one particular page, owing to some
archaic design choice. And yes, it may
be stupid, but it’s a design they’ve long
since grown accustomed to.
COATTA: Is there an integration story
among these different sites? Obviously,
if you’ve got all these data silos floating
around in the world, you would think
there would be some standards related
to information interchange. Or is it basically just the Wild West out there?
EVANS: The most promising global standard for this is FHIR, which
stands for fast healthcare interoperability resources. It’s a standard that
has gone through a number of iterations over the years, and it’s focused
It isn’t without its flaws, but FHIR is
definitely the best resource for interoperability between different systems
right now. Still, there’s room for lots
of data duplication, and it’s essentially
intended only for one-time transfers of
data. That is, there isn’t a FHIR network
in use now where all these different databases are kept in sync in any way. Or
at least there sure isn’t anything along
those lines that I’ve ever seen.
MCDONALD: Am I right to assume that
patients have very little say over what
happens with their information?
EVANS: That’s absolutely correct. If
you look at health-privacy standards,
they all emphasize patient consent. But
for all practical purposes, the only kind
of consent that actually seems to exist is implied consent, since, as things
currently stand, there’s no practical
mechanism patients can use either to
provide or withdraw consent as to how
their data is to be used. Which is to say
the patient is almost completely out of
The other side of this is that infor-
mation also isn’t shared among pro-
viders in any sort of way patients might
reasonably expect. If you always just go
to the same clinic, they already know
you and have your records readily at
hand. But if, for some reason, you find
you need to go to some other clinic or
end up in an emergency room, chances
are you’re a blank slate for anyone who
treats you there. They are not going to
know what allergies you have. They are
not going to know what prior condi-
tions you have. They are not going to
know what medications you’re on. So,
that means they are going to have to
scramble around to scrounge up all the
information they can from either you
or a family member.
COATTA: That sounds like a complete
mess, so what part of that problem are
you now trying to address?
EVANS: To borrow a term from the
blockchain world, we’re working to
deliver a shared ledger to the medical community. Our goal is to provide
a view of a patient’s records that not
only doctors and pharmacists are able
to share, but that can also be available
to the patient. This is something that’s
actually possible today.
One of our key objectives is to keep
track of all this information from the
patient’s perspective: What pharmacies do they use? What clinics do they
visit? Which doctors treat them at
those clinics? And how is it that each
of these participants in the patient’s
Circle of Care—as we call it—is authorized to access the patient’s records?
Moreover, can we provide transparency and some control for patients in
terms of how their data is being used
COATTA: That sounds like an admirable goal, but can you actually make
this happen? It sounds like you might
be trying to move the immovable object here.
EVANS: Actually, I wouldn’t say that
we’re working to move or replace anything. In fact, by regulation, we’re
precluded from replacing the existing EMR systems. We’re definitely not
aiming to capture all the data an EMR
needs to retain since each custodian
organization—that is, each healthcare
provider—remains legally responsible
for maintaining its own records.
However, as they continue to update the medication profile in some
amount of detail for each patient, what
we’re hoping to do is to integrate with
all those EMRs so they can keep shared
state about these patients’ records up
to date while also leveraging that information in such a way that everyone
within a patient’s Circle of Care can
readily review it.
We’re definitely not looking to
change the world here, but only to
make it possible for doctors, pharma-
cies, and patients to share a common
view of a patient’s prescription history
across all the different providers the
patient has used over time. That’s a big
enough challenge in its own right and
it’s an important goal, but it’s also a
lot more practical than attempting to
replace all the EMR systems out there.
MCDONALD: It would seem that one of
the keys to this problem has to do with
keeping track of who the patients are
and having some way to identify them
across all these different parties—
while also managing access control,
obviously. Can you elaborate a little on
this, especially the issues around digital identity?
EVANS: The system requires unique
identifiers, of course. So far, that
means we’re able to recognize patients by way of the healthcare card
numbers issued to them. By the same
token, we recognize doctors by their
license numbers and pharmacies by
their accreditation numbers. But we
also need to be able to accept some
of the other identifiers accepted out
in the world today, so we’re working to come up with a more robust
registration process. What this really comes down to, though, is taking whatever steps are necessary to
guard against having multiple profiles on the system for the same patient, doctor, or pharmacist.
Built on a foundation of proven, famil-
iar open-source software—Hyperledger
Fabric and Hyperledger Composer—
HealthChain presents no obvious im-
pediments to universal deployment.
Access to a network of patients and
providers formed using HealthChain,
however, is limited only to creden-
tialed participants, who in turn are
granted access only to certain informa-
tion assets on which they’re allowed to
perform a specific set of functions.
This means that, ultimately, access-control lists may prove to be the key to
resolving the longstanding healthcare
data-management stalemate, since
they’re not only the means by which
access to objects can be bound to each
participant type, but also the means for
defining the operations permitted on
any given object.
MCDONALD: Now that you have told us
what you set out to accomplish with
your application, tell us what you actually did.