Pioneering Ideas

Feeling Optimistic after SXSW

On Tuesday I had the privilege and the pleasure of serving as a judge on the health panel at the South by Southwest Accelerator. It's a great gig, getting to hear pitches from three startup companies that had been winnowed down from a field of dozens. And the companies were terrific—each is tackling an important challenge with imaginative solutions and great technology.

First up was Simplee. Simplee is positioning itself as the Mint.com of health care expenses. They've developed a service that reads through your insurance company accounts and displays your medical expenses in dashboard form—how much each family member has spent toward their annual deductible, for example.  They also explain the unexplainable—the "explanation" of benefits (EOB) statements we always get from our insurance companies, showing you what insurance covers, what you have to pay out of pocket, and why.  I'm on record (at last year’s Heath 2.0 conference) as saying that if Simplee can pull this off, co-founder Tomer Shomal should get a Nobel prize. While maybe not getting to that level of hyperbole, anyone who has regularly waded through these EOBs can relate to what I'm saying. They're starting with this basic problem of tracking and explaining expenses, but have the potential to go much further—facilitating bill payments, offering context-sensitive preventive reminders, and, as they get enough data, enabling comparison shopping for medical procedures. Best of all, you can try it today—Simplee.com.

Ginger.io then came on. As I've written before, I believe that the data we can capture about our day-to-day lives (observations of daily living) can greatly inform the care we receive and, as researchers start to mine it, it will become the source of new knowledge about what makes us healthy or sick. (Pioneer's Project HealthDesign has been focusing on this opportunity for the last couple of years and will soon have some research results to share.) Ginger.io focuses on a particular slice of this data—the data stored on our smartphones. They can capture social activity (calls, texts), geographic movement (GPS) and physical movement through accelerometers. Their twist is that they're really smart about processing the data and finding meaning in it. They're able to establish a behavioral signature and then identify any deviations from it, which can be important feedback for the user but also potentially for a clinician trying to improve a patient's health. Ginger.io is a spinoff from Sandy Pentland's lab at MIT (see his Pioneer- funded paper on reality mining) and by focusing on the smartphone data, they're avoiding the challenge of getting people to record anything—it's all passively collected. Currently, they're working through researchers and early adopter physicians.

The third and final presenter was Medify, which is working to bridge the gap between very high health information resources like WebMD and the medical literature. They have a robust natural language processing operation that is "reading" (crawling?) all the medical literature and coding it (i.e. intervention, disease, study population, conclusions and other attributes). The user can then get a summary along the lines of 15 studies on a total of 3,000 subjects, five of the studies showed that the intervention was safe and four showed that it was effective. You can then drill down and look at more detail and other dimensions like freshness of the research. I'm not doing it justice with this summary— you need to see it for yourself to get a feel for it. The key is that it takes the complexity of the literature on any given condition and/or treatment and starts to tame that complexity, which, as anyone who has gone Googling in search of deeper medical knowledge can attest, is a big deal. Like Simplee, you can use Medify today at medify.com.

Three exciting companies—each with the potential to bring great value to important challenges in health and health care in very different ways. In the end, the winner was... Ginger.io. Congrats to Anmol and Karan for the win, but also to the teams at Simplee and Medify for their creativity, ingenuity and terrific progress so far. These companies, along with five other excellent candidates, are cause for optimism.

OpenNotes: Mind the Gap

Last week, I contributed to The Health Care Blog about OpenNotes, a Pioneer grantee that is enabling patients to view the notes their doctors write after a medical visit. I wrote that it is a simple idea – but also a dangerous one.

OpenNotes recently completed a pre-survey published in the Annals of Internal Medicine that asked doctors and patients about their expectations of how the idea would play out in real life. What they found is fascinating. Doctors and patients are clearly divided. On a wide range of possible benefits, doctors are more skeptical than patients. But what really jumps out are the responses to questions of whether patients would find the notes more confusing than useful, and whether the notes would make them worry more. The gap is dramatic. In each case, most doctors said “yes” while less than one in six patients agreed.

Why this disconnect between doctors and their patients? Why the gap between what doctors believe their patients can handle, and what patients feel they are ready to see?

The post has generated a nice discussion on the blog, and in the comment responses you’ll find that the results of the survey are reflected in the dialogue. I recently added my own two cents to the conversation, and I’d love to see you post your thoughts, as well.

The survey results have also been covered by USA Today, MSNBC.com, and TIME’s Healthland Blog.

Blue Button: Driving a Patient-Centered Revolution in Health Care

Editor's note: This post originally appeared 9/13/2011 at the Huffington Post.

A lot can happen in a year.

Last October, I wrote about a promising new offering for people looking to take control of their own health and health care decisions. Known as "blue button," this simple (but rather revolutionary) technology offers individuals the ability to download their own health information with just the click of a mouse. They can then use and share this information however they may choose -- with doctors, care providers, or even third-party applications designed to help them track and make sense of their own personal data.

Born out of a collaborative working group convened by the Markle Foundation, the blue button was beta-tested and then implemented by the Department of Veterans Affairs, the Department of Defense, and the Centers for Medicare & Medicaid Services (CMS). The immediate demand from their patients and beneficiaries was inspiring.

Recognizing the disruptive potential of the blue button idea, we at the Robert Wood Johnson Foundation took an interest in it at an early stage of its growth. One of our aims is to help individuals understand, identify and receive high quality care. As such, exploring and supporting the development of technologies that enable people to make informed decisions is one way we hope to realize our vision of placing patients at the center of their care. Health data download capabilities modeled after the blue button approach can really move the ball forward in that regard. People can review their health records or claims information, educate themselves about conditions, procedures, medications, or test results found in their records, and share their information with family, friends and their health care providers. They can also point out errors they find and make sure that they are corrected.

I concluded my last post on the blue button idea by observing that the federal government had taken a strong step forward to give people access to their own health information, and that it was time for more in the private sector to do the same.

Not even a year later, I'm thrilled to look back and see that progress is being made -- in terms of both demand and implementation. Well over 400,000 veterans, members of the military, and Medicare beneficiaries have downloaded their data using the Department of Veterans Affairs', the Department of Defense's, and CMS' Blue Button, showing just how desired this functionality is by individuals. Equally inspiring is how much the private sector has taken up the challenge to make the health data they hold available to their patients and beneficiaries. Aetna, United Health Care, Walgreens and PatientsLikeMe are just a few of the major care providers, insurers and patient groups that have either implemented or committed to offering their consumers a blue button download capability.

Broader use of the blue button approach also offers opportunity to mobile app and software developers working in the burgeoning consumer e-health field. As the blue button download capability becomes more widespread, we expect to see more and more apps designed to take the data individuals can download and turn it into useful information and valuable tools used to manage one's health like reminders to get preventive services or refill a prescription, or a list of the lowest price outlets to order medications. To encourage these innovations, several organizations (including RWJF) have sponsored "developer challenges," and we expect more to be announced.

It's clear that we're in the middle of a health care quality revolution. But to improve outcomes on a broad scale, we need to empower individuals to become active participants in their care. Download capabilities like blue button can help do that, which is why we at RWJF are continuing to encourage their spread. Today, I am excited to announce the launch of bluebuttondata.org, a web site that is a one-stop-shop for anyone (individuals, providers, insurers, health care organizations, patient groups, mobile app/software developers) who is interested in finding out how they can join the revolution.

I encourage you to help us harness the early momentum blue button has made and turn it into a full-fledged movement. Spread the word. Or better yet, visit the site and commit to transforming health care as we know it. 

 

Putting It All Together

I've just spent two stimulating days in DC, first going to the HHS/IOM Health Data Initiative meeting (aka "Datapalooza") and then at an HHS/Kaiser-sponsored Health Innovation Summit (see #futurehealth on Twitter).  What's clear is that this is an exciting time for innovation:  we're seeing companies pledging to enable patient downloads of their data (I heard "bluebutton" used as a verb for the first time); more releases of federal population-level data; and a gaggle of companies leveraging these data to offer terrific wellness apps.  The future's bright, indeed.

In thinking about how the pieces come together, the challenge seems a bit different now.  A few years ago, a key question was how to leverage the data in one's own personal health record to build apps that would help people take care of themselves.  An insight of the Pioneer-funded Project HealthDesign was that the data needed to drive those apps came not so much from the medical record but rather from the flow of life: observations of daily living (ODLs) about diet, exercise, sleep, mood, pain and more.  So apps needed a platform that would integrate both medical record data and this new set of patient-generated ODLs. 

Now, as I look around, I see multiple categories of data that one might want somehow integrated:  in addition to medical record data and ODLs, there's the population-level data that HHS and others are releasing (that can provide useful context and benchmarking), environmental data (mashing up your own geo-location data with data sets about the environmental factors associated with those locations), genomic data, and, of course, if you're the competitive type, comparative data from your social network.

