Tuesday, February 22, 2011

Would You Like to be Engaged to Multiple EHRs?

There’s strong agreement that for meaningful use of Healthcare IT, patients and families need to be “engaged.” The question is, how? In Stage One, patients are to be given an electronic or paper copy of clinical summary information from office visits and hospitalizations, as well as discharge instructions from hospitals. This is good and a step beyond mere paper. But that raises the question of how this information can be used by the patient, other than viewed as if it were paper. What’s the “Quicken” equivalent of a “killer app” for patients to manage their health? (Oops, maybe “killer” app is not so good for healthcare…). The Stage 2 Meaningful Use Request for Comment proposes that patients be able to view and download their information from a “web based portal” that is a module of an EHR (considering that the incentive program is about “certified EHR technology”).

The best tool for any given patient depends on the individual, so speaking in statistical terms may gloss over some things. But let’s suppose that all EHRs were to offer such a portal. How many of them would a patient need to access? For myself last year, that would have been five (four EP and one hospital). According to an article by Peter Bach of Memorial Sloan Kettering Cancer Center, the typical Medicare patient sees seven doctors in four practices in one year, and those with multiple chronic conditions saw 11 docs in seven practices in one year. If you then throw in a hospital, that means eight healthcare organizations possibly with eight EHRs for one patient in a year. If you’re a family member caring for your children or aging relatives, multiply all that. You might have to interact with more than 20 web portal accounts, ending up with over 40 separate downloaded files (you’d probably need more than one per organization per year). Even if they’re in a standardized format, who or what will reconcile or pull them together? Will you be the one, using copy & paste with a spreadsheet or word processor? I don’t think so… If you use a web-based tool specifically designed to deal with those files, well then you’re probably using a Personal Health Record (PHR).
The problem with the current MU proposal for downloading from individual EHRs is that it implicitly endorses a “tethered PHR” model – a PHR extension of an EHR. But that PHR is not really patient-centric, because it contains the patient’s data from the organization using the EHR. Over time, the data will broaden as EHRs import and reconcile data from other sources such as EHRs or Healthcare Information Exchanges, but most EHRs aren’t there yet. And “tethered” is only one of several approaches to patient-centric health records. Other flavors include independent personally-controlled PHRs such as Microsoft’s and Google’s; health record banks; payer-based PHRs such as Blue Cross; employer-based PHRs such as Walmart’s using Dossia. They all have their strengths and weaknesses, but the personally controlled PHR has recently made some significant headway, associated in part with the Direct Project announcement re Dr. Paolo Andre’s EHR and Beth Israel’s Direct Push to Healthvault. If independent PHRs can be updated through a standard transport protocol (Direct secure SMTP), that securely pushes standardized content (e.g., CCD or CCR) into the PHR in a way that is structured and codified, as already required in Stage 1 MU, then the potential for a comprehensive, patient-centered record (not tied to any one provider) has increased. Now this isn’t magic – the patient or family member still needs to decide which data from the CCD are relevant and should be added to their master problem list, medication list, etc., as I’ve done with my own PHR. An ecosystem of innovative products (trackers, monitors) can then flourish around the PHR, given the influx of more robust data to operate on.

In preparing comments to the HIT Policy Committee’s Meaningful Use workgroup, I’ve recommended that MU should focus on the goal: getting the information to the patient or caregiver’s hands in a way that they can use, not on inadvertently favoring an EHR-centric “tethered PHR” flavor. Let the EHRs do the behind the scenes work of transmitting the patient’s info into PHRs of the patient’s choice (which might be any of the flavors I mentioned), enabling patients to interact productively with a tool they like, rather than making them “polygamously engaged” to multiple EHRs/portals and encumbered with lots of files to manually import into PHRs. For patients who don’t want to use a PHR, they can still get their electronic copies (as in MU Stage 1). But for those who want the opportunity to have their electronic information together in one place, there’s a great opportunity for MU, defined in a way that allows flexibility and innovation, to leverage the Direct Project and the existing and emerging clinical content standards like harmonized (I like that word) CDA Consolidation Project templates building upon CCD/C32.  

So, patient and family engagement: yes! Let this relationship between patients and HIT be productive, patient-centered, simple-to-use and... engaging, please!

Thursday, February 17, 2011

“Quickening” the Flow of Health Information – Part Two

Sometimes you just have to be quick. And I wasn’t, so Part Two is later than I planned. Here are some more analogies between Quicken (which I use for my “Electronic Financial Record”) and an Electronic Health Record inasmuch as both require information exchange. Of course, “money” and a relatively small data set are exchanged in financial transactions, rather than the vast array of health information. So with that caveat, here I go.

