Rest in Peace: Personal Health Records (PHRs)

While doing some research the other day on personal health records (PHRs), I came across this article, describing Revolution Health’s announcement — without much media attention — about dropping its PHR at the beginning of 2010. (Disclosure: I worked for Revolution Health in 2005-2006, and now have a business relationship with the company that acquired them, Everyday Health.)

The most interesting statement I found in this brief news article was, “The e-mail did not indicate why the company decided to terminate its PHR service. The company advised users to download their PHR as a .pdf file and save the document for their records.”

Ah, a PDF. Yes, that’ll make it extremely easy to get that data into some other PHR (sarcasm alert).

And that led me to understand the underlying problem with all PHRs today, and the problem PHRs have always had — nobody trusts the companies who offer them, and few people understand what they are or why they should care.

And that led me to understand the underlying problem with all PHRs today, and the problem PHRs have always had — nobody trusts the companies who offer them, and few people understand what they are or why they should care.

I kind of chuckle when I hear a company describe that a part of its business strategy is the personal health record. I first heard of a PHR back in 1999, when I worked for drkoop.com, at that time competing for the #1 spot as the leading consumer health website with WebMD (drkoop.com lost). Drkoop.com’s management had this brilliant idea that everybody would want — and pay for — a personal health record online. In fact, this was the founding principle of the company that eventually became drkoop.com (as seen in one of their SEC 10k filings from that time):

To say that the idea of a personal health record (or personal medical record, as they called it) has been kicking around the Internet for a long time would be an understatement. (Drkoop.com dropped the idea altogether after a falling out with their PHR development partner, HealthMagic.)

Our company was founded in July 1997 as Personal Medical Records, Inc. During 1997 our primary operating activities related to the development of software for Dr. Koop’s Personal Medical Record System.

But nobody pays much attention to history on the Internet. One of the most frustrating components of consulting for companies today is their inability and unwillingness to listen and to learn from the companies who’ve come before.

Had the folks who were running Revolution Health at the time really dug into the market for personal health records, they would’ve seen exactly what we’ve seen now for well over a decade — nobody’s clamoring for them. Nobody is going to their doctor and saying,

“Gee doc, if only I had some way where I could manually enter in all of this data and try and keep it updated on a regular basis, and ensure that the company I choose to enter all this data in with is (a) going to be around 5 years from now and (b) is going to allow me to export it in a way that is actually helpful, I would be so much happier and healthier!”

Of course, let’s assume that I’m wrong. Let’s assume the 2011 IDC Health Insights’ survey of 1,200 consumers done earlier this year showing only 7 percent of respondents ever having used a PHR, and less than half still using one — which is virtually unchanged from when they conducted the same survey 5 years ago. Let’s say consumers are chomping at the bit for this kind of personal tracking ability.

What will they find?

Poor quality systems that haven’t undergone real-world testing with real-world data. As we discussed back in April 2009, PHRs simply don’t work as intended with real-world data. If one of the largest companies in the technology world with some of the brightest engineering talent on Earth can’t get this right — Google Health, in this case — what hope do we have?

Well, it seems, not a whole lot.

A March 30 article over at eWeek describes how Google is unlikely to move forward with Google Health. Instead, it’s likely to be relegated to the backburner, along with all of its other failed experiments. Of course, Google won’t comment on this, and they never will. Every company trumpets to every media outlet possible when they release a product, but mum’s the word when it comes time to acknowledge the product was unsuccessful and they’re shutting it down (or worse, putting it into a static state with little future development).

This is exactly the lack of transparency and openness consumers are fed up with, and one of the primary reasons patients are leery of trusting their personal health data with a single company. You don’t know whether it’s going to go under, sell your health data (even in aggregate form), or simply decide to quietly stop supporting its service in any meaningful way (without actually shutting it down). You may get your data out, but it may only be as a PDF. Today, there are still no widely-implemented standards for sharing health data records (although that is changing, slowly).