What I've observed is many siloed apps.  I've got devices and web sites that either capture data (like the treadmill to the iPod to Nike+) or give me the opportunity to enter data (like putting my weight into Google Health), but nowhere that stores it all.  That doesn't bother me so much because at one level, where it's located isn't all that important as long as I'm confident that it's secure and reliable.  Missing are the abilities to a) present all of these disparately gathered data in some context that gives them meaning and b) to analyze these data to give me insights about patterns and correlations that would help me understand how to be healthier.  Where are the apps that will crawl across the different data stores and pull it all together? And what will they need to work? 

I'm curious to hear of companies that are trying to make this happen.  What are the best examples people have seen?

 

Whose ODLs Are They Anyway?

When it comes to deciding which “observations of daily living” should be collected, stored and reviewed, clinicians and patients have (surprise, surprise) different views.  And the overlap isn’t always that large.  This insight was a recurring them at last week’s meeting of Project HealthDesign teams.  So what’s to be done about that?

If you’re coming from a patient power or quantified self perspective, you might say that it doesn’t matter – people should track whatever they want, right?  Of course, but on the other hand, suppose you have limited patience for self-tracking and you really want your ODL data to influence how your clinical team understands your condition and how they make treatment decisions, then you probably want to make sure that some of your energy is devoted to collecting ODLs to which they might actually pay attention.  At the same time, it’s an opportunity for the patient to explain that ODLs she believes are relevant to her condition – either as triggers/exacerbators, or as symptoms of how her condition is manifesting itself in her life – are important to how she understands and manages her conditions.  So, for example, while a clinician might not be interested in a particular patient’s sleep pattern, that patient might feel strongly that sleep is inextricably linked to her health.

All this suggests a negotiation between person/patient and provider, which is the solution that the Crohnology.MD team is proposing.  They’ve mocked up an ODL prescription, which is, in effect, the outcome of the negotiation.  The prescription specifies the ODLs to be collected and the periods or durations during which they are to be collected.

We’re also discovering that there is a third party to the negotiation – the system developer.  In exploratory research project like the Project HealthDesign grants, this participation is explicit:  the developer talks with patients, talks with providers and ultimately builds a system around their requirements.  The developer’s design choices – from which ODLs to include to how they are defined to when they are collected – shape the patient’s participation in collecting them.  For example, the developers of FitBaby, a Project HealthDesign grant to investigate the use of ODLs to assist in the care of prematurely born infants coming home from the hospital, learned that understandably anxious parents in this situation often feel compelled to weigh the infants frequently.  However, weighing the baby too frequently not only lacks clinical value (as fluctuations are natural), but it’s also not healthy for the parents – particularly if the weighing becomes obsessive.  So the system developers set up the weight tracking routine to accept only one measurement of weight per week.  You can agree or disagree with that choice, but the point is that the developers shape the negotiation.  Outside of the research context, as developers build products for the marketplace, they are unwitting participants in the negotiation insofar as they are defining the solution space within which the negotiation between patient and clinician takes place.

From a policy perspective, this discrepancy between patient and clinician perspective on which ODLs matter has the effect of complicating the concept of “patient-generated” data, which appears in both the HIT Policy Committee’s proposed Stage 3 Meaningful Use criteria and the ONC’s recently released strategic plan. There are at least three types of “patient-generated” data: 

-          Data, such as blood pressure readings, already collected in a health care setting and typically found in a medical record.  The only difference is that the patient is collecting the reading.

-          Data, such as those found in a food diary, that are self-reported by the patient, but are not typically collected in a health care setting or documented in a medical record

-          Data, such as sensor data – or interpretations of sensor data, that are not self-reported by the patient but that are typically not found in the medical record.

As we go down the road of integrating “patient-generated” data – both traditional medical data and observations of daily living – into electronic medical records, then we’ll need to understand what types of data are really meant and which, under what circumstances, should be incorporated into one’s medical record.  To these questions, we hope to have more to say as the Project HealthDesign teams conduct their studies and experiment with different ways of providing all three of these data types to clinicians.

 

Focus on the Pig: The Making It Better Symposium at RISD

Aidan Petrie of Ximedica said it bluntly. In acknowledging that it is often tough for designers to break into health care institutions and play a significant role, Petrie, an industrial designer, gives this advice to designers: "remember, it’s not about putting lipstick on a pig – it’s about the pig." Exactly. In that stark phrase Petrie captured much of my thinking as I sat through the day-and-a-half long “symposium on art, design and the future of health care” at the Rhode Island School of Design (RISD), sponsored by RWJF’s Pioneer Portfolio. The intersection of art, design and health should not be about prettying up a lousy system—it should be about designing a better system to begin with.

The symposium was the brainchild of RISD president John Maeda, who wanted to connect the talents of RISD students and faculty with the urgent problems of health and health care. And there were many references to the strange bedfellows nature of the gathering. This awkwardness was most pronounced among the participants who came from the health care and public health side of the aisle. The artists and designers who spoke generally seemed quite comfortable with the notion that health does not exist independent of a broader life but is deeply intertwined, and as such, it is a natural subject for art and design.

So we saw Damon Rich apply a designer’s approach to understanding how to boost public participation in the decision-making processes that affect public spaces; the wonderful artist Mel Chin bring his passion and vision to the tragedy of lead poisoning in inner cities; and Amale Andraos’ brilliant visions of greener, healthier urban landscapes and buildings. But, by and large (there were exceptions), participants from the health sector looked at this question of how art and design could intersect with health with a slight puzzlement before defaulting to… communication. As one put it, we need great designers to help us communicate the complexity of health care in ways that people will understand. With all due respect to health communication, which is a vital field, health in the United States has even more fundamental challenges that demand the talents and skills of the design and artistic communities. Let’s get them to focus on the pig.

Within a couple of generations we have managed to engineer movement out of our work days, cooking out of our homes and play out of our children’s lives. We have designed communities and pioneered lifestyles whose only logical consequence is poor health. We have designed a health care system that has brought us the 7-minute office visit, the dreaded emergency room wait, the explanation of benefits statement, 50 million people without coverage and – oh yeah – a crushing federal deficit. I won’t be naïve and say that a few designers will turn this all around – these are deep-rooted problems – but I do believe strongly that we need the creativity and the perspective that they bring. We need people who look at these problems differently. Who are, as John Maeda put it, naturally curious. Whose views are not so deeply ingrained that they cannot ask “why not?” Who, like the delightfully unexpected Kelly Dobson, ask questions that stop us in our tracks, make us uncomfortable and make us remember that we’re human.

Data, Apps and the Rules that Bind Them

After several much-ballyhooed analyses projected an impending boom in the mobile health industry, it seems this once-geeky topic is being talked about everywhere lately. For those of us whom have been long-time advocates (and supporters) of the disruptive potential of mHealth technologies, we are thankful that this conversation is taking place.  We’re in an era where the technology enablers are moving into place and getting better every day, the business models are starting to develop and the policy environment is – well – unsettled at best.  At issue are the questions of whether and how much the FDA will regulate medical apps and how rules and regulations around privacy and security will circumscribe the kinds of products and services developers can offer.  This uncertainty in the policy environment raises the risks for developer/entrepreneurs and the investors that would back them.

FDA regulation is a thorny issue and is especially challenging when a “medical device” consists of a phone carried around by tens of millions of Americans and software that could have been written in a weekend.  The sheer volume of review that would have to take place is overwhelming in an era where agency budgets are going down not up.  An account of the recent FDA approval of the mobile radiology app Mobile MIM gives one a sense of the difficulty here.  And the lack of clarity has a chilling effect.

But aside from the question of how the technologies will be regulated, another fundamental concern needs to be addressed. That being, how will the data contained in an individual’s personal health record (PHR) – the “fuel” that powers the mHealth machine –be protected and secured? And what regulations will be put in place to do so? Until policy makers step in and provide guidance, the chilling effect on innovation will continue, furthering our reliance on an antiquated model of care that does little to empower patients to be active participants in their own health decisions.

But before providing the legal and regulatory framework app developers need to move forward, policy makers must recognize that PHRs are rapidly evolving.  This evolution is something I recently discussed as part of the Office of the National Coordinator for Health Information Technology’s public roundtable “Personal Health Records: Understanding the Evolving Landscape,” where I laid out three themes that have emerged from several Robert Wood Johnson Foundation Pioneer Portfolio-funded projects I’ve helped coordinate. These themes include: separating the apps from the data, expansion of the definition of health information, and the increasingly social nature of health care, all of which I explain in more detail below. It is my hope that policy makers take all three into consideration when crafting policies that will inevitably shape the industry.