  • Making payments is analogous to Direct Project “push” messages. I enter payments that are pushed to payees in a format that they can accept (some electronic, some paper checks). I can enter these payees’ addresses or have Quicken look some up in a directory. My interaction is simple, but Quicken and the financial network (like a healthcare HISP) securely route the transaction behind the scenes.
  • Quicken updates my register pulling relevant data from my bank. To me, the bank is analogous to a Healthcare Information Exchange in that it received and aggregated transactions from many payees, so I don’t have to connect to all the payees. I can download from the bank as needed. Of course, the bank has security and privacy protections for my data, similar to how health data must be protected.
  • Balancing my register, which occurs automatically, is comparable to data reconciliation in an EHR (though it is much much simpler for money than for health data!)
  • I can set up automatic actions such as reminders and payments, which is analogous to workflow automation features in healthcare, where events (like signing a letter or discharging a patient) can trigger data to be transmitted to others.
  • Quicken interfaces to tax preparation software such as Turbo Tax, which then files taxes with the IRS and state. This is analogous to healthcare transmissions to public health and quality reporting agencies.
The bottom line is that I use Quicken because it saves time and money and helps me do things that would be hard to do on my own. Just avoiding the late fee on one credit card bill pays for the software investment. I’ll surmise that healthcare providers would insist on those kinds of benefits and more. If a provider can have an EHR that can do more than just the electronic equivalent of FAX -- like quickly match and autofile incoming messages to the right patient, help reconcile the data, notify the user that new information is available, provide clinical decision support, improve efficiency  – that EHR would ultimately pay for itself with or without incentives. All of us who work on EHRs need to keep that in mind: if my professional livelihood depended on using this EHR, would I use it? Personal Health Records (PHRs) are an even closer analogy to Quicken, and I believe that they too need interoperability and Quicken-like value to deliver on their potential (rather than being something that most consumers don’t have time for).

Interestingly, Dr. Clem McDonald (an HIT pioneer) recently  wrote a similar analogy, that ideally healthcare data import ought to be as easy as importing bank statements to Quicken, in his commentary Clinical Decision Support and Rich Clinical Repositories in the Archives of Internal Medicine (sorry, the link used to be public but now requires membership).

To wrap up… securely pushing information, as simply as e-mail, is good progress. Liquidity of information, whether health or financial, enables value to be realized when software acts upon it. Financial software has progressed far beyond just moving information. Healthcare isn’t nearly as far, but it’s advancing and more options are becoming available. Not all EHRs are right for everyone, but I believe that they’re on the road to being as if they were “healthcare Quickens.”

Thursday, February 3, 2011

"Quickening" the Flow of Health Information -- Part 1

Yesterday was a big day for The Direct Project, which I’ve been pleased to be able to participate in. Kudos to those who have started demonstrating successful “pushing” of communication of clinical information across providers! As pointed out by many speakers, it’s great to see collaboration for the common good among many organizations that are competitors in the marketplace. There’s a quickening momentum as the trickle of information exchange grows to become a flood.

As ONC's Dr. Blumenthal aptly said, The Direct Project is “a lane in the information highway.” What he didn’t say was “but it’s not the cars, passengers, or cargo being transported down the highway.” That’s an important distinction that not everyone may realize, since the pilots really showcase much work that is outside the scope of Direct, such as “integrating patient data across provider settings and during transitions of care” as one pilot announcement said. Direct was used to transmit the data, but didn’t integrate it, because it’s the highway, not the cars or the place that the car is going to. It’s secure transport, not data content nor use of the content.

One beauty of Direct is that it is unbiased, neutral, toward the content it is transporting. It uses established standards to securely deliver any content to the right place so that only the intended recipient can read it (it’s an honest mailman who doesn’t read your mail). But the content had to exist in electronic form before Direct can encrypt and send it. The data could have come from EHR, or a scanner, or a person typing an e-mail message, but someone or something created it first, and someone else (not Direct itself) does something useful with it on the receiving end.

To the extent that something wonderful is done with the data that’s being sent securely through Direct, that’s done by innovative and usable applications of one sort or another, and while Direct didn’t make those applications, it sends them something to work with. Applications analogous to Quicken. Why do I say that? Well, I’ve often heard the healthcare industry unfavorably compared to other industries like banking, where everyone lauds the benefits of their interoperability through ATMs and the ability of banks to download data into personal financial software like Intuit's Quicken (which I’ve used for years). I'm sometimes skeptical of such comparisons because health data is so different than money. Nevertheless, there are some very useful parallels.

If my connection to the bank securely delivered images of paper bank statements as PDF files, and images of checks, it would have some value by saving postage, trees, drawer space, and a little time. But no one would be satisfied with only those benefits today. My account data is not just delivered as images of checks or my old monthly paper statements; it’s delivered in a discrete “consumable” form that I can use to write checks, transfer money, set up automatic payments, manage a budget, analyze my expenses, search, import into tax software – all while my ledger is automatically kept up to date and saves me from having to do the dreaded “reconcile the checkbook” chore that I thankfully abandoned long ago. The semantic standards are pretty basic compared to healthcare – we agree on dollars and cents and dates and times, not so much on the payees (I realize I’m oversimplifying).

So my message to end today’s post, which I’ll follow up with part 2 in another day or so, is that the flow of health information is quickening, and I’m happy that the Direct Project is playing an important role in doing that. But it’s health IT's Quicken counterpart at the other end that makes it really useful. I believe that to get the most value out of secure health transport, we need to increasingly send content that is human-readable but also appropriately structured and consumable at the receiving end. Deciding what that content should be is work for other projects ahead. Whatever it turns out to be, the Direct Project will help push it quickly where it needs to go.