There are some notable exceptions, which I should call out here so that we can end on a positive note. PatientsLikeMe.com — which just opened up to everyone for any health condition — is quite transparent about what they do with your data. They aggregate it, they sell it, that’s how they make their money. And apparently it’s working, since they’ve been around now for many years.

Electronic medical records, like Microsoft’s HealthVault, are also a different animal, because they have tended to focus on addressing more of the needs of the provider, health care systems and hospitals, rather than just consumers. Paying attention to both sides of the equation — how data enters the system and how data gets out — is so important, yet something a lot of products in this space underestimate or pay nothing but lip service to.

I think it’s time to finally put the idea of a global personal health record to bed, permanently. We already have lots of individual personal health records floating around out there, tied directly to our personal health interests. And while it would be nice and more convenient to have them all somehow communicate with one another, companies who own all these individual records seem reluctant to explore the possibilities of enabling such communication. There are a lot of reasons — some valid, some not — for this reluctance.

Perhaps this will change in a more open and transparent future. But I won’t be holding my breath.

Also check out Denise Amrich RN’s article on the Google Health rumor mill, Have rumors of Google Health’s death been greatly exaggerated?

John M. Grohol, Psy.D, is one of the pioneers in online mental health education and support, getting his start on the Internet in 1992 with depression support groups and advocacy efforts. He founded and is the publisher of the Internet’s leading mental health and psychology network, Psych Central.

17 replies »

  1. Direct Exchange may be a step in the right direction but it falls far short of our needs today and in the coming years. It requires that docs adopt EMRs, which I suspect is why the established EMR vendors support it. But some 80%± of the 600,000 or so active physicians haven’t adopted them so they it can’t possibly meet the needs of most patients and doctors — either now or in the foreseeable future. Moreover, Direct Exchange makes only selected parts of a patient’s record available, not all of it.

    If I remember correctly, ONC announced that through March 2011 some 27,000 docs adopted (or said they were going to adopt) EMR systems. (ONC’s goal is 100,000 adoptions by year end.) That is a drop in the proverbial bucket even when added to the 80,000 or 90,000 who already use EMR systems!

    The conclusion I draw is that it is a terrible mistake — if not delusional — to focus on trying to link provider EMR systems in massive networks as the answer to the problem of getting a patient’s medical record in the hands of their care provider at the point of care.

    I would much rather that we focus our efforts on a simpler, more practical and doable solution, namely, organize a patient’s records around the patient — in a true PHR containing copies of their physicians’ actual progress notes and test results, etc. After all, the one thing that a patient’s encounter shares with every other encounter that patient has is the patient! So why not put the patient at the center of his or her medical record universe?

    If we take the blinders off, we can do that now. We can make a patient’s records from ALL their providers available at the point of care now. That’s where I’m concentrating my effort.

  2. There is now standardized interoperability between doctors offices and patients – its called Direct Exchange, and the leading EHR and PHR vendors are supporting it.

    In the same manner that business data communications exploded with market adaption of standard data communications (the Internet), healthcare data communications is poised to explode with market adaption of standard health data communications ( Direct Exchange – directproject.org)

  3. There is now standardized interoperability between doctors offices and patients – its called Direct Exchange, and the leading EHR and PHR vendors are supporting it.

    In the same manner that business data communications exploded with market adaption of standard data communications (the Internet), healthcare data communications is poised to explode with market adaption of standard health data communications ( Direct Exchange – directproject.org)

  4. Thank you for sparking what has turned into a fascinating conversation. I hoped to add to it with a few more data points.

    It’s true that the most prominent PHRs have been marketed to consumers by entities outside of healthcare. Research published by the California Healthcare Foundation in 2010 placed this class of PHR at the low-end of the trust barometer.

    But the problem may be greater than trust alone. PHRs offered by trusted healthcare brands have been made broadly available as a component of health insurance and IDN offerings. Some have gained reasonable rates of adoption but, as you would expect from any consumer product, the market isn’t homogenous. There are segments who perceive value, and segments that do not. For those that do, including family care managers and individuals managing complex chronic diseases, the benefits outweigh the challenges.