Separating the Apps from the Data

When examining the first theme, we must take a look back in history and recognize that PHRs were designed to house – and give and individual access to –medical records, along with a few features that helped display and interpret the information.  Whoever maintained the data also provided the services that displayed, interpreted or otherwise helped you use the data.  Increasingly, however, we’re seeing a movement towards separating these functions:  the function of storing, maintaining, and providing access to the data serves as a platform while additional features and functions can be offered by third parties as apps.  By now we all recognize this as the way the computer industry works – operating systems, like Windows and the MacOS (and now iOS and Android) serve as platforms and third-party developers offer hundreds of thousands of programs or apps that run on those platforms. When it comes to PHRs, this model enables a patient to choose features and functions based upon their own individual preferences and circumstances, instead of being limited to the options of what a single PHR vendor builds into service.  The implication of this trend toward separating the apps from the data is that it makes it hard to regulate a “PHR service” or “PHR vendor,” unless one adopts a very narrow definition that focuses only on the storage, maintenance and control of access to the data.

Expansion of the Definition of Health Information

Individuals are increasingly becoming aware of the fact that health is not just what happens when you go to the doctor. Health happens where you live, work, learn and play.  It happens 24 hours a day, 365 days a year. It is based on the behavioral decisions you make every day, and it is based on the circumstances in which you live.  Apps that help people manage their health often draw from data one finds in a medical record (e.g. medications) and also data from people’s day-to-day experience that contribute to (or detract from) their health.   In our national program Project HealthDesign, which is supporting research on this topic, we refer to this latter set of data as Observations of Daily Living (ODLs), which can include how much exercise you got on a given day, what your text messages say about your mood, or the duration and intensity of headaches to name a few. You can even include in this category relevant environmental data on ambient temperature or the level of particulates in the air.  Regulations around the privacy of health information tend to focus on health information as information retained by health care service providers, but when you take a broader view of health and when app developers mix ODLs with traditional health data, it becomes very hard to draw a line between “health” information and other forms of personal information.  Any future regulations will have to wrestle with the meaning of an expanded definition of health information.  Perhaps the distinction between health information and any other information someone would like to keep private is no longer tenable.

The Increasingly Social Nature of Healthcare

Finally, policy makers must realize that individuals are increasingly sharing their data and health-related decisions with friends, family members, and an increasingly sophisticated care network. Health care is becoming more social in nature. And it’s not just social media or human to human interaction I’m talking about – many emergent mHealth technologies need to share data freely with software or other devices. A wireless blood pressure cuff talks to an app that then requests to deposit the data in a physician’s electronic health record (EHR).   A smartphone app for medication reminders sends a query to an EHR to get up-to-date prescription information.  Whether it’s person to person, device to device or app to app, sharing can add real value and regulations need to find a balance between helping people guard their privacy while also enabling the secure sharing of information.

Separating the apps from the data, expansion of the definition of health information, and the increasingly social nature of health care – all can potentially change how we view PHRs, and how we perceive health care at its very core. However, while mHealth shows tremendous promise, regulations could have a significant effect on how much of that promise is realized.  Policymakers will need to walk a fine line between providing enough clarity to encourage developers to act while offering enough flexibility to accommodate a rapidly evolving shift in how health information is understood and used.

Based upon their extensive experience in the field, several team members from Project HealthDesign submitted public comments to the ONC following the PHR Roundtable at which I spoke. These comments do an excellent job of laying out Project HealthDesign’s vision for the future of PHRs and related privacy and security concerns, providing relevant examples from their work where innovation could be impeded by impractical policy. For example, if data-encryption requirements are so stringent that a device cannot share data with a software program or another device, then we’d lose out on the benefits of technology that uses trackers to automatically monitor a whole slew of ODLs for patients (read Anind K. Dey’s public comment for a wonderful explanation). Or, if privacy/reporting requirements for mHealth devices are created with yesterday’s definition of health information in mind, a wide-range of health apps could be adversely affected (see Katherine Kim’s and Gillian Hayes’ comments). These very concerns are currently being echoed by developers throughout the mHealth industry.

Wading through all of these thorny issues promises to be difficult for policy makers and regulators, but the ultimate solution may be as simple as providing individuals with the ability to control and share their own data as they see fit. The new reality of health care rests on that very notion; that patients should be at the center of their care. Tomorrow’s mHealth policies need to be created with that philosophy in mind.  

 

It’s Starting to Add Up - A Few Observations from AMIA 2010

I’ve just finished three days at AMIA’s Annual Symposium – the geekfest gathering of informaticians (or informaticists, if you prefer).  It’s a big conference, with many themes and tracks, so it’s hard to draw general conclusions as any observations are largely functions of which sessions one chooses to attend.  So I’ll try not to generalize (too much) but offer a few thoughts on what I saw and heard.

Meaningful Use policy is a really delicate business.  As provider organizations are starting to translate Meaningful Use requirements into operational plans, the details are getting really tricky.  I saw a panel representing some real EHR pioneers (e.g. Intermountain Health Care, Marshfield Clinic) that showed how even for them, who’ve been using EHRs rather meaningfully for years, they will have to make substantial changes to their systems in order to qualify for the subsidies.  Meaningful Use is a great concept because it uses large scale EHR implementation as a vehicle to shift practice in the direction of better quality (good) and more prevention (good again), but doing so without being over-prescriptive is much harder than it looks.  Finding that balance of rewarding the principles while maintaining a fair degree of flexibility on the implementation details seems to be the key (and, to be fair, easier said than done).

Health IT raises interesting questions about roles.  Dan Masys, in his perennially fascinating review of the past year’s research findings and key developments, pointed out that now study after study about clinical decision support shows the same finding:  that decision alerts always have more impact if they are sent to anyone on the care team but the physician.  He also noted that the teachable moment of an order entry system is at the moment of entry (when, for example, a drug-drug interaction or contraindication alert is triggered) and that physicians don’t learn from these opportunities if they delegate the order entry task, as is common.  Dryly, he observed that studies also show that decision support is most effective when it’s not ignored.  On a serious note, he commented that if the trend of these findings continues, disintermediation of physicians will become inevitable.  Ouch.

Rapid Learning is really starting to happen.  I attended a terrific presentation by Nirav Shah, a researcher based at NYU and Geisinger.  He’s been mining Geisinger’s EHR data, which goes back many years now, and doing just the kind of studies one would hope to see when we can start to tap large clinical datasets.  After a preliminary rundown of the pros and cons of mining EHR data vs. running prospective clinical trials, he showed how he was able to compare the effectiveness (not a sanitized clinical trial of efficacy, but actual effectiveness in terms of real-world experience) of a next-generation brand-name drug against a generic (no difference except in cost).  He also showed that he could predict, in 50 percent of cases, which patients would develop congestive heart failure six to 24 months ahead of time.  Wow.  While it would be great to see the 50 percent number climb, that kind of result opens the door for preventive interventions that could save many lives.

Power to the Patients!  One panel brought together some of the true pioneers of patient engagement.  (Okay, I’m a bit biased as RWJF/Pioneer funds two of them currently) Warner Slack, who’s a great-grandmentor of some of the students in the audience (he trained the students that trained the current generation of faculty), gave a hilarious talk about just how strange his colleagues thought him when he was advocating for greater patient participation in clinical decisions 40 years ago.  Tom Delbanco discussed the OpenNotes project , and Patti Brennan discussed Project HealthDesign and the value of tracking observations of daily living (“ODLs,” also a frequent topic on this blog).  Patti gave a very clear explanation of the value of each: OpenNotes, she said, gives a window into what the clinician is thinking; ODLs give the clinician a window into the day-to-day life that a patient is living.  Both, she said, are necessary to improving care.  Charlie Safran made some important observations about the need to drop often stereotypic assumptions about how groups of patients will behave and e-Patient Dave was both passionate and entertaining as he spoke about the value of engaging patients in reviewing their records (not only will they be better informed, they will inevitably fins mistakes – some serious). Dave’s comments raised for me a key point, which is that clinicians should make sure that patients know what’s in their records – not for any moral or philosophical reasons – but for the simple practical reason that it can improve care and safety. 

Those were a few things I noticed.  You can check out the Twitter stream for many perspectives.   I’d love to hear from others who were there and get their take.  And for those who weren’t, are these observations consistent with what you’re seeing?

Why I Want a Blue Button

This article appeared originally in the Huffington Post on 10/19/2010.

This morning my colleague, Dr. Carol Diamond of the Markle Foundation, appeared on the Today Show to talk about an initiative called “blue button.” 

As many have experienced, getting a copy of your medical records is rarely as simple as it sounds.  The process often involves making multiple phone calls, having to fax in requests, paying photocopying charges and waiting – often as much as a few weeks.  And at the end of the process all you have is a stack of paper – good for reading and for filing away, but not much else.  But today, as more and more hospitals, pharmacies, and physician offices are adopting electronic medical records, the process should get better.  Health care institutions in the vanguard of information technology and customer service are making it possible for their patients to review their records online.  But not necessarily take them with them.  That’s where the blue button idea comes in.  It’s a simple practice:  when a patient logs in to her account, provide a highly visible, clickable button to download her records in digital form.  As Dr. Diamond noted, the federal government has taken a leadership role in modeling this practice.   The Department of Veterans Affairs started providing blue button downloads to all patients of the VA system back in August and the Center for Medicare and Medicaid Services has launched a similar blue button service for all Medicare beneficiaries.  And their efforts have not gone unnoticed – more than 60,000 veterans have downloaded their medical data since the launch.  The Markle Foundation has also worked with experts from consumer and patient groups, health care provider, privacy advocates and technology companies to develop a set of recommended blue button practices that ensure high levels of security and privacy. 

