What's Wrong with Microsoft HealthVault

This isn't a blow-by-blow product review of Microsoft's just-released HealthVault Personal Health Record (PHR) portal.  I'd have to share some personal health information with Microsoft to test the new site out, and I'm about as enthusiastic to do that as I was to put my checking and credit card account numbers into Microsoft Wallet.  Remember Microsoft Wallet?

Part of the geewhiz about MSHV is that it protects patient privacy, sort of ("We do not use your health information for commercial purposes unless we ask and you clearly tell us we may.")  According to Dr. Deborah Peel of the Patient Privacy Rights Foundation, "Microsoft is setting an industry standard for privacy." (link)

Um, no.  They may be setting an example, but they are not even close to setting an industry standard.  And that's the biggest part of what's wrong with HealthVault -- and it's the same thing that cripples all the competing PHR silos being erected across the country.

Dispensing with Privacy
Privacy standards are a whole lot different than technology standards, and, for the most part, they inhibit the adoption of same.  Basically, privacy -- or at least the perception of privacy -- is a prerequisite to widespread adoption of any health information technology, standardized or otherwise.  For the most part, privacy standards need to be encoded in regulatory and legal documents to a much greater degree than they need to be encoded in data files or communications protocols.

But once the bits start flying, and people start becoming aware of the routine uses of health information that take place in paper contexts, then the knickers knot up and people demand that the IT processes be much more difficult than the arguably riskier, costlier, and more error-prone folder-on-a-roller-cart processes they replace.

A true industry standard for privacy -- both legal protections and technical specifications -- would be great, but the federal government has been so scared of doing anything wrong, they have pretty much refused to do anything at all.  But that's not Microsoft's problem, it's everybody's problem.

Understandard?
The real standards that are missing are standards for the data itself.  PHR means whatevertheheckyawannittamean.  According to one commentator, "From what I can tell, [HealthVault] is simply a repository of uploaded documents. I see absolutely no structure or typical PHR data base elements at all."  Perhaps that's because, when it comes to PHR, there are no standards.  That means your personal health "record" is going to be about as organized as your digital photo collection -- and about as interesting to someone who isn't genetically related to you.

Okay, there's clearly too much wrong with PHR to go on this way.  Let me make it simple by bulleting it out, point-by-point.

• Providers can't trust what patients say about themselves -- it's too risky, legally and otherwise.
• Most patients won't keep their records up to date -- it's too much work, and there is no incentive unless and until a preponderance of providers will use it and act on it.  Or even then.
• Even if patients miraculously started putting data out there, providers won't get a login for 57 different portals (AHIMA actually lists 57 fee-based internet PHRs, plus 13 free options) to look at error-prone, outdated and incomplete patient data -- or 5 for that matter.  Or 3.
• Unstructured data is just about worthless -- or at least not much more worthwhile than a printed piece of paper.  It doesn't move, it doesn't graph, it doesn't aggregate and it definitely doesn't save time.
• Ubiquity among providers is key -- the patients who need healthcare the most typically interact with five or more providers.  If all the providers don't agree on which portal is "best,"  the patient is better off carrying a paper folder around, or a purse full of pill bottles.

Feed Me!
Microsoft is making at least a gesture in the right direction -- they're pushing partnerships with medical device manufacturers to download results to the site.  As an insulin pump user, I can see some value in that -- if I could get my doctor to look at Microsoft's site instead of one of the other 70....and if I could get my insulin pump manufacturer to  build a device driver specifically for MSHV, and get my glucose meter manufacturer to agree to a data structure that's compatible with all the other glucose meter manufacturers, and get them all to build interfaces to MSHV, etc.

With Circles and Arrows and a Paragraph on the Back of Each One
If I could get them all to do that, it would save me from printing about four pieces of paper every three months.  Of course, I would still need the paper if I wanted to look at it with my doctor  while I was sitting in his exam room.

It's an Ad, Ad, Ad, Ad World
I don't think the intrinsic pitfalls of PHR are news to the folks at Bellevue or any of the other 70 PHR developers.  So, if the whole value proposition to the patient is fundamentally flawed, why bother?  Um, I guess that would depend on whether you want to heal their bodies or just borrow their eyeballs.

Besides, who's going to be able to take a substandard software platform and turn it into a de facto standard computing environment?  I mean, really!

http://hittransition.com/flashpoint/ad_rta.htm

Comments

[Michael Christopher]

So does this mean that FierceHealthIT's editor, Anne Zeiger, is way off base saying we don't need Regional Health Information Exchanges (RHIOs), and stating the "Once they [PHRs] become more robust, they may solve many of the problems the more-expensive RHIOs want to address"?

For my money (and, of course, healthcare IS my money -- and yours), I'd rather have a standards-based, nonprofit, public trust-established, community service-oriented RHIO serving all of my providers from a single login system. If the RHIO wants to offer me a PHR, well, that's nice. But, like you say, I'm probably never going to use it. I'll let my providers and the RHIO deal with all that data shoveling.

[Marty]

Well, yeah. You know I'm an Anne fan, but her take on PHR vs. RHIO is amiss. And I don't agree with the statement about RHIOs being expensive, either. They're pretty cheap given what even a modicum of interchange between providers can create. And two often-overlooked things that the non-profit RHIO model provides is the establishment of trust which is necessary for the exchange, and a strong presence of public interest at the stakeholder table (by virtue of foundation sponsorship and non-provider board members).

[Norman Schroeder MD]

I think we have good reason to be very skeptical as providers of healthcare supporting such "health care data vaults." Remember how secure our credit card and other personal information has been, to wit, many recent lapses. Multiply that type of things with very personal health care records. Clearly not ready for prime time in my estimation.

A family physician.

[Steve]

I tried HealthVault this morning. I thought it might be a handy repository for some basic health data - test dates and results, etc - that I have a hard time remembering and my doc has trouble finding in my chart. After stumbling around the interface for a while, I discovered that there's no way to enter data directly into HealthVault. Instead, the user needs to turn to one of a bunch of add-on applications. At that point, I simply lost interest. Too much hassle for very little potential gain. I've heard Google has something similar under development. Hopefully, it will be better.