    Manhattan Research actually found that after years of stagnation, the number of PHR users grew from 7.4 million in 2009 and 10.2 million in 2010. Their estimates corroborate research findings from the Markle Foundation, Deloitte, PwC and others: consumers do understand and appreciate what electronic access to one’s clinical records could mean. And there is growing demand for this capability. At the same time, Markle’s research showed a marked increase in willingness among physicians to share clinical data with patients. It may simply be that the “personally controlled health record” – disconnected from the healthcare system – hasn’t been the best way to do it.

    As you note, Internet history is definitely worth reflection here. One key lesson learned is that for an online service to be successful, it must make a job consumers are trying to do more efficient and effective than the way they presently do it off-line.

    Communicating with the doctor’s office is a real-world job that patients are trying to do and it’s getting more challenging every day. The majority of Americans aren’t likely to adopt a PHR that’s disconnected from the provider’s workflow. When it is connected, it has the capability to dramatically increase consumer access to their physicians and their health records, as well as make easier the more mundane jobs they are trying to do like scheduling an appointment, requesting an RX refill, or paying a bill. When that physician’s EHR is connected to a Health Information Exchange, all that data – and all those tasks – can extend across multiple providers and health systems.

    When thought of in this way – as entities like HIMSS and KLAS have done – the value of a PHR takes on an entirely new meaning, and the definition does, too.

  5. Merle – the answer is an absolute Yes! to your question: can we design a simple, convenient, easy to use, affordable and financially sustainable system that makes a patient’s complete medical record available at the point of care, and enables doctors to coordinate their patient’s care, improve the quality of care they deliver and reduce the cost of care?

    MedeFile (www.medefile.com) has addressed this! They have pioneered one of the most robust, intuitive and easy-to-use iPHR systems on the market. The key is that they collect, digitize and index all actual medical records for the patient and store it. The patient can access all info through their secure web site, 24/7, or upload their files to one of several types of UBS thumb drive devices, or they can access it through a mobile, hand-held device, such as the iPhone. MedeFile believes that ONLY by centralizing the digital exchange of health information on the patient can universal interoperability be achieved. This is the right path — and in MedeFile’s opinion — the only way it will work.

  6. Margalit, first, let me say I agree wholeheartedly with you, John Halvorson and anyone else who says people want convenience — and in “people” I include docs who originate and keep patient records. And I agree that the primary reason they don’t use or want what’s out there today is that what’s out there simply is not convenient. They are too hard to use and aren’t worth the effort (or cost).

    Having said that, the question is would they want, support, use and pay for systems that are convenient? I think the answer is a rounding YES!

    I have yet to meet a doctor who wouldn’t like to have a simple EMR system that made it easier for him or her to create and access a patient’s records. Similarly, I haven’t met a consumer who doesn’t want their care providers to have access to all their records when they threat them (there are studies supporting this conclusion and the fact that a significant percentage will pay a $100 or so per year to control them themselves). Likewise, there are studies galore showing that consumers want their records to be secure and private — and that they don’t want them accessible over the Internet.

    I also agree that most patients rarely need access to all their records accumulated over a lifetime every time they need care. I do have friends in their 70s who wish they ahd records from their childhood to clarify problems that have emerged now, but I’m sure they represent a minority of patients. However, a great many people do need records accumulated over the most recent months or years and the only way to be certain you have them is to capture them as they are created. As time goes by, the byproduct of always having your most recent records available is your aggregated lifetime medical record — accumulated at no additional cost!

    So I come back to what I consider is the primary issue concerning medical records: can we design a simple, convenient, easy to use, affordable and financially sustainable system that makes a patient’s complete medical record available at the point of care, and enables doctors to coordinate their patient’s care, improve the quality of care they deliver and reduce the cost of care?

    My answer is yes — but not by pursuing the path we’re taking today.

  7. What Mint and PatientsLikeMe have in common is their heavy investment in user experience (UX). PLM founder Jamie Heywood mentioned at the health experience design conference (#hxdconf, http://www.healthcareexperiencedesign.com/) this week that they spent half of their seed money on a UX designer. Money well spent. MInt is used as an example of good UX that won out over Yodlee’s tough integration job in @EffectiveUI’s book by O’reilly.