So why does it matter?  For starters, reviewing your medical records is a step towards better engagement in your health.  You can learn more about your conditions, your lab results and the treatments you might be undergoing.  You can use the records as a springboard to conversations with your doctor.  But you can do that paper records.  Why does digital matter?  Two reasons:  sharing and apps.  Of course, when you get your paper records, you can always photocopy them and send them on to someone you’d like to review them.  But we all know that it’s so much easier to share digital information online.  So when you want your uncle the doctor to help you interpret a result, or when you and your sibling want to discuss Dad’s condition, it’s easily done. 

But the real power lies in the apps.  We now live in a world where there really is an app for just about everything.  There are 6,000 apps for the iPhone that focus on health and fitness.  They help  with day-to-day health-related tasks, like managing your diet, tracking your exercise, understanding your meds, or checking out your symptoms.  But these apps are, by and large, ignorant of the details of your health that could make them smarter and offer you more value.  Many of them, and many that will soon be invented, could be even better if they could draw upon the information contained in your medical records – to spot trends in your lab values and warn you when a trend is of concern; to correlate your moodiness with a new medication you’re taking; or to remind you it’s time order that prescription refill, to name a few.  And we’ve already seen that the data made available through blue button downloads will lead to new apps.  Earlier this month, the Markle Foundation and the Robert Wood Johnson Foundation announced the results of the blue button Challenge, held as part of the Health 2.0 Developer Challenge.  With only a modest prize, the challenge drew nine entries, including the winner, from Adobe Systems, which was a slick front-end for the VA’s data, making it easy for people to read, share and discuss the information.  It also included a set of tools and reminders related to medications.  Another significant entry came from Microsoft, which built a way for blue button downloads to go straight into its HealthVault personal health record service, which provides access to more than 50 different apps.  And this all happened over the course of a month.  As blue button becomes much more widespread, we can expect to see more and more apps designed to take the data you can download and turn it into useful information and valuable tools you can use to manage your health.

The federal government has taken a strong step forward to give people access to their medical records.  Now it’s time for more in the private sector to do the same.  I know I want a blue button on my doctor’s Website.  If you think this is a good idea – if you believe that people should be able to download their medical records, or if you would like your physician’s office to make yours available, or if you have a totally different take on this question, please join the conversation on Twitter – send a tweet with your opinion and be sure to use the hashtag #bluebutton.

And the Winners Are...

As I mentioned back in August, RWJF joined the Health 2.0 Developer Challenge by offering small prizes for three different challenges:  1) building apps that leveraged the Blue Button initiative; 2) apps that bring the data from County Health Rankings into everyday decisions; and 3) bringing Project HealthDesign designs to life by building apps that work on commercial PHR systems.  At the time I said that we wouldn’t know what to expect – that we might not get anything useful at all from this exercise.  WRONG!

The response was terrific.  I won’t say overwhelming, but given the modest amount of prize money (okay, Markle scored coffee with Clay Shirky, which is no small deal) and the relative short amount of time to respond (barely a month), I’d say pretty darn good.  And definitely useful.

I’m a big fan of the Blue Button initiative for two reasons:  1) it gets the data out of the health care system and into the hands of users, where a marketplace of translators, interpreters and other tools can grow around making data useful to people; and 2) it’s beautifully simple.  The response highlighted the potential of this market:  we had large companies like Adobe and Microsoft build really valuable utilities – to translate a largely unreadable ASCII text files into very nicely designed PDF documents and to import Blue Button data into HealthVault, respectively.  But we also saw smaller organizations (seven in all) like MedCommons and RememberItNow showcase more narrowly focused apps that deal with important tasks – like getting a second opinion on a radiology image or remembering to take your meds.  Adobe came out on top in the end – and deservedly so as they built a really nice app – but the key takeaway is that they’re just the tip of the iceberg of what could come.  Hats off to the Markle Foundation, CMS and the VA for bringing Blue Button so far along so quickly.

The County Health Rankings challenge drew a number of interesting submissions.  Where we live, work, learn and play dramatically affects our health.  So when you’re choosing a place to live, wouldn’t you want to look up health indicators the way parents look up data on the quality of the schools?  That’s what the challenge winner, Acsys Interactive, makes possible.  They embedded the county health rankings data into their mobile real estate app.  So now when you want to get the scoop on the house for sale you’re driving by, you point your phone at it and get both real estate data and easy access to county health data.  The challenge drew four other submissions, which also included a utility to have the data texted to your phone when you text in a zip code and a tool to help understand asthma risks in a given county.

The Project HealthDesign Developer Challenge was won by Ringful Health, which produced a really slick app for managing chronic pain.  With it, you can use your iPhone to jot down your pain levels and triggers, get feedback on frequent triggers and how effective your medications have been at controlling your pain, and generate reports that you can share with your doctor.  Ringful had been working on this app, but, inspired by the design from Roger Luckmann’s team at UMass, added several new features and then built back end integration with the HealthVault, Google Health and Dossia personal health record services.  Another competitor, CureTogether, extended their site, which helps people track their observations of daily living (ODLs) like sleep, exercise, and food intake, to include lab test data.  A third, and very intriguing, submission came from Fred Trotter, who built an open source utility to track ODLs and store them on Twitter (in a protected account) and then use Grafitter to analyze the data and display patterns.  Fred’s solution is especially interesting because he’s leveraging existing infrastructure and by building an open source tool, he’s inviting others to take it further.

So all in all, I’m quite pleased with how the challenges went.  And it makes me wonder what challenges we should put out next.  Any ideas?

Joining the Health 2.0 Developer Challenge

Today RWJF has joined the 2010 Health 2.0 Developer Challenge by supporting three new challenges.  Full details are available at health2challenge.org, but here’s a quick rundown:

 

First, we’ve launched a challenge related to the County Health Rankings work we’ve supported with the University of Wisconsin.  That project provides population health status information (such as infant mortality rates, obesity rates) on every county in America and shows how each county ranks within its state.  Despite all the attention the project has gotten, people don’t wake up each day and decide to look up county health.  The trick is figuring out when – in what context – this information is useful.  What types of decisions do people make where community health is highly relevant.  Deciding  where to live?  Where to locate a business?  What else?  That’s the heart of the challenge – we’re looking to see creative solutions for integrating the data into other apps, web sites, etc. that people might use to help them with decisions.

 

Second, we’ve joined with the Markle Foundation on a “Blue Button” challenge.  The Blue Button concept is about being able to download your medical records – why can’t we all log on to our providers’ sites and find a simple button that makes it so?  The VA and CMS have committed to making this possible and as part of the challenge, CMS is making available sample data sets of what you would get when you clicked their blue button.  The challenge we’ve posed is to develop apps that would run on top of the data sets that CMS will provide.

 

Separating the apps from the underlying personal health data is a key tenet of RWJF’s Project HealthDesign program, run by Patti Brennan’s team at the University of Wisconsin.  Building on the work of Project HealthDesign is the subject of the third challenge, for which we’ve partnered with the California HealthCare Foundation.  In the first round of the program, our grantees developed user-centered designs of different applications that used personal health data to support the day-to-day decisions and actions of people with a variety of health challenges, ranging from managing chronic pain to returning home from the hospital.  For this challenge, we’re asking developers to build apps that can run on commercial personal health record services.  In other words, take elements of the designs and make them real.

 

So … three challenges, and I’m really excited to see what we get.  That’s the fun of all this – of course we might get nothing useful (or nothing at all), but as far as philanthropic ventures go, this is a relatively low-cost experiment and I’m hopeful that we’ll be see a lot of creative responses that we hadn’t expected.  Moreover, we’re excited to learn about new ways of leveraging the investments we’ve already made and watch others add new value to them.  Who knows, maybe we’ll find we need to be doing much more of this sort of thing.  What do you think?

Introducing … DATA!

Next Wednesday at the IOM, HHS will do a big unveiling of its Community Health Data Initiative.  It will be a pretty big deal – HHS Secretary Kathleen Sebelius, White House CTO Aneesh Chopra and HHS CTO Todd Park will all be on hand and the expectation is that major tech companies will unveil prototype apps built off of some of the data sets that HHS will be making public.

The HHS/IOM event will be web cast, so check it out.  Either of these links should work:

1.    http://www.hhs.gov/open/
2.    http://videocast.nih.gov/summary.asp?live=9347

It’s an interesting headline when you step back and think about it.  HHS is making a major announcement – not about a new research breakthrough, a new vaccine, a new Medicare benefit or even a new grant opportunity.  It’s about **drumroll** … **drumroll** ... data!  Seriously.  The bet here is that the thousands (and I do mean thousands) of data sets that HHS maintains could actually support some useful applications – applications we can’t even imagine yet – in the same vein that the weather data produced by the National Weather Service generates so many services and businesses.  To some extent, these data have been available before, but they’ve been hard to get to.  The difference here is that HHS is planning to make access to the data easy and beyond that, make them available in ways that most lend themselves to application development.  It’s a conscious strategy to enable others to add value to these government data.

