Sunday, March 4, 2007

Open Source the Nationwide Health Information Network

I've been catching up on some missed news what with all the travel lately, and I found some bits and pieces that will make it into the next update of my consumer health information presentation.

I'm not sure whether we're looking at an age gap, a technology gap, or simple Ludditism, but whichever it is it needs to be addressed sooner rather than later.

More and more I'm seeing reports talking about what consumers want and expect from electronic medical records. Markle released a study in December of 2006 that reports "Two-thirds of the public (65%) is interested in accessing their own personal health information electronically".

Overall, the survey findings point to consumers wanting control over who has access to their records and being concerned about potential privacy issues.

In a February 2007 prepared statement to the Subcommittee on Oversight of Government Management, Markle's Connecting For Health Chair Carol Diamond underlines these concerns and neatly summarizes the problem: consumers want doctors to have access to their records, but consumers want control and oversight over the doctors who see their records.

Markle has an excellent paper on A Common Framework for Networked Personal Health Information which details the common, standards manner in which health information networks should be structured, including my own favourite; distributed data.

The first paragraph of the paper is especially informative:

"The average person’s ability to access data and communicate electronically is proliferating exponentially. Consumer adoption of digitally networked services has transformed the culture of many industries — often in ways
unimaginable barely a decade ago."

Unfortunately, the physician practice, and maybe the physician, hasn't kept up. The same physician who checks his stocks online, obtains electronic CME credit and can book a tee time from the course Web site still doesn't want his patients E-mailing him.

In the presentations I give, I hear push back from the older physicians and excitement from the younger ones. I think it's a simple fact of life, you can't change business practices wholesale. These docs have been doing business without electronic data exchange for decades, it's too much to ask them to lead the way.

As I say time and again, this is not necessarily a bad thing. On the one hand, health care is behind the times. On the other, the industry at large has a golden opportunity to learn from the mistakes of all the industries that *have* adopted electronic information networks as a core of their business.

We have this amazing opportunity to do it right the first time.

The gap, as I see it, is simply that consumers are now way more information-savvy than the health care industry; and the Nationwide Health Information Network (NHIN) doesn't seem to be going down the obvious path of not reinventing the wheel.

My major gripe is the whole *Regional* part of Regional Health Information Networks.

I know this is America and the land of free market and a (perceived) hands-off federal government, but the above survey bears out my belief that this is one instance where the feds should step in and say "this is how you're going to share data, this is how you're going to protect it".

Instead, we have dozens of almost-faceless organisations around the country trying to figure it out as they go, often formed by budget- and market-conscious hospital and physician groups who are the very people we're trying to change, the very people who have spent decades not sharing data with each other.

We have standards. We have lessons learned. We have vast repositories of open, transparent software that can be utilised. We have entire industries including finance and travel that have trodden this path. I don't feel like we're learning from them.

--

AHIMA just released a report that essentially says the same thing. RHIOs, state-level HIEs and the feds just aren't doing enough to coordinate the effort. Some highlights:
"Currently, there is little sharing of lessons learned, products (e.g., business agreements, policies, service contracts), and services between the NHIN contractors and the state-level HIEs beyond those states directly involved in the NHIN contract projects."

"There is no central authority that: (1) is accountable for ensuring that HIT is directed toward transforming healthcare, or measuring progress against that goal; or (2) makes key HIT adoption-related decisions, such as resolving disputes among collaborating entities."

"In summary, there is an understanding of how standards harmonization, certification compliance, security and privacy collaboration, and NHIN prototyping all relate strategically to the acceleration of HIT adoption. However, the disconnects among these tactical projects create the perception of multiple efforts directed at individual issues with no overarching strategic plan connecting them."

Linux and Perl figured this out a long time ago. The free and open source software community has a long established tradition of organised adhocracies and distributed development with benevolent dictator oversight.

And yet we seem to be building a national infrastructure like it's 1980; industry-facing, industry-led and industry-serving. I guess it's like those Microsoft / Apple ads. One's the stodgy business type, the other is user-friendly and cool.

AHIC has empanelled a Consumer Empowerment Workgroup, but my bet is the first hurdle will be showing a business case to the HIE/RHIO community that makes it worth their while. It's not anyone's fault, as long as we have disparate entities coming up with infrastructure there'll be no sustainable model for them to invest in providing the patient empowerment in the first place. You can't blame them for not doing it.

We're deep into Web 2.0 and the promise of a semantic Web that delivers on the promise of true user interaction and vastly improved user participation. Let's build a national health infrastructure that acknowledges this, that is people-facing, patient-focussed, open, transparent and accountable.

The culture clash of a closed-source industry such as health care and the open, transparent goals of a national health information network cannot be solved by throwing millions of dollars at closed-source vendors like Northrop Grumman.