    The user experience will ultimately drive the success of any Rs or Patient Portals. People will use these things with very clear value, and low risk (high trust) and minimal management. So far, except in apps, it doesn’t exist.

    I do use Mint for the very reason that I don’t have to coordinate all of the data myself. It’s simple and effective.

  8. Hi Merle, I don’t think we have a disagreement here. Patient Portals are not global PHRs (not yet), but they do have, or can have, the ability to make clinical data available to patients on the web. Many already display lab results, med lists, problem lists, immunizations, allergies, histories and some even have visit notes. And some allow you to download the information. Meaningful Use requires that they do.
    The only thing missing is the ability to display an aggregate of medical records from all providers. That will make the Portal a PHR.

    The big question, as Jonathan above also points out, is whether enough people want such aggregation, or are willing to pay, or do work, for such aggregation.
    I don’t think so. How often does an average person need an aggregate lifetime medical record? Not very often. Certainly less often than one would need an aggregated financial record. Yet, how many people have an aggregated financial record? Not very many.

    People want convenience. They want tools to make things they have to do now, easier, faster and cheaper. They don’t want new work and new responsibilities. Those who are not busy aggregating records on paper now, will find no use for aggregating records electronically either. So there may be a market for PHRs, but in my opinion, that market is rather smallish, compared to Patient Portals that are all about convenience.

  9. The problem PHRs have faced is that the work/reward ratio is too small. People are busy and don’t want to be told they have to do more work (real, right now) for a nebulous gain (someday, maybe).

    I don’t care whether we keep the term “PHR” or not, though I suspect it will recede in a few years. What matters is that the tool or tools make people’s lives easier and better, not more complicated and tedious.

    Seamless (from the user perspective) data interoperability will do that. Pre-population with provider and health plan data will do that. Integration of your health data with a convenient tool that keeps track of recommended screenings for you, and lets you immediately set up appointments online, will do that. A single place to track the care received by you and the people in your care–children, senior parents–will do that, when all you have to do is turn it on and the data gets filled in and organized. As others have said, the Record itself is the hub around which many tools will be built.

  10. Hi Margalit. While I greatly respect your understanding of healthcare IT, I must take issue with your writing about PHRs and Portals as if they are synonymous. They aren’t, certainly not as they exist and are used today.

    Most portals are used to refill prescriptions, schedule appointments, exchange e-mails with providers, etc. Few, if any, use them to exchange, capture, aggregate or otherwise make copies of a patient’s actual medical records, such as progress notes, available to care providers.

    The most widely used portal probably is Kaiser Permanente’s which reportedly is used by 3.5 million of their 8.6 million members for just those purposes. Most other portals, such as those provided by insurance companies, are used by fewer than 5% of the possible users.

    PHRs, on the other hand, typically contain a patient’s health records; they generally do not serve these other purposes.

    We can and should make portals available to patients. Even though they don’t use them widely yet, I suspect their acceptance and use will grow with time — especially as more and more docs use them to communicate with their patients. To me, these functions are where the analogy to bank portals fit.

    PHRs that exist today, on the other hand, don’t work and don’t fit your banking analogy. Each bank’s portal is the equivalent of a care provider’s record system and is every bit as much of a silo as the doctor’s system. A depositor’s records from all their financial accounts and institutions are not aggregated anywhere, just as a patient’s medical records are not aggregated anywhere.

    The challenge then is to develop a PHR system that can easily, cheaply and quickly aggregate a patient’s records so they are available to the patient and all their providers at the point of care. IMO, if we continue down the path we are on, I doubt that we’ll ever achieve that objective. We must change our approach to PHRs — but doing so will require a very different mind set from today’s conventional wisdom.

  11. “I think it’s time to finally put the idea of a global personal health record to bed, permanently.”