At RWJF, we’ve had a hand in one of the first major apps – the County Heath Rankings – which plots community health characteristics – for every county in America.  Go to the site and you can find both health outcome data, like premature death, and the social, behavioral and environmental factors that lead to those outcomes, like obesity, unemployment and air pollution.  And you can see how each county ranks on any of those factors compared to other counties in your state.  And the County Health Rankings data has even spawned an irreverent take on the data – the County Sin Rankings – winner of the Sunlight Labs Design for America contest for visualizing health data.  Check out the other contestants, who all offered imaginative ways to present health data to the public.  The point is that the government is not best suited to come up with creative ways to help people understand the health of their communities or the quality of the medical care they receive.  But creative designers, developers and activists, when given access to the data, can do it much better.  Once there’s a good platform, we always get happily surprised by the apps.

I’d love to hear what people think are the most exciting apps that get announced on Wednesday.  And while you’re at it, can you think of a more exciting name than “Community Heath Data Initiative?”

Project HealthDesign's Patti Brennan Emphasizes Value of Patient-Generated Data, ODLs at Meaningful Use Workgroup Hearing

Project HealthDesign’s national program director Patti Brennan testifies today at the HIT Policy Committee, Meaningful Use Workgroup’s hearing on patient and consumer engagement.  You can find Patti’s testimony here. 

Patti makes a key point that reflects Project HealthDesign’s current work: patient-generated data are not simply traditional clinical data (like blood pressure or glucose) collected by patients.  Instead, patient-generated data could include a whole host of observations about health behaviors, symptoms and environmental factors that are relevant to someone’s health.  As we’ve discussed over the past couple of years in various posts (see here or here), “observations of daily living” on diet, activity, sleep, pain, mood and others can help paint a clearer picture of one’s health and the factors that influence it and also drive an incredible variety of applications that give people valuable feedback.

Patti specifically offered the committee three things that she believes must be accomplished:

1. Health information that is selected and gathered by patients must be integrated into clinical care. The flow of information about an individual’s health should go both ways – not just from providers to patients – because patients are experts about their daily activities, and providers need their insights. 
2. Health information must be accessible to patients in a computable form. Project HealthDesign’s grantees and numerous private sector companies have been developing applications and services designed to let patients use health data in innovative ways, whether via PCs, mobile devices, online communities or other means.
3. Health information for patients must be actionable. Health information must be meaningful to patients as they make decisions about their own health care.

Project HealthDesign National Advisory Committee Chair Paul Tang co-chairs the Health IT Policy Committee Meaningful Use Workgroup; other speakers today included National Advisory Committee member M. Chris Gibbons of Johns Hopkins, first round grantee James Ralston of Group Health Research Institute, Dave deBronkart (better known as ePatient Dave), Eric Dishman of Intel and Scott Mackie of IDEO's health and wellness group. 

What is the Role of the Physician in a Data Rich World?

Last week’s Project HealthDesign workshop, held at the Vanderbilt Center for Better Health in Nashville, focused on how clinicians could use “observations of daily living,” or ODLs – data associated everyday experiences such as diet, exercise, sleep and pain – to provide better care to people with chronic diseases.  The five Project HealthDesign teams are refining their plans to integrate ODLS into the treatment of premature infants and their parents, obese teens at risk for depression, adults with Crohn’s disease and its complications, adults with asthma and depression or anxiety disorder, and elders with mild cognitive impairment.

A presentation by Kevin Johnson, vice chair of the Vanderbilt University Medical Center Department of Biomedical informatics and the project director of a previous Project HealthDesign grant, raised interesting questions about how to present information captured through ODLs and who should interpret the information.  Johnson showed this graph, which represented a self-report of medication usage over a month’s period as compared with a schedule. 

Clearly, any of us can tell at a glance that the patient adheres to this medication schedule extremely well. The presentation conveys this information quite clearly and, frankly, it doesn’t take a great deal of clinical training to interpret the information. But consider two other cases.  First, if this chart showed a much poorer rate of adherence, the challenge would be to look for patterns in the missed, late and on-time doses to see if there were behavioral triggers or environmental factors that explained the results (e.g. a change in work schedule means the noontime dose is problematic, a Thursday evening softball game makes it unlikely to remember). In the second case, one could overlay on the medication chart other data, such as pain level, mood or even clinical signs like blood pressure and look for patterns that might lead to inferences about correlations and interrelationships.

In each of these three cases, one could ask what skills and training are needed to review, interpret and act upon the information (one can even take it a step further and ask which of the cases requires human vs. algorithmic interpretation). Of course, at some level, the answer is “it depends,” but thinking about the question gives some insight into the broader question we’ve been asking of late in the Pioneer Portfolio:  “What is the role of the physician in a data rich world?” I’m not a clinician, so I’m on shaky ground here, but it seems to me that only the third case requires clinical training – because it’s a case that requires integrating clinical knowledge into a pattern sensing activity. It’s a form of clinical problem solving. The second case doesn’t seem to require clinical knowledge so much as an understanding of some concepts from consumer behavior, design or even behavioral economics and an ability to motivate – to work with the patient to find a solution.  And the first case seems to lend itself to automated processing to determine if the behavior is within some predefined range.

The question gets even more interesting when one factors in the patient’s own engagement in these cases. One would hope that the patient is looking at the same data and developing her own questions and hypotheses. With whom should she discuss them?

What do these scenarios and questions imply for the way we currently primary care and the health professions that make up that enterprise? How will those professions need to evolve?  Will we need new professions? And are the right skills being taught to the right students today?

Benchmarking Progress in Health IT

Being at TED last week led to some interesting conversations about data, health and the progress of health IT.  One conversation in particular stuck with me – a computer industry executive pointed out that the pace of innovation in the computer industry is orders of magnitude faster than in the health IT industry.  Orders of magnitude.  As in 10, 100, 1000 times faster.  A bold claim.  But then think about some of the advancements shown at TED:

 

- Microsoft’s integration of Photosynth and Sea Dragon technologies to create a Virtual Earth experience where you can now do a street level fly through of a city neighborhood and see the facades of the buildings around you.  And where there are web cams, seamlessly integrate live video into the view.

 

- A voyage through the Digital Universe, which is about what it sounds like – extending the Virtual Earth/Google Earth experience to all known objects in the universe.

 

- Google’s demo of an image recognition feature where the presenter took a photo of a postcard of a hotel with the Nexus One and Google (the omniscient Google – not the company) returned the name and address of the hotel.  And speech-to-speech translation through the Nexus One as well.

 

- John Underkoffler’s prototype interface in which people can gesture toward a screen to “pick up” a document, then walk across the room and drop the document onto the screen of a different computer.

 

When you step back and think about it, it’s truly extraordinary.  The gap between sci-fi and ship dates is closing rapidly.  Magic abounds.

So where are we with health IT?  Progress to be sure.  Pockets of excellence.  But as best as I can tell, we’re still struggling with threshold challenges around data exchange, interface design, workflow and deployment at scale.  I’m still processing all this and I’m probably missing something, so I’d really like to hear from people on this question – is the pace of innovation in health IT really that much slower than in the computer and software industries?  If so, then the implications for how we think about the integration of IT into health care are really serious.

The 90-minute TED

Much is has been written about TED 2010, so I thought I’d just chip in with a quick list of my five favorite TED talks that I urge people to watch online once they’re posted. (We’ll let you know via Twitter  — @pioneerrwjf —  when they are.) Before I get to the top 5, though, I do want to plug the talks by two of our Pioneer grantees – Nicholas Christakis and Philip Howard. If you haven’t had a chance to hear Christakis take you through the discoveries he made about the role of social networks in obesity and happiness or Howard put forth his ideas on how to change the legal system (and with it the way we approach malpractice in health care), please check them out on TED.com.

 

In no particular order, my five faves were as follows:

 

• Michael Sandel – a master teacher takes the audience through the meaning of justice.  Warm, funny and profound.

 

• Mark Roth – a natural storyteller tells a captivating tale of scientific inquiry as he seeks to understand a new and surprising way of saving people’s lives.  You get such an authentic sense of Roth’s curiosity and the joy with which he pursues his work.

• Sam Harris – a forceful argument for bringing expertise, knowledge and discoverable facts about how communities flourish into discussions of morality and values.

• George Whitesides – a thought-provoking and highly functional interpretation of simplicity.  “Simple” components are reliable, repeatable and predictable – thus they can be “stacked” to create remarkable systems

 

• Jake Shimabukuro – a virtuoso ukulele performance that you have to watch with the screen maximized and the sound way up.  Pure exhilaration.

