u n ity % 2 0 C a ll% 2 0 -% 2 0 2 0 0 9 0 1 1 5 . p d f
 HIE Technical Components
• End user applications
– Ambulatory EHRs
– Hospital EHRs
– Laboratory information systems
– Pharmacy Systems,
– Remote clinical viewer for providers
• Infrastructure (HIE) Services
– Provider Registry
– Enterprise Master Patient Index Services
– Data Repository
– Messaging Hub (document hub)
§ which supports granular and patient-centric privacy
§ which is efficient, scalable, secure

/ m e d sp h e re . o rg / se rvle t/ Jive S e rvle t/ d o w n lo a d / 1 2 8 6 -1 3 0 2 / E co syste m % 2 0 C o m m u n ity % 2 0 C a ll% 2 0 -% 2 0 2 0 0 9 0 1 1 5 . p d f

 MIRTH Results – Redwood
MedNet

courtesy of Will Ross, Redwood MedNet. used with permission

 MIRTH Results – Redwood
MedNet

courtesy of Will Ross, Redwood MedNet. used with permission

 Prototypical HIE Services

• Patient identification and patient

query
• Patient Record Locator
• Clinician Authorization
• Storage of Clinical Data
• Privacy and Security
• Consent management
• Secure messaging (provider to
provider, provider to hospital,
patient to provider?)
 HIE Architectural Models
• Repository Model
– centralized data storage
– participant data may be segmented but physically
stored in one database
– the more complex the healthcare community, the
harder it is to use a repository model
• Federated Model
– each provider retains control over their own data
– places a premium on harmonizing/standardizing data
elements
– NHIN and CDC’s proposed models are federated
• Hybrid – repository and federated combined
–

Public health and electronic health information exchange: A guide to local agency leadership.
nstitute for Public Health Informatics and Research
 Vendor based HIE’s

• Vendors are moving to interconnect

customers
– sold as an advantage to customers in
aggregating their data
– regionally close customers can
exchange data about patients they
might be co-managing
• Disadvantages
– continues to fragment and silo data in
healthcare – particularly in within a
region
– ignores the benefits of having a
 California HIE’s

• 12 operational health information

exchanges in California (2010)
• Examples
– Redwood MedNet
– Long Beach Network for Health
– Santa Cruz information exchange
– East Kern County integrated
technology association
 California HIEs – early 2010

courtesy of Will Ross, Redwood MedNet. Reproduced with permission

 Santa Cruz HIE

• Has been exchanging data since

1996
• Connects 350+ providers, 650 other
users, 7 EHRs
• Users include hospitals, doctors, labs,
imaging centers, surgery centers
• Exchanging hospital reports,
referrals, encounter data, lab
results, radiology reports, allergies,
and medication prescriptions
 Redwood MedNet
• Originally founded by Carl Henning and five
others including other Mendocino healthcare
providers non-physician members of the
Mendocino healthcare community
• Incorporated as a 501(c)(3) non-profit in
December 2005.
• Connects 6 regional laboratories, 2 regional
medical centers, and 5 provider practices
• Exchanging demographics, lab results, radiology
results today – e-prescribing planned for 2010
• Feb 2010 – Used NHIN Connect gateway to
demonstrate an exchange with Thayer County
Health Services in Nebraska
 Long Beach Network for
Health (LBNH)
• Established by Long Beach Public
Health Department in 2003
• Incorporated as a non-profit 501(c)
(3) in 2007
• Connects 4 hospitals, 35 community
clinics, Quest diagnostics, and
Wellpoint
• Exchanging demographics,
encounter data, lab results,
dictated notes, allergies, and
prescribed medications
 EKCITA

• East Kern County Integrated

Technology Association
• Established in rural California,
Tehachapi in Kern County
• Incorporated as a non-profit 501(c)
(3) in 2006
• Connects 22 providers including 1
hospital, a medical group, 3
regional health centers and 5
provider practices
 California and HIE funding

• In February 2009 California received

$32 million from ONC to build a
statewide health information
exchange
• HIMSS 2010 Interoperability
demonstration
– Santa Cruz Information Exchange
– Long Beach Network for Health
– East Kern County Integrated
Technology Association

tp :// w w w . h e a lth ca re itn e w s. co m / n e w s/ ca lifo rn ia -b u ild -h ie -n e a rly -4 0 -m illio n