    It depends what is meant by global personal health record. If it’s Matthew’s definition of all data about a person, than we need to put it aside for now. Not to bed, just aside, because such a global data collection (single repository, or federated) does not exist, and is not likely to exist for a few more years. Building applications that presume the existence of, or ability to assemble, such data is going to fail, again.

    If it’s Sherry’s definition of personal health record (sans global), then those do exist now and provide services to patients and communication channels between providers and patients. Those tethered PHRs or Patient Portals have utility for patients, staff and doctors, and therefore are being used.

    And just to beat a dead horse one more time, folks are using Bank websites almost routinely now to check balances and pay bills. Those portals, tethered to a financial institution, are a convenient replacement for a paper checkbook, envelopes and stamps. This is exactly what Patient Portals are in health care. They make small hassles go away – making appointments, paying bills, printing something for school, etc.

    So before we start talking about Mint (anybody know anybody that uses that?), how about just plain, useful portals that provide value to customers, instead of pro-forma value to start-ups?

  12. That’s what the self-anointed cognoscenti said about MP3 players — until Apple launched the iPod. And, going back a few years, what they said about the horseless carriage, computers, PCs, instant photography, cell phones and iPads — and many other disruptive innovations that met previously unmet needs!

    Nearly everyone I have talked to about having all their records from all their providers on a device they own and control, waxes ecstatic. Their first question is “where can I get one.” While my sample is far from statistically valid, one can only conclude that the need and demand are there.

    The lesson is to be learned is that you have to do it right. You have to meet the need. So far, no one has. But that doesn’t mean we should abandon the effort. On the contrary, we should try harder!

  13. Sounds like PHRs are a lousy product. Maybe if we paid each patient $44,000 they would use one?

  14. John is right and Matthew nailed it.. but perhaps it is time for everyone to shift the conversation from a noun (PHR) service (application) or even a verb – a process (communication) to a value (patient centered care).

    What patients (especially those with chronic conditions as well as those who are busy working parents) value is not simply access to their data and information but communication with their health care team that fits into their life and workflow, repsects their time and adds value.

    At large integrated systems like Group Health Coop ( 620,000 members) over 50% of all members use the patient portal. Now in many of the family practice clinics 35% of all provider encounters are now happening via email, online or telephone encounters. This advanced use of a patient centered EHR system has allowed them to shift to the medical home model with 30 minute office visits and the patient involvement via the patient portal to their EHR was critical. It also let patients ask simple questions via email, make appts, see their labs, and get prescriptions refilled without a visit or call to their providers. Distant parents can see ER visits in almost real time and caregivers can (with permission) monitor their elderly parents care.

    How we duplicate this system of “communication” in the non integrated systems will be a real challenge (docs on salary have a vested interest in answering email and minimizing office visit time) unless we bring the payers to the table. No no one wants just a data repository – (who enters their financial data into mint by hand) they want real time, asynchronous, tools that enable them to move their “relationship” with their health care team out of the office.

    I believe that the way you get there is by using patient centered systems of design at the policy level as well as in the communication tools and processes you implement.

  15. From personal experience, to me the problem with adoption is resources to set up interoperability.

    Every doctor’s office has to work on authorizing and setting up a link to every ancillary system they work with to pass data to Healthvault or similar tools. It requires patient authorizations for each instance, and is cumbersome. Doctor’s just dont have the resources to set this up for each entity, each patient.

    Until there is a way for patients and doctors to check a box to send the info automatically, PHRs aren’t going to work.

  16. You point out some of the key issues around PHRs (and any kind of Rs). Trust (nobody trusts these companies), they lack access to the most relevant data (largely due to lack of, and willingness for, integration) and overall user experience (nobody wants to manage or manually input this data).

    Perhaps naively, I still believe that with the right combination that fixes these three things, there’s still a big opportunity. To Matthew’s point, nobody’s done it yet. Trust may be the biggest hurdle.

  17. PHRs dont exist. Neither do EMRs. There is data; there are applications. Period. As discussed here endlessly since about 2007.

    And despite the intransigence of the industry data wants to be liberated and will increasingly become easier to access and move around–into those applications. An those applications include communities (like Psych Central) which will make even better use of that data.