 

Oh – and check out blippy.com – a site where you stream your credit card transactions to a social network, which sounds like the definition of conspicuous consumption but it might not be as crazy as it sounds – you could eliminate expense reports, for one thing.

Health IT: What’s the Future?

I’ve just come back from the "Discovery and Innovation in Health IT" workshop. You can find more on the workshop’s web site.  You can also follow the twitter stream by using the #dihit hash tag.  This workshop pulled together about 100 people – mostly researchers from academia, government and corporate labs – to lay out the key health IT research challenges going forward.  I’ve attempted to summarize some of the key themes and points that were raised, but I’ve by no means captured it comprehensively and I’m certain that I’ve taken a lot of license in my interpretation – so please add your voices – I’d love to get other perspectives.

The workshop started with some tone-setting presentations that illustrated the gap between the potential of health IT and where we are now.  Zak Kohane, in a brief introduction, noted that back when he was a grad student (more than a few years ago), clinical decision support was said to be right around the corner and while it’s still not reached its promise, Apple’s managed to generate 86,000 apps in its app store, many of which focus on health, in less than a year.  “Why?” he asked.  Bill Stead, recapping the work he led for the National Research Council’s report on Computational Technology for Effective Health Care, laid out the challenge starkly, saying (and I’m paraphrasing) that the path we’re on with health IT will not solve the problems in our system and could even make them worse.  Most hospitals have implemented IT in ways that rate a 5 on a scale of 100, where 100 is ideal and the best are at 15-20, he went on.  Today’s health IT systems “chain us” to the realm of transactions not decisions and often focus on post-hoc documentation.  “Would you ever get into an airplane where the pilots go through the checklist after they’ve taken off?” he quipped.  He focused a lot on the need for cognitive support, showing a hockey stick graph of the number of facts that will be relevant to a given clinical decision over time (this theme reappeared several times over the two days).  The number is expected to reach 1000 by 2020, while the number of facts that a human can contemplate while making a decision remains stuck at um, five.

 

Other presentations, notably from Dietrich Stephan of Navigenics and Craig Feied from Microsoft, extrapolated from recent progress in genomics, proteomics and systems biology to sketch out a future of increasing precision in our understanding of diseases (hence the hockey stick of medical knowledge).  In these visions, the increasingly ability to subclassify diseases at the molecular level, to understand the biological processes that cause them and to design laser-like therapies to target them, leads to precision diagnoses, early detection and better understanding of which therapies will work and which won’t.

Opening Physicians' Notes to Patients

Today’s Boston Globe ran a story (page one, no less!) announcing our grant to Beth Israel Deaconess Medical Center run a three-site demonstration of opening up physicians’ notes to patients.  That’s not just making labs, drugs, allergies, etc. available to patients – it’s giving them access to the actual notes that the physician records about a visit.  Now these notes are technically available now – under HIPAA each of us has a right to our full medical records (of which physician notes are a part), but the processing for obtaining them is often slow, cumbersome and even expensive in some cases.  Under this project, called Open Notes, patients will receive a secure email after the note has been completed and they can see it right away.  They’ll also be prompted to review the note prior to their next visit.  So instead of limiting access to the very determined, access will be easy for anyone who’s mildly interested.

Why would we fund this?  Several reasons, really.  First, is that at the Pioneer Portfolio, we’re very interested in patient-centered innovation.  Let’s face it:  virtually every trend suggests that people are going to have to become much more engaged in their care and in taking care of themselves.  And, as the pioneers of shared decisionmaking, patient centeredness, patient activation, online support groups and the health 2.0 community have shown us, real benefits come from this engagement.  So much of the energy and excitement in health care today is coming from the patient/consumer side of the equation.  So it’s a space where we believe we will find many innovations that can ultimately transform health.

Second, it’s an elegant system tweak that could really affect behavior.  It’s just a hypothesis at this point, but think about how the knowledge that a patient will read a note will affect how the physician writes the note.  Of course the effect will vary by individual, but this little tweak – with a very small marginal cost – takes a task that physicians do dozens of times a day and reframes it.  As my colleague Paul Tarini has pointed out, it says that this note is for the patient more so than about the patient.  Subtle but important.  My own bet is that this change could do more for influencing how physicians see their relationship with their patients than years of training on how to be more patient centered.  Of course I could be wrong!

And that brings me to the third reason.  This is a controversial idea that needs to be tested.  As we – and Tom Delbanco, the principal investigator on the grant – talked to people, we found strong opinions on both sides of the issue.  There are many people with a patient advocacy perspective who think this is so obvious, such a right and there should be no question about it.  And there are many physicians who think – for very plausible reasons – that this is a terrible idea.  As I said in the Globe article, there’s a bit of a religious character to the debate – you either believe one thing or the other and you believe it strongly.  But there’s been precious little evidence to inform that debate.  That’s why it’s important to do the study and do it on a large scale.

So we’re all excited about this project.  It’s a simple but powerful idea that deserves a real test.  And we hope it sparks enough discussion to raise some interesting and fundamental questions.

More on the App Store for EHRs

For those of you interested in this idea – of electronic health records working as platforms that support substitutable applications (see earlier posts here and here ) – Ken Mandl and Zak Kohane convened a small working group to come up with principles for fostering the development of an “iPhone-like” platform for healthcare information technology.  Check it out and come back here for discussion.

Catalyzing the app store for EHRs

Posted by Steve Downs and John Lumpkin, Senior Vice President, Health Care Group

Recently, Steve posted about the idea, floated by Ken Mandl and Zak Kohane, that EHRs (or health IT more broadly) could move to a model of competitive, substitutable applications running off a platform that would provide secure medical record storage.  In other words, the iPhone app model, but, for example, you could have an e-prescribing app that runs over an EHR instead of the Yelp restaurant review app on your iPhone.  We’re thinking about the provider side of the market here, as Google Health and Microsoft HealthVault are already doing this on the consumer side.

It’s nice to ponder these “what ifs,” but we’re a bit more action-oriented here and we’ve turned our attention to asking what it would take to make this happen.  It seems that there are two things that are needed. First, we need the platform.  Some of the most notable platforms started out as proprietary that were then opened up.  The IBM PC comes to mind as an example. Some were designed from the beginning to be open platforms with limited functionality until the market started developing applications.  A recent example is the development of iGoogle and the tons of applications that are available for free.  Finally, there was the purely public domain development from the beginning to end that we've seen in the Linux world.  Or perhaps we don’t need a common platform and maybe what is needed is to stimulate the market for health IT products that have open application programming interfaces (APIs) that allow for third-party application development?  Several ideas come to mind.

Overshoots and Apps: Disruptive Innovation and Health IT

As Paul Tarini just discussed, we had a thought-provoking discussion with Clayton Christensen about disruptive innovations and health IT.  One idea that interested me in particular was the potential for electronic health record (EHR) systems that are offered on the Software as a Service (SaaS) model to serve as a disruptive innovation.

Let me back up for a minute. Christensen talked about looking at the pre-conditions for disruptive innovations. One of them is when companies “overshoot” the market in terms of performance. For example, their product adds more and more features (each of which adds value to fewer and fewer customers) and becomes very expensive. They target the high-end of the market, where they make the highest margins, and as a result, they offer more than the lower end needs at a price that the lower end can’t afford. They’re “overshooting” that part of the market. This then creates opportunities for new entrants, with a new approach – that gives them a cost advantage – to make inroads at the low end of the market. The dynamic that follows is what ultimately transforms the market: the incumbent happily cedes the low end of the market because they make higher margins at the upper levels, giving the new entrant some traction. The new entrant then seeks the next rung up in the market, which the incumbent again, gladly cedes so they can focus on their most profitable customers, and so on, until Toyota, which started with cheap subcompacts in the 70s, introduces Lexus and starts taking on Mercedes.

So that got us thinking about the EHR marketplace and especially small practices.  A lot of people experience real cognitive dissonance when they think of a three-doc practice installing a traditional EHR where they install and maintain the hardware and software on site.  Systems designed for larger practices (with dedicated IT support) can be cost-prohibitive for small practices.  Sounds like overshoot to me.  Enter SaaS-based EHRs, which, by offering a very different technical and business model and (presumably) a real cost advantage, ought to be primed to take on the low end of the market, away from which the incumbents might happily walk.

What am I missing here?  Is anyone seeing signs of this happening?  Are there SaaS EHR vendors that look particularly promising?

The other key Christensen concept that came into the discussion is the idea Paul mentioned that customers have “jobs” to do, as opposed to systems they need.  In my mind, “job” relates quite directly to “app,” as in “there’s an app for that.”  (There I go quoting Apple ads again.)  This gets back to my earlier post on EHRs and apps, which I won’t rehash other than to say that adoption of EHRs would likely be enhanced if they offered the apps that help providers do the many jobs they need to get done and that the best way to ensure that is to open up app development to 3rd parties.

An App Store for Your EHR? Why Not?

