Presented at HxRefactored 2015, Boston, MA. Key takeaways:
* Any enterprise app which acts like a consumer app that doesn’t integrate well into hospital or ambulatory systems and workflows is doomed
* There’s nothing unique about health IT data that justifies complex, expensive, or special technology.
* There’s a lot unique about healthcare workflows that require common technologies to be adapted properly.
The No BS Guide to Developing an Enterprise
Integration Strategy for your Digital Health Product
Stop dreaming about interoperability and focus on
tactical enterprise integration
By Shahid N. Shah, CEO
Who is Shahid?
• 20+ years of software engineering and multi-
discipline complex IT implementations (Gov.,
defense, health, finance, insurance)
• 12+ years of healthcare IT and medical
devices experience (blog at
• 15+ years of technology management
experience (government, non-profit,
• 10+ years as architect, engineer, and
implementation manager on various EMR
and EHR initiatives (commercial and non-
Author of Chapter 13, “You’re
the CIO of your Own Office”
What’s this talk about?
• Many enterprise apps are being built
these days, most are designed to
work as a stand alone system similar
to consumer apps
• Healthcare-specific software
engineering and integration tools
are going to do more harm than
good (industry-neutral is better).
• Any enterprise app which acts like
a consumer app that doesn’t
integrate well into hospital or
ambulatory systems and workflows
• There’s nothing unique about health
IT data that justifies complex,
expensive, or special technology.
• There’s a lot unique about
healthcare workflows that require
common technologies to be
There is no interoperability
crisis in healthcare.
Even if there was, there’s nothing you can do about it so stop complaining.
There is, however, a vendor management
and incentives alignment crisis.
Why do health IT systems
Because customers don’t know how
to effectively punish vendors that
don’t integrate well.
But, that’s changing. Slowly.
Because app developers don’t have
a systems engineering culture where
we think of data integration as a
discipline our customers will buy.
But, that’s easy to fix. Be disciplined.
Because we want to wait for others
to create a new standard or magical
API that makes integration
But, that’s easy to fix. Follow what other industries are doing.
The tactical issues
• We don't support shared
identities, single sign on (SSO),
authentication and authorization
• We're too focused on "structured
data integration" instead of
"practical app integration" in our
early project phases
• We focus more on "pushing"
versus "pulling" data than is
warranted early in projects
• We have “Inside out”
architecture, not “Outside in”
• We're too focused on
formats instead of lightweight or
• Data emitted is not tagged using
semantic markup, so it's not
securable or searchable by
• When health IT systems produce
and other common outputs, it's
not done in a security- and
So what do we do?
Copy features and enhance (everything is separate)
Connect directly to existing data, but copy features and enhance
Create API between applications, integrate data, create new data
Create common services and have all applications use them
Start cataloging and
formalizing use of
You’re not the first (or second) to see
these problems. Read. Hire experts.
Learn about MDM, ESB, ETL, and
BPM – grab open source or
commercial implementations and
build around them.
Don’t hand code things.
Create a formal Enterprise
Integration Group (EIG).
Even get a cool logo and team mascot.
Popular cloud data exchange approach
DHP Protected Environment
CDO’s Protected Environment *
* EHR could be a standalone or hosted by a service provider
DHP = Digital Health Product
DHPSB = Digital Health Product Service Bus
In cloud, data filter strategy is crucial
Encrypt and Transmit
Customize to send only
Send data for all CDO’s
Send all data and let
the DHP Gateway
perform the filter
Pre-filter data from
each practice system
before it reaches DHP
Alternate option requires more work
on the CDO’s part; the preferred
option requires more work on the
Encrypt and Transmit
Data architecture questions to pose
1. Kinds of data to collect (e.g. demographics, medications, labs, physician
notes, etc.) - we need this in excruciating, painful detail. Lots of arguments will
emerge here and that's really good. This means we give specific field names,
field types, data lengths, etc. as part of the requirements.
2. Frequency of data to collect (e.g. real-time, hourly, daily, etc.)
3. Data code types (e.g. LOINC, CPT, ICD9, ICD10, etc.)
4. Discrete and structured data standards (e.g. CCR, CCD, BB+, etc.)
5. Directionality (unidirectional from EHR to DHP
, bidirectional between both
EHR and DHP)
6. File types and transport types (HL7, CSV, etc.)
7. Broad source types (BlueButton, BlueButton+, Ambulatory EHR, Health
System EHR, Private HIE, Public HIE, etc.)
8. Specific sources (e.g. named customers, named EHRs, named HIE, etc.)
Healthcare integration RACI
• Name the EHR or HIE (or any other source named in #7/#8
• Coordinate with the legal and executive team to ensure that you
have legal rights to the data
• Coordinate with the IT team at the system owner at the client
(usually a covered entity but could be a systems integration firm
or consulting outfit)
• Coordinate with the EHR or HIE vendor see how cooperative they
• Send the EHR or HIE vendor the Enterprise Integration document
and have them fill that out in collaboration with your team
• Get physical access to the vendor’s database or system
Be disciplined, use tools like
Caristix to document
requirements and visualize
Lets see what all
of this looks like in
You can do this in less than 40 man-
hours of work.
Ambulatory MU2 Market Share (2014)
Inpatient MU2 Market Share (2014)
Healthcare fears open source
• Only the government spends more per
user on antiquated software than we do
• There is a general fear that open source
means unsupported software or lower
quality solutions or unwanted security
Open source can save health IT
• Other industries save billions by using
• Commercial vendors give better pricing,
service, and support when they know
they are competing with open source.
• Open source is sometimes more secure,
higher quality, and better supported
than commercial equivalents.
• Don’t dismiss open source, consider it
the default choice and select commercial
alternatives when they are known to be
Rely first on open source, then proprietary
“Free” is not as important as open source, you should pay for software but require openness
E-mail [email protected]
Need to develop an enterprise integration
strategy for your digital health product?
Call or write to us.