Open source is what built and maintains the World Wide Web you're reading this article on, it's the foundation of the Internet, and we manage to keep it up just about 24/7. Everyone seems to like how it works. Open source delivers your E-mail, uploads your photos, gets you your credit card statement and let's you pay bills online.

I'm not saying we should hand the NHIN over to Silicon Valley or the open source community, but we might want to ask them to join the discussion.

--

Further reading: Open source vs. closed source (Wikipedia) and The Cathedral and the Bazaar.‎

7 comments:

Jaz said...

Update: I completely missed this item in GHIT last week, wherein HHS will ask for proposals next month for a new version of a health care "network of networks" capable of giving consumers unprecedented control over the dissemination of their personal health care information.

Anonymous said...

Jaz,
Not about RHIOs but about personal health records (but their design and use are related). Do you know about Project Health Design? You can read more about it on Pioneering Ideas, here.
Kate

Jaz said...

I think the link above may be broken, it doesn't work for me, so here it is again.

http://www.projecthealthdesign.org/

My only worry - if I put my software hat on - would be that these sorts of programs, while stimulating and obviously beneficial, can often lead to purpose-built data structures.

My chief worry is getting the data flowing across regional and state borders in a manner which allows these awesome ideas and the eight million other ideas that will be thought of later to be played out.

When data drives policy, we fail before we begin. The PHR policy needs to be open data, open architectures, standards-based systems that allow this kind of research and user activity to occur without having to think of it first.

I'm not saying it's the case, but it's my fear. Without an open architecture, we'll be back to "porting" applications such as the Personal Health Application for Diabetes Self-Management to work with each RHIO's version of the PHR data set.

Let's pretend HL7 will actually work as a standard in the marketplace. Without open architecture how can we build annex systems to access the HL7 data? What policies will govern how a doctor in Connecticut gets my health data stored in New York? What ability will *I* have to dictate to the nameless, faceless entities that are suddenly unassigned stewards of my most sensitive data how I want my information shared.

How do I stop the Diabetes application team from also acquiring my mental health record? Or my AIDS data. Or my STD data. Or any other data I feel is not relevant.

Without top-down leadership from the ONCHIT, we're segmenting and building the barriers on a system we haven't even built yet.

The best analogy I can think of is that we're building the on- and off-ramps, *and* the toll-booths, to a highway we haven't built, only mapped where we think it will be.

Worse, we're only planning on laying the HOV lane (for the RHIOs) for a while. The public three lanes won't be there for quite some time.

At risk of stretching the analogy way too far, it's hard to get from the HOV lane to the off-ramp without the three lanes in between.

Nonetheless, the project looks like a very cool bunch of people and ideas, I'll explore in more detail later.

Jaz said...

For the love of Pete... here's a *clickable* version:

http://www.projecthealthdesign.org/

Unknown said...

Yes I am 9 months behind in some cases. :-)

But this is an issue that I think is important so I'll reply now anyway.

Semantic interoperability is key to making any of this work.

The openEHR Foundation has specifications that are based on almost two decades of research. They are being implemented and they are working.

Many of us in the US seem to have a "not invented here" mentality. This is really futile.

NOT using these well engineered specifications is also futile.

Please visit http://www.openehr.org

Or form more questions ask me here or at LinkedIn http://www.linkedin.com/in/timothywaynecook

Unknown said...

Yes I am 9 months behind in some cases. :-)

But this is an issue that I think is important so I'll reply now anyway.

Semantic interoperability is key to making any of this work.

The openEHR Foundation has specifications that are based on almost two decades of research. They are being implemented and they are working.

Many of us in the US seem to have a "not invented here" mentality. This is really futile.

NOT using these well engineered specifications is also futile.

Please visit http://www.openehr.org

Or form more questions ask me here or at LinkedIn http://www.linkedin.com/in/timothywaynecook

Jaz said...

Hi Timothy, seeing as how this comment will be hard to find buried this far back, I'd be happy to post your guest blog here if you want to put some notes together.

I know of openEHR through my work with the CMS DOQ-IT program, and I support the FOSS movement in general. Do you have some input on why the physician office is so averse to "free" software?

Disclosures and Disclaimers

Disclosures

My employer is compensated through funding to provide analytical research, technology solutions, and Web-based public and private health care performance reports by the State of New York, the State of Illinois, the Centers for Medicare & Medicaid Services, the Agency for Healthcare Research and Quality, the Commonwealth Fund and Bridges to Excellence. I am not being compensated by any of these organisations to create articles for or make edits to this Web site or any other medium; and all posts authored by me are as an individual and do not represent my employer or the agencies I work for.