Up on the Project HealthDesign blog, Lygeia Ricciardi calls attention to Ken Mandl and Zak Kohane’s perspective article in the New England Journal on the need for a flexible information infrastructure in health care.  In the article, Mandl and Kohane offer a simple prescription – that the infrastructure be designed as a platform upon which many competitive, substitutable applications can be offered.  They cite the Apple iPhone as an example where this has worked successfully:

 

“The platform separates the system from the functionality provided by the applications. And the applications are substitutable: a consumer can download a calendar reminder system, reject it, and then download another one. The consumer is committed to the platform, but the applications compete on value and cost.”

 

The idea of separating the infrastructure from the applications has been vital to innovation in the computer industry and was the key to our approach on Project HealthDesign (see my post introducing the program).  We were focused on personal health records, but the notion of applying the idea to the design of the infrastructure that supports the business of health care is getting some attention.  Peter Neupert, who leads Microsoft’s health solutions business, has been calling for this, which should be no surprise since Microsoft knows full well the value of a software platform.  The National Research Council (NRC) released a report in January on Computational Technology for Effective Health Care: Immediate Steps and Strategic Directions, in which they also recommend that health care institutions “insist that vendors supply IT that permits the separation of data from applications.” (Disclosure:  Robert Wood Johnson Foundation was a partial funder of the report).

 

This discussion comes at a critical time for two reasons.  First, the American Recovery and Reinvestment Act includes approximately $19 billion in investments in health IT, so to the extent that the incentives work, we’ll see a tremendous amount of investment in health IT over the next several years.  Second, and perhaps even more importantly, we are on the brink of health reform, which suggests that health care institutions need to be poised for changes in the ways that they work.  Mandl and Kohane make the point that macro trends such as the aging population and work force shortages will put pressure on the system to adapt and therefore the information infrastructure needs to be sufficiently flexible.  The NRC committee, led by Bill Stead and shepherded by Herb Lin, noted that

 

“IT is often implemented in systems in a monolithic fashion that makes even small changes hard to introduce ...IT applications appear designed largely to automate tasks or business processes”

 

and recommended that

A few takeaways from the Project HealthDesign conference

Over the last week and half I’ve had a chance to talk with people about Project HealthDesign’s New Frontiers in Personal Health Records conference and follow the blog discussions on the event. James Ralston, the principal investigator for the University of Washington’s Project HealthDesign grant, pointed out to me recently that we went the full range from the user-centered design issues (e.g. how to get the most out of a small cell-phone screen) all the way to policy implications such as the need to alter reimbursement policies. And the conference had a similar range – there were a lot of concepts crammed into an 8-hour session. So here’s my shot at distilling them.

It’s not the record, it’s what you do with it. It’s a simple mantra, but it seems to be catching on. The previously dominant idea of a PHR as a window onto a medical record seems to be fading as more and more people (including policymakers) are recognizing that applications building on the record hold more promise to improve people’s health.

When end users are engaged in the design, the designs look different. At the start of this program, we postulated that putting the end users (patients, consumers, pick your term) at the center of the design process, they would come up with a very different set of PHRs and that’s exactly what happened. The designs are much more mobile, action-oriented, and unobtrusive than earlier visions of PHRs. The grantee teams were constantly challenged to fit their designs into the flow of people’s lives, to free the user from their desktop PC, to work within an existing calendar, rather than creating a “medical calendar,” to make use of the devices people already carry. The nine project videos underscored an important theme – that health is but an enabler (or disabler) of the lives we lead – not an end in itself.

Not everyone gets the PHR ecosystem concept. We (and others) have been pushing this idea for a while now, but there was a fair degree of confusion about it at the conference. The jargon isn’t so important and different people use different terms, but there are basically three types of actors in this ecosystem: 1) the source data providers (e.g. pharmacies, physician offices, hospitals, insurers) that hold parts of a person’s medical record; 2) the PHR platform providers (e.g. Google Health, Microsoft HealthVault, Dossia) that can assemble and maintain a person’s record from multiple sources and that offer application programming interfaces (APIs) for 3rd-party application developers; and 3) the 3rd-party application developers that build the very specific applications (such as those designed by the Project HealthDesign teams) that people can use to take care of their health needs. There are plenty of nuances – for example, the same organization could participate in all three levels – but the key is to allow for the separation of these three functions. Separating the functions enables competition and innovation in the space that markets traditionally operate very well – in meeting the diverse needs of end users.

The policy implications are far-reaching. We could have spent a full day and then some on the policy implications of next-generation PHRs and the PHR ecosystem described above. There are issues about stewardship of personal health data (made even more complicated by data that are user-generated, such as a data on medication usage, diet, exercise, etc.), issues of asymmetric regulation, where different types of organizations providing the same services face different regulatory schemes (e.g. some are covered by HIPAA and others are not), and, most fundamentally, implications for how health care is delivered and financed. The applications demonstrated at the conference make possible a very different patient-clinician relationship, one in which a good deal of data exchange and communication takes place between visits, which could become fewer and further between. That kind of relationship is not well supported in most arrangements today.

We’ve come a long way in two years. Much has happened since the launch of Project HealthDesign two years ago. We’ve seen the emergence of HealthVault, Google Health, Dossia, the iPhone, the iPhone apps store, Android, PatientsLikeMe and the Health 2.0 movement. The key elements of the ecosystem are in place – more and more providers with electronic record systems, platform offerings from major companies with huge consumer brands, and a rapidly growing developer community.

But there’s a long way to go. The reality is that the pieces are in place, but only a precious few have access to them in a truly connected way. Only a few leading health care providers have announced links to GoogleHealth or HealthVault – and we’re not at the kind of plug and play stage where my apps run seamlessly on my handheld, retrieving data from my consolidated record which is constantly updated by my various health care providers.

All in all, though, it was a day for optimism. A day that showed what could be, a day that displayed creativity and ingenuity, and above all, a day that brought together a lot of people who want to make the vision a reality.

Are electronic health records the answer?

Part of our role in the Pioneer Portfolio is to keep an ear open for those who challenge the conventional wisdom – those who don’t jump on the bandwagon, but wonder openly about where it’s headed. So when Gordon Moore came to us a few years back, when there was so much excitement about the prospects for a nationwide health information network of interconnected electronic health records in every practice, and suggested that EHRs, as we knew them, missed the point, we listened. We gave Gordon a small grant to develop his ideas and the results are now posted – in the form of two white papers – on our site.

Gordon’s central concern is that the emphasis should not be placed on the documentation of information about a patient so that it can be retrieved later but rather on the workflow and transactions of care – the entry of orders and the follow up activities. He gives examples of how care can be diagrammed as a series of loops: a test is ordered, the patient then (one hopes) shows up for the test, the result is communicated to the physicians and the patient, and then follow-up action is planned and taken, as appropriate. He also notes that in his own experience, and the experience of a small practice he visited, documentation of notes in the EHR is resulting in added time. This theme is echoed in a recent viewpoint paper – “Promoting Electronic Health Record Adoption. Is it the Correct Focus?” – in the Journal of the American Medical Informatics Association by Donald Simborg. Simborg discusses the trade-off between speed and accuracy in documenting notes. The methods for achieving speed – notably using templates or copying and pasting from previous notes – raise questions of accuracy.

Moore’s and Simborg’s papers open up an interesting discussion, which I would argue is not simply about whether EHRs as a concept are valuable and desirable. After all EHRs are systems that involve many functions to a greater or lesser extent depending on the implementation. Instead, they suggest the importance of identifying priority functions that ought to be implemented within (or even outside of) EHRs. The Center for Information Technology Leadership's work has provided some important insights in this regard – their study on Ambulatory CPOE analyzed value by different levels of functionality and their research into IT and disease management (that we supported) showed that of all the IT interventions typically applied in diabetes management, disease registries had far and away the most value. Add to that Tom Frieden and Farzad Mostashari’s recent JAMA commentary, in which they argue that most EHRs are missing many of the functions necessary for quality preventive care, and we have the seeds of an important discussion.

Medical records make WIRED's list of "things that suck"

You’ve got to hand it to Wired for calling things as they see them.  In their February cover story they explore the banes of our existence – the “33 Things That Drive Us Crazy”.  Not far from the top is the header:  “Medical Records – They’re a Mess!”  It’s a pretty strong condemnation, especially when you consider that other items on the list include junk mail, air travel, commuter traffic, magazine subscription cards, and my personal pet peeve:  the hard plastic packaging that can only be opened by an arsenal of dangerous kitchen implements.  It’s a good rant and it raises some cautionary points about believing that electronic medical records will solve everything.  It also offers an exhortation for people to pressure their doctors to accept a more transparent system. 

So this of course leads to me ask what else about health and health care drives you crazy?  EOB statements?  Having to pay more for your drugs when they go from prescription to over the counter?  Send us your rants.

Network-Centric Warfare and Health IT

After my laptop gave out on a long flight to San Diego last week, I caught up on some reading. Wired offered an intriguing article about the failure of network-centric warfare in Iraq. The article described the basic concepts underlying the Pentagon’s approach of network-centric warfare.

The central idea is that an IT-enabled, highly networked force, that can communicate instantly in a peer-to-peer fashion can have the information it needs, when it needs it, to make rapid decisions. Such a force – that can pinpoint targets with breathtaking accuracy and eliminate them with air strikes within minutes – could overwhelm much larger, less technologically advanced forces. The article provided examples: small Special Forces teams defeating Iraqi Army units outnumbering them by as much as 500:1, a decrease in the time from target identification to target elimination from 3 days in the first Gulf War to under 10 minutes in the current conflict.

But then it goes on to discuss how this strategy, while highly effective in overrunning Saddam’s army, was entirely inappropriate for the years of fighting the insurgency that followed. The fascinating insight from the article was that the network-centric strategy failed because it excluded the most important nodes, or sources of information, from the network: many of the US troops on the ground, local policemen, Iraqi army officers, and tribal leaders. The counter-insurgency strategy that has apparently had some success in Iraq involves much more of a low-tech, messy, patient, trust-building social network approach.

So now let’s consider health IT, where we hear discussions of a nationwide interoperable network of electronic health records that could give clinicians the information they need, when they need it, to make the right clinical decisions.

In the same way that network-centric warfare is remarkably good at killing the enemy (as opposed to persuading the enemy not to be an enemy), network-centric health care (as so often envisioned) might be remarkably good at caring for acute conditions, where relatively little independent patient compliance is required (e.g., show up for surgery, adhere to a short course of medications), but for the messy world of chronic disease – perhaps not so much. In the same sense that in Iraq the network didn’t benefit from its most important participants, one could argue that network-centric healthcare, by not more deeply engaging the patient, fails in the same sense.

Which brings me to the meeting to which I flew. The Markle Foundation convened about 100 people to talk about key health IT issues (see David Kibbe’s post at the Health 2.0 blog for an insightful report on the meeting). At the meeting, Jamie Heywood gave a presentation of his site, PatientsLikeMe, which pulls together data from patients with ALS, Parkinson’s, MS, and HIV/AIDS. They’re getting extensive self-reported data about how each disease is progressing in different people, the symptoms they’re experiencing, the medications and treatments that people are taking, and the effects that they’re having.

Their numbers have grown to the point that they have more participants than most clinical trials on any of these diseases and they offer patients a very different body of information about their disease than what they’ll find in the literature. It’s the street intelligence: the day-to-day understanding of what’s really going on at the ground level in the war (in this case) against certain chronic diseases.

So to me, the parallel is striking: just as it is now with some hindsight that one can understand the flaws of the network-centric warfare strategy in Iraq, it would seem that a network-centric health system needs to move sooner rather than later to figure out how to incorporate the vital information assets that each person brings to the network.

The Tip of the Iceberg

One of the aims of Project HealthDesign was to turn the concept of personal health records (PHRs) on its head.  The idea was to start with end user applications (i.e. the tools people can use to manage their health) and see how they could build off access to one’s medical record rather than start with the medical record itself and see how it could be enhanced.  We anticipated that many of the applications our grantees would design would involve capturing data from end users in the flow of their lives, outside of their contact with the medical care system. 

In a previous post, I raised the question of what constitutes one’s PHR – what’s the relationship between the data in one’s “official” medical record and the data one generates outside of the care system.  The most recent design workshop, which brought our grantee design teams together, gave us a glimpse of the different nature of these two types of data. 

While most PHR offerings today provide an option for some “patient-sourced” data as a complement to the care-generated data that dominates the PHR, our experience at the workshop suggested that the “official” data is merely the tip of the iceberg.

Games as a Diagnostic Device?

In my December post on personal health records, I wrote about health data generated in the home and how it might fit into a PHR.  A story in the Cleveland Plain Dealer highlights another intriguing possibility -- using games to track cognitive ability.  By looking at how well people perform on certain games, you could potentially spot early signs of cognitive decline.  (This idea is also illustrated in a video we supported by the Center for Aging Services Technologies that Al Shar described on this blog a couple of months ago.)  The challenge, of course, is in the analysis, interpretation, and communication of the data.  There's another question, which is whether the game then becomes a diagnostic device, in the legal sense.  Will Halo 4 require FDA approval??

PHRs: They’re Hot, They’re Sexy, But What Are They, Exactly? (Part 2)

In my earlier post on this topic, I raised the question of what constitutes a personal health record and then went on to suggest, as Patti Brennan has said, that we should free ourselves from the idea that each of us should ever have one repository that contains all of our health information.  Instead, I would argue that we should think of the personal health record as existing to serve a set of functions.  For example, knowing a person’s drug allergies is only valuable at the time of prescribing or administering a drug.  The record is only useful to the extent that you can retrieve and act on important information when you need it.

This idea – that the value of the record lies in the functions it supports, such as taking meds on time, saving money on health care costs, motivating behavior changes, or enriching dialogue with health care provider, than in the information itself – is the underlying premise of Project HealthDesign.  In my October 18 post on the program, I explained how we’d be funding different teams to work closely with different sets of end users to design and then build prototypes of personal health applications that help them manage a wide range of health challenges.  We’re asking these teams to assume that a PHR platform – core data and technical services that can support the varied applications – exists so that they can focus on providing the functions that users need, rather than having to struggle with all the thorny issues of how to deliver the actual personal health record (e.g. the meds list, allergies, visit histories – all the usual PHR data).

In talking about Project HealthDesign, we’ve often used the term “common platform” and there’s been a fair amount of confusion about what we mean by that term. I’d suggest that a PHR platform is really about how to record, gain access to, and share data – it’s not one monolithic standard PHR.  For example, data generated by a home monitoring device are likely to be maintained in a different place than claims records, prescription history or provider’s list of diagnoses).  The key is that any application be able to draw on these data as needed – and also be able to store data that can be called up by other applications that have a need for it.  So your claims-based PHR may contain your prescription-filling history, your primary provider’s EHR portal may hold your current problem list, and your iPod may have your running log.  This all works long as your cell phone (for example) can get at those data when you need them.

We’re asking the teams we’ve funded (see the Project HealthDesign site for a list) to use their design processes to identify the core PHR components that would support their designs.   My guess is that when they scope it out, it will look quite different than what most of us currently think of as a PHR.

PHRs: They’re Hot, They’re Sexy, But What Are They, Exactly?

With apologies to Rolling Stone and Jim Morrison for the headline, it’s time to acknowledge the obvious: PHRs, or personal health records, are definitely in vogue – well at least among the health IT crowd, at any rate. I’ve just come back from the annual Connecting Americans to Their Health Care: Empowered Consumers, Personal Health Records, and Emerging Technologies conference (go here for the archived web cast of the plenary sessions), which we co-sponsored with the Markle Foundation and a number of HHS agencies, and among this substantially-sized (500 plus) crowd, PHRs are all the rage. We had representatives from consumer organizations, policy makers, PHR developers, researchers, and, in a sign of the times, some major tech industry players. Google, Microsoft, Intel, and Intuit (the Quicken and TurboTax people) were there and there in numbers. And among this crowd, there’s a lot of excitement and enthusiasm to move forward with PHRs. Rep. Patrick Kennedy, one of the keynote speakers, is even sponsoring legislation that would provide incentive payments to physicians for each patient of theirs who uses a PHR.

So we’re all in favor of PHRs, but it’s not at all clear if we’re talking about the same thing. Generally speaking, people refer to PHRs as online access to your medical record. It gets interesting and a bit confusing when one

Rethinking Personal Health Records

Project HealthDesign, the first RWJF national program to come out the Pioneer portfolio, had its proposal deadline in September and the response was tremendous. We launched Project HealthDesign as a way to stimulate more innovation in the area of personal health records. We used the tag line “Rethinking the Power and Potential of Personal Health Records” because we–and Patti Brennan, the national program director for Project HealthDesign–thought that the real power of a PHR lay not in assembling and presenting a person’s medical record, but in helping people transform that information into action. In other words, it’s not the record itself, but what you do with the record that matters. The basic construct of the program is a common platform–representing a person’s record and some common technical services like authentication – that is then layered with a variety of personal health applications (think Mac OS widgets as an example) that would help people manage certain health tasks.

A hypothetical example we often offer is an application that uses a person’s current medication list (provided by a common PHR platform) to crawl the web and find the lowest price sources for those meds (or there generics). That’s a simple example, but the point is that these applications should be diverse and numerous–reflecting the diversity of people’s health challenges. Having a common platform (or platforms) with stand application programming interfaces (APIs) would enable developers to create innovative applications without having to build a whole PHR infrastructure.  In Project HealthDesign, we’ll fund 8-10 organizations to work with end users to design and then build prototypes of a variety of personal health applications.  The grantees will also work together to develop requirements for a common platform that would support their designs.

So...getting back to the proposal deadline, we got...